Healthcare support system and recording medium

Patent metadata

Field	Value
Publication number	US-2021125726-A1
Application number	US-202017016740-A
Country	US
Kind code	A1
Filing date	Sep 10, 2020
Priority date	Oct 23, 2019
Publication date	Apr 29, 2021
Grant date	—

Abstract

According to one embodiment, a healthcare support system includes a memory and a hardware processor connected to the memory. The hardware processor predicts a risk value of a disease based on medical checkup data for a medical examinee. The hardware processor sets a reduction target for the risk value of the disease, and sets a plurality of second factors constituting search targets among a plurality of first factors relating to the disease and a search range for each of the second factors. The hardware processor searches, by using a predetermined search method, in the search range for each of the second factors, for a target value candidate of each of the second factors so that the risk value of the disease is brought close to the reduction target.

First claim

What is claimed is:

1 . A healthcare support system configured by a computer comprising a memory and a hardware processor connected to the memory,
wherein the hardware processor: predicts a risk value of a disease based on medical checkup data for a medical examinee; sets a reduction target for the risk value of the disease, and sets a plurality of second factors constituting search targets among a plurality of first factors relating to the disease and a search range for each of the second factors; and searches, by using a predetermined search method, in the search range for each of the second factors, for a target value candidate of each of the second factors so that the risk value of the disease is brought close to the reduction target.

2 . The healthcare support system according to claim 1 ,
wherein the hardware processor comprises:
a calculation formula that adds together a first loss and a second loss using fixed weighting, the first loss being calculated from the difference between a current risk value and a risk value constituting the reduction target, and the second loss being calculated from the difference between an examination value for each of the second factors and the target value candidate of each of the second factors,
the hardware processor searches for the target value candidate of each of the second factors for which a third loss obtained using the calculation formula is equal to or less than a preset threshold value.

3 . The healthcare support system according to claim 2 ,
wherein the calculation formula comprises a normalization parameter configured to normalize the value of each of the second factors.

4 . The healthcare support system according to claim 2 ,
wherein the calculation formula comprises an individual parameter configured to reflect an individual intention regarding a plurality of items relating to lifestyle.

5 . The healthcare support system according to claim 1 ,
wherein the hardware processor ends the search processing if the number of candidate searches using the predetermined search method reaches a preset number.

6 . The healthcare support system according to claim 1 ,
wherein the predetermined search method includes a random method.

7 . The healthcare support system according to claim 1 ,
wherein the predetermined search method includes Bayesian method.

8 . The healthcare support system according to claim 1 ,
wherein the hardware processor presents the target value for each of the second factors which is obtained as a search result by a target search unit.

9 . The healthcare support system according to claim 8 ,
wherein the hardware processor presents a current risk value predicted from the medical checkup data and a post-reduction risk value so as to enable comparison between the risk values.

10 . A recording medium storing a program executed by a computer,
the program being configured to cause the computer to operate as: a prediction unit that predicts a risk value of a disease based on medical checkup data for a medical examinee; a setting unit that sets a reduction target for risk value of the disease and sets a plurality of second factors constituting search targets among a plurality of first factors relating to the disease and a search range for each of the second factors; and a target search unit that searches, by using a predetermined search method, in the search range for each of the second factors, for a target value candidate of each of the second factors so that the risk value of the disease is brought close to the reduction target.

11 . The recording medium according to claim 10 ,
wherein the target search unit comprises: a calculation formula that adds together a first loss and a second loss using fixed weighting, the first loss being calculated from the difference between a current risk value and a risk value constituting the reduction target, and the second loss being calculated from the difference between an examination value for each of the second factors and the target value candidate of each of the second factors, and the target search unit searches the target value candidate of each of the second factors in which a third loss obtained using the calculation formula is equal to or less than a preset threshold value.

Assignees

• Toshiba Kk

Inventors

• Doniwa Kenichi
• Haruki Kosuke
• Tanaka Takahiro

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G16H50/30 Primary

  for calculating health indices; for individual health risk assessment · CPC title

• A61B5/7275

  Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor · CPC title

• A61B5/7435

  Displaying user selection data, e.g. icons in a graphical user interface · CPC title