A Multi-Modal AI Framework for Lead Discovery and Lesion Detection in Multiple Sclerosis
Abstract
This repository contains the paper and Python source code for a multi-modal AI framework for Multiple Sclerosis (MS) research. The framework includes: 1. DeepDTI: Drug-target interaction prediction using deep learning 2. Graph-MS: Molecular graph analysis with betweenness centrality 3. NeuroVision: Automated MRI lesion segmentation 4. ChemScreen: Molecular screening with Lipinski's rules Key results: - Top binding energy: -9.71 kcal/mol for S1P1 receptor - MRI Dice score: 0.85 on clinical data - Pearson correlation: 0.84 for binding prediction
Fields of Study & Subjects
Computational Drug Discovery Methods
Multiple Sclerosis Research Studies
Machine Learning in Materials Science
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 27 February 2026
- Language
- EN
- Document type
- Preprint