Julien Cohen-Adad

Pierre Louis Benveniste

PhD - Polytechnique Montréal

Co-supervisor :

PhD - Polytechnique Montréal

Thomas Dagonneau

Master's Research - Polytechnique Montréal

Naga Karthik Enamundram

PhD - Polytechnique Montréal

Co-supervisor :

Master's Research - Polytechnique Montréal

Youssef Laatar

Master's Research - Polytechnique Montréal

Julien Laborde-Peyré

Research Intern - Polytechnique Montréal

Nilser Laines Medina

PhD - Polytechnique Montréal

Nathan Molinier

PhD - Polytechnique Montréal

Kalum Ost

Collaborating researcher

Arthur Toulouse

Master's Research - Polytechnique Montréal

Publications

Effectiveness of regional diffusion MRI measures in distinguishing multiple sclerosis abnormalities within the cervical spinal cord

Haykel Snoussi

Julien Cohen‐Adad

Olivier Commowick

Benoit Combes

Elise Bannier

Slimane Tounekti

Anne Kerbrat

Christian Barillot

Emmanuel Caruyer

Multiple sclerosis (MS) is an inflammatory disorder of the central nervous system. Although conventional magnetic resonance imaging (MRI) is… (see more) widely used for MS diagnosis and clinical follow‐up, quantitative MRI has the potential to provide valuable intrinsic values of tissue properties that can enhance accuracy. In this study, we investigate the efficacy of diffusion MRI in distinguishing MS lesions within the cervical spinal cord, using a combination of metrics extracted from diffusion tensor imaging and Ball‐and‐Stick models.

2023-09-28

Brain and Behavior (published)

Influence of preprocessing, distortion correction and cardiac triggering on the quality of diffusion MR images of spinal cord

Kurt G. Schilling

Anna Combes

Karthik Ramadass

François Rheault

Grace Sweeney

Logan Prock

Subramaniam Sriram

Julien Cohen‐Adad

John C. Gore

Bennett A. Landman

Seth A. Smith

Kristin P. O’Grady

Diffusion MRI of the spinal cord (SC) is susceptible to geometric distortion caused by field inhomogeneities, and prone to misalignment acro… (see more)ss time series and signal dropout caused by biological motion. Several modifications of image acquisition and image processing techniques have been introduced to overcome these artifacts, but their specific benefits are largely unproven and warrant further investigations. We aim to evaluate two specific aspects of image acquisition and processing that address image quality in diffusion studies of the spinal cord: susceptibility corrections to reduce geometric distortions, and cardiac triggering to minimize motion artifacts. First, we evaluate 4 distortion preprocessing strategies on 7 datasets of the cervical and lumbar SC and find that while distortion correction techniques increase geometric similarity to structural images, they are largely driven by the high-contrast cerebrospinal fluid, and do not consistently improve the geometry within the cord nor improve white-to-gray matter contrast. We recommend at a minimum to perform bulk-motion correction in preprocessing and posit that improvements/adaptations are needed for spinal cord distortion preprocessing algorithms, which are currently optimized and designed for brain imaging. Second, we design experiments to evaluate the impact of removing cardiac triggering. We show that when triggering is foregone, images are qualitatively similar to triggered sequences, do not have increased prevalence of artifacts, and result in similar diffusion tensor indices with similar reproducibility to triggered acquisitions. When triggering is removed, much shorter acquisitions are possible, which are also qualitatively and quantitatively similar to triggered sequences. We suggest that removing cardiac triggering for cervical SC diffusion can be a reasonable option to save time with minimal sacrifice to image quality.

2023-09-26

bioRxiv (preprint)

Pontomedullary junction as a reference for spinal cord cross-sectional area: validation across neck positions

