Detection of Pcos/Pcod and Prediction of Infertility at the Early Stage Based on RF-ELM Technique

When infertility does not happen during ovulation, over 70% of cases are due to polycystic ovarian syndrome. Hyperandrogenism and polycystic ovary morphology are the hallmarks of ovarian dysfunction syndrome. Olig- or anovulation, hyperandrogenemia (clinical and/or biochemical), and ultrasound-verified polycystic ovaries are the diagnostic criteria for polycystic ovarian syndrome (PCOS). The following steps must be carried out in the precise sequence: preprocessing, data segmentation, feature extraction, and model training. Preprocessing is an essential step before adding data to the feature extraction stage. Segmentation analysis was used to eliminate those that did not correspond to actual follicles. Feature extraction retrieves the original statistical forms of factors like homogeneity, contrast, and standard deviation. Features are retrieved first in the training of RF-ELM models. The proposed method outperforms two state-of-the-art alternatives, RF and ELM. Following the implementation of the technique, the results demonstrated an accuracy boost of 93.50%.