Health condition evaluation based on temporal analysis of dietary habits

Patent metadata

Field	Value
Publication number	US-12354748-B2
Application number	US-202016991020-A
Country	US
Kind code	B2
Filing date	Aug 12, 2020
Priority date	Aug 12, 2020
Publication date	Jul 8, 2025
Grant date	Jul 8, 2025

Abstract

Evaluation of a health condition includes obtaining a correlation between dietary habits and health issues of a plurality of reference users. The correlation is determined based on reference patterns. Each of the reference patterns indicates temporal characteristics of at least one dietary habit and at least one health issue of a reference user of the plurality of reference users. A target pattern of a target user is determined at least in part based on a profile for the target user. The target pattern at least indicates temporal characteristics of at least one dietary habit of the target user. An evaluation of a health condition of the target user is generated based on the target pattern and the correlation.

First claim

What is claimed is:

1. A computer-implemented method comprising:
using a collection of historical profiles associated with multiple reference users, generating, by one or more processors, correlations between dietary habits and health issues of the multiple reference users, each of the correlations being determined based on a respective temporal relation sequence of a respective reference pattern associated with a respective reference user of the multiple reference users, each of the reference patterns indicating temporal characteristics of multiple dietary habits and at least one health issue of the respective reference user of the multiple reference users, the temporal relation sequence indicating a dietary habit of the multiple dietary habits and the at least one health issue of the respective reference user were aligned in time by occurring at least partially simultaneously;
determining, by the one or more processors, a target pattern of a target user at least in part based on a profile of the target user, the target pattern at least indicating temporal characteristics of multiple dietary habits of the target user;
generating, by the one or more processors, an evaluation of a health condition of the target user based on the target pattern and the correlations and based on the target pattern having a similarity to a reference pattern of at least one of the reference users that exceeds a predetermined threshold value, the similarity being determined based on the multiple dietary habits being weighted based on a ratio of the overlap of their respective time period of occurrence compared to corresponding time period of occurrence for corresponding dietary habits of the reference users, the overlap being determined for a relative time scale, wherein the evaluation is further based on a first correlation from the at least one of the reference users; and
sending, by the one or more processors, a dietary habit suggestion and a health issue notification to a target user device based on the evaluation of the health condition of the target user, wherein the health issue notification indicates a potential health issue for the target user and a recommended physical examination that relates to the potential health issue.

2. The computer-implemented method of claim 1 , wherein the temporal characteristics of the multiple dietary habits of the respective reference user comprise a respective time period corresponding to the respective dietary habit of the respective reference user,
wherein the temporal characteristics of the at least one health issue of the respective reference user comprise a time period corresponding to the at least one health issue of the respective reference user, and
wherein the temporal characteristics of the multiple dietary habits of the target user comprise a respective time period corresponding to the respective dietary habit of the target user.

3. The computer-implemented method of claim 1 , wherein the profile of the target user further indicates a health condition of the target user, and wherein the target pattern is further determined via:
determining, by the one or more processors and based on the correlations, that at least some of the multiple dietary habits of the multiple reference users are correlated to the health condition of the target user;
obtaining, by the one or more processors, temporal characteristics of the at least some of the multiple dietary habits of the multiple reference users; and
constructing, by the one or more processors and before the evaluation of the health condition is generated, the target pattern further based on the obtained temporal characteristics and the correlated at least some of the multiple dietary habits of the multiple reference users.

4. The computer-implemented method of claim 1 , further comprising:
comparing, by the one or more processors, the multiple dietary habits of the target user to the multiple dietary habits in the collection of the historical profiles so that the potential health issue for the target user is identified from the collection of the historical profiles, the identifying being based on matches of one or more of the multiple dietary habits of the target user to one or more of the dietary habits of the multiple reference users;
generating, via the one or more processors, one or more automated questions for the target user regarding the identified potential health issues to inquire whether the target user has or had the identified potential health issue; and
in response to a determination that the target user has or had the identified potential health issue, adding via the one or more processors temporal characteristics of the identified potential health issue to the target pattern before the evaluation of the health condition is generated.

5. The computer-implemented method of claim 4 , further comprising:
in further response to the determination that the target user has or had the identified potential health issue, generating, via the one or more processors, one or more automated questions for the target user regarding the temporal characteristics of the identified potential health issue; and
wherein in response to receiving a target user response to the one or more automated questions for the target user regarding the temporal characteristics of the identified potential health issue, the adding of the temporal characteristics to the target pattern is performed based on the target user response.

6. The computer-implemented method of claim 1 , wherein the potential health issue is identified from the at least one of the reference users whose reference pattern had the similarity to the target pattern that exceeded the predetermined threshold value.

7. A system comprising:
one or more processors;
a memory coupled to the one or more processors; and
computer-readable instructions stored in the memory, the instructions, when executed by the one or more processors, causing the one or more processors to perform a method comprising:
using a collection of historical profiles associated with multiple reference users, generating correlations between dietary habits and health issues of the multiple reference users, each of the correlations being determined based on a respective temporal relation sequence of respective reference patterns associated with a respective reference user of the multiple reference users, each of the reference patterns indicating temporal characteristics of multiple dietary habits and at least one health issue of the respective reference user of the multiple reference users, the temporal relation sequence indicating a dietary habit of the multiple dietary habits and the at least one health issue of the respective reference user were aligned in time by occurring at least partially simultaneously;
determining a target pattern of a target user at least in part based on a profile of the target user, the target pattern at least indicating temporal characteristics of multiple dietary habits of the target user;
generating an evaluation of a health condition of the target user based on the target pattern and the correlations and based on the target pattern having a similarity to a reference pattern of at least one of the reference users that exceeds a predetermined threshold value, the similarity being determined based on the multiple dietary habits being weighted based on a ratio of the overlap of their respective time period of occurrence compared to corresponding time period of occurrence for corresponding dietary habits of the reference users, the overlap being determined for a relative time scale, wherein the evaluation is further based on a first correlation from the at least one of the reference users; and
sending a dietary habit suggestion and a health issue notification to a

Assignees

• IBM

Inventors

• Mei Lijun
• Ma Jie
• Zhou Xin
• Li qi cheng
• Chen Hao

Classifications

• G16H10/60

  for patient-specific data, e.g. for electronic patient records · CPC title

• G16H50/30

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

• G16H50/70

  for mining of medical data, e.g. analysing previous cases of other patients · CPC title

• G16H20/60 Primary

  relating to nutrition control, e.g. diets · CPC title

• G16H50/20 Primary

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title