Sandrine Bédard

Maxime Bouthillier

Spinal cord cross-sectional area (CSA) is an important MRI biomarker to assess spinal cord atrophy in various neurodegenerative and traumati… (see more)c spinal cord diseases. However, the conventional method of computing CSA based on vertebral levels is inherently flawed, as the prediction of spinal levels from vertebral levels lacks reliability, leading to considerable variability in CSA measurements. Computing CSA from an intrinsic neuroanatomical reference, the pontomedullary junction (PMJ), has been proposed in previous work to overcome limitations associated with using a vertebral reference. However, the validation of this alternative approach, along with its variability across and within participants under variable neck extensions, remains unexplored. The goal of this study was to determine if the variability of CSA across neck flexions/extensions is reduced when using the PMJ, compared to vertebral levels. Ten participants underwent a 3T MRI T2w isotropic scan at 0.6 mm3 for 3 neck positions: extension, neutral and flexion. Spinal cord segmentation, vertebral labeling, PMJ labeling, and CSA were computed automatically while spinal segments were labeled manually. Mean coefficient of variation for CSA across neck positions was 3.99 ± 2.96% for the PMJ method vs. 4.02 ± 3.01% for manual spinal segment method vs. 4.46 ± 3.10% for the disc method. These differences were not statistically significant. The PMJ method was slightly more reliable than the disc-based method to compute CSA at specific spinal segments, although the difference was not statistically significant. This suggests that the PMJ can serve as a valuable alternative and reliable method for estimating CSA when a disc-based approach is challenging or not feasible, such as in cases involving fused discs in individuals with spinal cord injuries.

2023-08-18

Scientific Reports (published)

An 8-channel Tx dipole and 20-channel Rx loop coil array for MRI of the cervical spinal cord at 7 Tesla

Nibardo Lopez-Rios

Kyle M. Gilbert

Daniel Papp

Gaspard Cereza

Alexandru Foias

Deshpande Rangaprakash

Markus W. May

Bastien Guerin

Lawrence L. Wald

Boris Keil

Jason P. Stockmann

Robert L. Barry

The quality of cervical spinal cord images can be improved by the use of tailored radiofrequency coil solutions for ultra-high field imaging… (see more); however, very few commercial and research 7 Tesla radiofrequency coils currently exist for the spinal cord, and in particular those with parallel transmit capabilities. This work presents the design, testing and validation of a pTx/Rx coil for the human neck and cervical/upper-thoracic spinal cord. The pTx portion is composed of 8 dipoles to ensure high homogeneity over this large region of the spinal cord. The Rx portion is made of 20 semi-adaptable overlapping loops to produce high Signal-to-noise ratio (SNR) across the patient population. The coil housing is designed to facilitate patient positioning and comfort, while being tight fitting to ensure high sensitivity. We demonstrate RF shimming capabilities to optimize B1+ uniformity, power efficiency and/or specific absorption rate (SAR) efficiency. B1+ homogeneity, SNR and g-factor was evaluated in adult volunteers and demonstrated excellent performance from the occipital lobe down to the T4-T5 level. We compared the proposed coil with two state-of-the-art head and head/neck coils, confirming its superiority in the cervical and upper-thoracic regions of the spinal cord. This coil solution therefore provides a convincing platform for producing the high image quality necessary for clinical and research scanning of the upper spinal cord.

2023-07-12

NMR in Biomedicine (published)

Enhancing Medical Image Segmentation with TransCeption: A Multi-Scale Feature Fusion Approach

Reza Azad

Yiwei Jia

Ehsan Khodapanah Aghdam

Dorit Merhof

While CNN-based methods have been the cornerstone of medical image segmentation due to their promising performance and robustness, they suff… (see more)er from limitations in capturing long-range dependencies. Transformer-based approaches are currently prevailing since they enlarge the reception field to model global contextual correlation. To further extract rich representations, some extensions of the U-Net employ multi-scale feature extraction and fusion modules and obtain improved performance. Inspired by this idea, we propose TransCeption for medical image segmentation, a pure transformer-based U-shape network featured by incorporating the inception-like module into the encoder and adopting a contextual bridge for better feature fusion. The design proposed in this work is based on three core principles: (1) The patch merging module in the encoder is redesigned with ResInception Patch Merging (RIPM). Multi-branch transformer (MB transformer) adopts the same number of branches as the outputs of RIPM. Combining the two modules enables the model to capture a multi-scale representation within a single stage. (2) We construct an Intra-stage Feature Fusion (IFF) module following the MB transformer to enhance the aggregation of feature maps from all the branches and particularly focus on the interaction between the different channels of all the scales. (3) In contrast to a bridge that only contains token-wise self-attention, we propose a Dual Transformer Bridge that also includes channel-wise self-attention to exploit correlations between scales at different stages from a dual perspective. Extensive experiments on multi-organ and skin lesion segmentation tasks present the superior performance of TransCeption compared to previous work. The code is publicly available at https://github.com/mindflow-institue/TransCeption.

