Algorithm for Automated Segmentation and Feature Extraction of Thermal Images

Abstract

Medical infrared thermal imaging techniques can provide the highquality images for monitoring and pre-clinical diagnostic of the diseases by showing the thermal abnormalities available in the body. Its biggest advantage is non-contact, non-invasive and very fast way of use. However, the incorrect interpretation of the thermal images via simple observation or manual analysis and detection approximate regions of interest lead to the numerous false positive results. The main objective of this research work is to develop an algorithm for automated image processing and analysis of the extracted features on thermal images for screening or pre-diagnosis of the diseases. This work presents the results of our previously developed Copyright algorithm applied to thermal images obtained from the patients with Axial Spondyloarthritis. This processing and analysis of the extracted features from thermal images can offer a novel quick and non-invasive tool for diagnosis and monitoring of rheumatoid diseases.