Granted Patent B2

US 12,417,836 · App. 18/090,411 · Granted Sep 16, 2025

Inventor: Kenneth Neumann (Lakewood, CO)

Assignee: KPN INNOVATIONS LLC

G16H20/60G06F18/2415G16H10/60

Abstract

Claims (78)

1. An apparatus for scoring a nutrient, comprising:

at least a processor; and

a memory communicatively connected to the at least a processor, the memory containing instructions configuring the at least a processor to:

receive user data and edible data;

extract, from the edible data, a plurality of nutrients, wherein extracting the plurality of nutrients comprises performing a web search configured to generate a plurality of weights to a plurality of semantic elements of the edible data, wherein extracting the plurality of nutrients further comprises using at least an optical character recognition (OCR) process by converting the edible data into machine-encoded text by the at least an OCR process, wherein converting the edible data into the machine-encoded text comprises converting images of text in the edible data into the machine-encoded text and further comprises:

pre-processing image components of the images, wherein pre-processing the image components comprises:

de-skewing at least one of the image components by applying a homography transform to the at least one of the image components;

converting at least a portion of one of the images from color or greyscale to a binary image format; and

normalizing an aspect ratio of at least one of the image components;

implementing an OCR algorithm comprising a matrix matching process, wherein implementing the OCR algorithm comprises:

comparing pixels of at least one of the pre-processed images to pixels of a stored glyph on a pixel-by-pixel basis; and

ascertaining a similar font and scale therebetween based on the comparison; and

post-processing an output of the matrix matching process to increase OCR accuracy by constraining the output to a lexicon containing a set of words whose occurrence is permitted;

classify the plurality of nutrients, extracted from the edible data using the at least an OCR process, to a plurality of impact factors utilizing a nutrient classifier, wherein the nutrient classifier is configured to receive a meal ID and recipe data as an input and output the plurality of impact factors matched to a plurality of nutrition data elements of the meal ID and the recipe data;

classify a user to a profile cluster as a function of at least an element of the user data, wherein the profile cluster comprises a grouping of phenotypes;

assign the user one or more cohort labels as a function of the profile cluster;

score at least a nutrient of the plurality of nutrients as a function of the cohort label and the plurality of impact factors, wherein scoring the at least a nutrient comprises:

generating training data correlating phenotype data to nutrient scores, wherein the nutrient scores comprise the plurality of weights derived by the web search;

iteratively training a nutrient score machine learning model using the training data, wherein iteratively training the nutrient score machine learning model further comprises:

using training data applied to an input layer of nodes comprising at least one input node, one or more intermediate layers of nodes, and an output layer of nodes;

adjusting one or more connections and one or more weights between nodes in adjacent layers of the machine learning model;

detecting additional correlations between the output layer of nodes and the input layer of nodes;

identifying a plurality of nutrients as a function of edible data;

updating the training data based on user input comprising additional training data;

retraining the nutrient score machine learning model using the detected additional correlations between the output layer of nodes and the input layer of nodes; and

generating, as a function of the nutrient score machine learning model, a nutrient score; and

generate a nutrient chain by comparing one or more impact factors of the plurality of impact factors with the at least a nutrient of the plurality of nutrients using an objective function, wherein:

the objective function comprises an optimization criterion; and

the optimization criterion assigns weights to each of the one or more impact factors, wherein the one or more impact factors are a metric of influence of one or more nutrients on an individual's biological system.

2. The apparatus of claim 1 , wherein the memory contains instructions further configuring the at least a processor to score the at least a nutrient based on the one or more impact factors.

3. The apparatus of claim 1 , wherein the memory contains instructions configuring the at least a processor to generate a plurality of nutrient chain combinations.

4. The apparatus of claim 1 , wherein scoring the at least a nutrient further comprises utilizing at least one objective function to score the at least a nutrient.

5. The apparatus of claim 1 , wherein assigning the user one or more cohort labels further comprises assigning at least a first cohort label associated with a first cluster and assigning a second cohort label associated with at least a second cluster.

6. The apparatus of claim 1 , wherein the at least a processor is further configured to:

receive additional user data;

reclassify the user to additional groupings of phenotypes as a function of the additional user data; and

assign the user one or more additional cohort labels as a function of the additional user data.

7. The apparatus of claim 6 , wherein the at least a processor is further configured to score the at least a nutrient as a function of the one or more additional cohort labels.

8. The apparatus of claim 1 , wherein the memory contains instructions further configuring the at least a processor to determine a nutrition target range.