2023-01-24

ArXiv (preprint)

Label fusion and training methods for reliable representation of inter-rater uncertainty

Andréanne Lemay

Charley Gros

Enamundram Naga Karthik

Medical tasks are prone to inter-rater variability due to multiple factors such as image quality, professional experience and training, or g… (see more)uideline clarity. Training deep learning networks with annotations from multiple raters is a common practice that mitigates the model's bias towards a single expert. Reliable models generating calibrated outputs and reflecting the inter-rater disagreement are key to the integration of artificial intelligence in clinical practice. Various methods exist to take into account different expert labels. We focus on comparing three label fusion methods: STAPLE, average of the rater's segmentation, and random sampling of each rater's segmentation during training. Each label fusion method is studied using both the conventional training framework and the recently published SoftSeg framework that limits information loss by treating the segmentation task as a regression. Our results, across 10 data splittings on two public datasets, indicate that SoftSeg models, regardless of the ground truth fusion method, had better calibration and preservation of the inter-rater rater variability compared with their conventional counterparts without impacting the segmentation performance. Conventional models, i.e., trained with a Dice loss, with binary inputs, and sigmoid/softmax final activate, were overconfident and underestimated the uncertainty associated with inter-rater variability. Conversely, fusing labels by averaging with the SoftSeg framework led to underconfident outputs and overestimation of the rater disagreement. In terms of segmentation performance, the best label fusion method was different for the two datasets studied, indicating this parameter might be task-dependent. However, SoftSeg had segmentation performance systematically superior or equal to the conventionally trained models and had the best calibration and preservation of the inter-rater variability.

2023-01-17

Machine Learning for Biomedical Imaging (published)

HiFormer: Hierarchical Multi-scale Representations Using Transformers for Medical Image Segmentation

Moein Heidari

Amirhossein Kazerouni

Milad Soltany

Reza Azad

Ehsan Khodapanah Aghdam

Dorit Merhof

Convolutional neural networks (CNNs) have been the consensus for medical image segmentation tasks. However, they suffer from the limitation … (see more)in modeling long-range dependencies and spatial correlations due to the nature of convolution operation. Although transformers were first developed to address this issue, they fail to capture low-level features. In contrast, it is demonstrated that both local and global features are crucial for dense prediction, such as segmenting in challenging contexts. In this paper, we propose HiFormer, a novel method that efficiently bridges a CNN and a transformer for medical image segmentation. Specifically, we design two multi-scale feature representations using the seminal Swin Transformer module and a CNN-based encoder. To secure a fine fusion of global and local features obtained from the two aforementioned representations, we propose a Double-Level Fusion (DLF) module in the skip connection of the encoder-decoder structure. Extensive experiments on various medical image segmentation datasets demonstrate the effectiveness of HiFormer over other CNN-based, transformer-based, and hybrid methods in terms of computational complexity, and quantitative and qualitative results. Our code is publicly available at: https://github.com/amirhossein-kz/HiFormer

2023-01-01

2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) (published)

Biomedical image analysis competitions: The state of current participation practice

Matthias Eisenmann

Annika Reinke

Vivienn Weru

Minu Dietlinde Tizabi

Fabian Isensee

T. Adler

PATRICK GODAU

Veronika Cheplygina

Michal Kozubek

Sharib Ali

Anubha Gupta

Jan. Kybic

Alison Professor Noble

Carlos Ortiz de Sol'orzano

Samiksha Pachade

Caroline Petitjean

Daniel Sage

Donglai Wei

Elizabeth Wilden

Deepak Alapatt … (see 334 more)

