Improved inter-scanner ms lesion segmentation by adversarial training on longitudinal data

Mattias Billast*, Maria Ines Meyer, Diana M. Sima, David Robben

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

Abstract

The evaluation of white matter lesion progression is an important biomarker in the follow-up of MS patients and plays a crucial role when deciding the course of treatment. Current automated lesion segmentation algorithms are susceptible to variability in image characteristics related to MRI scanner or protocol differences. We propose a model that improves the consistency of MS lesion segmentations in inter-scanner studies. First, we train a CNN base model to approximate the performance of icobrain, an FDA-approved clinically available lesion segmentation software. A discriminator model is then trained to predict if two lesion segmentations are based on scans acquired using the same scanner type or not, achieving a 78% accuracy in this task. Finally, the base model and the discriminator are trained adversarially on multi-scanner longitudinal data to improve the inter-scanner consistency of the base model. The performance of the models is evaluated on an unseen dataset containing manual delineations. The inter-scanner variability is evaluated on test-retest data, where the adversarial network produces improved results over the base model and the FDA-approved solution.
Original languageEnglish
Title of host publicationBrainlesion : Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - Part 1
Volume11992
PublisherSpringer
Publication date2020
Pages98-107
ISBN (Print)978-3-030-46639-8
ISBN (Electronic)978-3-030-46640-4
DOIs
Publication statusPublished - 2020
Event5th International MICCAI Brainlesion Workshop, BrainLes 2019, held in conjunction with the Medical Image Computing for Computer Assisted Intervention, MICCAI 2019 - Shenzhen, China
Duration: 17 Oct 201917 Oct 2019

Conference

Conference5th International MICCAI Brainlesion Workshop, BrainLes 2019, held in conjunction with the Medical Image Computing for Computer Assisted Intervention, MICCAI 2019
Country/TerritoryChina
CityShenzhen
Period17/10/201917/10/2019
SeriesLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11992
ISSN0302-9743

Keywords

  • Deep learning
  • Inter-scanner
  • Lesion segmentation
  • Adversarial training
  • Longitudinal data
  • Multiple sclerosis

Fingerprint

Dive into the research topics of 'Improved inter-scanner ms lesion segmentation by adversarial training on longitudinal data'. Together they form a unique fingerprint.

Cite this