Using Machine Learning to Assess Cancer Care Needs in Virginia; Addressing Cancer Inequities in Virginia: A Community-Based Approach 3 views

The technical project analyzes county-level healthcare demand and supply for the state of Virginia. Using publicly available APCD data, the project aims to identify hotspots and predict future healthcare demand. Extrapolating insurance data, it utilizes ICD-10 Diagnosis, ICD-10 Procedural Codes, and CPT Procedural Codes to aggregate and predict county-level cancer counts. It uses LSTM as a baseline and more complex models (this is in progress) to predict cancer counts per county several months in advance. Furthermore, it utilizes SaTScan’s built-in models for spatial anomaly detection, identifying anomalous counties based on the year’s cancer counts. These analytics models help determine areas of elevated cancer burden and predict cancer counts. The sociotechnical research explores how social groups in Virginia have responded to cancer disparities. As federal and state governments have fallen short, further burdening healthcare systems in medically underserved areas, community organizations, nonprofits, and peer-led groups have emerged to help bridge the gap between isolated communities and care. Through education, outreach, trust-building, and reducing cost barriers, social groups have helped expand access to critical healthcare services, such as screening and treatments, thereby helping mitigate cancer disparities.

BS (Bachelor of Science)

community, Virginia, cancer, disparities, healthcare, infrastructure, access

School of Engineering and Applied Science Bachelor of Science in Computer Science Technical Advisor: Anil Vullikanti STS Advisor: Peter Norton

All rights reserved by the author (no additional license for public reuse)

2025-12-12