MamaCare AI: A GeoAI Framework for Maternal Healthcare Accessibility Analysis in Underserved Communities

Ayinde Abdul-Sobur

Keywords: GeoAI, Maternal Healthcare, Accessibility Analysis, Spatial Clustering, Healthcare Equity, Underserved Communities, Machine Learning for Health

TL;DR: A GeoAI framework using DBSCAN clustering and composite risk scoring to identify maternal healthcare accessibility gaps and guide intervention planning in Northern Nigeria.

Abstract:

Maternal mortality remains a critical public health challenge
in Northern Nigeria, where geographic barriers significantly
delay access to essential healthcare services. In Kebbi State,
approximately 62% of women of reproductive age reside more
than 5 km from the nearest health facility, representing a
severe and measurable accessibility gap. This paper presents
MamaCare AI, a GeoAI framework integrating geospatial
intelligence and machine learning to support maternal
healthcare accessibility analysis and evidence-based
intervention planning in underserved communities. Using
GRID3 facility data (1,207 facilities), WorldPop population
estimates, and Nigeria DHS-derived indicators, we conduct
a spatial accessibility analysis across 1,510 grid points
in Kebbi State. DBSCAN spatial clustering (silhouette
score: 0.83) identifies priority underserved population
zones, and a composite risk scoring model across all 21
Local Government Areas identifies Argungu, Augie, and Dandi
as critical intervention priorities. Based on these findings,
we propose a phased AI-assisted mobile clinic deployment
strategy targeting 310,000 women across three priority LGAs.
This work demonstrates how GeoAI methods can translate
spatial health data into actionable intervention frameworks
for low-resource maternal healthcare environments.

Track: Track 1: ML Research Addressing Challenges Faced by Muslim Communities

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Non Archival Confirmation: I understand that submissions to MusIML are non-archival and can be submitted to other venues.

Submission Number: 60