Detection of Ovarian Malignancy from Combination of CA125 in Blood and TVUS Using Machine Learning
Carcinoma of the ovary keeps on being one of the main causes of death from gynecologic malignancies on the earth. Women determined to have beginning phase (1/2) ovarian disease have considerably better endurance rates contrasted with those analyzed in later stages. The speculation behind screening for ovarian malignant growth is that previous recognition of the illness will bring about more women being analyzed at less propelled stages and that this will convert into lower generally death rates for the ovarian disease. Both transvaginal ultrasound and the Cancer Antigen 125 in blood have been used to screen for early ovarian malignant growth. The aims of our study are to the detection of ovarian malignancy in primary stage and thereby reduce the mortality due to ovarian cancer. In this study, we have used a real-world dataset, namely PLCO to anticipate the discoveries from CA-125 and TVUS joined screening utilizing machine learning models. We obtained a critical presentation from the model as far as accuracy, precision, and recall with the estimations of 80, 86, and 90.25% individually. Our work will support the doctors and the suspected to analyze ovarian threats in the beginning and decreased the complication and mortality due to ovarian malignancy.