A Machine Learning-Aided Platform for Point-of-Care Pregnancy Risk Assessment from 2D Ultrasound

Overview

These AI-powered tools are designed to identify early warning signs of high-risk pregnancies, including the detection and classification of uterine fibroids. By supporting timely risk identification and continuous pregnancy health monitoring, the tools aim to reduce maternal mortality and improve outcomes for expectant mothers, particularly in settings with limited access to specialist care.

Responsible AI practices

The classification model incorporates an explainable AI tool that provides visual segmentations, helping users understand which regions the model considered relevant in its decisions. The project emphasizes responsible AI practices through inclusive data collection, ensuring participation from mothers across diverse regions of Uganda to improve fairness and accuracy.