Fetal Health Risk Detection using Machine Learning

Abraham Obianke

Abraham Obianke

https://github.com/abrahamoaks/Fetal-Health-Risk-Detection-Web-App

The neonatal period is the most fragile and critical time for a child’s survival. The United Nations Children’s Fund (UNICEF) reports that in 2022 alone, 2.3 million children worldwide lost their lives within the first month of birth equating to 6,300 neonatal deaths every single day. These heartbreaking statistics underscore the urgent need for innovative solutions to improve early detection and intervention for neonatal health risks.

Drawing on a rich dataset from the University of California, UC Irvine, data archive featuring over 2,126 fetal Cardiotocograms (CTGs), my model leverages biometric data such as fetal heart rate (FHR) and uterine contractions (UC). These measurements, expertly classified and labeled by Obstetricians, form the foundation of an AI-powered system designed to identify potential fetal health risks. By equipping healthcare professionals and expectant mothers with timely and actionable insights, this solution aims to transform neonatal care, saving lives and fostering healthier beginnings. Ultimately, the project supports the attainment of the United Nations Sustainable Development Goal (SDG) 3 specifically on “Ending preventable deaths of newborns and children under 5 years old.”

Tools Used:
– Python
– Coding Tools: VS Code
– ML Libraries: Scikit-Learn, Pandas, Numpy, Matplotlib, Seaborn, Lime, Joblib, Comet ML
– Web App Frameworks: Flask, Bootstrap, HTML, CSS, & JavaScript

Q4 Status: Not yet deployed due to cost issues

Project Demo on LinkedIn: https://www.linkedin.com/posts/abraham-obianke-269112197_ai-ml-fetalhealth-activity-7239981184624033792-jnGo/?utm_source=share&utm_medium=member_android