9. The apparatus of claim 1 , wherein the memory contains instructions further configuring the at least a processor to display, through a graphical user interface, the score of the at least a nutrient.

10. A method of scoring a nutrient using a computing device, comprising:

receiving, using at least a processor, user data from a user;

classifying, by the at least a processor, the user to a profile cluster as a function of the user data, wherein the profile cluster comprises a grouping of phenotypes;

assigning, by the at least a processor, the user one or more cohort labels as a function of the user data;

receiving, by the at least a processor, edible data;

extracting, by the at least a processor, from the edible data, a plurality of nutrients, wherein extracting the plurality of nutrients comprises performing a web search configured to generate a plurality of weights to a plurality of semantic elements of the edible data, wherein extracting the plurality of nutrients further comprises using at least an optical character recognition (OCR) process by converting the edible data into machine-encoded text by the at least an OCR process, wherein converting the edible data into the machine-encoded text comprises converting images of text in the edible data into the machine-encoded text and further comprises:

pre-processing image components of the images, wherein pre-processing the image components comprises:

de-skewing at least one of the image components by applying a homography transform to the at least one of the image components;

converting at least a portion of one of the images from color or greyscale to a binary image format; and

normalizing an aspect ratio of at least one of the image components;

implementing an OCR algorithm comprising a matrix matching process, wherein implementing the OCR algorithm comprises:

comparing pixels of at least one of the pre-processed images to pixels of a stored glyph on a pixel-by-pixel basis; and

ascertaining a similar font and scale therebetween based on the comparison; and

post-processing an output of the matrix matching process to increase OCR accuracy by constraining the output to a lexicon containing a set of words whose occurrence is permitted;

classifying, by the at least a processor, the plurality of nutrients, extracted from the edible data using the at least an OCR process, to a plurality of impact factors utilizing a nutrient classifier, wherein the nutrient classifier is configured to receive a meal ID and recipe data as an input and output the plurality of impact factors matched to a plurality of nutrition data elements of the meal ID and the recipe data;

scoring, by the at least a processor, at least a nutrient of the plurality of nutrients as a function of the profile cluster of the user and the plurality of impact factors, wherein scoring the at least a nutrient comprises:

generating training data correlating phenotype data to nutrient scores, wherein the nutrient scores comprise the plurality of weights derived by the web search;

iteratively training a nutrient score machine learning model using the training data, wherein iteratively training the nutrient score machine learning model comprises:

updating the training data based on user input comprising additional training data;

using training data applied to an input layer of nodes comprising at least one input node, one or more intermediate layers of nodes, and an output layer of nodes;

adjusting one or more connections and one or more weights between nodes in adjacent layers of the machine learning model;

detecting additional correlations between the output layer of nodes and the input layer of nodes;

identifying a plurality of nutrients as a function of edible data; and

generating, as a function of the nutrient score machine learning model, a nutrient score; and

generating, by the at least a processor, a nutrient chain by comparing one or more impact factors of the plurality of impact factors with the at least a nutrient of the plurality of nutrients using an objective function, wherein:

the objective function comprises an optimization criterion; and

the optimization criterion assigns weights to each of the one or more impact factors, wherein the one or more impact factors are a metric of influence of one or more nutrients on an individual's biological system.

11. The method of claim 10 , wherein scoring the at least a nutrient further comprises scoring, by the at least a processor, the at least a nutrient of the edible data based on the one or more impact factors.

12. The method of claim 10 , further comprising generating, by the at least a processor, a plurality of nutrient combinations.

13. The method of claim 10 , wherein scoring the at least a nutrient further comprises utilizing at least one objective function to generate the score of the at least a nutrient.

14. The method of claim 10 , wherein assigning the user one or more cohort labels further comprises assigning at least a first cohort label associated with a first cluster and assigning a second cohort label associated with at least a second cluster.

15. The method of claim 10 , further comprising:

receiving, by the at least a processor, additional user data;

reclassifying, by the at least a processor, the user to additional groupings of phenotypes as a function of the additional user data; and

assigning, by the at least a processor, the user one or more additional cohort labels as a function of the additional user data.

16. The method of claim 15 , wherein scoring the at least a nutrient comprises scoring the at least a nutrient as a function of the one or more additional cohort labels.

17. The method of claim 10 , further comprising determining, by the at least a processor, a nutrition target range.

18. The method of claim 10 , further comprising displaying, through a graphical user interface, the score of the at least a nutrient.

Assignments (1)

Continuity (1)

References Cited (26)