Polycystic Ovary Syndrome (PCOS) diagnostic methods in machine learning: a systematic literature review

Polycystic Ovarian Syndrome, also known as PCOS, is a major hormonal imbalance affecting women primarily in their reproductive age. Women with PCOS may have either infrequent or extended menstrual cycles or sometimes excess male hormone i.e. androgen levels. The ovaries may grow with number of slight collections of fluid, called follicles that fail to release eggs every month regularly. It is also seen that PCOS not only affects a woman’s fertility but also indirectly causes various other health issues like type 2 Diabetes, obesity, blood pressure and other metabolic disorders. Recent researches have focussed on the use of different algorithms in Machine Learning to diagnose PCOS using structured or unstructured data. Therefore, in this study, a considerable Literature Review has been carried out to provide a detailed analysis of various algorithms that have been used to detect PCOS with their comparative study. Also some of the work particularly focuses on health issues, the food and dietary patterns and the ways to manage PCOS. Further, these algorithms are critically analysed to understand the framework and limitations that should be considered while putting forward the solutions relevant to the diagnosis of PCOS in an effective way.