Retired Gen. Robert Neller ordered the creation of a &quot;secure, centralized&quot; Marine Corps wargaming center in 2017.

Marine Wargaming Center, Allowing High-Tech Prep for War, Unveiled at Quantico | Military.com

cQuant.io

cquantio

Introduction This is a companion guide to (the "Part 1 guide"). I wrote the Part 1 guide on how to finetune Meta AI's model on Chilean Spanish. It is assumed that you have completed that guide and have generated your own finetuned XLS-R model. This guide will explain the steps to run on your finetuned XLS-R model via a . Working With wav2vec2 Part 1 - Finetuning XLS-R for Automatic Speech Recognition wav2vec2 XLS-R ("XLS-R") inference Kaggle Notebook Prerequisites and Before You Get Started To complete the guide, you will need to have: A finetuned XLS-R model for the Spanish language. An existing . Kaggle account Intermediate knowledge of Python. Intermediate knowledge of working with Kaggle Notebooks. Intermediate knowledge of ML concepts. Basic knowledge of ASR concepts. Building the Inference Notebook Step 1 - Setting Up Your Kaggle Environment Step 1.1 - Creating a New Kaggle Notebook Log in to Kaggle. Create a new Kaggle Notebook. The name of the notebook can be changed as desired. This guide uses the notebook name . spanish-asr-inference Step 1.2 - Adding the Test Datasets This guide uses the as its source for test data. Like the , the Peruvian speakers dataset also consists of two sub-datasets: 2,918 recordings of male Peruvian speakers and 2,529 recordings of female Peruvian speakers. Peruvian Spanish Speech Data Set Chilean Spanish Speech Data Set This dataset has been uploaded to Kaggle as 2 distinct datasets: Recordings of Peruvian male speakers Recordings of Peruvian female speakers Add both of these datasets to your Kaggle Notebook by clicking on . Add Input Step 1.3 - Adding the Finetuned Model You should have saved your finetuned model in of the guide as a . Step 4 Working With wav2vec2 Part 1 - Finetuning XLS-R for Automatic Speech Recognition Kaggle Model Add your finetuned model to your Kaggle Notebook by clicking on . Add Input Step 2 - Building the Inference Notebook The following 16 sub-steps build each of the inference notebook's 16 cells in order. You will note that many of the same utility methods from the Part 1 guide are used here. Step 2.1 - CELL 1: Installing Packages The first cell of the inference notebook installs dependencies. Set the first cell to: ### CELL 1: Install Packages ###
!pip install --upgrade torchaudio
!pip install jiwer Step 2.2 - CELL 2: Importing Python Packages The second cell imports required Python packages. Set the second cell to: ### CELL 2: Import Python packages ###
import re
import math
import random
import pandas as pd
import torchaudio
from datasets import load_metric
from transformers import pipeline Step 2.3 - CELL 3: Loading WER Metric The third cell imports the HuggingFace WER evaluation metric. Set the third cell to: ### CELL 3: Load WER metric ###
wer_metric = load_metric("wer") WER will be used to measure the performance of the finetuned model on test data. Step 2.4 - CELL 4: Setting Constants The fourth cell sets constants that will be used throughout the notebook. Set the fourth cell to: ### CELL 4: Constants ###
# Testing data
TEST_DATA_PATH_MALE = "/kaggle/input/google-spanish-speakers-peru-male/"
TEST_DATA_PATH_FEMALE = "/kaggle/input/google-spanish-speakers-peru-female/"
EXT = ".wav"
NUM_LOAD_FROM_EACH_SET = 3

# Special characters
SPECIAL_CHARS = r"[\d\,\-\;\!\¡\?\¿\।\‘\’\"\–\'\:\/\.\“\”\৷\…\‚\॥\\]"

# Sampling rates
ORIG_SAMPLING_RATE = 48000
TGT_SAMPLING_RATE = 16000 Step 2.5 - CELL 5: Utility Methods for Reading Index Files, Cleaning Text, and Creating Vocabulary The fifth cell defines utility methods for reading the dataset index files, as well as for cleaning transcription text and generating a random set of samples from test data. Set the fifth cell to: ### CELL 5: Utility methods for reading index files, cleaning text, random indices generator ###
def read_index_file_data(path: str, filename: str):
    data = []
    with open(path + filename, "r", encoding = "utf8") as f:
        lines = f.readlines()
        for line in lines:
            file_and_text = line.split("\t")
            data.append([path + file_and_text[0] + EXT, file_and_text[1].replace("\n", "")])
    return data

def clean_text(text: str) -> str:
    cleaned_text = re.sub(SPECIAL_CHARS, "", text)
    cleaned_text = cleaned_text.lower()
    return cleaned_text

def get_random_samples(dataset: list, num: int) -> list:
    used = []
    samples = []
    for i in range(num):
        a = -1
        while a == -1 or a in used:
            a = math.floor(len(dataset) * random.random())
        samples.append(dataset[a])
        used.append(a)
    return samples The method reads a dataset index file and produces a list of lists with audio filename and transcription data, e.g.: read_index_file_data line_index.tsv [
    ["/kaggle/input/google-spanish-speakers-chile-male/clm_08421_01719502739", "Es un viaje de negocios solamente voy por una noche"]
    ...
] The method is used to strip each text transcription of the characters specified by the regular expression assigned to in . These characters, inclusive of punctuation, can be eliminated as they don't provide any semantic value when training the model to learn mappings between audio features and text transcriptions. clean_text SPECIAL_CHARS Step 2.4 The method returns a set of random test samples with the quantity set by the constant in . get_random_samples NUM_LOAD_FROM_EACH_SET Step 2.4 Step 2.6 - CELL 6: Utility Methods for Loading and Resampling Audio Data The sixth cell defines utility methods using to load and resample audio data. Set the sixth cell to: torchaudio ### CELL 7: Utility methods for loading and resampling audio data ###
def read_audio_data(file):
    speech_array, sampling_rate = torchaudio.load(file, normalize = True)
    return speech_array, sampling_rate

def resample(waveform):
    transform = torchaudio.transforms.Resample(ORIG_SAMPLING_RATE, TGT_SAMPLING_RATE)
    waveform = transform(waveform)
    return waveform[0] The method loads a specified audio file and returns a multi-dimensional matrix of the audio data along with the sampling rate of the audio. All the audio files in the training data have a sampling rate of Hz. This "original" sampling rate is captured by the constant in . read_audio_data torch.Tensor 48000 ORIG_SAMPLING_RATE Step 2.4 The method is used to downsample audio data from a sampling rate of to the target sampling rate of . resample 48000 16000 Step 2.7 - CELL 7: Reading Test Data The seventh cell reads the test data index files for the recordings of male speakers and the recordings of female speakers using the method defined in . Set the seventh cell to: read_index_file_data Step 2.5 ### CELL 7: Read test data ###
test_data_male = read_index_file_data(TEST_DATA_PATH_MALE, "line_index.tsv")
test_data_female = read_index_file_data(TEST_DATA_PATH_FEMALE, "line_index.tsv") Step 2.8 - CELL 8: Generating Lists of Random Test Samples The eighth cell generates sets of random test samples using the method defined in . Set the eighth cell to: get_random_samples Step 2.5 ### CELL 8: Generate lists of random test samples ###
random_test_samples_male = get_random_samples(test_data_male, NUM_LOAD_FROM_EACH_SET)
random_test_samples_female = get_random_samples(test_data_female, NUM_LOAD_FROM_EACH_SET) Step 2.9 - CELL 9: Combining Test Data The ninth cell combines the male test samples and female test samples into a single list. Set the ninth cell to: ### CELL 9: Combine test data ###
all_test_samples = random_test_samples_male + random_test_samples_female Step 2.10 - CELL 10: Cleaning Transcription Test The tenth cell iterates over each test data sample and cleans the associated transcription text using the method defined in . Set the tenth cell to: clean_text Step 2.5 ### CELL 10: Clean text transcriptions ###
for index in range(len(all_test_samples)):
    all_test_samples[index][1] = clean_text(all_test_samples[index][1]) Step 2.11 - CELL 11: Loading Audio Data The eleventh cell loads each audio file specified in the list. Set the eleventh cell to: all_test_samples ### CELL 11: Load audio data ###
all_test_data = []

for index in range(len(all_test_samples)):
    speech_array, sampling_rate = read_audio_data(all_test_samples[index][0])
    all_test_data.append({
        "raw": speech_array,
        "sampling_rate": sampling_rate,
        "target_text": all_test_samples[index][1]
    }) Audio data is returned as a and stored in as a list of dictionaries. Each dictionary contains the audio data for a particular sample, the sampling rate, and the text transcription of the audio. torch.Tensor all_test_data Step 2.12 - CELL 12: Resampling Audio Data The twelfth cell resamples audio data to the target sampling rate of . Set the twelfth cell to: 16000 ### CELL 12: Resample audio data and cast to NumPy arrays ###
all_test_data = [{"raw": resample(sample["raw"]).numpy(), "sampling_rate": TGT_SAMPLING_RATE, "target_text": sample["target_text"]} for sample in all_test_data] Step 2.13 - CELL 13: Initializing Instance of Automatic Speech Recognition Pipeline The thirteenth cell initializes an instance of the HuggingFace library class. Set the thirteenth cell to: transformer pipeline ### CELL 13: Initialize instance of Automatic Speech Recognition Pipeline ###
transcriber = pipeline("automatic-speech-recognition", model = "YOUR_FINETUNED_MODEL_PATH") The parameter must be set to the path to finetuned model added to the Kaggle Notebook in , e.g.: model your Step 1.3 transcriber = pipeline("automatic-speech-recognition", model = "/kaggle/input/xls-r-300m-chilean-spanish/transformers/hardy-pine/1") Step 2.14 - CELL 14: Generating Predictions The fourteenth cell calls the initialized in the previous step on the test data to generate text predictions. Set the fourteenth cell to: transcriber ### CELL 14: Generate transcriptions ###
transcriptions = transcriber(all_test_data) Step 2.15 - CELL 15: Calculating WER Metrics The fifteenth cell calculates WER scores for each prediction as well as an overall WER score for all predictions. Set the fifteenth cell to: ### CELL 15: Calculate WER metrics ###
predictions = [transcription["text"] for transcription in transcriptions]
references = [transcription["target_text"][0] for transcription in transcriptions]

wers = []
for p in range(len(predictions)):
    wer = wer_metric.compute(predictions = [predictions[p]], references = [references[p]])
    wers.append(wer)

zipped = list(zip(predictions, references, wers))
df = pd.DataFrame(zipped, columns=["Prediction", "Reference", "WER"])
wer = wer_metric.compute(predictions = predictions, references = references) Step 2.16 - CELL 16: Printing WER Metrics The sixteenth and final cell simply prints the WER calculations from the previous step. Set the sixteenth cell to: ### CELL 16: Output WER metrics ###
pd.set_option("display.max_colwidth", None)
print(f"Overall WER: {wer}")
print(df) WER Analysis Since the notebook generates predictions on random samples of test data, the output will vary each time the notebook is run. The following output was generated on a run of the notebook with set to for a total of 6 test samples: NUM_LOAD_FROM_EACH_SET 3 Overall WER: 0.013888888888888888
                                                                              Prediction  \
0                                     quiero que me reserves el mejor asiento del teatro   
1                                         el llano en llamas es un clásico de juan rulfo   
2         el cuadro de los alcatraces es una de las pinturas más famosas de diego rivera   
3                           hay tres cafés que están abiertos hasta las once de la noche   
4  quiero que me recomiendes una dieta pero donde uno pueda comer algo no puras verduras   
5        cuántos albergues se abrieron después del terremoto del diecinueve de setiembre   

