This study addresses the complex challenges of Chronic Venous Insufficiency \(\left(CVI\right)\) by proposing an innovative approach that integrates machine learning, specifically the Naive Bayes Classifier \(\left(NBC\right),\) with Jellyfish Search Optimizer \(\left(JSO\right)\) and Flying Foxes Optimization (FFO). By using a large dataset that includes demographic data, baseline severity indicators, and specifics of yoga practices, the research seeks to predict the effects of yoga on CVI. Machine learning algorithms predict results like changes in symptom intensity and gains in general well-being through feature engineering and model selection. Personalized CVI treatment plans might be revolutionized by this prediction algorithm, which can optimize therapy techniques and provide customized suggestions for certain yoga practices. The integration of machine learning aids in healthcare resource allocation, guiding efficient interventions and enhancing patient education for informed decision-making. While considering economic implications, the study suggests potential healthcare cost reduction and improved quality of life. For a responsible translation into clinical settings, safety, patient preferences, and ethical issues are prioritized. Personalized \(CVI\) management is significantly enhanced by the unique and promising Naive Bayes Classifier with optimization techniques. This research informs current practice and guides future studies on the specific benefits of Yoga for CVI. Based on the results, the NBJS model registered superior performance in predicting and categorizing VCSS-PRE with 91.5% for Precision, 90% for Accuracy, Recall, and F1-Score. On the other hand, in the case of \(VCSS-1\) , the NBFO model performed better by possessing 90% for Accuracy and Recall, 89% for \(Precision\) and \(F1-Score\) .
