AIHealthcare Analytics

Core Applications

Generative AI, particularly LLMs, has emerged as a promising tool for processing unstructured clinical data in infection prevention and control. LLMs have been evaluated for healthcare-associated infection (HAI) surveillance including central line-associated bloodstream infections (CLABSI), surgical site infections (SSI), and catheter-associated urinary tract infections (CAUTI), with pooled sensitivities exceeding 90% across studies. Additional applications include diagnostic stewardship, risk assessment for multidrug-resistant organism (MDRO) exposure, public health surveillance for avian influenza, and central line necessity auditing.

Performance Characteristics

Models perform best when used to augment rather than replace expert review. Common limitations include reduced specificity, sensitivity to prompt framing, and dependence on completeness of clinical data provided to the model. GenAI applications show greatest promise when aligned with their core capability: natural language processing of unstructured clinical text.

Public Health Surveillance

An equity-aware multimodal GenAI copilot for digital public health surveillance has been developed to coordinate multiple data streams — routine case reporting, contextual regional indicators, environmental measurements, and digital signals — addressing the fragmentation of existing disconnected tools for forecasting, outbreak flagging, fairness auditing, and interpretation.

Implementation Requirements

Current evidence supports GenAI use as screening and decision-support tools with human oversight. Further validation across diverse settings and integration within electronic health records are needed before widespread adoption. The combination of high sensitivity and moderate specificity makes these tools best suited for initial screening with human expert confirmation.