Vincent Andrearczyk

Ujjwal Baid

Spyridon Bakas

Niranjan Balu

Sophia Bano

Vivek Singh Bawa

Jorge Bernal

Sebastian Bodenstedt

Alessandro Casella

Jinwook Choi

Olivier Commowick

M. Daum

Adrien Depeursinge

Reuben Dorent

J. Egger

H. Eichhorn

Sandy Engelhardt

Melanie Ganz

Gabriel Girard

Lasse Donovan Hansen

Mattias Paul Heinrich

Nicholas Heller

Alessa Hering

Arnaud Huaulm'e

Hyunjeong Kim

Bennett Landman

Hongwei Bran Li

Jianning Li

Junfang Ma

Anne L. Martel

Carlos Mart'in-Isla

Bjoern Menze

Chinedu Innocent Nwoye

Valentin Oreiller

Nicolas Padoy

Sarthak Pati

Kelly Payette

Carole H. Sudre

K. V. Wijnen

Armine Vardazaryan

Tom Kamiel Magda Vercauteren

Martin Wagner

Chuanbo Wang

Moi Hoon Yap

Zeyun Yu

Chuner Yuan

Maximilian Zenk

Aneeq Zia

David Zimmerer

Rina Bao

Chanyeol Choi

Andrew Cohen

Oleh Dzyubachyk

Adrian Galdran

Tianyuan Gan

Tianqi Guo

Pradyumna Gupta

M. Haithami

Edward Ho

Ikbeom Jang

Zhili Li

Zheng Luo

Filip Lux

Sokratis Makrogiannis

Dominikus Muller

Young-Tack Oh

Subeen Pang

Constantin Pape

Gorkem Polat

Charlotte Rosalie Reed

Kanghyun Ryu

Tim Scherr

Vajira L. Thambawita

Haoyu Wang

Xinliang Wang

Kele Xu

H.-I. Yeh

Doyeob Yeo

Yi Yuan

Yan Zeng

Xingwen Zhao

Julian Ronald Abbing

Jannes Adam

Nagesh Adluru

Niklas Agethen

S. Ahmed

Yasmina Al Khalil

Mireia Alenya

Esa J. Alhoniemi

C. An

Talha E Anwar

Tewodros Arega

Netanell Avisdris

D. Aydogan

Yi-Shi Bai

Maria Baldeon Calisto

Berke Doga Basaran

Marcel Beetz

Cheng Bian

Hao-xuan Bian

Kevin Blansit

Louise Bloch

Robert Bohnsack

Sara Bosticardo

J. Breen

Mikael Brudfors

Raphael Brungel

Mariano Cabezas

Alberto Cacciola

Zhiwei Chen

Yucong Chen

Dan Chen

Minjeong Cho

Min-Kook Choi

Chuantao Xie Chuantao Xie

Dana Cobzas

Jorge Corral Acero

Sujit Kumar Das

Marcela de Oliveira

Hanqiu Deng

Guiming Dong

Lars Doorenbos

Cory Efird

Di Fan

Mehdi Fatan Serj

Alexandre Fenneteau

Lucas Fidon

Patryk Filipiak

Ren'e Finzel

Nuno Renato Freitas

C. Friedrich

Mitchell J. Fulton

Finn Gaida

Francesco Galati

Christoforos Galazis

Changna Gan

Zheyao Gao

Sheng Gao

Matej Gazda

Beerend G. A. Gerats

Neil Getty

Adam Gibicar

Ryan J. Gifford

Sajan Gohil

Maria Grammatikopoulou

Daniel Grzech

Orhun Guley

Timo Gunnemann

Chun-Hai Guo

Sylvain Guy

Heonjin Ha

Luyi Han

Ilseok Han

Ali Hatamizadeh

Tianhai He

Ji-Wu Heo

Sebastian Hitziger

SeulGi Hong

Seungbum Hong

Rian Huang

Zi-You Huang

Markus Huellebrand

Stephan Huschauer

M. Hussain

Tomoo Inubushi

Ece Isik Polat

Mojtaba Jafaritadi

Seonghun Jeong

Bailiang Jian

Yu Jiang

Zhifan Jiang

Yu Jin

Smriti Joshi

A. Kadkhodamohammadi

R. A. Kamraoui

Inhak Kang

Jun-Su Kang

Davood Karimi

April Ellahe Khademi

Muhammad Irfan Khan

Suleiman A. Khan

Rishab Khantwal

Kwang-Ju Kim

Timothy Lee Kline

Satoshi Kondo

Elina Kontio

Adrian Krenzer

Artem Kroviakov

Hugo J. Kuijf

Satyadwyoom Kumar

Francesco La Rosa

Abhishek Lad

Doohee Lee

Minho Lee

Chiara Lena

Hao Li

Ling Li

Xingyu Li

F. Liao

Kuan-Ya Liao

Arlindo L. Oliveira

Chaonan Lin

Shanhai Lin

Akis Linardos

M. Linguraru

Han Liu

Tao Liu

Dian Liu

Yanling Liu

Joao Lourencco-Silva

Jing Lu

Jia Lu

Imanol Luengo

Christina Bach Lund

Huan Minh Luu

Yingqi Lv

Uzay Macar

Leon Maechler

L. SinaMansour

Kenji Marshall

Moona Mazher

Richard McKinley

Alfonso Medela

Felix Meissen