                                                                               Reference  \
0                                     quiero que me reserves el mejor asiento del teatro   
1                                         el llano en llamas es un clásico de juan rulfo   
2         el cuadro de los alcatraces es una de las pinturas más famosas de diego rivera   
3                           hay tres cafés que están abiertos hasta las once de la noche   
4  quiero que me recomiendes una dieta pero donde uno pueda comer algo no puras verduras   
5       cuántos albergues se abrieron después del terremoto del diecinueve de septiembre   

        WER  
0  0.000000  
1  0.000000  
2  0.000000  
3  0.000000  
4  0.000000  
5  0.090909 As can be seen, the model did an excellent job! It only made one error with the sixth sample (index ), mis-spelling the word as . Of course, running the notebook again with different test samples and, more importantly, a larger number of test samples, will produce different and more informative results. Nonetheless, this limited data suggests the model can perform well on different dialects of Spanish - i.e. it was trained on Chilean Spanish, but appears to perform well on Peruvian Spanish. 5 septiembre setiembre Conclusion If you are just learning how to work with wav2vec2 models, I hope that the guide and this guide were useful for you. As mentioned, the finetuned model generated by the Part 1 guide is not quite state-of-the-art, but should still prove useful for many applications. Happy building! Working With wav2vec2 Part 1 - Finetuning XLS-R for Automatic Speech Recognition

Working with wav2vec2 Part 2 - Running Inference on Finetuned ASR Models

wav2vec2 পার্ট 2-এর সাথে কাজ করা - Finetuned ASR মডেলগুলিতে অনুমান চালানো

Arbeiten mit wav2vec2 Teil 2 - Ausführen von Inferenzen auf fein abgestimmten ASR-Modellen

Trabajar con wav2vec2 Parte 2: Ejecución de inferencia en modelos ASR ajustados

Travailler avec wav2vec2 Partie 2 - Exécution d'inférence sur des modèles ASR affinés

wav2vec2 के साथ कार्य करना भाग 2 - फ़ाइनट्यून्ड ASR मॉडल पर इन्फ़रेंस चलाना

wav2vec2 の操作パート 2 - 微調整された ASR モデルでの推論の実行

wav2vec2 작업 2부 - 미세 조정된 ASR 모델에서 추론 실행

Trabalhando com wav2vec2 Parte 2 - Executando inferência em modelos ASR ajustados

Работа с wav2vec2. Часть 2. Выполнение вывода на точно настроенных моделях ASR

wav2vec2 ile Çalışmak Bölüm 2 - İnce Ayarlı ASR Modellerinde Çıkarım Çalıştırmak

Làm việc với wav2vec2 Phần 2 - Chạy suy luận trên các mô hình ASR đã được tinh chỉnh

使用 wav2vec2 第 2 部分 - 在经过微调的 ASR 模型上运行推理

How Much Does It Cost to Start a Video Game Business?

একটি ভিডিও গেম ব্যবসা শুরু করতে কত খরচ হয়?

Wie viel kostet es, ein Videospielunternehmen zu gründen?

¿Cuánto cuesta iniciar un negocio de videojuegos?

Combien coûte la création d’une entreprise de jeux vidéo ?

वीडियो गेम व्यवसाय शुरू करने में कितना खर्च आता है?

ビデオゲームビジネスを始めるにはどれくらいの費用がかかりますか?

비디오 게임 사업을 시작하는 데 비용이 얼마나 드나요?

Quanto custa para iniciar um negócio de videogame?

Сколько стоит начать бизнес по производству видеоигр?

Bir Video Oyunu İşine Başlamanın Maliyeti Ne Kadardır?

Chi phí bao nhiêu để bắt đầu kinh doanh trò chơi điện tử?

启动视频游戏业务需要多少钱？

Decoding my dive into the IBM Quantum Challenge 2024 Greetings from the sunny shores of Durban! Having just braved the IBM Quantum Challenge 2024 & still harboring a serious case of quantum fever, I‘m here to spill the Qiskit Kool-Aid on the good, the not-so-bad and the downright mind-blowing. While the focus this year was broader, the last time I took part was IBM Quantum Challenge Africa 2021, which was all about tackling issues that hit close to home. Teaming up with Wits University, the IBM Research Lab in South Africa developed a challenge that targeted real-world African problems with crop logistics in Agriculture, Financial optimization and quantum Drug discovery. It was all about empowering us as African problem-solvers with the skills and knowledge to leverage the power of quantum computing for our communities. 2K24 & I’m levelling up my Quantum IQ: Lets be honest, quantum computing can feel like its spoken in a language only qubits understand. Going into this challenge, I was a solid 5 outta 10 on the quantum knowledge scale. By the end? Lets just say I graduated to a confident 6😉 The challenge wasn’t just informative, it made these mind-bending concepts feel tangible. So, Challenge Accepted (with a few Nitpicks) the whole experience gets a thumbs up from me - a solid 10 outta 10. The content was dense (think brain food buffet) but the learning curve never really felt like a sheer cliff. Our real-time forum for troubleshooting was epic, all of our challenge parties in different corners of the globe, not forgetting our Dev humor keeping us sane. Tools that stole the show this challenge came loaded with some seriously cool tools, and a few that deserve special recognition: Qiskit Code Assistant This code-whisperer was a godsend, suggesting optimizations and catching errors like a quantum ninja, in the Beta version! And being the first ones to try him out! t’was super cool to be able to do that!
Qiskit Serverless This auto backend selection had me deploying quantum tasks like a pro. Plus, the seamless integration and potential for parallel workflows? Quantum Devs, rejoice! A small caveat: more documentation on custom backend integration would be thee cherry on top.
AI Transpiler This one showed promise, but had a few hiccups with a shopping list of errors (think gremlins in the lab) 🥴
Circuit Knitting Toolbox more like untangling the Quantum yarn, this lab showed me how to break down complex circuits into bite-sized pieces.
VQC & Error Mitigation building a Variational Quantum Classifier, this was a such a cool lab. This lab threw me into the heart of practical quantum machine learning.
Scaling VQC to 50+ Qubits had me taking Quantum leaps & bounds, pushing both hardware and my understanding of quantum simulations to the limit. It wasn’t  just about building a bigger classifier, this was a glimpse into the future. These last 10 days have certainly been a game changer for quantum engineers and I have come out with so much new knowledge. Major props to the Team of Developers for the crystal clear instructions & the encouragement. Imagine bouncing ideas off fellow quantum Devs/engineers, it was a perfect blend of theory, practice and future-focused challenges - now that’s how we conquer collaboratively. For my setup, I primarily tested my Labs locally before running them in the Cloud, giving me the flexibility to switch environments. Sure, there were a few code-based hiccups, but nothing a little debugging couldn’t fix. Each lab built upon the previous one, creating a beautiful symphony of learning. Some Advice to Sharpen your Quantum Sword Before diving in, brush up on the basics, online courses are great places to start.
Practice makes perfect, find platforms that offer tutorials and playgrounds to experiment with quantum circuits.
Don’t go it alone, find Quantum forum’s and communities to join & connect with other quantum enthusiasts and get help when you are stuck.
Most Important know your coding environment, decide how you’ will be running your labs, locally or leveraging the cloud. Never be afraid to experiment and seek help, learning curves are steep, but the journey is always worth half the fun. My Quantum Reflections. Its clear IBM is on a mission to democratize quantum computing for developers and the tools I got to explore are a giant leap forward that left me feeling prepped to tackle the quantum frontier. The possibilities for innovation are 🤯 as these tools keep evolving. This was a transformative experience. I left feeling empowered, yet both humbled by the complexity and energized by the possibilities. I’m always incredibly grateful for these opportunities to be at the forefront exploring cutting-edge technology and putting the insights I gain to good use. Decoding my dive into the IBM Quantum Challenge 2024 Decoding my dive into the IBM Quantum Challenge 2024 Greetings from the sunny shores of Durban! Having just braved the IBM Quantum Challenge 2024 & still harboring a serious case of quantum fever, I‘m here to spill the Qiskit Kool-Aid on the good, the not-so-bad and the downright mind-blowing. While the focus this year was broader, the last time I took part was IBM Quantum Challenge Africa 2021, which was all about tackling issues that hit close to home. Teaming up with Wits University, the IBM Research Lab in South Africa developed a challenge that targeted real-world African problems with crop logistics in Agriculture, Financial optimization and quantum Drug discovery. It was all about empowering us as African problem-solvers with the skills and knowledge to leverage the power of quantum computing for our communities. 2K24 & I’m levelling up my Quantum IQ: Lets be honest, quantum computing can feel like its spoken in a language only qubits understand. Going into this challenge, I was a solid 5 outta 10 on the quantum knowledge scale. By the end? Lets just say I graduated to a confident 6😉 The challenge wasn’t just informative, it made these mind-bending concepts feel tangible. 2K24 & I’m levelling up my Quantum IQ: So, Challenge Accepted ( with a few Nitpicks ) the whole experience gets a thumbs up from me - a solid 10 outta 10. The content was dense ( think brain food buffet ) but the learning curve never really felt like a sheer cliff. Our real-time forum for troubleshooting was epic, all of our challenge parties in different corners of the globe, not forgetting our Dev humor keeping us sane. Challenge Accepted ( with a few Nitpicks ) think brain food buffet Tools that stole the show this challenge came loaded with some seriously cool tools, and a few that deserve special recognition: Tools that stole the show Qiskit Code Assistant This code-whisperer was a godsend, suggesting optimizations and catching errors like a quantum ninja, in the Beta version! And being the first ones to try him out! t’was super cool to be able to do that! Qiskit Serverless This auto backend selection had me deploying quantum tasks like a pro. Plus, the seamless integration and potential for parallel workflows? Quantum Devs, rejoice! A small caveat: more documentation on custom backend integration would be thee cherry on top. AI Transpiler This one showed promise, but had a few hiccups with a shopping list of errors (think gremlins in the lab) 🥴 Circuit Knitting Toolbox more like untangling the Quantum yarn, this lab showed me how to break down complex circuits into bite-sized pieces. VQC & Error Mitigation building a Variational Quantum Classifier, this was a such a cool lab. This lab threw me into the heart of practical quantum machine learning. Scaling VQC to 50+ Qubits had me taking Quantum leaps & bounds, pushing both hardware and my understanding of quantum simulations to the limit. It wasn’t  just about building a bigger classifier, this was a glimpse into the future. Qiskit Code Assistant This code-whisperer was a godsend, suggesting optimizations and catching errors like a quantum ninja, in the Beta version! And being the first ones to try him out! t’was super cool to be able to do that! Qiskit Code Assistant Qiskit Serverless This auto backend selection had me deploying quantum tasks like a pro. Plus, the seamless integration and potential for parallel workflows? Quantum Devs, rejoice! A small caveat: more documentation on custom backend integration would be thee cherry on top. Qiskit Serverless AI Transpiler This one showed promise, but had a few hiccups with a shopping list of errors (think gremlins in the lab) 🥴 AI Transpiler Circuit Knitting Toolbox more like untangling the Quantum yarn, this lab showed me how to break down complex circuits into bite-sized pieces. Circuit Knitting Toolbox VQC & Error Mitigation building a Variational Quantum Classifier, this was a such a cool lab. This lab threw me into the heart of practical quantum machine learning. VQC & Error Mitigation Scaling VQC to 50+ Qubits had me taking Quantum leaps & bounds, pushing both hardware and my understanding of quantum simulations to the limit. It wasn’t  just about building a bigger classifier, this was a glimpse into the future. Scaling VQC to 50+ Qubits These last 10 days have certainly been a game changer for quantum engineers and I have come out with so much new knowledge. Major props to the Team of Developers for the crystal clear instructions & the encouragement. Imagine bouncing ideas off fellow quantum Devs/engineers, it was a perfect blend of theory, practice and future-focused challenges - now that’s how we conquer collaboratively. Team of Developers For my setup, I primarily tested my Labs locally before running them in the Cloud, giving me the flexibility to switch environments. Sure, there were a few code-based hiccups, but nothing a little debugging couldn’t fix. Each lab built upon the previous one, creating a beautiful symphony of learning. Some Advice to Sharpen your Quantum Sword Some Advice to Sharpen your Quantum Sword Before diving in, brush up on the basics, online courses are great places to start. Practice makes perfect, find platforms that offer tutorials and playgrounds to experiment with quantum circuits. Don’t go it alone, find Quantum forum’s and communities to join & connect with other quantum enthusiasts and get help when you are stuck. Most Important know your coding environment, decide how you’ will be running your labs, locally or leveraging the cloud. Before diving in, brush up on the basics, online courses are great places to start. Practice makes perfect, find platforms that offer tutorials and playgrounds to experiment with quantum circuits. Don’t go it alone, find Quantum forum’s and communities to join & connect with other quantum enthusiasts and get help when you are stuck. Most Important know your coding environment, decide how you’ will be running your labs, locally or leveraging the cloud. Most Important Never be afraid to experiment and seek help, learning curves are steep, but the journey is always worth half the fun. My Quantum Reflections. Its clear IBM is on a mission to democratize quantum computing for developers and the tools I got to explore are a giant leap forward that left me feeling prepped to tackle the quantum frontier. The possibilities for innovation are 🤯 as these tools keep evolving. This was a transformative experience. I left feeling empowered, yet both humbled by the complexity and energized by the possibilities. I’m always incredibly grateful for these opportunities to be at the forefront exploring cutting-edge technology and putting the insights I gain to good use.

