Next-Gen Midwifery Support: Designing an Artificial Intelligence (AI) Enhanced Mobile App for Pregnancy Risk...

Area of Science:

• Digital Health
• Maternal Healthcare
• Artificial Intelligence in Medicine

Background:

• Limited medical professionals in rural areas hinder effective patient treatment.
• Rapid prenatal risk assessment is crucial for improving pregnancy care in resource-constrained settings.

Purpose of the Study:

• To develop and evaluate an innovative digital system for midwives to recognize prenatal risks.
• To assist in clinical decision-making, particularly in rural maternity hospitals.

Main Methods:

• A smartphone application utilizing AI and the modified Alberta perinatal risk score was developed.
• A pilot study involved training midwives and screening 1010 pregnant women in rural primary health centers.
• Data was gathered, integrated with Firebase, and processed by an AI model for risk classification.

Main Results:

• The app effectively classified prenatal cases: 37.33% low risk, 37.82% intermediate risk, 24.85% high risk.
• High-risk cases were referred; midwives managed low and moderate-risk cases, collaborating with medical officers.
• The system achieved 99.0% accuracy in forecasting newborn fatalities using the Gradient Boost algorithm.

Conclusions:

• An integrated Android application with an AI antenatal risk assessment system enhances midwives' skills.
• This leads to timely interventions for pregnant women, contributing to improved birthing outcomes.
• The digital system shows potential for improving patient care in underserved areas.