University Hospital of Liège
NEUROLOGY

Author Of 2 Presentations

Machine Learning/Network Science Poster Presentation

P0005 - Decoding the EDSS Scores of Multiple Sclerosis Patients from MRI Biomarkers (ID 1620)

Speakers
Presentation Number
P0005
Presentation Topic
Machine Learning/Network Science

Abstract

Background

Treatment response in multiple sclerosis (MS) is frequently suboptimal and, in many cases, different disease-modifying therapies need to be tested. The expanded disability status scale (EDSS) score and other markers of disease activity (such as relapses, new lesions, brain atrophy, etc.) are crucial for treatment decision making. However, EDSS suffers from poor reliability, repeatability, and high inter-rater variability. Therefore, automatic and objective disability scoring using MRI information could help to monitor disease progression reliably and optimize treatment.

Objectives

Develop a machine-learning model that learns relations between MRI-based brain volumes and clinical disability measured by EDSS.

Methods

Multi-center FLAIR and T1 MRI data from 325 MS patients were used. Individuals were rated in each center using EDSS within 0-89 days before or after the MRI scan. Automated image analysis was performed using icobrain, providing volumetric quantification of gray matter, white matter, whole brain, lateral ventricles, T1 hypointense and FLAIR hyperintense lesions. Moreover, other features, such as age, sex, and center, were available. A machine-learning model based on random forest regression was built for estimating EDSS automatically from these features. The model’s performance was assessed by means of mean squared error (MSE) and mean absolute error (MAE) evaluated overall, and on two EDSS subgroups, <=4 and >4,in a 100 repetition 10-fold-cross-validation fashion. Subsequently, the percentage of cases for which the automatic EDSS was within 1.5, 1 or 0.5 points, respectively, from the clinically reported EDSS was computed.

Results

The proposed automatic EDSS estimation model obtained MSE=2.36±0.03, MAE=1.24±0.89 for the overall interval, with MSE=1.83±0.04, MAE=1.11±0.76 for EDSS<=4 (N=200) and MSE=3.26±0.06, MAE=1.46±1.05 for EDSS>4 (N=118). The percentage of cases with absolute error strictly below 1.5, 1 and 0.5 EDSS points was 67%, 46% and 24%, respectively.

Conclusions

A good match between the automatic EDSS and the measured EDSS is only possible up to a certain extent, suggesting that the MRI-based EDSS score might also capture complementary information on disease activity compared to the clinically measured EDSS. Understanding such differences is a prerequisite for predicting future disability progression in MS.

Collapse
Machine Learning/Network Science Poster Presentation

P0015 - Predicted brain age as a cognitive biomarker in Multiple Sclerosis (ID 1388)

Speakers
Presentation Number
P0015
Presentation Topic
Machine Learning/Network Science

Abstract

Background

Brain-predicted age difference (BPAD) is the difference between chronological and brain age, the latter being decoded from brain MR images by a machine learning model. BPAD has been established as a biomarker for physical deterioration in multiple sclerosis (MS).

Objectives

The objective was to examine the value of brain-predicted age difference as a biomarker for cognitive deterioration in MS.

Methods

In the first phase we used supervised machine learning (ridge regression, 7-fold cross-validation) to predict chronological age from brain volumes. This was achieved on a sample of 1743 healthy controls (HC) with T1-weighted MR images from a range of public collections, using the icobrain software to compute normalized brain volumes (whole brain, white matter, (cortical) grey matter and lateral ventricles). The age range of this sample was 8 to 94 years. In the second phase we applied this algorithm to decode brain age in 231 T1-weighted MR images from MS patients from two clinical centers. The age range of this sample was 14 to 77 years. BPAD computed for HC (BPADHC) and MS (BPADMS) were compared with an independent student t-test. Cognitive functioning was assessed through the symbol digit modalities test (SDMT), available for all 231 included MS patients, the brief visuospatial memory test revised (BVMT-R) and California verbal learning test II (CVLT-II), the latter two being available for a subset of 97 patients from one center. Pearson correlation was computed between the BPAD values and scores on each cognitive test. P-values were corrected for multiple comparisons by the Benjamini/Hochberg method.

Results

After training, the mean (SD) BPADHC in the best-performing of seven folds was -0.5 (9.2) years, opposed to 9.7 (12.0) years for BPADMS. This difference was statistically significant (p < 0.001). Correlations between BPADMS and SDMT, BVMT-R and CVLT-II were -0.19 (p = 0.005), -0.14 (p = 0.157) and -0.30 (p = 0.005) respectively.

Conclusions

Our results showed a significant association between BPADMS and two cognitive measures in MS, information processing speed and verbal memory. These findings support a potential role for BPADMS as a cognitive biomarker in MS.

Collapse