Ovarian Cancer Detection and Classification Using Deep Learning Techniques

Abstract

One of the most prevalent and deadly diseases that affect women globally is ovarian cancer (OC). Detecting OC in its early phases in daily life is still a difficult endeavor. One common factor contributing to the highest mortality rates among women is ovarian cancer (OC). The malignant development of ovarian cells is ovarian cancer cells that reproduce quickly and have the ability to invade and damage healthy human tissue. With the best cytoreductive surgery, patients have the best chance of controlling their disease or being cured. The classification of the various stages of OC has shown to perform better when using deep learning (DL) approaches. However, sub-optimal resection offers no advantage over chemotherapeutic and raises the possibility of complications following surgery. There are currently no clear standards for interpretation, despite the fact that there is a substantial body of literature comparing performance to that of surgery and laparoscopy. However, because of inadequate feature representation, the majority of the aforementioned approaches offer poor performance. Due to the rising expense of computing, certain models still do not have optimization processes. Although there are several established DL (Deep Learning) approaches to classification used for OC detection, they have several drawbacks, including the inability to pinpoint the precise location of the tumors and greater complexity newline