Diagnosis and treatment auxiliary device

Abstract

The invention provides a diagnosis and treatment support device which evaluates the performance of a pre-detection model for preventing harmful phenomena. A diagnosis and treatment auxiliary device according to an embodiment includes a detection part, a classification part, and a calculation part. The detection part outputs a pre-detection of the presence or absence of a pre-detection at each of aplurality of times of application of a pre-detection model that applies a harmful phenomenon to time-series diagnosis and treatment data. The classification part classifies the detection event of thepre-detection model related to the plurality of application timings into a plurality of patterns defined by the combination of the presence or absence of the pre-detection and the presence or absenceof a diagnosis and treatment event related to a detrimental phenomenon. The calculation part calculates a performance index for evaluating the pre-detection model on the basis of the number of piecesof each of the plurality of patterns.

Description

[0001] Related Application Cross Reference: This application is based on Japanese submitted by Japan on September 2, 2019, and the next application is incorporated herein by reference. Technical field [0002] The present invention relates to a diagnostic auxiliary device. Background technique [0003] Inadvertent medical incidents generated by the patient, it is collectively referred to as harmful phenomena. Harmful phenomena includes, for example, a heart failure that needs hospitalization, the deterioration of a patient caused by a disease caused by a disease, from the treatment of the liver, renal dysfunction caused by the treatment by treatment. In this case, there is known to use the actual diagnosis and treatment data to construct a pre-detection model and pre-detected the harmful phenomenon. In pre-detection of harmful phenomena, intervention (treatment intervention) for preventing harmful phenomena is implemented by doctors. [0004] However, depending on the accuracy of ...

AI Technical Summary

Problems solved by technology

[0004] However, depending on the accuracy of pre-detection, harmful phenomena may not be prevented or unnecessary intervention may be performed

For example, there is a problem that the same patient cannot be observed simultaneously with and without pre-testing.

Also, there is a problem that it is not possible to simultaneously observe the occurrence of the harmful phenomenon and the case of the absence of the harmful phenomenon in the same patient.

Therefore, in the case where no harmful phenomenon has occurred, it is impossible to distinguish the case where no harmful phenomenon has occurred due to intervention in accordance with the pre-detection, and the case where no harmful phenomenon has occurred due to inappropriate prior detection and independent of the implementation of the intervention.

That is, there is a problem that it is impossible to verify whether the pre-detection based on the pre-detection model is appropriate for each patient.

In other words, there is a problem that the performance of pre-detection models for preventing harmful phenomena cannot be evaluated

Images