A Novel PAD Temperature Index Using Infrared Thermography and Fuzzy Clustering

Peripheral arterial disease (PAD) is a widespread circulation disorder that affects 202 million persons globally, including more than 20 percent of adults over 70. PAD is a slow- developing disease that is asymptomatic in the early stages, but it progresses to critical limb ischemia (CLI) and non-healing wounds. Endovascular treatments such as peripheral angioplasty and conventional bypass surgery have been extensively adopted to treat PAD. Despite promising initial results, surgical intervention failure occurs after two years in about 50 percent of patients. Therefore, there is an unmet need for an early warning system that enables frequent surveillance and can detect early signs of failure that may improve patient outcomes. One of the PAD indications identified during the physical examination is a decrease in the patient's lower leg or foot skin temperature compared to the other leg or the rest of the body. This highlights the potential of using infrared thermography as a suitable screening approach for PAD diagnosing. Infrared thermography is a non-invasive, radiation- free, pain-free capable of detecting changes in skin temperature with high spatial accuracy. The proposed monitoring system is based on fuzzy C-means (FCM) clustering and calculates an individualized PAD index based on the foot temperature distribution. Results on a pilot dataset obtained from ten patients monitored over two months showed that the proposed fuzzy temperature index provides stable and repeatable results across different scenarios.