AI-assisted sensor system for real-time monitoring of surgical hand disinfectant healthcare efficacy

Surgical hand disinfection is one of the most effective methods of preventing Surgical Site Infections (SSIs). However, the traditional methods of monitoring SSIs are microbiological culture testing and monitoring staff compliance, both of which can be subjective and delayed. For this purpose, this research introduces an AI-assisted sensor system that retrieves information about the surgical hand disinfectant's effectiveness in real time through machine learning, biosensors, and IoT-based analytics. The methodology uses a novel approach by employing phytochemical biosensors that can assess residual disinfectant coverage and subsequent bacterial contamination through fluorescence spectroscopy and impedance microbiology. The sensor data using clinical bacterial culture datasets is used for post-disinfection bacterium persistence assessment during deep learning. When the AI systems detect insufficient disinfection measures, the system alerts the personnel. This feedback is provided immediately to foster constant compliance with WHO hand hygiene recommendations. The Hybrid AI-sensor framework proposed in this research provided additional detection accuracy using Temporal Convolutional Networks (TCNs) and Hypergraph Neural Networks (HGNNs) to detect sensor data streams.