University of British Columbia

Author Of 2 Presentations

Imaging Late Breaking Abstracts

LB1197 - Myelin water imaging provides evidence for unique anatomical-functional relationships between myelin damage and different cognitive domains in MS (ID 2022)

Abstract

Background

Background: An improved understanding of the impact of demyelination on multiple sclerosis (MS) related cognitive impairment is crucial for targeting and testing therapies with the potential to slow cognitive decline. Demyelination can be assessed using myelin water imaging, a quantitative magnetic resonance imaging (MRI) technique that measures signal from water in the myelin bilayers, providing a specific measure of myelin (myelin water fraction, MWF).

Objectives

Objective: To determine if there is an anatomical-functional relationship between myelin content and location with cognitive performance.

Methods

Methods: 76 MS participants (mean (SD) age:50.4y(10.6y), 51F) underwent T2 relaxation imaging to calculate MWF maps and cognitive testing (Symbol Digit Modalities Test (SDMT); Selective Reminding Test (SRT); Controlled Oral Word Association Test (COWAT); Brief Visuospatial Memory Test-Revised (BVMT-R)). Nonparametric permutation testing with FSL Randomise was used to determine which white matter (WM) MWF voxels were associated with cognitive test performance for each test (p<0.01, after multiple comparisons correction), creating test-specific maps of associated WM areas. Pearson ́s correlations assessed relationships between mean MWF in the cognitive test-specific WM areas and respective test scores. MS patients were categorized into cognitively impaired, mildly impaired and cognitively preserved groups based on published norms. Kruskal Wallis ANOVA with post hoc pairwise comparisons investigated mean MWF differences between cognitive groups.

Results

Results: MWF in several WM areas was significantly associated with SDMT, SRT and BVMT-R scores but not the COWAT. All tests found voxels within the corona radiata, posterior thalamic radiation and parts of the corpus callosum significant. Unique WM areas were the inferior longitudinal fasciculus and anterior cingulum for SDMT and the retrolenticular part of the internal capsule for the BVMT-R. Mean MWF in the test-specific WM areas correlated significantly with performance on the SDMT (r=0.58, p= 4.11 x 10-8), SRT (r=0.56, p= 4.14 x 10-7) and BVMT-R (r=0.56, p= 1.0 x 10-6). Mean MWF in the test-specific WM areas was significantly lower in the cognitively impaired group relative to the cognitively preserved group (p<0.01).

Conclusions

Conclusions: There is an anatomical-functional relationship between myelin damage and cognitive performance in MS with unique WM patterns for different cognitive domains.

Collapse
Machine Learning/Network Science Late Breaking Abstracts

LB1282 - Machine learning of deep grey matter volumes on MRI for predicting new disease activity after a first clinical demyelinating event (ID 2182)

Speakers
Presentation Number
LB1282
Presentation Topic
Machine Learning/Network Science

Abstract

Background

Deep grey matter (DGM) atrophy is a feature in all multiple sclerosis (MS) phenotypes. Studies have shown a strong relationship between DGM atrophy and clinical worsening but the utility of DGM volumes for predicting disease activity is largely unexplored, especially in early disease. Machine learning (ML) is a computational approach that can identify patterns that predict disease outcomes. In ML, the study dataset is divided into training and test subsets. The training set contains known outcomes, which the ML algorithm uses to form a prediction model, which is then evaluated on the test set.

Objectives

To develop an ML model for predicting new disease activity (clinical or MRI) within 2 years of a first clinical demyelinating event, using baseline DGM volumes. The motivation is to identify individuals at higher risk of new disease activity.

Methods

3D T1-weighted MRIs acquired within 90 days of a first clinical event in 140 subjects from a completed placebo-controlled trial of minocycline were used. Eighty subjects had new disease activity within 2 years, 28 were stable, and 32 withdrew early (unknown outcome). The stable and unknown groups were combined into 1 for ML training. Advanced Normalization Tools and FMRIB Software Library were used to segment the thalami, putamina, globi pallidi, and caudate nuclei. A random forest ML model was trained to predict new disease activity with feature vectors composed of individual DGM nuclei volumes and several other variables (e.g., minocycline vs. placebo, mono-focal vs. multi-focal CIS, normalized brain volume, and sex). Model performance was evaluated using 3-fold cross-validation, with 80% of the data used for training, and the rest for testing.

Results

Sequential elimination of variables ranked the least important by the trained model resulted in improved classification accuracy. Therefore, the less predictive variables were pruned from the feature vector. The best model used DGM volumes alone and achieved 82.1% accuracy, 87% precision, 81% recall and F1-score of 0.84 with area under the curve (AUC) of 0.76.

Conclusions

ML can learn patterns predictive of new disease activity within 2 years after a first clinical demyelinating event from baseline DGM volumes. This approach can potentially augment the many other clinical and demographic variables used in a typical MS clinical work up. Further investigation with larger data sets is warranted to determine generalizability of the approach.

Collapse