<![CDATA[This study explores the application of the Gaussian Naive Bayes (GNB) classifier to predict the conversion from Clinically Isolated Syndrome (CIS) to Multiple Sclerosis (MS) among Mexican mestizo patients. Utilizing a dataset gathered from the National Institute of Neurology and Neurosurgery in Mexico City, which included patients diagnosed with CIS between 2006 and 2010, we employed a prospective cohort study design. Our approach involved preprocessing the data to handle missing values and scale normalization followed by splitting it into training and testing subsets. The GNB model's performance was assessed through a 5-fold cross-validation, focusing on accuracy, precision, recall, and F1-score. Results demonstrated the model's capability to predict MS conversion with reasonable precision, highlighted by a significant peak in performance metrics in the third fold of the validation. The study addresses a gap in predictive diagnostics for MS within a specific demographic group, providing valuable insights for early intervention strategies. Despite some limitations such as the model's sensitivity to data heterogeneity and the demographic specificity of the cohort, the findings underscore the potential for predictive models in clinical settings. Recommendations for future research include the use of more sophisticated algorithms and broader demographic studies to enhance predictive accuracy and generalizability.]]>
Copyrights © 2024
