Prototyping Intelligent Software Solutions for Global Health Engagement

Area of Science:

• Military Medicine
• Global Health Security
• Data Science in Healthcare

Background:

• Global Health Engagements (GHE) are crucial for military planning, aiming to build partner nation capacity and ensure care for U.S. service members.
• Effective GHE planning is hindered by challenges in data availability, standardization, integration, and visualization.
• Improved data curation, analysis, and user interface design are essential for optimizing GHEs and care delivery.

Purpose of the Study:

• To enhance the Center for Global Health Engagement's (CGHE) capabilities in planning, executing, and tracking GHEs.
• To leverage Machine Learning (ML), including Large Language Models (LLMs), for processing large and diverse GHE datasets.
• To develop a data-driven approach for more effective military health engagements.

Main Methods:

• Developed a prototype ML algorithm utilizing LLMs for tactical-level hospital data ingestion, curation, and analysis.
• Automated data ingestion from unstructured GHE surveys into a unified, standardized format.
• Created interactive analytic tools and user-specific visualizations, including a 3D geospatial mapping tool for crisis scenarios.

Main Results:

• A functional ML prototype capable of ingesting, curating, and analyzing global health data.
• A 3D geospatial visualization tool displaying patient distribution in crisis scenarios.
• An adaptive user interface providing human-readable explanations of ML algorithm outputs.

Conclusions:

• Demonstrated the practical application of AI in global health data processing and analysis.
• Highlighted implications for biosurveillance, medical countermeasures, and medical logistics.
• Emphasized the importance of integrating partner healthcare capabilities into military plans for improved decision-making in operational planning, crisis management, and conflict healthcare delivery.