Machine learning blood glucose monitoring system and method based on CGM sensor

The invention discloses a machine learning blood glucose monitoring system and method based on a CGM sensor. The system comprises a server terminal, a medical care APP terminal, the CGM sensor and a patient APP terminal. Real-time data interaction is achieved, blood glucose data is collected uninterruptedly from medicine taking, blood glucose sampling time and medicine effect time are highly synchronous in real time, real-time data interaction is achieved, compared with traditional blood glucose data monitoring, reliability is higher, data are related to medicine taking, diet and exercise of a patient under continuous collection and have a causal relationship, and the accuracy of data interaction is improved. Therefore, the causal relationship can be conveniently corresponded from the time dimension, the blood glucose data of the patient can be conveniently evaluated and monitored, in addition, through automatic OGTT testing, the real-time data acquisition characteristics of CMG and a timing reminding functio