A Systematic Review based on the Detection of PCOS using Machine Learning Techniques

Women lacking appropriate nutrients in their meals are having the most rising non-communicable illness called Polycystic Ovary Syndrome (PCOS). The major reason behind this disease is unhealthy diet. At their reproductive age, 8–10% of women influenced with endocrine disorder is PCOS it leads to lot of health issues especially miscarriage and infertility with ovarian dysfunction. The male hormones abnormality production and female hormone levels variation characterizes an endocrine disorder. The PCOS symptoms are hirsutism, acne, male hormone excessive production, irregular menstrual cycle and obesity. Because of symptoms associated with heterogeneity, PCOS diagnose is extremely difficult. The PCOS patients met the major challenge is ovary scanning and innumerous clinical tests with cost as well as time. By addressing these challenges, the existing authors suggested different kinds of machine learning (ML) algorithms with respect to artificial intelligence models. This research study reviews the ML scheme to analyze PCOS diagnosis with its advantages, limitations and experimental analysis and datasets. By this investigation, the tables and figures are summarized for readers’ convenience. Still, there are several problem to tackle and possibility to pursue growing popularity of ML models in the future.