Related Experiment Video
Updated: Jul 26, 2026
Whole-brain Segmentation and Change-point Analysis of Anatomical Brain MRI—Application in Premanifest Huntington's Disease
Published on: June 9, 2018
Deep learning detection of acute and sub-acute lesion activity from single-timepoint conventional brain MRI in
Quentin Spinat1, Benoit Audelan1, Xiaotong Jiang2
1TheraPanacea, Paris, France.
Abstract:
Multiple sclerosis (MS) is a chronic inflammatory disease characterized by demyelinating lesions in the central nervous system. Cross-sectional measurements of acute inflammatory lesion activity are typically obtained by detecting the presence of gadolinium enhancement in lesions, which typically lasts 3-6 weeks. We formulate the novel and clinically relevant task of quantification of recent acute lesion activity from the past 24 weeks (6 months) using single-timepoint conventional brain magnetic resonance imaging (MRI). We develop and compare several deep learning (DL) methods for estimating this brain-level acuteness score and show that a 2D-UNet can accurately predict acute disease activity at the patient-level while outperforming transformers and ensemble approaches. In the context of identifying subjects with acute (less than 6 months-old) lesion activity, our 2D-UNet achieves an area under the receiver-operating curve in the range 80-84% on independent relapsing-remitting MS cohorts. When used in conjunction with measurements of gadolinium-enhancing lesion activity, our model significantly improves the prognostication of future acute lesion activity (over the next 6 months). This model could thus be leveraged for population recruitment in clinical trials to identify a higher number of patients with acute inflammatory activity than current standard approaches (e.g., gadolinium positivity) with a predictable precision/recall trade-off.
Insights
This study introduces a deep learning model to quantify recent inflammatory activity in multiple sclerosis (MS) using brain MRI. The 2D-UNet accurately predicts acute lesions, improving clinical trial recruitment.
Area of Science:
- Neurology
- Medical Imaging
- Artificial Intelligence
Background:
- Multiple sclerosis (MS) is a chronic central nervous system inflammatory disease.
- Current methods for assessing acute inflammatory lesion activity rely on gadolinium enhancement, which has a limited detection window of 3-6 weeks.
- There is a need for methods to quantify recent acute lesion activity over longer periods using conventional MRI.
Purpose of the Study:
- To develop and evaluate deep learning models for quantifying recent acute lesion activity in multiple sclerosis (MS) using single-timepoint conventional brain MRI.
- To assess the performance of a 2D-UNet model in predicting acute disease activity within the past 24 weeks.
- To determine if the developed model can improve the prognostication of future acute lesion activity.
Main Methods:
- Formulation of a novel task: quantification of recent acute lesion activity (past 24 weeks) from conventional brain MRI.
- Development and comparison of several deep learning (DL) methods, including 2D-UNet, transformers, and ensemble approaches.
- Evaluation of model performance using area under the receiver-operating curve (AUC) on independent relapsing-remitting MS cohorts.
Main Results:
- A 2D-UNet model accurately predicts acute disease activity at the patient-level, outperforming transformers and ensemble methods.
- The 2D-UNet achieved an AUC of 80-84% for identifying subjects with acute lesion activity (<6 months).
- Integration of the model with gadolinium-enhancing lesion data significantly improved the prognostication of future acute lesion activity.
Conclusions:
- Deep learning, specifically a 2D-UNet, can accurately quantify recent acute lesion activity in MS using conventional brain MRI.
- This approach offers a more comprehensive assessment of disease activity compared to traditional gadolinium enhancement.
- The model holds potential for optimizing patient recruitment in clinical trials by identifying individuals with higher recent inflammatory activity.
