An Improved Infectious Disease Risk Prediction Model Based on Attention Mechanism
Yingshuai Wang, Yanli Wan, Qingkun Chen, Xingyun Lei, Yan Wang, Guoqiang Sun, Xiaoze Li, Hongpu Hu · 2024
Most patients infected with COVID-19, caused by the coronavirus, exhibit mild to moderate respiratory symptoms and have the ability to recover on their own. However, for the elderly and those with underlying conditions such as cardiovascular disease, diabetes, chronic respiratory diseases, and cancer, infection with COVID-19 can lead to severe illness and even pose a life-threatening risk. In this global pandemic, the scarcity and rational allocation of medical resources have become particularly critical. Therefore, accurately predicting the mortality risk of patients and accordingly allocating medical resources under limited conditions is crucial for effectively reducing the burden on the healthcare system and ensuring the health and safety of the general public. This paper aims to develop an improved machine learning model capable of predicting infectious disease risk. The model will assess whether COVID-19 patients are at risk of death or in a high-risk state based on their real-time symptoms, current health status, and medical history.