Study of chronic health influencing factors and their preventive measures for divers based on big data analysis in diving medicine

Abstract

Diving decompression sickness is an occupational chronic disease that seriously threatens the life and safety of divers, which is hidden, lasts for a long time, and causes more serious harm. The clinical manifestations are complex, including joint pain, muscle pain, rash, fatigue, headache, coma, and so on. This paper explores the value formation mechanism of chronic disease health management based on big data based on information ecology chains and dissipative structures. A total of 1036 divers engaged in diving and fishing operations are being taken as experimental subjects in the chronic disease health management big data platform using a stratified whole-group sampling method. XGBoost algorithm was used to establish the risk assessment model of chronic decompression sickness for divers. And it was applied to the clinic for example, to analyze the effectiveness of the model for chronic health prevention in divers. The results show that the model has a predictive performance AUC value of 0.8102 in 5-fold cross-validation, which can be used for chronic health risk assessment of large-scale diving populations by medical personnel to guide early diagnosis, treatment, and prevention of chronic health in divers.

Keywords

• chronic diseases
• big data
• health management
• xgboost algorithm
• risk assessment models
• 97p10