PCOS diagnosis with confluence CNN: A revolution in women's health

Abstract

Polycystic Ovary Syndrome (PCOS) emerges as a prevalent endocrine aberration afflicting women in their reproductive prime. This disorder shows up as a complicated messing up of the production of androgens in the ovaries, which are normally present in small amounts in a woman's body. The hallmark of PCOS lies in its pronounced hormonal imbalance, a feature conspicuously absent in conventional ovarian cysts. Recent studies posit that approximately 15 percent of women of reproductive age grapple with this condition, a substantial contributor to female infertility. Despite its global pervasiveness, the diagnostic conundrum surrounding PCOS persists, posing a formidable challenge. The worldwide discourse on this matter remains inconclusive, and the elusive nature of an accurate diagnosis compounds the predicament. Notably, the complexity of PCOS is exacerbated by its symptomatic overlap with other medical conditions, further confounding the diagnostic process. Our research endeavors are driven by an ardent interest in unraveling the intricacies of this enigmatic syndrome, employing sophisticated models such as Long Short-Term Memory (LSTM), Bidirectional LSTM (BI LSTM), Convolutional Neural Network (CNN) with LSTM, Convolutional Neural Network with Bidirectional LSTM (CNN+BI LSTM), and CNN to illuminate novel insights into this pervasive health challenge, where CNN came up with an accuracy of 97.74%.