IoMT-Based Point-of-Care Testing for PCOS Diagnosis Using Dempster-Shafer-Theory of Evidence

Polycystic ovary syndrome (PCOS) is a significant health problem common in many women globally in their fertile ages. Also, PCOS is associated with greater risks of developing life-threatening and long-term health issues, as the body is unresponsive to the produced insulin properly. Hence, early identification and appropriate medical help at the outset will lower the risk of developing long-term complications related to PCOS. Diagnosing PCOS is not always simple and straightforward. The complexity associated with the diagnostic process has increased due to multifarious symptoms and probable deviation in the conditions. Thus, depending solely on pelvic ultrasound is not adequate to make a complete diagnosis of PCOS. Other detection criteria are also required to secure a comprehensive diagnosis of PCOS. Our research in this paper addresses smart diagnosis of PCOS from various data with the integration of Internet of Medical Things (IoMT) based point-of-care testing (POCT) technologies. With the help of this architecture, remote patient monitoring and “on-spot” support of patients is possible in real time. In this work, we have adopted the Dempster-Shafer-Theory of Evidence (DST-E) for fusing data accumulated from several sources to derive a conclusion about PCOS diagnosis. The result reveals a substantially high percentage of accuracy (up to 96.67%).