PCOS DETECTION USING DEEP LEARNING

Prediction High levels of androgens in women cause polycystic ovary syndrome (PCOS), a collection of symptoms. PCOS is caused by a combination of genes and environment variables that are common conditions commonly associated with atherosclerosis, hypertrichosis, acne and hyperandrogenism, and persistent infertility. According to a recent study, about 18% of Indian women suffer from this disease. Doctors manually examined an ultrasound scan to determine which ovary was damaged, but could not determine if it was a benign cyst, PCOS, or a malignant cyst. This study proposes a DCNN-based algorithm, where the PCOS classification is coded in Python programming and filled with blood or fluid using ultrasound images. To classify the PCOS in the dataset, this study uses DCNN-based image processing feature extraction. That is, the study is done using a trained dataset of the same PCOS-related disorders. Finally, use the test dataset to perform a feature extraction and evaluate the accuracy against the performance parameters. PCOS (Polycystic Ovary Syndrome) is an endocrine disorder that affects many women in the childbearing age group and is associated with infertility, diabetes, and cardiovascular disease. Most imaging functions are used to diagnose illness. Ultrasound diagnostic imaging has become an important tool in diagnosing PCOS. The typical appearance of an image is follicle overlap, equipment inherent noise, and primarily empirical operation, which is increasingly difficult due to the lack of understanding of the operator due to the time-consuming diagnostic process. The above situation affects the accuracy of cyst detection. Early and accurate detection of female reproductive system abnormalities is essential to avoid infertility prior to the treatment process. To achieve maximum accuracy in cyst identification in a short period of time, this task reviews the various approaches proposed so far for speckle noise removal, segmentation extraction of areas of interest,