Bits to Qubits: Decoding my dive into the IBM Quantum Challenge 2024

বিট টু কিউবিটস: আইবিএম কোয়ান্টাম চ্যালেঞ্জ 2024-এ আমার ডাইভ ডিকোডিং

Von Bits zu Qubits: Dekodierung meines Eintauchens in die IBM Quantum Challenge 2024

De Bits a Qubits: decodificando mi inmersión en el IBM Quantum Challenge 2024

Bits to Qubits : décoder ma plongée dans l'IBM Quantum Challenge 2024

बिट्स टू क्यूबिट्स: आईबीएम क्वांटम चैलेंज 2024 में मेरे गोते को डिकोड करना

ビットから量子ビットへ: IBM Quantum Challenge 2024 への私の挑戦を解読

비트를 큐비트로: IBM Quantum Challenge 2024에 대한 나의 다이빙 디코딩

Bits para Qubits: decodificando meu mergulho no IBM Quantum Challenge 2024

Биты в кубиты: расшифровка моего погружения в IBM Quantum Challenge 2024

Bitlerden Qubitlere: IBM Quantum Challenge 2024'e yaptığım dalışın kodunu çözüyorum

Bits to Qubits: Giải mã việc tôi tham gia Thử thách lượng tử IBM 2024

从比特到量子比特：解读我对 2024 年 IBM 量子挑战赛的理解

Deep Neural Networks to Detect and Quantify Lymphoma Lesions: Results

ডিপ নিউরাল নেটওয়ার্ক লিম্ফোমা ক্ষত সনাক্ত এবং পরিমাপ করতে: ফলাফল

Tiefe neuronale Netzwerke zur Erkennung und Quantifizierung von Lymphomläsionen: Ergebnisse

Redes neuronales profundas para detectar y cuantificar lesiones de linfoma: resultados

Réseaux de neurones profonds pour détecter et quantifier les lésions de lymphome : résultats

लिम्फोमा घावों का पता लगाने और मात्रा निर्धारित करने के लिए डीप न्यूरल नेटवर्क: परिणाम

リンパ腫病変を検出し定量化するディープニューラルネットワーク：結果

림프종 병변을 탐지하고 정량화하기 위한 심층 신경망: 결과

Redes Neurais Profundas para Detectar e Quantificar Lesões de Linfoma: Resultados

Глубокие нейронные сети для обнаружения и количественной оценки поражений лимфомой: результаты

Lenfoma Lezyonlarını Tespit Etmek ve Ölçmek İçin Derin Sinir Ağları: Sonuçlar

Mạng lưới thần kinh sâu để phát hiện và định lượng tổn thương ung thư hạch: Kết quả

深度神经网络检测和量化淋巴瘤病变：结果

Deep Neural Networks to Detect and Quantify Lymphoma Lesions: Materials and Methods

ডিপ নিউরাল নেটওয়ার্ক লিম্ফোমা ক্ষত সনাক্ত এবং পরিমাপ করার জন্য: উপাদান এবং পদ্ধতি

Tiefe neuronale Netzwerke zur Erkennung und Quantifizierung von Lymphomläsionen: Materialien und Methoden

Redes neuronales profundas para detectar y cuantificar lesiones de linfoma: materiales y métodos

Réseaux de neurones profonds pour détecter et quantifier les lésions de lymphome : matériels et méthodes

लिम्फोमा घावों का पता लगाने और मात्रा निर्धारित करने के लिए डीप न्यूरल नेटवर्क: सामग्री और विधियाँ

リンパ腫病変を検出し定量化するディープニューラルネットワーク：材料と方法

림프종 병변을 검출하고 정량화하기 위한 심층 신경망: 재료 및 방법

Redes Neurais Profundas para Detectar e Quantificar Lesões de Linfoma: Materiais e Métodos

Глубокие нейронные сети для обнаружения и количественной оценки поражений лимфомой: материалы и методы

Lenfoma Lezyonlarını Tespit Etmek ve Ölçmek İçin Derin Sinir Ağları: Malzemeler ve Yöntemler

Mạng lưới thần kinh sâu để phát hiện và định lượng tổn thương ung thư hạch: Vật liệu và phương pháp

