Insulin control method, device and equipment

Abstract

The invention provides an insulin control method, device and equipment, and relates to the field of blood glucose control. The method comprises the steps: receiving a blood glucose control target instruction for a human body; detecting physiological parameters in the human body to obtain a human body physiological parameter set St at the moment t; inputting the physiological parameter set St to aninsulin control model, and obtaining insulin dosage Nt needed by the human body at the moment t; and iteratively repeating the above steps at the t+1 moment. The insulin control model in the scheme is obtained by training a deep deterministic strategy gradient model; according to the measured glucose distribution, the daily basic insulin rate and the ratio of insulin to carbohydrate of each humanbody are optimized, human body characteristics in a daily blood glucose curve can be learned in real time, and a personalized insulin regulation scheme is provided; and the method can adapt to humanbody blood glucose values which are rapidly and dynamically changed, so that the blood glucose is maintained within a reasonable range.

Description

technical field [0001] The present invention relates to the field of blood sugar control, and more specifically, to insulin control methods, devices and equipment. Background technique [0002] In type 1 diabetes, the pancreas of the human body does not produce insulin hormone due to autoimmune destruction of pancreatic beta cells, which leads to excessive rise in blood sugar levels, which leads to various complications. At present, it is more and more common to use insulin pumps to simulate basal doses and high doses to regulate blood sugar in people with type 1 diabetes. Among them, the basic dose and the high dose are calculated according to the physiological parameters of the human body. Insulin pump injection can effectively control the patient's blood sugar and reduce some of the burden on the patient. However, this method requires many links of manual operation, and there is a large influence of human factors. [0003] At present, researchers in some related fields ...

AI Technical Summary

Problems solved by technology

[0004] However, blood glucose changes in the human body are affected by various factors such as diet, daily activities, exercise stress and hormones, and are often complex, non-linear and time-varying in nature

Some of the control algorithms mentioned above are either based on linear system design, or are used for open-loop system optimization, or are based on rules that rely on the designer’s experience. They are often only applicable to specific blood glucose models, and are not good Adapt to dynamic changes in blood glucose levels in the human body, so all have limitations

Images