Multiple Sclerosis Classification and Disability Prediction with Clinical and MRI Data

Paola Valsasina's research dataset, published on figshare in April 2026, contains demographic, clinical, and MRI-derived features for 1,554 multiple sclerosis patients and 520 healthy controls from the Italian Neuroimaging Network Initiative. The data was used to train machine learning models for patient classification and disability prediction. The dataset is a 352.9 KB PDF file describing the study and its results.

Use Cases

• Classifying multiple sclerosis patients versus healthy controls based on T2 lesion volume and brainstem/cerebellar grey matter volumes.
• Differentiating relapsing from progressive MS phenotypes based on EDSS score, age, and thalamic/cortical grey matter volumes.
• Predicting disability severity (EDSS score) using T2 lesion volume, sex, and cortical/cerebellar/thalamic grey matter volumes.

Strengths

• Includes data from 2,074 total subjects (1,554 patients, 520 controls).
• Integrates multiple data types: demographic, clinical assessment (EDSS), and derived MRI features.
• Models described achieved high accuracy (89–96%) for classification tasks.

Limitations

• The dataset is a PDF summary (352.9 KB); the underlying raw data is not directly accessible.
• Column-level documentation and sample data are unavailable, requiring manual inspection after download.
• Row count for the underlying data tables is unknown, which may limit suitability assessment.

Provenance

Source

Italian Neuroimaging Network Initiative repository.

Collection Method

Data likely gathered from neurological assessments and brain T2-/3D T1-weighted MRI scans of included subjects.

Time Range

null

Freshness

Last updated 2026-04-10 05:29:23.

Geography

Italy (inferred from the source repository).