Granted Patent B2

US 12,633,410 · App. 17/106,588 · Granted May 19, 2026

Inventor: Kenneth Neumann (Lakewood, CO)

Assignee: KPN INNOVATIONS, LLC

G16H50/20G06N20/00

Abstract

Claims (59)

1 . A system for determining a predictive intervention using biomarkers, the system comprising a computing device, wherein the computing device is configured to:

receive physiological data from a subject, wherein the physiological data is received from at least a physiological sensor wherein the physiological data comprises multi-dimensional biomarker data comprising one or more of a genetic, metabolic, immunological, and blood panel data;

identify a malady, as a function of the physiological data, wherein identifying the malady further comprises:

indicating a figure of merit for a plurality of maladies as they correlate to at least a biomarker, wherein the figure of merit is a quantity used to characterize a fit of a malady for at least a biomarker detected in the physiological data, wherein the figure of merit includes a percentile that indicates a likelihood the malady fits a particular pattern of biomarkers, wherein the computing device determines a match between the malady and physiological data as a function of the percentile, wherein the figure of merit includes a quantifier derived from a machine learning model that informs the system on selecting the malady from the physiological data;

selecting the malady as a function of the figure of merit;

classify the malady to a plurality of intervention categories for the malady, wherein each intervention category of the plurality of intervention categories is a distinct classification based on physiological data utilizing an intervention machine-learning model which comprises:

receiving intervention training data, wherein the intervention training data correlates a plurality of biomarker data to a plurality of intervention category data;

iteratively training the intervention machine-learning model with the intervention training data and includes retraining the intervention machine-learning model with feedback from previous iterations of the intervention machine-learning model, wherein the intervention machine-learning model is trained further to derive a numerical scale for providing numerical values to determine an efficacy value of an intervention category and to learn:

upper and lower limits to the scale;

increments to determining the efficacy value; and

a criteria for increasing and decreasing the efficacy value, wherein training the intervention machine-learning model with the training data comprises:

applying the training data to input nodes of the intervention machine learning model;

creating connections between a plurality of nodes of the intervention machine-learning model; and

adjusting the connections and weights between the plurality of nodes of the intervention machine-learning model to produce desired values at output nodes of the intervention machine learning model;

identifying, using the intervention machine-learning model a plurality of intervention categories correlated to the identified malady;

generate at least a combination of the plurality of intervention categories using a machine learning model comprising:

receiving training data, wherein the training data correlates a plurality of relationship data to a plurality of intervention category data for the malady generated from the trained intervention machine-learning model;

iteratively training the machine learning model with the training data which includes retraining the machine learning model with feedback from previous iterations of the machine learning model;

assign efficacy values, using the intervention machine-learning model, to the at least a combination of the plurality of intervention categories as a function of addressing the malady; and

display, to the subject, a predictive intervention by generating a representation, via a graphical user interface, which includes a combination as a function of the assigned values, wherein the combination comprises, ordering, the plurality of interventions based of a variety of criteria including at least a difficulty of adaption criterion.

2 . The system of claim 1 , wherein classifying the malady to at least the intervention category further comprises identifying a plurality of predictive interventions as a function of the physiological data and the intervention category.

3 . The system of claim 1 , wherein assigning efficacy values further comprises determining the efficacy for each combination of the plurality of intervention categories in addressing malady.

4 . The system of claim 1 , wherein assigning efficacy values further comprises:

training an intervention machine-learning model with training data that includes a plurality of data entries wherein each data entry correlates intervention efficacy to numerical value scales; and

assigning efficacy values to the plurality of predictive interventions as a function of the intervention machine-learning model.

5 . The system of claim 4 , wherein determining efficacy further comprises retrieving the plurality of predictive interventions from each intervention category of the combination and assigning an efficacy value as a function of the intervention classifier.

6 . The system of claim 1 , wherein assigning values for at least the combination further comprises:

determining a plurality of predictive intervention combinations; and

generating an objective function of the plurality of predictive intervention combinations as a function of efficacy in addressing the malady, wherein maximizing the objective function maximizes the efficacy value of the combination.

