I have deleted some repeated photos and utilized the following dataset for this project: 2000 Images of Ultrasound for PCOS.
This project focuses on detecting Polycystic Ovary Syndrome (PCOS) using ultrasound images of ovaries. The goal is to classify whether an ovary is infected with PCOS or not based on the image.
Size: The dataset contains a total of 2,000 ovarian ultrasound images, with 787 infected and 1,145 normal images.
Data folder consists of:
- 'train' and 'test' subfolders containing 2 categories of data: 'infected' and 'notinfected'
- infected: Images of ovaries having PCOS
- notinfected: Images of healthy ovaries
Polycystic Ovary Syndrome (PCOS) disrupts women's health, impacting fertility and metabolism. Characterized by hormonal imbalances and ovarian issues, it affects 5-10% of women. PCOS increases the risk of diabetes, heart disease, and mental health disorders. Early diagnosis and treatment mitigate long-term complications. Originating in 1935, research on PCOS has evolved, with prevalence notably high among Indian women. Artificial Intelligence (AI) revolutionizes PCOS detection and treatment. Machine Learning (ML) algorithms analyze medical data, aiding accurate diagnosis and personalized treatment plans. Using ovary ultrasound scans, a novel ML technique improves PCOS prediction accuracy, offering efficient and precise care.