深度神经网络检测和量化淋巴瘤病变：材料和方法

Authors:
(1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com);
(2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA;
(3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada;
(4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea;
(5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada;
(6) Don Wilson, BC Cancer, Vancouver, BC, Canada;
(7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada;
(8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada;
(9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada;
(10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA;
(11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA;
(12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA;
(13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada;
(14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Table of Links Abstract and Intro
Related Work
Materials and Methods
Results
Discussion
Conclusion and References VI. CONCLUSION In this study, we assessed various neural network architectures for automating lymphoma lesion segmentation in PET/CT images across multiple datasets. We examined the reproducibility of lesion measures, revealing differences among networks, highlighting their suitability for specific clinical uses. Additionally, we introduced three lesion detection criteria to assess network performance at a per-lesion level, emphasizing their clinical relevance. Lastly, we discussed challenges related to ground truth consistency and stressed the importance of having well-defined protocol for segmentation. This work provides valuable insights into deep learning’s potentials and limitations in lymphoma lesion segmentation and emphasizes the need for standardized annotation practices to enhance research validity and clinical applications. REFERENCES [1] S. F. Barrington et al. “FDG PET for therapy monitoring in Hodgkin and non-Hodgkin lymphomas”. In: European Journal of Nuclear Medicine and Molecular Imaging 44.1 (Aug. 2017), pp. 97–110. ISSN: 1619- 7089. [2] K. Okuyucu et al. “Prognosis estimation under the light of metabolic tumor parameters on initial FDG-PET/CT in patients with primary extranodal lymphoma”. en. In: Radiol. Oncol. 50.4 (Dec. 2016), pp. 360–369. [3] N. Wu et al. “Deep Neural Networks Improve Radiologists’ Performance in Breast Cancer Screening”. In: IEEE Transactions on Medical Imaging 39.4 (2020), pp. 1184–1194. [4] C. Yuan et al. “Diffuse large B-cell lymphoma segmentation in PET-CT images via hybrid learning for feature fusion”. In: Medical Physics 48.7 (2021), pp. 3665– 3678. [5] H. Hu et al. “Lymphoma Segmentation in PET Images Based on Multi-view and Conv3D Fusion Strategy”. In: 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI). 2020, pp. 1197–1200. [6] H. Li et al. “DenseX-Net: An End-to-End Model for Lymphoma Segmentation in Whole-Body PET/CT Images”. In: IEEE Access 8 (2020), pp. 8004–8018. [7] L. Liu et al. “Improved Multi-modal Patch Based Lymphoma Segmentation with Negative Sample Augmentation and Label Guidance on PET/CT Scans”. In: Multiscale Multimodal Medical Imaging. Ed. by X. Li et al. Cham: Springer Nature Switzerland, 2022, pp. 121–129. ISBN: 978-3-031-18814-5. [8] C. S. Constantino et al. “Evaluation of Semiautomatic and Deep Learning–Based Fully Automatic Segmentation Methods on [18F]FDG PET/CT Images from Patients with Lymphoma: Influence on Tumor Characterization”. In: Journal of Digital Imaging 36.4 (Aug. 2023), pp. 1864–1876. ISSN: 1618-727X. [9] A. J. Weisman et al. “Comparison of 11 automated PET segmentation methods in lymphoma”. English. In: Physics in medicine & biology 65.23 (2020), pp. 235019–235019. [10] A. J. Weisman et al. “Convolutional Neural Networks for Automated PET/CT Detection of Diseased Lymph Node Burden in Patients with Lymphoma”. In: Radiology: Artificial Intelligence 2.5 (2020), e200016. [11] C. Jiang et al. “Deep learning–based tumour segmentation and total metabolic tumour volume prediction in the prognosis of diffuse large B-cell lymphoma patients in 3D FDG-PET images”. In: European Radiology 32.7 (July 2022), pp. 4801–4812. ISSN: 1432-1084. [12] P. Blanc-Durand et al. “Fully automatic segmentation of diffuse large B cell lymphoma lesions on 3D FDGPET/CT for total metabolic tumour volume prediction using a convolutional neural network.” In: European Journal of Nuclear Medicine and Molecular Imaging 48.5 (May 2021), pp. 1362–1370. ISSN: 1619-7089. [13] S. Ahamed et al. “A U-Net Convolutional Neural Network with Multiclass Dice Loss for Automated Segmentation of Tumors and Lymph Nodes from Head and Neck Cancer PET/CT Images”. In: Head and Neck Tumor Segmentation and Outcome Prediction. Ed. by V. Andrearczyk et al. Cham: Springer Nature Switzerland, 2023, pp. 94–106. ISBN: 978-3-031-27420-6. [14] S. Gatidis et al. “A whole-body FDG-PET/CT Dataset with manually annotated Tumor Lesions”. In: Scientific Data 9.1 (Oct. 2022), p. 601. ISSN: 2052-4463. [15] M. Pop et al. “Left-Ventricle Quantification Using Residual U-Net”. English. In: vol. 11395. Statistical Atlases and Computational Models of the Heart. Atrial Segmentation and LV Quantification Challenges. Switzerland: Springer International Publishing AG, 2019, pp. 371–380. ISBN: 0302-9743. [16] A. Myronenko. “3D MRI Brain Tumor Segmentation Using Autoencoder Regularization”. English. In: Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. Cham: Springer International Publishing, 2019, pp. 311–320. ISBN: 0302-9743. [17] F. Isensee et al. “nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation”. In: Nature Methods 18.2 (Dec. 2020), pp. 203–211. [18] A. Hatamizadeh et al. “Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images”. English. In: (2022). [19] M. J. Cardoso et al. MONAI: An open-source framework for deep learning in healthcare. 2022. arXiv: 2211. 02701 [cs.LG]. [20] S. Ahamed et al. “Towards enhanced lesion segmentation using a 3D neural network trained on multiresolution cropped patches of lymphoma PET images”. In: Journal of Nuclear Medicine 64.supplement 1 (2023), P1360–P1360. ISSN: 0161-5505. [21] J. L. Fleiss. “Measuring nominal scale agreement among many raters”. English. In: Psychological bulletin 76.5 (1971), pp. 378–382. [22] A. K. Jha et al. “Nuclear Medicine and artificial intelligence: Best practices for evaluation (the RELAINCE guidelines)”. en. In: J. Nucl. Med. 63.9 (Sept. 2022), pp. 1288–1299. [23] N. Hasani et al. “Artificial Intelligence in Lymphoma PET Imaging: A Scoping Review (Current Trends and Future Directions)”. In: PET clinics 17.1 (Jan. 1), pp. 145–174. ISSN: 1556-8598. [24] S. K. Warfield et al. “Simultaneous truth and performance level estimation (STAPLE): an algorithm for the validation of image segmentation”. en. In: IEEE Trans. Med. Imaging 23.7 (July 2004), pp. 903–921. This paper is available on arxiv under CC 4.0 license. Authors: (1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com); (2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA; (3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada; (4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea; (5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada; (6) Don Wilson, BC Cancer, Vancouver, BC, Canada; (7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada; (8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada; (9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada; (10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA; (11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA; (12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA; (13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada; (14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Authors: Authors: (1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com); (2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA; (3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada; (4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea; (5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada; (6) Don Wilson, BC Cancer, Vancouver, BC, Canada; (7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada; (8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada; (9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada; (10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA; (11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA; (12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA; (13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada; (14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Table of Links Abstract and Intro Related Work Materials and Methods Results Discussion Conclusion and References Abstract and Intro Abstract and Intro Related Work Related Work Materials and Methods Materials and Methods Results Results Discussion Discussion Conclusion and References Conclusion and References VI. CONCLUSION In this study, we assessed various neural network architectures for automating lymphoma lesion segmentation in PET/CT images across multiple datasets. We examined the reproducibility of lesion measures, revealing differences among networks, highlighting their suitability for specific clinical uses. Additionally, we introduced three lesion detection criteria to assess network performance at a per-lesion level, emphasizing their clinical relevance. Lastly, we discussed challenges related to ground truth consistency and stressed the importance of having well-defined protocol for segmentation. This work provides valuable insights into deep learning’s potentials and limitations in lymphoma lesion segmentation and emphasizes the need for standardized annotation practices to enhance research validity and clinical applications. REFERENCES [1] S. F. Barrington et al. “FDG PET for therapy monitoring in Hodgkin and non-Hodgkin lymphomas”. In: European Journal of Nuclear Medicine and Molecular Imaging 44.1 (Aug. 2017), pp. 97–110. ISSN: 1619- 7089. [2] K. Okuyucu et al. “Prognosis estimation under the light of metabolic tumor parameters on initial FDG-PET/CT in patients with primary extranodal lymphoma”. en. In: Radiol. Oncol. 50.4 (Dec. 2016), pp. 360–369. [3] N. Wu et al. “Deep Neural Networks Improve Radiologists’ Performance in Breast Cancer Screening”. In: IEEE Transactions on Medical Imaging 39.4 (2020), pp. 1184–1194. [4] C. Yuan et al. “Diffuse large B-cell lymphoma segmentation in PET-CT images via hybrid learning for feature fusion”. In: Medical Physics 48.7 (2021), pp. 3665– 3678. [5] H. Hu et al. “Lymphoma Segmentation in PET Images Based on Multi-view and Conv3D Fusion Strategy”. In: 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI). 2020, pp. 1197–1200. [6] H. Li et al. “DenseX-Net: An End-to-End Model for Lymphoma Segmentation in Whole-Body PET/CT Images”. In: IEEE Access 8 (2020), pp. 8004–8018. [7] L. Liu et al. “Improved Multi-modal Patch Based Lymphoma Segmentation with Negative Sample Augmentation and Label Guidance on PET/CT Scans”. In: Multiscale Multimodal Medical Imaging. Ed. by X. Li et al. Cham: Springer Nature Switzerland, 2022, pp. 121–129. ISBN: 978-3-031-18814-5. [8] C. S. Constantino et al. “Evaluation of Semiautomatic and Deep Learning–Based Fully Automatic Segmentation Methods on [18F]FDG PET/CT Images from Patients with Lymphoma: Influence on Tumor Characterization”. In: Journal of Digital Imaging 36.4 (Aug. 2023), pp. 1864–1876. ISSN: 1618-727X. [9] A. J. Weisman et al. “Comparison of 11 automated PET segmentation methods in lymphoma”. English. In: Physics in medicine & biology 65.23 (2020), pp. 235019–235019. [10] A. J. Weisman et al. “Convolutional Neural Networks for Automated PET/CT Detection of Diseased Lymph Node Burden in Patients with Lymphoma”. In: Radiology: Artificial Intelligence 2.5 (2020), e200016. [11] C. Jiang et al. “Deep learning–based tumour segmentation and total metabolic tumour volume prediction in the prognosis of diffuse large B-cell lymphoma patients in 3D FDG-PET images”. In: European Radiology 32.7 (July 2022), pp. 4801–4812. ISSN: 1432-1084. [12] P. Blanc-Durand et al. “Fully automatic segmentation of diffuse large B cell lymphoma lesions on 3D FDGPET/CT for total metabolic tumour volume prediction using a convolutional neural network.” In: European Journal of Nuclear Medicine and Molecular Imaging 48.5 (May 2021), pp. 1362–1370. ISSN: 1619-7089. [13] S. Ahamed et al. “A U-Net Convolutional Neural Network with Multiclass Dice Loss for Automated Segmentation of Tumors and Lymph Nodes from Head and Neck Cancer PET/CT Images”. In: Head and Neck Tumor Segmentation and Outcome Prediction. Ed. by V. Andrearczyk et al. Cham: Springer Nature Switzerland, 2023, pp. 94–106. ISBN: 978-3-031-27420-6. [14] S. Gatidis et al. “A whole-body FDG-PET/CT Dataset with manually annotated Tumor Lesions”. In: Scientific Data 9.1 (Oct. 2022), p. 601. ISSN: 2052-4463. [15] M. Pop et al. “Left-Ventricle Quantification Using Residual U-Net”. English. In: vol. 11395. Statistical Atlases and Computational Models of the Heart. Atrial Segmentation and LV Quantification Challenges. Switzerland: Springer International Publishing AG, 2019, pp. 371–380. ISBN: 0302-9743. [16] A. Myronenko. “3D MRI Brain Tumor Segmentation Using Autoencoder Regularization”. English. In: Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries. Cham: Springer International Publishing, 2019, pp. 311–320. ISBN: 0302-9743. [17] F. Isensee et al. “nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation”. In: Nature Methods 18.2 (Dec. 2020), pp. 203–211. [18] A. Hatamizadeh et al. “Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images”. English. In: (2022). [19] M. J. Cardoso et al. MONAI: An open-source framework for deep learning in healthcare. 2022. arXiv: 2211. 02701 [cs.LG]. [20] S. Ahamed et al. “Towards enhanced lesion segmentation using a 3D neural network trained on multiresolution cropped patches of lymphoma PET images”. In: Journal of Nuclear Medicine 64.supplement 1 (2023), P1360–P1360. ISSN: 0161-5505. [21] J. L. Fleiss. “Measuring nominal scale agreement among many raters”. English. In: Psychological bulletin 76.5 (1971), pp. 378–382. [22] A. K. Jha et al. “Nuclear Medicine and artificial intelligence: Best practices for evaluation (the RELAINCE guidelines)”. en. In: J. Nucl. Med. 63.9 (Sept. 2022), pp. 1288–1299. [23] N. Hasani et al. “Artificial Intelligence in Lymphoma PET Imaging: A Scoping Review (Current Trends and Future Directions)”. In: PET clinics 17.1 (Jan. 1), pp. 145–174. ISSN: 1556-8598. [24] S. K. Warfield et al. “Simultaneous truth and performance level estimation (STAPLE): an algorithm for the validation of image segmentation”. en. In: IEEE Trans. Med. Imaging 23.7 (July 2004), pp. 903–921. This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv available on arxiv

Deep Neural Networks to Detect and Quantify Lymphoma Lesions: Conclusion and References

ডিপ নিউরাল নেটওয়ার্ক লিম্ফোমা ক্ষত সনাক্ত এবং পরিমাপ করার জন্য: উপসংহার এবং রেফারেন্স

Tiefe neuronale Netzwerke zur Erkennung und Quantifizierung von Lymphomläsionen: Schlussfolgerung und Referenzen

Redes neuronales profundas para detectar y cuantificar lesiones de linfoma: conclusiones y referencias

Réseaux de neurones profonds pour détecter et quantifier les lésions de lymphome : conclusion et références

लिम्फोमा घावों का पता लगाने और मात्रा निर्धारित करने के लिए डीप न्यूरल नेटवर्क: निष्कर्ष और संदर्भ

リンパ腫病変の検出と定量化のためのディープニューラルネットワーク：結論と参考文献

림프종 병변을 검출하고 정량화하기 위한 심층 신경망: 결론 및 참고 문헌

Redes Neurais Profundas para Detectar e Quantificar Lesões de Linfoma: Conclusão e Referências

Глубокие нейронные сети для обнаружения и количественной оценки поражений лимфомой: заключение и ссылки

Lenfoma Lezyonlarını Tespit Etmek ve Ölçmek İçin Derin Sinir Ağları: Sonuç ve Referanslar

Mạng lưới thần kinh sâu để phát hiện và định lượng tổn thương ung thư hạch: Kết luận và tài liệu tham khảo