7 . The system of claim 1 , wherein providing the predictive intervention further comprises generating an ordering of the plurality of combinations based on efficacy value.

8 . A method for determining a predictive intervention using biomarkers, the method comprising:

receiving, by a computing device, physiological data from a subject, wherein the physiological data is received from at least a physiological sensor wherein the physiological data comprises multi-dimensional biomarker data comprising one or more of a genetic, metabolic, immunological, and blood panel data;

identifying, by the computing device, a malady, as a function of the physiological data, wherein identifying the malady further comprises:

indicating a figure of merit for a plurality of maladies as they correlate to at least a biomarker, wherein the figure of merit is a quantity used to characterize a fit of a malady for at least a biomarker detected in the physiological data, wherein the figure of merit includes a percentile that indicates a likelihood the malady fits a particular pattern of biomarkers, wherein the computing device determines a match between the malady and physiological data as a function of the percentile, wherein the figure of merit includes a quantifier derived from a machine learning model that informs a system on selecting the malady from the physiological data;

selecting the malady as a function of the figure of merit;

classifying, by the computing device, the malady to a plurality of intervention categories for the malady, wherein each intervention category of the plurality of intervention categories is a distinct classification based on physiological data utilizing an intervention machine-learning model which comprises:

receiving intervention training data, wherein the intervention training data correlates a plurality of biomarker data to a plurality of intervention category data;

iteratively training the intervention machine-learning model with the intervention training data and includes retraining the intervention machine-learning model with feedback from previous iterations of the intervention machine-learning model, wherein the intervention machine-learning model is trained further to derive a numerical scale for providing numerical values to determine an efficacy value of an intervention category and to learn: upper and lower limits to the scale;

increments to determining the efficacy value; and

a criteria for increasing and decreasing the efficacy value, wherein training the intervention machine-learning model with the training data comprises:

applying the training data to input nodes of the intervention machine learning model;

creating connections between a plurality of nodes of the intervention machine-learning model; and

adjusting the connections and weights between the plurality of nodes of the intervention machine-learning model to produce desired values at output nodes of the intervention machine learning model;

identifying, using the intervention machine-learning model a plurality of intervention categories correlated to the identified malady;

generating, by the computing device, at least a combination of the plurality of intervention categories using a machine learning model comprising:

receiving training data, wherein the training data correlates a plurality of relationship data to a plurality of intervention category data for the malady generated from the trained intervention machine-learning model;

iteratively training the machine learning model with the training data which includes retraining the machine learning model with feedback from previous iterations of the machine learning model;

assigning, using the intervention machine-learning model, by the computing device, efficacy values to the at least a combination of the plurality of intervention categories as a function of addressing the malady; and

displaying, by the computing device, to the subject, a predictive intervention by generating a representation, via a graphical user interface, which includes a combination as a function of the assigned values, wherein providing the combination comprises, ordering, the plurality of interventions based of a variety of criteria including at least a difficulty of adaption criterion.

9 . The method of claim 8 , wherein classifying the malady to at least the intervention category further comprises identifying a plurality of predictive interventions as a function of the physiological data and the intervention category.

10 . The method of claim 8 , wherein assigning efficacy values further comprises determining the efficacy for each combination of the plurality of intervention categories in addressing malady.

11 . The method of claim 8 , wherein assigning efficacy values further comprises:

training an intervention machine-learning model with training data that includes a plurality of data entries wherein each data entry correlates intervention efficacy to numerical value scales; and

assigning efficacy values to the plurality of predictive interventions as a function of the intervention machine-learning model.

12 . The method of claim 11 , wherein determining efficacy further comprises retrieving the plurality of predictive interventions from each intervention category of the combination and assigning an efficacy value as a function of the intervention classifier.

13 . The method of claim 8 , wherein assigning values for at least the combination further comprises:

determining a plurality of predictive intervention combinations; and

generating an objective function of the plurality of predictive intervention combinations as a function of efficacy in addressing the malady, wherein maximizing the objective function maximizes the efficacy value of the combination.

14 . The method of claim 8 , wherein providing the predictive intervention further comprises generating an ordering of the plurality of combinations based on efficacy value.

Assignments (1)

Continuity (1)

References Cited (18)