OPTIMIZATION OF PRE-HOSPITAL FIRST AID MANAGEMENT STRATEGIES FOR PATIENTS WITH INFECTIOUS DISEASES IN HUIZHOU CITY USING DEEP LEARNING ALGORITHM.

The aim of the study was to optimize the pre-hospital first aid management strategy for patients with infectious diseases in Huizhou city, which is expected to provide a basis for the epidemic prevention and control, to save lives, and increase the pre-hospital first aid efficiency. At the Department of Emergency, Huizhou Third People's Hospital as the research subject, the common pre-hospital first aid procedure for infectious diseases was identified. The Petri net was used to model and determine the execution time of each link of the pre-hospital first aid process. The isomorphic Markov chain was used to optimize the pre-hospital first aid procedure for infectious diseases. In terms of the emergency path, deep learning was combined with the reinforcement learning model to construct the reinforcement learning model for ambulance path planning. Isomorphic Markov chain analysis revealed that the patient status when returning to the hospital, the time needed for the ambulance to come to designated location, and the on-site treatment were the main problems in the first aid process, and the time needed for the pre-hospital first aid process was reduced by 25.17% after optimization. In conclusion, Petri net and isomorphic Markov chain can optimize the pre-hospital first aid management strategies for patients with infectious diseases, and the use of deep learning algorithm can effectively plan the emergency path, achieving intelligent and informationalized pre-hospital transfer, which provides a basis for reducing the suffering, mortality, and disability rate of patients with infectious diseases.