Varicose veins are a common vascular disorder associated with chronic venous insufficiency and may lead to severe complications if not monitored effectively. Conventional diagnostic approaches are limited to periodic clinical assessments and fail to capture continuous physiological variations. This study presents a wearable, non-invasive monitoring system for continuous varicose vein risk assessment using multi-modal sensing and intelligent data analysis. The proposed system integrates photoplethysmography (PPG), skin temperature sensing, and inertial measurement-based posture detection to acquire real-time physiological and behavioral data. A hybrid risk prediction framework combining threshold-based clinical evaluation and a Random Forest classifier is employed to improve reliability. The system is implemented on an embedded platform with wireless communication for remote monitoring. Experimental results demonstrate a classification accuracy of 94.6% with low latency and extended operational capability, indicating the effectiveness of the proposed approach for continuous and real-world monitoring of varicose vein risk.
L et al. (Thu,) studied this question.