Interpretation of polycystic ovarian syndrome (PCOS) employing computational neural network CNN

Polycystic ovarian syndrome is one of the major diseases hampering the fertility of women between the age of 15 to 50. The PCOS prone women are affected both mentally and physically, which claims difficulty to conceive. There are several ways to detect this disease, but untimely detection can reduce the effect of it. we are presenting an established model for the early detection of PCOS applying CNN. The previous models built utilizing machine learning produces an accuracy of 89%, Fuzzy TOPSIS provided an accuracy of 98% and SVM (Support Vector Machine) algorithm provided an accuracy of 94%. But the newly established model utilizes the ultrasound images of the ovary to detect the cysts that are formed and classifies into infected and non-infected. The conducted research extracts the feature from the ultrasound images which are taken out from the Kaggle repository and the outcome is measured in terms of accuracy since classification is the problem basis. This emphasized model provides the best accuracy among all the previously developed models. The research is done with the motto of enhancing the sexual health of women and to provide better methods for detection in the region of biomedical technology.