Mingyuan Meng

Dylan Bradley Miller

S. Mirjahanmardi

Arnab Kumar Mishra

Samir Mitha

Hassan Mohy-ud-Din

Tony C. W. Mok

Gowtham Krishnan Murugesan

Enamundram Naga Karthik

Sahil Nalawade

Jakub Nalepa

M. Naser

Ramin Nateghi

Hammad Naveed

Quang-Minh Nguyen

Cuong Nguyen Quoc

Brennan Nichyporuk

Bruno Oliveira

David Owen

Jimut Bahan Pal

Junwen Pan

Wei-Dong Pan

Winnie Pang

Bogyu Park

Vivek G. Pawar

Kamlesh Pawar

Michael Peven

Lena Philipp

Tomasz Pieciak

Szymon S Płotka

Marcel Plutat

Fattane Pourakpour

Domen Prelovznik

K. Punithakumar

Abdul Qayyum

Sandro Queir'os

Arman Rahmim

Salar Razavi

Jintao Ren

Mina Rezaei

Jonathan Adam Rico

ZunHyan Rieu

Markus Rink

Johannes Roth

Yusely Ruiz-gonzalez

Numan Saeed

Anindo Saha

Mostafa M. Sami Salem

Ricardo Sanchez-matilla

Kurt G Schilling

Weizhen Shao

Zhiqiang Shen

Ruize Shi

Pengcheng Shi

Daniel Sobotka

Th'eodore Soulier

Bella Specktor Fadida

D. Stoyanov

Timothy Sum Hon Mun

Xiao-Fu Sun

Rong Tao

Franz Thaler

Antoine Th'eberge

Felix Thielke

Helena R. Torres

K. Wahid

Jiacheng Wang

Yifei Wang

Wei David Wang

Xiong Jun Wang

Jianhui Wen

Ning Wen

Marek Wodziński

Yehong Wu

Fangfang Xia

Tianqi Xiang

Cheng Xiaofei

Lizhang Xu

Tingting Xue

Yu‐Xia Yang

Lingxian Yang

Kai Yao

Huifeng Yao

Amirsaeed Yazdani

Michael Yip

Hwa-Seong Yoo

Fereshteh Yousefirizi

Shu-Fen Yu

Lei Yu

Jonathan Zamora

Ramy A. Zeineldin

Dewen Zeng

Jianpeng Zhang

Bokai Zhang

Jiapeng Zhang

Fangxi Zhang

Huahong Zhang

Zhongchen Zhao

Zixuan Zhao

Jia Zhao

Can Zhao

Qiuyue Zheng

Yuheng Zhi

Ziqi Zhou

Baosheng Zou

Klaus Maier-Hein

PAUL F. JÄGER

Annette Kopp-Schneider

Lena Maier-Hein

2022-12-15

ArXiv (preprint)

Histology-informed automatic parcellation of white matter tracts in the rat spinal cord

Harris Nami

Christian S. Perone

The white matter is organized into “tracts” or “bundles,” which connect different parts of the central nervous system. Knowing where… (see more) these tracts are located in each individual is important for understanding the cause of potential sensorial, motor or cognitive deficits and for developing appropriate treatments. Traditionally, tracts are found using tracer injection, which is a difficult, slow and poorly scalable technique. However, axon populations from a given tract exhibit specific characteristics in terms of morphometrics and myelination. Hence, the delineation of tracts could, in principle, be done based on their morphometry. The objective of this study was to generate automatic parcellation of the rat spinal white matter tracts using the manifold information from scanning electron microscopy images of the entire spinal cord. The axon morphometrics (axon density, axon diameter, myelin thickness and g-ratio) were computed pixelwise following automatic axon segmentation using AxonSeg. The parcellation was based on an agglomerative clustering algorithm to group the tracts. Results show that axon morphometrics provide sufficient information to automatically identify some white matter tracts in the spinal cord, however, not all tracts were correctly identified. Future developments of microstructure quantitative MRI even bring hope for a personalized clustering of white matter tracts in each individual patient. The generated atlas and the associated code can be found at https://github.com/neuropoly/tract-clustering.

