Apparatus and method for estimating actual dietary content of meal of user

Patent metadata

Field	Value
Publication number	US-11024198-B2
Application number	US-201615259896-A
Country	US
Kind code	B2
Filing date	Sep 8, 2016
Priority date	Sep 8, 2016
Publication date	Jun 1, 2021
Grant date	Jun 1, 2021

Abstract

Provided are an apparatus and method for acquiring information designating a plurality of menu items provided in a meal; reading a characteristic of each of the plurality of menu items provided in the meal, from a menu item database that stores a characteristic for each menu item; generating an classification model for classifying a specified menu item from among the plurality of menu items, based on characteristics of the menu items; acquiring measurement data measured by a sensor attached to a utensil during the meal; and classifying a menu item corresponding to the measurement data, from a characteristic corresponding to the measurement data, using the classification model.

First claim

What is claimed is:

1. An apparatus for calorie estimation for a user, the apparatus comprising:
one or more computer processors; and
one or more computer-readable mediums collectively including instructions executable by the one or more computer processors to:
acquire information designating, from a set of possible menu items, a plurality of menu items actually included in a meal and further designating measurable characteristics of the plurality of menu items, wherein each of the plurality of menu items is shared by a plurality of users including the user;
generate, based on machine learning using the information as training data, a classification model for classifying a given measurable characteristic as being of a specified menu item from among the plurality of menu items, wherein using the information as training data results in the classification model having a greater degree of accuracy in classifying the plurality of menu items as compared to using an entirety of the set of possible menu items as training data, wherein the classification model includes at least one of a support vector machine and a neural network;
measure, by a first sensor attached to a utensil of the user as the first sensor comes into contact with a respective portion of each menu item of the plurality of menu items as consumed by the user during the meal, at least one measurable characteristic of the respective portion in order to acquire measurement data;
classify, based on the measurement data and the classification model, each consumed portion as being of a respective menu item of the plurality of menu items; and
upon determining that the meal is complete, based on detecting a level of usage of the utensil by a second sensor attached to the utensil, generate, by a processing element of the utensil, dietary advice for the user based on the measurement data and each classified portion, wherein the dietary advice is output to the user.

2. The apparatus according to claim 1 , wherein
the classifying each portion as a respective menu item includes selecting a menu item having a highest confidence rating among the plurality of menu items as the menu item corresponding to the measurement data.

3. The apparatus according to claim 1 , wherein the measurable characteristics include a respective measurable characteristic of each of the plurality of menu items, wherein the respective measurable characteristic is read from a menu item database that stores the respective measurable characteristic for each menu item;
wherein the reading of the measurable characteristic of each of the plurality of menu items includes, for each menu item, reading from the menu item database at least one measurable characteristic that includes a variance based on at least one of an identity of a particular chef, a particular cooking style, and a particular cooking inconsistency; and
wherein the generating the classification model includes performing a learning process to classify food as being of a corresponding menu item, wherein the food is classified based on each of the at least one measurable characteristic of each of the plurality of menu items.

4. The apparatus according to claim 1 , wherein the instructions further cause the processor to:
estimate dietary content of the meal of the user, based on time-series data of the measurement data.

5. The apparatus according to claim 1 , wherein the instructions further cause the one or more computer processors to:
acquire the measurement data measured by the first sensor attached to the utensil during cooking or after cooking; and
register the measurable characteristic corresponding to the measurement data in a menu item database in association with a new menu item that has not been registered in the menu item database, in response to the measurement data measured by the first sensor attached to the utensil during cooking or after cooking relating to the new menu item.

6. The apparatus according to claim 1 , wherein the first sensor includes at least one of a salt concentration sensor, a sugar concentration sensor, a temperature sensor, an electrical resistance sensor, and a warping sensor, wherein the second sensor includes at least one of an acceleration sensor and a gyro sensor.

7. The apparatus of claim 1 , wherein the dietary advice comprises a recommended modification to a eating habit of the user, whereafter the recommended modification to the eating habit is adopted by the user;
wherein classifying each consumed portion as being of a respective menu item includes selecting a menu item having a highest confidence rating among the plurality of menu items as the menu item corresponding to the measurement data;
wherein the at least one measurable characteristic includes at least one of salt concentration, sugar concentration, temperature, electrical resistance, warping, and pH, wherein the first and second sensors are of a plurality of types.

8. The apparatus according to claim 7 , wherein the measurable characteristics include a respective measurable characteristic of each of the plurality of menu items, wherein the respective measurable characteristic is read from a menu item database that stores the respective measurable characteristic for each menu item;
wherein reading the measurable characteristic of each of the plurality of menu items includes, for each menu item, reading from the menu item database at least one measurable characteristic that includes a variance based on at least one of an identity of a particular chef, a particular cooking style, and a particular cooking inconsistency;
wherein generating the classification model includes performing a learning process to classify food as being of a corresponding menu item, wherein the food is classified based on each of the at least one measurable characteristic of each of the plurality of menu items.

9. The apparatus according to claim 8 , wherein the plurality of menu items consist of a subset of available menu items, wherein the classification model is generated based on the subset rather than based on the available menu items;
wherein the classification model includes a support vector machine (SVM) and a neural network, wherein the instructions further cause the processor to estimate dietary content of the meal of the user, based on time-series data of the measurement data, wherein estimating the dietary content of the meal of the user includes estimating an amount of each of salt and sugar ingested by the user.

10. The apparatus according to claim 9 , wherein the measurement data includes a measurement value indicating a measurable characteristic and further includes a measurement value indicating an amount of a menu item that has been eaten by the user, as measured by the first sensor during the meal;
wherein the recommended modification is generated by evaluating the dietary content of the meal of the user, wherein the recommended modification specifies an order in which the plurality of menu items should be eaten by the user to attain a specified health benefit.

11. The apparatus according to claim 10 , wherein the instructions further cause the one or more computer processors to:
acquire the measurement data measured by the first sensor during cooking or after cooking; and
register the measurable characteristic corresponding to the measurement data in the menu item database in association with a new menu item that has not been registered in the menu item database, in response to the measurement data measured by the first sensor during cooking or after cooking relating to the new menu item.

12. The apparatus according to claim 11 , wherein the first sensor includes a salt concentration sensor

Assignees

• IBM

Inventors

• Hara Kazuhiro
• Nakano Hiroki
• Nishiyama Hiroki
• Okada Ai
• Yoshida Tadayuki
• Yoshitake Mayumi

Classifications

• A47G21/00

  Table-ware (crumb trays A47L13/52; table knives B26B) · CPC title

• G09B19/0092 Primary

  Nutrition · CPC title

• B26B3/02

  Table-knives (B26B9/02 takes precedence) · CPC title