GENERATION METHOD FOR ABNORMALITY DETERMINATION MODEL, ABNORMALITY DETERMINATION DEVICE, ABNORMALITY DETERMINATION METHOD, AND LEARNT MODEL

Japanese Patent JP2023125820

A

To provide a method of generating an abnormality determination model capable of excellent abnormality determination of a subject based on load detection.SOLUTION: A generation method for an abnormality determination model includes: detecting load of a subject by a load sensor arranged on a bed; creating teacher data associating each of many kinds of feature amounts based on the detected load with an abnormal state; creating a model classifying a state of the subject into the abnormal condition on the basis of the many kinds of feature amounts by machine learning with a teacher using the teacher data; determining at least one of the many kinds of feature amounts on the basis of the model as an explanatory variable; and generating an abnormality determination model determining by machine learning using the explanatory variable that the subject is in the abnormal state on the basis of the explanatory variable. The many kinds of feature amounts include frequency feature amounts calculated by performing Fourier transformation of the load of the subject for a short time.SELECTED DRAWING: Figure 8

WO/2021/028928	SYSTEM AND METHOD FOR DETERMINING ONSET AND DISEASE PROGRESSION
JP2014501544	Prompts initiated by ventilator in response to planning adjustments
JP2013123495	RESPIRATORY SOUND ANALYSIS DEVICE, RESPIRATORY SOUND ANALYSIS METHOD, RESPIRATORY SOUND ANALYSIS PROGRAM, AND RECORDING MEDIUM

FUSE TORU
ZAITSU YUSUKE

JP2022030136A

September 07, 2023

February 28, 2022

Click for automatic bibliography generation  

MINEBEAMITSUMI INC

A61B5/08; A61B5/11; G06N20/00

Kijuro Kawakita
Masahiro Fujita
Ikuo Kawamura

Previous Patent: Print control program, information processing device, and print control method

Next Patent: Receiving device, broadcasting device, broadcasting system, receiving method and program