Towards Multimodal Machine Learning Prediction of Individual Cognitive Evolution in Multiple Sclerosis

Area of Science:

• Neurology
• Artificial Intelligence
• Biomedical Informatics

Background:

• Multiple sclerosis (MS) presents with varied symptoms, making individual prognosis difficult.
• Current prognostic methods using biomarkers lack personalization for disease course prediction.
• Cognitive deterioration in MS is common and impactful but often overlooked in predictive research.

Purpose of the Study:

• To advance machine learning applications for predicting cognitive decline in MS.
• To provide an overview of machine learning techniques, study design considerations, and existing literature.
• To explore emerging AI trends for future cognitive prognosis in MS.

Main Methods:

• Review of current literature on machine learning for MS prognosis, focusing on cognitive aspects.
• Discussion of machine learning principles, potential challenges, and essential study design elements.
• Exploration of novel machine learning trends applicable to MS research.

Main Results:

• The review synthesizes existing studies on machine learning for cognitive prognosis in MS.
• Identifies gaps in research, particularly the underrepresentation of cognitive deterioration prediction.
• Highlights the potential of multimodal data integration and personalized prediction models.

Conclusions:

• Machine learning holds significant promise for improving individual-level prediction of cognitive outcomes in MS.
• Further research integrating advanced AI techniques is needed to address cognitive heterogeneity in MS.
• This review serves as a foundation for future studies aiming to enhance cognitive prognosis in MS through AI.