深度神经网络检测和量化淋巴瘤病变：结论和参考文献

Authors:
(1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com);
(2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA;
(3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada;
(4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea;
(5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada;
(6) Don Wilson, BC Cancer, Vancouver, BC, Canada;
(7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada;
(8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada;
(9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada;
(10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA;
(11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA;
(12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA;
(13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada;
(14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Table of Links Abstract and Intro
Related Work
Materials and Methods
Results
Discussion
Conclusion and References II. RELATED WORK Numerous works have explored the application of deep learning methods for segmenting lymphoma in PET/CT images. Yuan et al. [4] developed a feature fusion technique to utilize the complementary information from multi-modality data. Hu et al. [5] proposed fusing a combination of 3D ResUNet trained on volumetric data and three 2D ResUNet trained on 2D slices from three orthogonal directions for enhancing segmentation performance. Li et al. [6] proposed DenseX-Net trained in an end-to-end fashion integrating supervised and unsupervised methods for lymphoma detection and segmentation. Liu et al. [7] introduced techniques such as patch-based negative sample augmentation and label guidance for training a 3D Residual-UNet for lymphoma segmentation. A major limitation of all these works was that they were developed on relatively smaller-sized datasets (less than 100 images). Moreover, most of these methods did not compare the performance of their proposed methods with other baselines or with the performance of physicians. Constantino et al. [8] compared the performances of 7 semi-automated and 2 deep learning segmentation methods, while Weisman et al. [9] compared 11 automated segmentation techniques, although both these studies were performed on smaller datasets of sizes 65 and 90 respectively. Weisman et al. [10] compared the segmentation performances of automated 3D Deep Medic method with that of physician although even this study included just 90 lymphoma cases. Except for [10], none of these studies reported model generalization on out-ofdistribution dataset (such as on data collected from different centers), limiting their robustness quantification and external validity. Jiang et al. [11] used a relatively larger dataset as compared to the above studies with 297 images to train a 3D UNet. They even performed out-of-distribution testing on 117 images collected from a different center. To the best of our knowledge, the largest lymphoma PET/CT dataset for deep learning-based lesion segmentation ever reported is the work by Blanc-Durand et al. [12] who used 639 images for model development and 94 for external testing; however, this study only used standard segmentation evaluation metrics and assessed their model’s ability for predicting accurate TMTV. Both the studies [11] and [12] are limited by the fact that their datasets exclusively consisted of patients diagnosed with diffuse large B-cell lymphoma (DLBCL), representing only a single subtype of lymphoma. Most of the existing studies on deep learning-based lymphoma segmentation report their performances on generic segmentation metrics such as Dice similarity coefficient (DSC), intersection-over-union (IoU), sensitivity, etc. In the presence of large segmented lesions, very small missed lesions or small false positives do not contribute much to the DSC value. Hence, there is a need to report the volumes of false positives and false negatives. It will also be beneficial to evaluate the detection performances on a per-lesion basis (number of connected components detected vs missed), since automated detection of even a few voxels of all lesions can help physicians quickly locate the regions of interest, even if the DSC is low. Moreover, the difficulty of the segmentation/detection task is often not assessed via inter- or intra-observer agreement analysis. Our study aims to address these limitations. We trained and validated four deep neural networks on lymphoma PET/CT datasets from three cohorts, encompassing two distinct subtypes of lymphoma: DLBCL and primary mediastinal large B-cell lymphoma (PMBCL). (i) We performed both in (images coming from same cohorts as the training/validation set) and out-of-distribution or external (images from a fourth cohort not used for training/validation) testing to evaluate the robustness of our models. (ii) We reported the performance using DSC, volumes of false positives and negatives, and evaluated the performance dependence on six different types of lesion measures. (iii) We also evaluated the ability of our networks to reproduce these ground truth lesion measures and computed networks’ error in predicting them. (iv) We proposed three types of detection criteria for our use-case and evaluate model’s performance on these metrics. (v) Finally, we evaluated the intra- and inter-observer agreement to give a measure of the difficulty of lesion segmentation task on our datasets. This paper is available on arxiv under CC 4.0 license. Authors: (1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com); (2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA; (3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada; (4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea; (5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada; (6) Don Wilson, BC Cancer, Vancouver, BC, Canada; (7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada; (8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada; (9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada; (10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA; (11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA; (12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA; (13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada; (14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Authors: Authors: (1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com); (2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA; (3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada; (4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea; (5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada; (6) Don Wilson, BC Cancer, Vancouver, BC, Canada; (7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada; (8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada; (9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada; (10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA; (11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA; (12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA; (13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada; (14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Table of Links Abstract and Intro Related Work Materials and Methods Results Discussion Conclusion and References Abstract and Intro Abstract and Intro Related Work Related Work Materials and Methods Materials and Methods Results Results Discussion Discussion Conclusion and References Conclusion and References II. RELATED WORK Numerous works have explored the application of deep learning methods for segmenting lymphoma in PET/CT images. Yuan et al. [4] developed a feature fusion technique to utilize the complementary information from multi-modality data. Hu et al. [5] proposed fusing a combination of 3D ResUNet trained on volumetric data and three 2D ResUNet trained on 2D slices from three orthogonal directions for enhancing segmentation performance. Li et al. [6] proposed DenseX-Net trained in an end-to-end fashion integrating supervised and unsupervised methods for lymphoma detection and segmentation. Liu et al. [7] introduced techniques such as patch-based negative sample augmentation and label guidance for training a 3D Residual-UNet for lymphoma segmentation. A major limitation of all these works was that they were developed on relatively smaller-sized datasets (less than 100 images). Moreover, most of these methods did not compare the performance of their proposed methods with other baselines or with the performance of physicians. Constantino et al. [8] compared the performances of 7 semi-automated and 2 deep learning segmentation methods, while Weisman et al. [9] compared 11 automated segmentation techniques, although both these studies were performed on smaller datasets of sizes 65 and 90 respectively. Weisman et al. [10] compared the segmentation performances of automated 3D Deep Medic method with that of physician although even this study included just 90 lymphoma cases. Except for [10], none of these studies reported model generalization on out-ofdistribution dataset (such as on data collected from different centers), limiting their robustness quantification and external validity. Jiang et al. [11] used a relatively larger dataset as compared to the above studies with 297 images to train a 3D UNet. They even performed out-of-distribution testing on 117 images collected from a different center. To the best of our knowledge, the largest lymphoma PET/CT dataset for deep learning-based lesion segmentation ever reported is the work by Blanc-Durand et al. [12] who used 639 images for model development and 94 for external testing; however, this study only used standard segmentation evaluation metrics and assessed their model’s ability for predicting accurate TMTV. Both the studies [11] and [12] are limited by the fact that their datasets exclusively consisted of patients diagnosed with diffuse large B-cell lymphoma (DLBCL), representing only a single subtype of lymphoma. Most of the existing studies on deep learning-based lymphoma segmentation report their performances on generic segmentation metrics such as Dice similarity coefficient (DSC), intersection-over-union (IoU), sensitivity, etc. In the presence of large segmented lesions, very small missed lesions or small false positives do not contribute much to the DSC value. Hence, there is a need to report the volumes of false positives and false negatives. It will also be beneficial to evaluate the detection performances on a per-lesion basis (number of connected components detected vs missed), since automated detection of even a few voxels of all lesions can help physicians quickly locate the regions of interest, even if the DSC is low. Moreover, the difficulty of the segmentation/detection task is often not assessed via inter- or intra-observer agreement analysis. Our study aims to address these limitations. We trained and validated four deep neural networks on lymphoma PET/CT datasets from three cohorts, encompassing two distinct subtypes of lymphoma: DLBCL and primary mediastinal large B-cell lymphoma (PMBCL). (i) We performed both in (images coming from same cohorts as the training/validation set) and out-of-distribution or external (images from a fourth cohort not used for training/validation) testing to evaluate the robustness of our models. (ii) We reported the performance using DSC, volumes of false positives and negatives, and evaluated the performance dependence on six different types of lesion measures. (iii) We also evaluated the ability of our networks to reproduce these ground truth lesion measures and computed networks’ error in predicting them. (iv) We proposed three types of detection criteria for our use-case and evaluate model’s performance on these metrics. (v) Finally, we evaluated the intra- and inter-observer agreement to give a measure of the difficulty of lesion segmentation task on our datasets. This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv

Deep Neural Networks to Detect and Quantify Lymphoma Lesions: Related Work

ডিপ নিউরাল নেটওয়ার্ক লিম্ফোমা ক্ষত সনাক্ত এবং পরিমাপ করতে: সম্পর্কিত কাজ

Tiefe neuronale Netzwerke zur Erkennung und Quantifizierung von Lymphomläsionen: Verwandte Arbeiten

Redes neuronales profundas para detectar y cuantificar lesiones de linfoma: trabajo relacionado

Réseaux de neurones profonds pour détecter et quantifier les lésions de lymphome : travaux connexes

लिम्फोमा घावों का पता लगाने और उनकी मात्रा निर्धारित करने के लिए डीप न्यूरल नेटवर्क: संबंधित कार्य

リンパ腫病変の検出と定量化のためのディープニューラルネットワーク: 関連研究

림프종 병변을 탐지하고 정량화하기 위한 심층 신경망: 관련 연구

Redes Neurais Profundas para Detectar e Quantificar Lesões de Linfoma: Trabalho Relacionado

Глубокие нейронные сети для обнаружения и количественной оценки поражений лимфомой: соответствующая работа

Lenfoma Lezyonlarını Tespit ve Ölçmek İçin Derin Sinir Ağları: İlgili Çalışma

Mạng lưới thần kinh sâu để phát hiện và định lượng tổn thương ung thư hạch: Công việc liên quan

