Intelligent generation method of infection risk map and management system in hospital waiting room for respiratory infectious diseases

Abstract

Respiratory infectious diseases, such as seasonal influenza, pose significant risks to public health safety through airborne transmission. This study aims to explore methods for controlling airborne infection risk in waiting rooms, a hotspot for cross-infection. A database of 990 samples was established by integrating the results of computational fluid dynamics and social force model simulations into the Wells-Riley equation. We introduced two indicators, Pr (Particle Removal Efficiency) and Hr (High-risk Area Ratio), to evaluate the cleanliness performance of waiting room spaces. Additionally, the Sci (Space Cleanliness Index) was proposed as a criterion for evaluating architectural designs and fine-tuning fresh air system operational parameters. We also unveiled the mapping relationship between spatial morphology parameters and ventilation design parameters in relation to the cleanliness performance of waiting rooms. An infection-risk prediction proxy model was established based on machine learning methods, suitable for rapid prediction of spatial infection risks in waiting rooms and optimization of fresh air system operational parameters. The findings can be applied to the selection of hospital design schemes and the optimization of ventilation operational control strategies, aiding in strengthening infectious disease control and reducing the public health safety risks of cross-infection from respiratory infections during peak seasons in hospitals.