2022-11-28

Frontiers in Neuroanatomy (published)

Shimming toolbox: An open‐source software toolbox for <scp>B0</scp> and <scp>B1</scp> shimming in MRI

Alexandre D'Astous

Gaspard Cereza

Daniel Papp

Kyle M. Gilbert

Jason P. Stockmann

Eva Alonso‐Ortiz

Julien Cohen‐Adad

Introduce Shimming Toolbox ( https://shimming‐toolbox.org), an open‐source software package for prototyping new methods and performing s… (see more)tatic, dynamic, and real‐time B0 shimming as well as B1 shimming experiments. Shimming Toolbox features various field mapping techniques, manual and automatic masking for the brain and spinal cord, B0 and B1 shimming capabilities accessible through a user‐friendly graphical user interface. Validation of Shimming Toolbox was demonstrated in three scenarios: (i) B0 dynamic shimming in the brain at 7T using custom AC/DC coils, (ii) B0 real‐time shimming in the spinal cord at 3T, and (iii) B1 static shimming in the spinal cord at 7T. Shimming Toolbox provides an open‐source platform where researchers can collaborate, prototype and conveniently test B0 and B1 shimming experiments. Future versions will include additional field map preprocessing techniques, optimization algorithms, and compatibility across multiple MRI manufacturers.

2022-11-27

Magnetic Resonance in Medicine (unknown)

Automatic measure and normalization of spinal cord cross-sectional area using the pontomedullary junction

Sandrine Bédard

Julien Cohen‐Adad

Spinal cord cross-sectional area (CSA) is a relevant biomarker to assess spinal cord atrophy in neurodegenerative diseases. However, the con… (see more)siderable inter-subject variability among healthy participants currently limits its usage. Previous studies explored factors contributing to the variability, yet the normalization models required manual intervention and used vertebral levels as a reference, which is an imprecise prediction of the spinal levels. In this study we implemented a method to measure CSA automatically from a spatial reference based on the central nervous system (the pontomedullary junction, PMJ), we investigated factors to explain variability, and developed normalization strategies on a large cohort (N = 804). Following automatic spinal cord segmentation, vertebral labeling and PMJ labeling, the spinal cord CSA was computed on T1w MRI scans from the UK Biobank database. The CSA was computed using two methods. For the first method, the CSA was computed at the level of the C2–C3 intervertebral disc. For the second method, the CSA was computed at 64 mm caudally from the PMJ, this distance corresponding to the average distance between the PMJ and the C2–C3 disc across all participants. The effect of various demographic and anatomical factors was explored, and a stepwise regression found significant predictors; the coefficients of the best fit model were used to normalize CSA. CSA measured at C2–C3 disc and using the PMJ differed significantly (paired t-test, p-value = 0.0002). The best normalization model included thalamus, brain volume, sex and the interaction between brain volume and sex. The coefficient of variation went down for PMJ CSA from 10.09 (without normalization) to 8.59%, a reduction of 14.85%. For CSA at C2–C3, it went down from 9.96 to 8.42%, a reduction of 15.13 %. This study introduces an end-to-end automatic pipeline to measure and normalize cord CSA from a neurological reference. This approach requires further validation to assess atrophy in longitudinal studies. The inter-subject variability of CSA can be partly accounted for by demographics and anatomical factors.

2022-11-01

Frontiers in Neuroimaging (published)

Segmentation of Multiple Sclerosis Lesions across Hospitals: Learn Continually or Train from Scratch?

Enamundram Naga Karthik

Anne Kerbrat

Pierre Labauge

Tobias Granberg

Jason F. Talbott

Daniel S Reich

Massimo Filippi

Rohit Bakshi

Virginie Callot

A. Chandar

Segmentation of Multiple Sclerosis (MS) lesions is a challenging problem. Several deep-learning-based methods have been proposed in recent y… (see more)ears. However, most methods tend to be static, that is, a single model trained on a large, specialized dataset, which does not generalize well. Instead, the model should learn across datasets arriving sequentially from different hospitals by building upon the characteristics of lesions in a continual manner. In this regard, we explore experience replay, a well-known continual learning method, in the context of MS lesion segmentation across multi-contrast data from 8 different hospitals. Our experiments show that replay is able to achieve positive backward transfer and reduce catastrophic forgetting compared to sequential fine-tuning. Furthermore, replay outperforms the multi-domain training, thereby emerging as a promising solution for the segmentation of MS lesions. The code is available at this link: https://github.com/naga-karthik/continual-learning-ms

2022-10-26

ArXiv (preprint)