Real-Time Personal Protective Equipment (PPE) Compliance and Clinical Tool Monitoring Using Generative AI: A Novel...

Area of Science:

• Medical Informatics
• Artificial Intelligence
• Patient Safety Engineering

Background:

• Hospital-acquired infections (HAIs) are a significant patient safety issue, with non-compliance to personal protective equipment (PPE) protocols being a key preventable factor.
• Current monitoring systems for PPE compliance, including manual audits and CCTV, suffer from delays, inconsistency, and reactivity.
• Traditional AI systems lack flexibility, requiring extensive retraining when safety protocols are updated.

Purpose of the Study:

• To develop and assess a generative AI-powered system for real-time monitoring of hospital compliance.
• The system leverages Google Gemini on Raspberry Pi hardware to translate natural language rules into enforcement logic without retraining.
• The goal is to create an adaptable and efficient solution for monitoring evolving safety protocols.

Main Methods:

• The system integrated Google Gemini, OpenCV, and Streamlit to process natural language rules into executable logic.
• Performance evaluation involved 168 mannequin-based trials simulating diverse conditions (skin tones, orientations, object presence).
• Key performance metrics included accuracy, recall, specificity, F1 score, and Cohen's Kappa, compared against reference labels.

Main Results:

• The generative AI system achieved high performance metrics: 95.8% accuracy, 91.0% recall, 100% specificity, F1 score of 0.95, and Cohen's Kappa of 0.92.
• Consistent performance was observed across different mannequin skin tones and between rulebook-derived and free-text prompts.
• Notably, the system recorded zero false positives, indicating high reliability.

Conclusions:

• The developed generative AI system offers a highly accurate, adaptable, and cost-effective solution for monitoring hospital compliance.
• Its integration into hospital workflows can facilitate proactive, real-time oversight of dynamic safety protocols.
• This approach promises improved compliance rates, enhanced patient safety, and reduced operational costs compared to existing methods.