e-ISSN: 2582-5208 International Research Journal of Modernization in Engineering Technology and Science ( Peer-Reviewed, Open Access, Fully Refereed International Journal ) Volume:04/Issue:01/January-2022 Impact Factor- 6.752 www.irjmets.com www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science [195] PCOS (POLYCYSTIC OVARIAN SYNDROME) DETECTION USING DEEP LEARNING Shubham Bhosale *1 , Lalit Joshi *2 , Arun Shivsharan *3 *1,2,3 Research Scholar, Dr D Y Patil School Of Engineering Academy Ambi, India. ABSTRACT Predicting High levels of androgens in women produce polycystic ovarian syndrome (PCOS), a collection of symptoms. PCOS is caused by a combination of genetic and environmental variables that are frequent illnesses that are commonly associated with atherosclerosis, hirsutism, acne, and hyperandrogenism, as well as persistent infertility. According to recent studies, approximately 18% of Indian women suffer from this illness. Doctors manually examined ultrasound scans to determine which ovary was damaged, but they were unable to determine if it was a benign cyst, PCOS, or malignant cyst. In this research, DCNN-based algorithms are proposed, and coding for PCOS classification is produced in Python programming, and they are filled with blood or fluid using ultrasound pictures. To classify PCOS in the dataset, the study uses DCNN-based image processing feature extraction. That is, the research is conducted utilising a trained dataset of the same PCOS-related disorders. Finally, the test dataset is used to perform feature extraction and assess accuracy using performance parameters. PCOS (Polycystic Ovary Syndrome) is an endocrine illness that affects many women in their reproductive age groups and is linked to infertility, diabetes, and cardiovascular disease. The majority of imaging characteristics are used to diagnose the illness. Ultrasound imaging has emerged as a critical tool in the diagnosis of PCOS. The typical appearance of the image gets increasingly hard due to overlapping of the follicles, intrinsic noise of the equipment, and lack of operator understanding as it is mostly an experience based operation, making the diagnosing process time consuming. The accuracy of cyst detection is harmed as a result of the aforementioned circumstances. To avoid infertility, early and accurate detection of abnormalities in the female reproductive system is essential prior to the treatment process. In order to obtain maximum accuracy in cyst identification in a short period of time, this work reviews various approaches proposed so far in terms of removing speckle noise, extracting region of interest using segmentation, and classification of images. Keywords: Pcos Disease, Ultrasound Images, Deep Learning Cnn (Convolution Neural Network). Classification, Enhancement, Poly Cystic Ovary Syndrome, Segmentation, Speckle Noise, Ultrasound Images. I. INTRODUCTION Deep Learning is a rapidly evolving technology that can help solve problems in a variety of fields. Deep learning is supporting healthcare practitioners and researchers in identifying hidden opportunities in data, allowing the medical sector to function more effectively. It also aids doctors in precisely analysing any type of ailment and better medicating patients, resulting in improved medical decisions. A medical condition like Polycystic Ovarian Syndrome (PCOS) lacks a reliable diagnosis and therapeutic alternatives. It's a common endocrine condition that causes ovarian cysts to form in child-bearing women, which can lead to infertility.