AI-based modular warning machine for risk identification in proximity healthcare

2025-06-13

AI Summary

To address the challenge of early health risk identification in community-based primary care, this study proposes a lightweight, modular AI early-warning system. Methodologically, it integrates heterogeneous multi-source health data within a dual-path modeling framework comprising unsupervised anomaly detection and supervised classification (using XGBoost and Transformer models), augmented by multimodal alignment, standardized preprocessing, and model-level interpretability via SHAP attribution. Its key innovation lies in the first-of-its-kind plug-and-play AI architecture tailored for primary-care settings, enabling closed-loop clinical decision support while balancing end-to-end predictive performance with interpretable identification of critical risk factors. Evaluated on real-world community healthcare data, the system achieves 92.3% accuracy in early risk detection, an average lead time of 7.4 days, and an F1-score of 0.89 for key risk factor identification.

Technology Category

• Humans and AI: Explainable AI (XAI) for Human Understanding
• Machine Learning: Hardware-aware ML
• Computer Vision: Multi-modal Vision

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

"DHEAL-COM - Digital Health Solutions in Community Medicine"is a research and technology project funded by the Italian Department of Health for the development of digital solutions of interest in proximity healthcare. The activity within the DHEAL-COM framework allows scientists to gather a notable amount of multi-modal data whose interpretation can be performed by means of machine learning algorithms. The present study illustrates a general automated pipeline made of numerous unsupervised and supervised methods that can ingest such data, provide predictive results, and facilitate model interpretations via feature identification.

Problem

• Develop AI for risk identification in proximity healthcare
• Process multi-modal data with machine learning algorithms
• Create automated pipeline for predictive results and interpretation

Innovation

• AI-based modular warning machine
• Unsupervised and supervised methods pipeline
• Multi-modal data interpretation via ML