Multiple Sclerosis Severity Estimation and Progression Prediction Based on Machine Learning Techniques.

Area of Science:

• Neurology
• Medical Informatics
• Artificial Intelligence

Background:

• Multiple Sclerosis (MS) poses challenges in severity estimation and progression prediction.
• Accurate prediction aids in personalized treatment strategies and patient management.
• Existing methods may not fully leverage the potential of machine learning for MS data analysis.

Purpose of the Study:

• To apply machine learning (ML) techniques for Multiple Sclerosis (MS) severity estimation using the EDSS score.
• To predict the future progression of MS utilizing ML approaches.
• To evaluate the efficacy of different ML models in analyzing MS patient data.

Main Methods:

• Data collected from 30 MS patients at three time points from the University Hospital of Ioannina.
• Features included demographic, clinical, test results, treatment, and comorbidity data.
• Implementation and evaluation of several machine learning algorithms for prediction tasks.

Main Results:

• High accuracy achieved in MS severity estimation (94.87%).
• Significant accuracy in predicting MS disease progression (83.33%).
• Demonstrated the effectiveness of ML in analyzing complex neurological data.

Conclusions:

• Machine learning models show strong potential for accurate MS severity estimation.
• ML-based prediction of MS progression is feasible and effective.
• This approach can support clinical decision-making in managing Multiple Sclerosis.