深度神经网络检测和量化淋巴瘤病变：相关工作

Authors:
(1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com);
(2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA;
(3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada;
(4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea;
(5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada;
(6) Don Wilson, BC Cancer, Vancouver, BC, Canada;
(7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada;
(8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada;
(9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada;
(10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA;
(11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA;
(12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA;
(13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada;
(14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Table of Links Abstract and Intro
Related Work
Materials and Methods
Results
Discussion
Conclusion and References V. DISCUSSION In this work, we trained and evaluated four distinct neural network architectures to automate the segmentation of lymphoma lesions from PET/CT datasets sourced from three different cohorts. To assess models performance, we conducted comprehensive evaluations on internal test set originating from these three cohorts and showed that SegResNet and UNet outperformed DynUNet and SwinUNETR on the DSC (mean and median) and median FPV metrics, while SwinUNETR had the best median FNV. In addition to internal evaluations, we extended our analysis to encompass an external outof-distribution testing phase on a sizable public lymphoma PET/CT dataset. On this external test set as well, SegResNet emerged as the top performer in terms of DSC and FPV metrics, underscoring its robustness and effectiveness, while UNet displayed the best performance on FNV. It is important to highlight that SegResNet and UNet were trained on patches of larger sizes, specifically (224, 224, 224) and (192, 192, 192) respectively, while DynUNet and SwinUNETR were trained using relatively smaller patches, namely (160, 160, 160) and (128, 128, 128) respectively. Utilizing larger patch sizes during training allows the neural networks to capture a more extensive contextual understanding of the data, thereby enhancing its performance in segmentation tasks [17]. This observation aligns with our results, where the superior performance of SegResNet and UNet can be attributed to their exposure to larger patch sizes during training. Moreover, larger batch sizes enable robust training by accurately estimating the gradients [17], but with our chosen training patch sizes, we could not train SegResNet, DynUNet and SwinUNETR with nb > 1 due to memory limitations (although we could accommodate nb = 8 for UNet). Hence, for a fair comparison between networks, all networks were trained with nb = 1. It is worth noting that our inability to train DynUNet and SwinUNETR on larger patch and mini-batch sizes was primarily due to computational resource limitations. However, this limitation presents an avenue for future research, where training these models with larger patches and batch sizes could potentially yield further improvements in segmentation accuracy. We assessed the reproducibility of lesions measures and found that on the internal test set, TMTV and TLG were reproducible across all networks, while Dmax was not reproducible by any network. SUVmean was reproducible by all networks except UNet, SUVmax by only SegResNet and number of lesions by only UNet and SegResNet. On the external test set, reproducibility was more limited, with only SUVmean being reproducible by both SegResNet and SwinUNETR, number of lesions by SegResNet, and TLG by DynUNet (Fig. 3 and 4). Furthermore, we quantified the networks’ error in estimating the value of lesion measures using MAPE and found that MAPE generally decreases as a function of lesion measure values (for all lesion measures) on the combined internal and external test set (Fig. 5). The networks generally made significant errors in the accurate prediction when the ground truth lesion measures were very small. We also showed that, in general, on a set of images with larger patient level lesion SUVmean, SUVmean, TMTV, and TLG, a network is able to predict a higher median DSC, although for very high values of these lesion measures, the performance generally plateaus. On the other hand, the DSC performance is not much affected by the number of lesions, while for a set of images with higher Dmax, the performance generally decreases for all networks (Fig. 7). As much of PET/CT data is privately owned by healthcare institutions, it poses significant challenges for researchers in accessing diverse datasets for training and testing deep learning models. In such a scenario, to improve the interpretability of models, it is crucial for researchers to investigate how the performance of their models depend on dataset characteristics. By studying how model performance correlates with the image/lesion characteristics, researchers can gain insights into the strengths and limitations of their models [13]. Alongside the evaluation of segmentation performance, we also introduced three distinct detection criteria, denoted as Criterion 1, 2, and 3. These criteria served a specific purpose: to evaluate the networks’ performance on a per-lesion basis. This stands in contrast to the segmentation performance assessment, which primarily focuses on the voxel-level accuracy of the networks. The rationale behind introducing these detection criteria lies in the need to assess how well the networks identify and detect lesions within the images, as opposed to merely evaluating their ability to delineate lesion boundaries at the voxel level. The ability to detect the presence of lesions (Criterion 1) is crucial, as it directly influences whether a potential health concern is identified or missed. Detecting even a single voxel of a lesion could trigger further investigation or treatment planning. Lesion count and accurate localization (Criterion 2) are important for treatment planning and monitoring disease progression. Knowing not only that a lesion exists but also how many there are and where they are located can significantly impact therapeutic decisions. Criterion 3 which focused on segmenting lesions based on lesion metabolic characteristics (SUVmax), adds an additional layer of clinical relevance. Using these detection metrics, we assessed the sensitivities and FP detections for all networks and showed that depending on the detection criteria, a network can have very high sensitivity even when the DSC performance was low. Given these different detection criteria, a trained model can be chosen based on specific clinical use cases. For example, some use cases might involve being able to detect all lesions without being overly cautious about segmenting exact lesion boundary, while some other use cases might be looking for more robust boundary delineations. Furthermore, we assessed the intra-observer variability of a physician in segmenting both “easy” and “hard” cases, noting challenges in consistent segmentation of cases from the “hard” subset. In lymphoma lesion segmentation, cases can vary in difficulty due to factors like size, shape, and location of lesions, or image quality. By identifying which cases are consistently difficult for even an experienced physician to segment, we gained insights into the complexities and nuances of the segmentation task. Finally, we also assessed the interobserver agreement between three physicians. Although, we inferred that there was substantial level of agreement between the three physicians, the assessment was performed only on 9 cases, resulting in low statistical power. To improve the consistency of ground truth in medical image segmentation, a well-defined protocol is essential. This protocol should engage multiple expert physicians independently in delineating regions of interest (ROIs) or lesions within PET/CT images. Instead of a single physician segmenting a cohort independently, multiple annotators should segment the same images without knowledge of each other’s work. Discrepancies or disagreements among physicians can be resolved through structured approaches such as facilitated discussions, clinical information reviews, or image clarification. This robust ground truth process enhances inter-observer agreement accuracy and strengthens the validity of research findings and clinical applications relying on these annotations. This paper is available on arxiv under CC 4.0 license. Authors: (1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com); (2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA; (3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada; (4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea; (5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada; (6) Don Wilson, BC Cancer, Vancouver, BC, Canada; (7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada; (8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada; (9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada; (10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA; (11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA; (12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA; (13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada; (14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Authors: Authors: (1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com); (2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA; (3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada; (4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea; (5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada; (6) Don Wilson, BC Cancer, Vancouver, BC, Canada; (7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada; (8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada; (9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada; (10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA; (11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA; (12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA; (13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada; (14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Table of Links Abstract and Intro Related Work Materials and Methods Results Discussion Conclusion and References Abstract and Intro Abstract and Intro Related Work Related Work Materials and Methods Materials and Methods Results Results Discussion Discussion Conclusion and References Conclusion and References V. DISCUSSION In this work, we trained and evaluated four distinct neural network architectures to automate the segmentation of lymphoma lesions from PET/CT datasets sourced from three different cohorts. To assess models performance, we conducted comprehensive evaluations on internal test set originating from these three cohorts and showed that SegResNet and UNet outperformed DynUNet and SwinUNETR on the DSC (mean and median) and median FPV metrics, while SwinUNETR had the best median FNV. In addition to internal evaluations, we extended our analysis to encompass an external outof-distribution testing phase on a sizable public lymphoma PET/CT dataset. On this external test set as well, SegResNet emerged as the top performer in terms of DSC and FPV metrics, underscoring its robustness and effectiveness, while UNet displayed the best performance on FNV. It is important to highlight that SegResNet and UNet were trained on patches of larger sizes, specifically (224, 224, 224) and (192, 192, 192) respectively, while DynUNet and SwinUNETR were trained using relatively smaller patches, namely (160, 160, 160) and (128, 128, 128) respectively. Utilizing larger patch sizes during training allows the neural networks to capture a more extensive contextual understanding of the data, thereby enhancing its performance in segmentation tasks [17]. This observation aligns with our results, where the superior performance of SegResNet and UNet can be attributed to their exposure to larger patch sizes during training. Moreover, larger batch sizes enable robust training by accurately estimating the gradients [17], but with our chosen training patch sizes, we could not train SegResNet, DynUNet and SwinUNETR with nb > 1 due to memory limitations (although we could accommodate nb = 8 for UNet). Hence, for a fair comparison between networks, all networks were trained with nb = 1. It is worth noting that our inability to train DynUNet and SwinUNETR on larger patch and mini-batch sizes was primarily due to computational resource limitations. However, this limitation presents an avenue for future research, where training these models with larger patches and batch sizes could potentially yield further improvements in segmentation accuracy. We assessed the reproducibility of lesions measures and found that on the internal test set, TMTV and TLG were reproducible across all networks, while Dmax was not reproducible by any network. SUVmean was reproducible by all networks except UNet, SUVmax by only SegResNet and number of lesions by only UNet and SegResNet. On the external test set, reproducibility was more limited, with only SUVmean being reproducible by both SegResNet and SwinUNETR, number of lesions by SegResNet, and TLG by DynUNet (Fig. 3 and 4). Furthermore, we quantified the networks’ error in estimating the value of lesion measures using MAPE and found that MAPE generally decreases as a function of lesion measure values (for all lesion measures) on the combined internal and external test set (Fig. 5). The networks generally made significant errors in the accurate prediction when the ground truth lesion measures were very small. We also showed that, in general, on a set of images with larger patient level lesion SUVmean, SUVmean, TMTV, and TLG, a network is able to predict a higher median DSC, although for very high values of these lesion measures, the performance generally plateaus. On the other hand, the DSC performance is not much affected by the number of lesions, while for a set of images with higher Dmax, the performance generally decreases for all networks (Fig. 7). As much of PET/CT data is privately owned by healthcare institutions, it poses significant challenges for researchers in accessing diverse datasets for training and testing deep learning models. In such a scenario, to improve the interpretability of models, it is crucial for researchers to investigate how the performance of their models depend on dataset characteristics. By studying how model performance correlates with the image/lesion characteristics, researchers can gain insights into the strengths and limitations of their models [13]. Alongside the evaluation of segmentation performance, we also introduced three distinct detection criteria, denoted as Criterion 1, 2, and 3. These criteria served a specific purpose: to evaluate the networks’ performance on a per-lesion basis. This stands in contrast to the segmentation performance assessment, which primarily focuses on the voxel-level accuracy of the networks. The rationale behind introducing these detection criteria lies in the need to assess how well the networks identify and detect lesions within the images, as opposed to merely evaluating their ability to delineate lesion boundaries at the voxel level. The ability to detect the presence of lesions (Criterion 1) is crucial, as it directly influences whether a potential health concern is identified or missed. Detecting even a single voxel of a lesion could trigger further investigation or treatment planning. Lesion count and accurate localization (Criterion 2) are important for treatment planning and monitoring disease progression. Knowing not only that a lesion exists but also how many there are and where they are located can significantly impact therapeutic decisions. Criterion 3 which focused on segmenting lesions based on lesion metabolic characteristics (SUVmax), adds an additional layer of clinical relevance. Using these detection metrics, we assessed the sensitivities and FP detections for all networks and showed that depending on the detection criteria, a network can have very high sensitivity even when the DSC performance was low. Given these different detection criteria, a trained model can be chosen based on specific clinical use cases. For example, some use cases might involve being able to detect all lesions without being overly cautious about segmenting exact lesion boundary, while some other use cases might be looking for more robust boundary delineations. Furthermore, we assessed the intra-observer variability of a physician in segmenting both “easy” and “hard” cases, noting challenges in consistent segmentation of cases from the “hard” subset. In lymphoma lesion segmentation, cases can vary in difficulty due to factors like size, shape, and location of lesions, or image quality. By identifying which cases are consistently difficult for even an experienced physician to segment, we gained insights into the complexities and nuances of the segmentation task. Finally, we also assessed the interobserver agreement between three physicians. Although, we inferred that there was substantial level of agreement between the three physicians, the assessment was performed only on 9 cases, resulting in low statistical power. To improve the consistency of ground truth in medical image segmentation, a well-defined protocol is essential. This protocol should engage multiple expert physicians independently in delineating regions of interest (ROIs) or lesions within PET/CT images. Instead of a single physician segmenting a cohort independently, multiple annotators should segment the same images without knowledge of each other’s work. Discrepancies or disagreements among physicians can be resolved through structured approaches such as facilitated discussions, clinical information reviews, or image clarification. This robust ground truth process enhances inter-observer agreement accuracy and strengthens the validity of research findings and clinical applications relying on these annotations. This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv available on arxiv

Deep Neural Networks to Detect and Quantify Lymphoma Lesions: Discussion

লিম্ফোমা ক্ষত সনাক্ত এবং পরিমাপ করার জন্য গভীর নিউরাল নেটওয়ার্ক: আলোচনা

Tiefe neuronale Netzwerke zur Erkennung und Quantifizierung von Lymphomläsionen: Diskussion

Redes neuronales profundas para detectar y cuantificar lesiones de linfoma: discusión

Réseaux de neurones profonds pour détecter et quantifier les lésions de lymphome : discussion

लिम्फोमा घावों का पता लगाने और मात्रा निर्धारित करने के लिए डीप न्यूरल नेटवर्क: चर्चा

リンパ腫病変を検出し定量化するディープニューラルネットワーク：議論

림프종 병변을 탐지하고 정량화하기 위한 심층 신경망: 토론

Redes Neurais Profundas para Detectar e Quantificar Lesões de Linfoma: Discussão

Глубокие нейронные сети для обнаружения и количественной оценки поражений лимфомой: обсуждение

Lenfoma Lezyonlarını Tespit Etmek ve Ölçmek İçin Derin Sinir Ağları: Tartışma

Mạng lưới thần kinh sâu để phát hiện và định lượng tổn thương ung thư hạch: Thảo luận

