Artificial intelligence in multiple sclerosis detection: current landscape and future trends

Hishanudin, Aiman Farhan and Abdul Razak, Abdul Hadi and Mohd Yassin, Ahmad Ihsan and Ahmad, Adizul and Saaid, Mohammad Farid and Hashim, Hilwati (2026) Artificial intelligence in multiple sclerosis detection: current landscape and future trends. Journal of Electrical and Electronic Systems Research (JEESR), 29 (1): 3. pp. 26-37. ISSN e-ISSN : 3030-640X
Identification Number (DOI): 10.24191/jeesr.v29i1.003
Abstract

This study conducts a bibliometric analysis of Artificial Intelligence (AI) in Multiple Sclerosis (MS) detection, examining 1502 peer-reviewed publications from 1994 to 2024 using ScientoPy, VOSviewer and Biblioshiny to understand its rapid evolution, current landscape, and emerging trends. A significant surge in research activity has been observed since 2017, driven by breakthroughs in machine learning and neuroimaging. The United States, Germany, China, and Italy are the most active countries by author appearances, while the most productive individual institutions are located in Europe, particularly Italy, Spain, and the United Kingdom. Key research clusters focus on gait analysis, lesion segmentation, biomarker discovery, and advanced neuroimaging. The analysis highlights the growing application of convolutional neural networks (CNNs), support vector machines (SVMs), and generative models to enhance diagnostic precision, monitor disease progression, and predict symptom trajectories in MS. AI models have demonstrated diagnostic accuracies exceeding 95% in some cases, providing personalized prognostic insights and real-time clinical decision support, thereby serving as a valuable complement to conventional neuroimaging and clinical assessments. The study also indicates a paradigm shift towards multimodal data integration, regulatory-compliant model design, and clinically explainable AI systems, with an increasing emphasis on transparency, reproducibility, and standardized validation frameworks such as AutoML and federated learning. Furthermore, the integration of biomarkers like cerebrospinal fluid analysis and brain atrophy metrics into predictive models reflects a holistic approach to MS care

Item Details
Edit Item
Edit Item
Downloads & Files
[thumbnail of 149107.pdf]
Text
149107.pdf
Download (1MB)
Indexing & Metrics