A novel approach to venous clinical severity score prediction: combining metaheuristic algorithm and random forest classification

作者：Hao Zhu, Nianyun Zhang, Yuan-Zhen Ni, Qiang Sun · 发表于：Computer Methods in Biomechanics & Biomedical Engineering · 年份：2025 · DOI：10.1080/10255842.2025.2514133 · 被引用次数：1 · 研究领域：Artificial Intelligence in Healthcare、Quality and Safety in Healthcare、Imbalanced Data Classification Techniques

Varicose veins stem from valve failure, with conventional treatments offering limited relief. Yoga, along with lifestyle and dietary changes, may help prevent and improve the condition. This study used Random Forest Classification to predict VCSS, a standard measure of chronic venous insufficiency severity. BWO and IAOA optimizers enhanced model performance, evaluated across four VCSS categories: absent, mild, moderate, and severe. The RFBW hybrid model, combining RFC and BW, showed the highest accuracy, supported by high precision scores of 0.917, 0.952, 0.976, and 1.000, highlighting its efficiency and reliability. Notably, the RFIA model showed results similar to the RFBW model.