深度神经网络检测和量化淋巴瘤病变：讨论

Author:
(1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com);
(2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA;
(3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada;
(4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea;
(5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada;
(6) Don Wilson, BC Cancer, Vancouver, BC, Canada;
(7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada;
(8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada;
(9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada;
(10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA;
(11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA;
(12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA;
(13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada;
(14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Table of Links Abstract and Intro
Related Work
Materials and Methods
Results
Discussion
Conclusion and References Abstract This study performs comprehensive evaluation of four neural network architectures (UNet, SegResNet, DynUNet, and SwinUNETR) for lymphoma lesion segmentation from PET/CT images. These networks were trained, validated and tested on a diverse, multi-institutional dataset of 611 cases. Internal testing (88 cases; total metabolic tumor volume (TMTV) range [0.52, 2300] ml) showed SegResNet as the top performer with a median Dice similarity coefficient (DSC) of 0.76 and median false positive volume (FPV) of 4.55 ml; all networks had a median false negative volume (FNV) of 0 ml. On the unseen external test set (145 cases with TMTV range: [0.10, 2480] ml), SegResNet achieved the best median DSC of 0.68 and FPV of 21.46 ml, while UNet had the best FNV of 0.41 ml. We assessed reproducibility of six lesion measures, calculated their prediction errors, and examined DSC performance in relation to these lesion measures, offering insights into segmentation accuracy and clinical relevance. Additionally, we introduced three lesion detection criteria, addressing the clinical need for identifying lesions, counting them, and segmenting based on metabolic characteristics. We also performed expert intra-observer variability analysis revealing the challenges in segmenting “easy” vs. “hard” cases, to assist in the development of more resilient segmentation algorithms. Finally, we performed inter-observer agreement assessment underscoring the importance of a standardized ground truth segmentation protocol involving multiple expert annotators. Code is available at: https://github.com/microsoft/lymphoma-segmentationdnn. Index Terms— Positron emission tomography, computed tomography, deep learning, segmentation, detection, lesion measures, intra-observer variability, inter-observer variability I. INTRODUCTION F LUORODEOXYGLUCOSE (18F-FDG) PET/CT imaging is the standard of care for lymphoma patients, providing accurate diagnoses, staging, and therapy response evaluation. However, traditional qualitative assessments, like Deauville scores [1], can introduce variability due to observer’s subjectivity in image interpretation. Using quantitative PET analysis that incorporates lesion measures such as mean lesion standardized uptake value (SUVmean), total metabolic tumor volume (TMTV), and total lesion glycolysis (TLG) offers a promising path to more reliable prognostic decisions, enhancing our ability to predict patient outcomes in lymphoma with greater precision and confidence [2]. Quantitative assessment in PET/CT imaging often relies on manual lesion segmentation, which is time-consuming and prone to intra- and inter-observer variabilities. Traditional thresholding-based automated techniques can miss low-uptake disease and produce false positives in regions of physiological high uptake of radiotracers. Therefore, deep learning offers promise for automating lesion segmentation, reducing variability, increasing patient throughput, and potentially aiding in the detection of challenging lesions [3]. Although promising, deep learning methods face challenges of their own. Convolutional neural networks (CNNs) require large, well-annotated datasets that can be difficult to obtain. Models trained on small datasets may not be generalizable. Moreover, lymphoma lesions vary significantly in size, shape, and metabolic activity, making training deep networks accurately challenging in the absence of well-defined priors. Deep learning aims to reduce observer variability, but inconsistent manual annotations used for training can lead to error perpetuation. Understanding these challenges is crucial towards harnessing the full potential of these methods in PET/CT quantitative analysis. This paper is available on arxiv under CC 4.0 license. Author: (1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com); (2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA; (3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada; (4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea; (5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada; (6) Don Wilson, BC Cancer, Vancouver, BC, Canada; (7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada; (8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada; (9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada; (10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA; (11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA; (12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA; (13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada; (14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Author: Author: (1) Shadab Ahamed, University of British Columbia, Vancouver, BC, Canada, BC Cancer Research Institute, Vancouver, BC, Canada. He was also a Mitacs Accelerate Fellow (May 2022 - April 2023) with Microsoft AI for Good Lab, Redmond, WA, USA (e-mail: shadabahamed1996@gmail.com); (2) Yixi Xu, Microsoft AI for Good Lab, Redmond, WA, USA; (3) Claire Gowdy, BC Children’s Hospital, Vancouver, BC, Canada; (4) Joo H. O, St. Mary’s Hospital, Seoul, Republic of Korea; (5) Ingrid Bloise, BC Cancer, Vancouver, BC, Canada; (6) Don Wilson, BC Cancer, Vancouver, BC, Canada; (7) Patrick Martineau, BC Cancer, Vancouver, BC, Canada; (8) Franc¸ois Benard, BC Cancer, Vancouver, BC, Canada; (9) Fereshteh Yousefirizi, BC Cancer Research Institute, Vancouver, BC, Canada; (10) Rahul Dodhia, Microsoft AI for Good Lab, Redmond, WA, USA; (11) Juan M. Lavista, Microsoft AI for Good Lab, Redmond, WA, USA; (12) William B. Weeks, Microsoft AI for Good Lab, Redmond, WA, USA; (13) Carlos F. Uribe, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada; (14) Arman Rahmim, BC Cancer Research Institute, Vancouver, BC, Canada, and University of British Columbia, Vancouver, BC, Canada. Table of Links Abstract and Intro Related Work Materials and Methods Results Discussion Conclusion and References Abstract and Intro Abstract and Intro Related Work Related Work Materials and Methods Materials and Methods Results Results Discussion Discussion Conclusion and References Conclusion and References Abstract Abstract This study performs comprehensive evaluation of four neural network architectures (UNet, SegResNet, DynUNet, and SwinUNETR) for lymphoma lesion segmentation from PET/CT images. These networks were trained, validated and tested on a diverse, multi-institutional dataset of 611 cases. Internal testing (88 cases; total metabolic tumor volume (TMTV) range [0.52, 2300] ml) showed SegResNet as the top performer with a median Dice similarity coefficient (DSC) of 0.76 and median false positive volume (FPV) of 4.55 ml; all networks had a median false negative volume (FNV) of 0 ml. On the unseen external test set (145 cases with TMTV range: [0.10, 2480] ml), SegResNet achieved the best median DSC of 0.68 and FPV of 21.46 ml, while UNet had the best FNV of 0.41 ml. We assessed reproducibility of six lesion measures, calculated their prediction errors, and examined DSC performance in relation to these lesion measures, offering insights into segmentation accuracy and clinical relevance. Additionally, we introduced three lesion detection criteria, addressing the clinical need for identifying lesions, counting them, and segmenting based on metabolic characteristics. We also performed expert intra-observer variability analysis revealing the challenges in segmenting “easy” vs. “hard” cases, to assist in the development of more resilient segmentation algorithms. Finally, we performed inter-observer agreement assessment underscoring the importance of a standardized ground truth segmentation protocol involving multiple expert annotators. Code is available at: https://github.com/microsoft/lymphoma-segmentationdnn . https://github.com/microsoft/lymphoma-segmentationdnn Index Terms — Positron emission tomography, computed tomography, deep learning, segmentation, detection, lesion measures, intra-observer variability, inter-observer variability Index Terms I. INTRODUCTION F LUORODEOXYGLUCOSE (18F-FDG) PET/CT imaging is the standard of care for lymphoma patients, providing accurate diagnoses, staging, and therapy response evaluation. However, traditional qualitative assessments, like Deauville scores [1], can introduce variability due to observer’s subjectivity in image interpretation. Using quantitative PET analysis that incorporates lesion measures such as mean lesion standardized uptake value (SUVmean), total metabolic tumor volume (TMTV), and total lesion glycolysis (TLG) offers a promising path to more reliable prognostic decisions, enhancing our ability to predict patient outcomes in lymphoma with greater precision and confidence [2]. Quantitative assessment in PET/CT imaging often relies on manual lesion segmentation, which is time-consuming and prone to intra- and inter-observer variabilities. Traditional thresholding-based automated techniques can miss low-uptake disease and produce false positives in regions of physiological high uptake of radiotracers. Therefore, deep learning offers promise for automating lesion segmentation, reducing variability, increasing patient throughput, and potentially aiding in the detection of challenging lesions [3]. Although promising, deep learning methods face challenges of their own. Convolutional neural networks (CNNs) require large, well-annotated datasets that can be difficult to obtain. Models trained on small datasets may not be generalizable. Moreover, lymphoma lesions vary significantly in size, shape, and metabolic activity, making training deep networks accurately challenging in the absence of well-defined priors. Deep learning aims to reduce observer variability, but inconsistent manual annotations used for training can lead to error perpetuation. Understanding these challenges is crucial towards harnessing the full potential of these methods in PET/CT quantitative analysis. This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv

Deep Neural Networks to Detect and Quantify Lymphoma Lesions: Abstract and Intro

লিম্ফোমা ক্ষত সনাক্ত এবং পরিমাপ করার জন্য গভীর নিউরাল নেটওয়ার্ক: বিমূর্ত এবং ভূমিকা

Tiefe neuronale Netzwerke zur Erkennung und Quantifizierung von Lymphomläsionen: Zusammenfassung und Einführung

Redes neuronales profundas para detectar y cuantificar lesiones de linfoma: resumen e introducción

Réseaux de neurones profonds pour détecter et quantifier les lésions de lymphome : résumé et introduction

लिम्फोमा घावों का पता लगाने और मात्रा निर्धारित करने के लिए डीप न्यूरल नेटवर्क: सार और परिचय

リンパ腫病変を検出し定量化するディープニューラルネットワーク：概要と概要

림프종 병변을 탐지하고 정량화하기 위한 심층 신경망: 개요 및 소개

Redes Neurais Profundas para Detectar e Quantificar Lesões de Linfoma: Resumo e Introdução

Глубокие нейронные сети для обнаружения и количественной оценки поражений лимфомой: краткое содержание и введение

Lenfoma Lezyonlarını Tespit Etmek ve Ölçmek İçin Derin Sinir Ağları: Özet ve Giriş

Mạng lưới thần kinh sâu để phát hiện và định lượng các tổn thương ung thư hạch: Tóm tắt và giới thiệu

