Emergency treatment grading auxiliary method and device for atypical critical patients

Abstract

The invention discloses an emergency treatment grading auxiliary method and device for atypical critical patients, and the method comprises the steps: obtaining the treatment information of patients whose initial grades are III / IV grades, wherein the treatment information comprises the population informatics data, key signs, consciousness states and diagnosis mode information of the patients; performing feature cleaning and processing on the doctor seeing information to obtain graded identification features of the patient; and inputting the grading identification features as input into a pre-trained intelligent emergency treatment grading model for identification to obtain an emergency treatment supplementary grading result of the patients whose initial grades are III / IV grades. According to the invention, factors are considered comprehensively, the prediction accuracy is high, and the workload of emergency workers is reduced to a great extent; a solid reference foundation is provided for emergency treatment grading of atypical critical patients by medical staff; the method is wide in application range, and is also suitable for juvenile patients, and the application range of the intelligent emergency treatment grading model is expanded.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, in particular to an emergency grading assistance method and device for atypical critically ill patients. Background technique [0002] At present, more and more patients visit emergency departments due to their convenience, and emergency department congestion and delays in diagnosis and treatment caused by the increase in the number of emergency department visits are widespread all over the world. Delayed diagnosis and treatment will lead to increased mortality, and the emergency triage system can be used to distinguish the severity of patients, so as to rationally allocate medical resources. There are many kinds of emergency triage systems in the world. At present, the ESI-5 level triage system is widely used in the United States. China is currently using the Chinese 4-level triage system determined in the "Guidelines for Classification of Emergency Patients" issued by...

AI Technical Summary

Problems solved by technology

However, there are three problems: 1) Clinically, there will always be atypical critically ill patients who cannot be identified as being in critical condition by existing standards; 2) The condition of critically ill patients is changing, but they may When grading, the performance is not serious, and it does not meet the critical state defined by the grading standard; 3) Some characteristics related to critical illness identification, such as age, gender, way of coming to the emergency room, time of emergency room visit, etc., these characteristics are related to the presence of specific signs of the patient non-linear association and cannot be written directly into a rule-based ranking system

[0004] These issues pose great challenges to existing rule-based grading systems

Images