Early Detection of Fetal Congenital Heart Disease Using Ultrasound AI CSR Proposal

Early detection of Fetal Congenital Heart Disease

Objective

• Develop a machine-learning-based software tool to diagnose fetal heart abnormalities through ultrasound imaging, aiming to improve early detection of congenital heart disease (CHD) in fetuses and reduce neonatal and infant mortality rates in India.

Description

• Problem: High neonatal and infant mortality rates in India, with about 10% of these deaths attributed to CHD., Solution: Develop a software tool that combines ultrasound images with machine learning algorithms to detect CHD as early as possible and refer high-risk cases for further treatment., Methodology: Collect fetal ultrasound data from Mediscan hospital, Chennai, label and verify it with expert clinicians, and using deep learning techniques for image analysis. Incorporate Doppler data to detect abnormal flow patterns and develop a real-time diagnostic tool., Beneficiaries: Infants in India, especially in rural areas, with limited access to cardiac care facilities, will benefit from early detection and timely intervention for CHD.

Impact

• This project aims to significantly reduce infant mortality rates in India by improving the early detection of CHD. The machine-learned software tool will be accessible in rural areas, allowing non-expert ultrasound specialists to provide early diagnoses and refer cases for specialized care. It also contributes to medical research by studying factors contributing to CHD development.