深度神经网络检测和量化淋巴瘤病变：摘要和简介

Authors:
(1) Xiao-Yang Liu, Hongyang Yang, Columbia University (xl2427,hy2500@columbia.edu);
(2) Jiechao Gao, University of Virginia (jg5ycn@virginia.edu);
(3) Christina Dan Wang (Corresponding Author), New York University Shanghai (christina.wang@nyu.edu). Table of Links Abstract and 1 Introduction 2 Related Works and 2.1 Deep Reinforcement Learning Algorithms 2.2 Deep Reinforcement Learning Libraries and 2.3 Deep Reinforcement Learning in Finance 3 The Proposed FinRL Framework and 3.1 Overview of FinRL Framework 3.2 Application Layer 3.3 Agent Layer 3.4 Environment Layer 3.5 Training-Testing-Trading Pipeline 4 Hands-on Tutorials and Benchmark Performance and 4.1 Backtesting Module 4.2 Baseline Strategies and Trading Metrics 4.3 Hands-on Tutorials 4.4 Use Case I: Stock Trading 4.5 Use Case II: Portfolio Allocation and 4.6 Use Case III: Cryptocurrencies Trading 5 Ecosystem of FinRL and Conclusions, and References ABSTRACT Deep reinforcement learning (DRL) has been envisioned to have a competitive edge in quantitative finance. However, there is a steep development curve for quantitative traders to obtain an agent that automatically positions to win in the market, namely to decide where to trade, at what price and what quantity, due to the error-prone programming and arduous debugging. In this paper, we present the first open-source framework FinRL as a full pipeline to help quantitative traders overcome the steep learning curve. FinRL is featured with simplicity, applicability and extensibility under the key principles, full-stack framework, customization, reproducibility and hands-on tutoring Embodied as a three-layer architecture with modular structures, FinRL implements fine-tuned state-of-the-art DRL algorithms and common reward functions, while alleviating the debugging workloads. Thus, we help users pipeline the strategy design at a high turnover rate. At multiple levels of time granularity, FinRL simulates various markets as training environments using historical data and live trading APIs. Being highly extensible, FinRL reserves a set of user-import interfaces and incorporates trading constraints such as market friction, market liquidity and investor’s risk-aversion. Moreover, serving as practitioners’ stepping stones, typical trading tasks are provided as step-by-step tutorials, e.g., stock trading, portfolio allocation, cryptocurrency trading, etc. 1 INTRODUCTION Deep reinforcement learning (DRL), that balances exploration (of uncharted territory) and exploitation (of current knowledge), is a promising approach to automate trading in quantitative finance [50][51][47][54][21][13]. DRL algorithms are powerful in solving dynamic decision-making problems by learning through interactions with an unknown environment and offer two major advantages of portfolio scalability and market model independence [6]. In quantitative finance, algorithmic trading is essentially making dynamic decisions, namely, to decide where to trade, at what price and what quantity, in a highly stochastic and complex financial market. Incorporating many financial factors, as shown in Fig. 1, a DRL trading agent builds a multi-factor model to trade automatically, which is difficult for human traders to accomplish [4, 53]. Therefore, DRL has been envisioned to have a competitive edge in quantitative finance. Many existing works have applied DRL in quantitative financial tasks. Both researchers and industry practitioners are actively designing trading strategies fueled by DRL, since deep neural networks are significantly powerful at estimating the expected return of taking a certain action at a state. Moody and Saffell [33] utilized a policy search for stock trading; Deng et al. [9] showed that DRL can obtain more profits than conventional methods. More applications include stock trading [35, 47, 51, 54], futures contracts [54], alternative data (news sentiments) [22, 35], high frequency trading [15], liquidation strategy analysis [3], and hedging [6]. DRL is also being actively explored in the cryptocurrency market, e.g., automated trading, portfolio allocation, and market making. However, designing a DRL trading strategy is not easy. The programming is error-prone with tedious debugging. The development pipeline includes preprocessing market data, building a training environment, managing trading states, and backtesting trading performance. These steps are standard for implementation but yet time consuming especially for beginners. Therefore, there is an urgent demand for an open-source library to help researchers and quantitative traders to overcome the steep learning curve. In this paper, we present a FinRL framework that automatically streamlines the development of trading strategies, so as to help researchers and quantitative traders to iterate their strategies at a high turnover rate. Users specify the configurations, such as picking data APIs and DRL algorithms, and analyze the performance of trading results. To achieve this, FinRL introduces a three-layer framework. At the bottom is an environment layer that simulates financial markets using actual historical data, such as closing price, shares, trading volume, and technical indicators. In the middle is the agent layer that implements fine-tuned DRL algorithms and common reward functions. The agent interacts with the environment through properly defined reward functions on the state space and action space. The top layer includes applications in automated trading, where we demonstrate several use cases, namely stock trading, portfolio allocation, cryptocurrency trading, etc. We provide baseline trading strategies to alleviate debugging workloads. Under the three-layer framework, FinRL is developed with three primary principles: • Full-stack framework. To provide a full-stack DRL framework with finance-oriented optimizations, including market data APIs, data preprocessing, DRL algorithms, and automated backtesting. Users can transparently make use of such a development pipeline. • Customization. To maintain modularity and extensibility in development by including state-of-the-art DRL algorithms and supporting design of new algorithms. The DRL algorithms can be used to construct trading strategies by simple configurations. • Reproducibility and hands-on tutoring. To provide tutorials such as step-by-step Jupyter notebooks and user guide to help users walk through the pipeline and reproduce the use cases. This leads to a unified framework where developers are able to efficiently explore ideas through high-level configurations and specifications, and to customize their own strategies at request. Our contributions are summarized as follows: • FinRL is the first open-source framework that demonstrates the great potential of applying DRL algorithms in quantitative finance. We build an ecosystem around the FinRL framework, which seeds the rapidly growing AI4Finance community. • The application layer provides interfaces for users to customize FinRL to their own trading tasks. Automated backtesting module and performance metrics are provided to help quantitative traders iterate trading strategies at a high turnover rate. Profitable trading strategies are reproducible and hands-on tutorials are provided in a beginner-friendly fashion. Adjusting the trained models to the rapid changing markets is also possible. • The agent layer provides state-of-the-art DRL algorithms that are adapted to finance with fine-tuned hyperparameters. Users can add new DRL algorithms. • The environment layer includes not only a collection of historical data APIs, but also live trading APIs. They are reconfigured into standard OpenAI gym-style environments [5]. Moreover, it incorporates market frictions and allows users to customize the trading time granularity. The remainder of this paper is organized as follows. Section 2 reviews related works. Section 3 presents the FinRL framework. Section 4 demonstrates benchmark trading tasks using FinRL. We conclude this paper in Section 5. This paper is available on arxiv under CC BY 4.0 DEED license. Authors: (1) Xiao-Yang Liu, Hongyang Yang, Columbia University (xl2427,hy2500@columbia.edu); (2) Jiechao Gao, University of Virginia (jg5ycn@virginia.edu); (3) Christina Dan Wang (Corresponding Author), New York University Shanghai (christina.wang@nyu.edu). Authors: Authors: (1) Xiao-Yang Liu, Hongyang Yang, Columbia University (xl2427,hy2500@columbia.edu); (2) Jiechao Gao, University of Virginia (jg5ycn@virginia.edu); (3) Christina Dan Wang ( Corresponding Author ), New York University Shanghai (christina.wang@nyu.edu). Corresponding Author Table of Links Abstract and 1 Introduction Abstract and 1 Introduction 2 Related Works and 2.1 Deep Reinforcement Learning Algorithms 2 Related Works and 2.1 Deep Reinforcement Learning Algorithms 2.2 Deep Reinforcement Learning Libraries and 2.3 Deep Reinforcement Learning in Finance 2.2 Deep Reinforcement Learning Libraries and 2.3 Deep Reinforcement Learning in Finance 3 The Proposed FinRL Framework and 3.1 Overview of FinRL Framework 3 The Proposed FinRL Framework and 3.1 Overview of FinRL Framework 3.2 Application Layer 3.2 Application Layer 3.3 Agent Layer 3.3 Agent Layer 3.4 Environment Layer 3.4 Environment Layer 3.5 Training-Testing-Trading Pipeline 3.5 Training-Testing-Trading Pipeline 4 Hands-on Tutorials and Benchmark Performance and 4.1 Backtesting Module 4 Hands-on Tutorials and Benchmark Performance and 4.1 Backtesting Module 4.2 Baseline Strategies and Trading Metrics 4.2 Baseline Strategies and Trading Metrics 4.3 Hands-on Tutorials 4.3 Hands-on Tutorials 4.4 Use Case I: Stock Trading 4.4 Use Case I: Stock Trading 4.5 Use Case II: Portfolio Allocation and 4.6 Use Case III: Cryptocurrencies Trading 4.5 Use Case II: Portfolio Allocation and 4.6 Use Case III: Cryptocurrencies Trading 5 Ecosystem of FinRL and Conclusions, and References 5 Ecosystem of FinRL and Conclusions, and References ABSTRACT Deep reinforcement learning (DRL) has been envisioned to have a competitive edge in quantitative finance. However, there is a steep development curve for quantitative traders to obtain an agent that automatically positions to win in the market, namely to decide where to trade, at what price and what quantity, due to the error-prone programming and arduous debugging. In this paper, we present the first open-source framework FinRL as a full pipeline to help quantitative traders overcome the steep learning curve. FinRL is featured with simplicity, applicability and extensibility under the key principles, full-stack framework, customization, reproducibility and hands-on tutoring Embodied as a three-layer architecture with modular structures, FinRL implements fine-tuned state-of-the-art DRL algorithms and common reward functions, while alleviating the debugging workloads. Thus, we help users pipeline the strategy design at a high turnover rate. At multiple levels of time granularity, FinRL simulates various markets as training environments using historical data and live trading APIs. Being highly extensible, FinRL reserves a set of user-import interfaces and incorporates trading constraints such as market friction, market liquidity and investor’s risk-aversion. Moreover, serving as practitioners’ stepping stones, typical trading tasks are provided as step-by-step tutorials, e.g., stock trading, portfolio allocation, cryptocurrency trading, etc. 1 INTRODUCTION Deep reinforcement learning (DRL), that balances exploration (of uncharted territory) and exploitation (of current knowledge), is a promising approach to automate trading in quantitative finance [50][51][47][54][21][13]. DRL algorithms are powerful in solving dynamic decision-making problems by learning through interactions with an unknown environment and offer two major advantages of portfolio scalability and market model independence [6]. In quantitative finance, algorithmic trading is essentially making dynamic decisions, namely, to decide where to trade, at what price and what quantity, in a highly stochastic and complex financial market. Incorporating many financial factors, as shown in Fig. 1, a DRL trading agent builds a multi-factor model to trade automatically, which is difficult for human traders to accomplish [4, 53]. Therefore, DRL has been envisioned to have a competitive edge in quantitative finance. Many existing works have applied DRL in quantitative financial tasks. Both researchers and industry practitioners are actively designing trading strategies fueled by DRL, since deep neural networks are significantly powerful at estimating the expected return of taking a certain action at a state. Moody and Saffell [33] utilized a policy search for stock trading; Deng et al. [9] showed that DRL can obtain more profits than conventional methods. More applications include stock trading [35, 47, 51, 54], futures contracts [54], alternative data (news sentiments) [22, 35], high frequency trading [15], liquidation strategy analysis [3], and hedging [6]. DRL is also being actively explored in the cryptocurrency market, e.g., automated trading, portfolio allocation, and market making. However, designing a DRL trading strategy is not easy. The programming is error-prone with tedious debugging. The development pipeline includes preprocessing market data, building a training environment, managing trading states, and backtesting trading performance. These steps are standard for implementation but yet time consuming especially for beginners. Therefore, there is an urgent demand for an open-source library to help researchers and quantitative traders to overcome the steep learning curve. In this paper, we present a FinRL framework that automatically streamlines the development of trading strategies, so as to help researchers and quantitative traders to iterate their strategies at a high turnover rate. Users specify the configurations, such as picking data APIs and DRL algorithms, and analyze the performance of trading results. To achieve this, FinRL introduces a three-layer framework. At the bottom is an environment layer that simulates financial markets using actual historical data, such as closing price, shares, trading volume, and technical indicators. In the middle is the agent layer that implements fine-tuned DRL algorithms and common reward functions. The agent interacts with the environment through properly defined reward functions on the state space and action space. The top layer includes applications in automated trading, where we demonstrate several use cases, namely stock trading, portfolio allocation, cryptocurrency trading, etc. We provide baseline trading strategies to alleviate debugging workloads. Under the three-layer framework, FinRL is developed with three primary principles: • Full-stack framework. To provide a full-stack DRL framework with finance-oriented optimizations, including market data APIs, data preprocessing, DRL algorithms, and automated backtesting. Users can transparently make use of such a development pipeline. • Full-stack framework. • Customization. To maintain modularity and extensibility in development by including state-of-the-art DRL algorithms and supporting design of new algorithms. The DRL algorithms can be used to construct trading strategies by simple configurations. • Customization. • Reproducibility and hands-on tutoring. To provide tutorials such as step-by-step Jupyter notebooks and user guide to help users walk through the pipeline and reproduce the use cases. • Reproducibility and hands-on tutoring. This leads to a unified framework where developers are able to efficiently explore ideas through high-level configurations and specifications, and to customize their own strategies at request. Our contributions are summarized as follows: • FinRL is the first open-source framework that demonstrates the great potential of applying DRL algorithms in quantitative finance. We build an ecosystem around the FinRL framework, which seeds the rapidly growing AI4Finance community. • The application layer provides interfaces for users to customize FinRL to their own trading tasks. Automated backtesting module and performance metrics are provided to help quantitative traders iterate trading strategies at a high turnover rate. Profitable trading strategies are reproducible and hands-on tutorials are provided in a beginner-friendly fashion. Adjusting the trained models to the rapid changing markets is also possible. • The agent layer provides state-of-the-art DRL algorithms that are adapted to finance with fine-tuned hyperparameters. Users can add new DRL algorithms. • The environment layer includes not only a collection of historical data APIs, but also live trading APIs. They are reconfigured into standard OpenAI gym-style environments [5]. Moreover, it incorporates market frictions and allows users to customize the trading time granularity. The remainder of this paper is organized as follows. Section 2 reviews related works. Section 3 presents the FinRL framework. Section 4 demonstrates benchmark trading tasks using FinRL. We conclude this paper in Section 5. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

Deep Reinforcement Learning Framework to Automate Trading in Quantitative Finance

পরিমাণগত অর্থায়নে ট্রেডিং স্বয়ংক্রিয় করার জন্য গভীর শক্তিবৃদ্ধি শেখার কাঠামো

Deep Reinforcement Learning Framework zur Automatisierung des Handels im quantitativen Finanzwesen

Marco de aprendizaje por refuerzo profundo para automatizar el comercio en finanzas cuantitativas

Cadre d'apprentissage par renforcement profond pour automatiser le trading en finance quantitative

क्वांटिटेटिव फाइनेंस में ट्रेडिंग को स्वचालित करने के लिए डीप रीइनफोर्समेंट लर्निंग फ्रेमवर्क

定量金融における取引を自動化する深層強化学習フレームワーク

양적 금융 거래를 자동화하기 위한 심층 강화 학습 프레임워크

Estrutura de aprendizagem por reforço profundo para automatizar negociações em finanças quantitativas

Структура глубокого обучения с подкреплением для автоматизации торговли в области количественных финансов

Kantitatif Finansta Ticareti Otomatikleştirmek için Derin Takviyeli Öğrenme Çerçevesi

Khung học tập củng cố sâu để tự động hóa giao dịch trong tài chính định lượng

深度强化学习框架实现量化金融交易自动化

Stories, News & Mentions about CQUANTIO

Cquantio’s Stories on HackerNoon

Cquantio's News & Mentions