Sepsis early warning equipment, method, device and storage medium

Abstract

The embodiment of the invention relates to the technical field of medical treatment, and discloses sepsis early warning equipment, method, device and a storage medium. A processor of the equipment isused for sampling acquired monitoring data after a patient object enters an intensive care unit to obtain sampled monitoring data; when a data missing field is detected to exist in the sampling monitoring data, missing data in the data missing field is extracted; the missing data are filled into the data missing field to obtain target monitoring data; a plurality of feature data extracted from thetarget monitoring data are inputted into a risk prediction model to obtain the risk probability of suffering from sepsis of the patient object; and when the risk probability is detected to be greaterthan a preset threshold, early warning prompt information is generated, and the early warning prompt information is sent to the medical care terminal. In the way, the accuracy of sepsis prediction can be improved, and the occurrence rate of sepsis is reduced. The method relates to a block chain technology. Monitoring data can be stored in a block chain.

Description

technical field [0001] The invention relates to the field of medical technology, in particular to a sepsis early warning device, method, device and storage medium. Background technique [0002] According to statistics, there are more than 18 million cases of severe sepsis every year in the world, and the mortality rate is as high as 50%. Early detection and timely treatment of sepsis play a crucial role in the outcome of patients with sepsis, and the mortality rate increases by 4%-8% for every hour of delay. A survey report showed that 86% of physicians believed that the symptoms of sepsis were atypical, leading to delays in diagnosis and treatment. At present, most of the prediction schemes for sepsis use traditional machine learning methods, such as random forest, xgboost, etc., and the prediction accuracy is poor. At the same time, the interpretability of the model is poor, it cannot explain the decision-making process, and the results cannot convince clinicians. There...

AI Technical Summary

Problems solved by technology

A survey reported that 86% of physicians believe that the symptoms of sepsis are atypical, leading to delays in diagnosis and treatment

At present, most of the prediction schemes for sepsis use traditional machine learning methods, such as random forest, xgboost, etc., and the prediction accuracy is poor

At the same time, the interpretability of the model is poor, it cannot explain the decision-making process, and the results cannot convince clinicians

Images