Dynamic data-driven biological state analysis
In some implementations, a system is capable of obtaining and processing both actively monitored and passively monitored data in parallel in order to improve the accuracy and the specificity by which pathological risks are identified for a user. Data indicating measured levels of one or more metabolic biomarkers and activity data associated with a user is obtained. A biological state for the user is determined based on the measured levels of the one or more metabolic biomarkers. One or more user inputs indicated within the activity data, and scores reflecting respective likelihoods that a particular user input indicates a change to one or more aspects of the biological state for the user for each of the one or more user inputs is determined. Data corresponding to the biological state for the user is then adjusted. A communication that is generated based on the adjusted data is then provided for output.
1. A method performed by one or more computers, the method comprising:
determining a biological state for a user during a particular time period, wherein:
the biological state is determined based on biomarker data collected for the user during the particular time period, the biomarker data comprising at least a measured glucose level and a measured ketone level, and
the biological state is associated with a nutritional plan that the user participates in during the particular time period;
identifying, based on the biological state, activity data representing (i) one or more terms included in messages submitted by the user on a computing device during the particular time period and (ii) one or more external actions performed by the user on the computing device during the particular time period;
classifying, using one or more machine learning models trained to process historical account data of the user, the activity data as corresponding to a particular conversational pattern from among multiple conversational patterns associated with the biological state, wherein:
the particular conversational pattern is classified based on the one or more external actions performed by the user on the computing device during the particular time period and a context associated with the user when performing the one or more external actions—an emotional sentiment of the user during the particular time period based on the particular conversational;
predicting a preference or an aversion to the nutritional plan by the user based on the one or more terms included in messages submitted by the user on the computing device during the particular time period;
determining a change to a user-specific requirement associated with the nutritional plan based at least on the preference or the aversion and the emotional sentiment of the user during the particular time period;
selecting content to provide to the user based on determining the change to the user-specific requirement associated with the nutritional plan; and
providing, for output to a computing device associated with the user, a communication that includes the selected content.
2. The method of claim 1 , wherein the biomarker data indicating the measured levels of the one or more metabolic biomarkers comprises at least a measured level of a glucose biomarker and a measured level of a ketone biomarker.
3. The method of claim 2 , further comprising:
determining that at least one of the measured level of glucose biomarker and the measured level of ketone biomarker exceeds a corresponding predetermined threshold value within the one or more samples of the user; and
in response to determining that at least one of the measured level of glucose biomarker or the measured level of ketone biomarker exceeds a corresponding predetermined threshold value within the one or more samples of the user, automatically determining that a carbohydrate intake of the user is above a metabolic tolerance for the user.
4. The method of claim 1 , wherein the biomarker data indicating the measured levels of the one or more metabolic biomarkers comprises at least a measured level of palmitoleic acid and a measured level of dihomo-γ-linolenic acid within the one or more samples of the user.
5. The method of claim 1 , wherein determining the change to implement in the monitoring plan comprises determining a change to at least one of a user-specific nutritional requirement to sustain a carbohydrate remission for the user, a weight loss goal specified for the user, or a communication provided to the user corresponding to the biological state for the user.
6. The method of claim 1 , wherein:
the one or more machine learning models comprise a neural network trained to identify an overall sentiment indicated by the one or more terms within the text messages; and
the emotional sentiment to the user is determined based on the overall sentiment identified by the neural network.
7. The method of claim 6 , wherein identifying the overall sentiment indicated by the one or more terms within the text messages comprises:
obtaining, by the neural network, data indicating (i) a collection of terms, and (ii) a predetermined sentiment corresponding to each term included within the collection of terms; and
determining that a particular set of terms within the text messages are included within the collection of terms; and
determining the overall sentiment based on predetermined sentiments of terms within the particular set of terms.
8. A system comprising:
one or more computing devices; and
one or more storage devices storing instructions that are executable by the one or more computing devices and cause the one or more computing devices to perform operations comprising:
determining a biological state for a user during a particular time period, wherein:
the biological state is determined based on biomarker data collected for the user during the particular time period, the biomarker data comprising at least a measured glucose level and a measured ketone level, and
the biological state is associated with a nutritional plan that the user participates in during the particular time period;
identifying, based on the biological state, activity data representing (i) one or more terms included in messages submitted by the user on a computing device during the particular time period and (ii) one or more external actions performed by the user on the computing device during the particular time period;
classifying, using one or more machine learning models trained to process historical account data of the user, the activity data as corresponding to a particular conversational pattern from among multiple conversational patterns associated with the biological state, wherein:
the particular conversational pattern is classified based on the one or more external actions performed by the user on the computing device during the particular time period and a context associated with the user when performing the one or more external actions;
determining an emotional sentiment of the user during the particular time period based on the particular conversational pattern;
predicting a preference or an aversion to the nutritional plan by the user based on the one or more terms included in messages submitted by the user on the computing device during the particular time period;
determining a change to a user-specific requirement associated with the nutritional plan based at least on the preference or the aversion and the emotional sentiment of the user during the particular time period;
selecting content to provide to the user based on determining the change to the user-specific requirement associated with the nutritional plan; and
providing, for output to a computing device associated with the user, a communication that includes the selected content.
9. The system of claim 8 , wherein the biomarker data indicating the measured levels of the one or more metabolic biomarkers comprises at least a measured level of a glucose biomarker and a measured level of a ketone biomarker.
10. The system of claim 9 , wherein the operations further comprise:
determining that at least one of the measured level of glucose biomarker and the measured level of ketone biomarker exceeds a corresponding predetermined threshold value within the one or more samples of the user; and
in response to determining that at least one of the measured level of glucose biomarker or the measured level of ketone biomarker exceeds a corresponding predetermined threshold value within the one or more samples of the user, automatically determining that a carbohydrate intake of the user is above a metabolic tolerance for the user.
11. The system of claim 8 , wherein the biomarker data indicating the measured levels of the one or more metabolic biomarkers comprises at least a measured level of palmitoleic acid and a measured level of dihomo-γ-linolenic acid within the one or more samples of the user.
12. The system of claim 8 , wherein determining the change to implement in the monitoring plan comprises determining a change to at least one of a user-specific nutritional requirement to sustain a carbohydrate remission for the user, a weight loss goal specified for the user, or a communication provided to the user corresponding to the biological state for the user.
13. The system of claim 8 , wherein:
the one or more machine learning models comprise a neural network trained to identify an overall sentiment indicated by the one or more terms within the text messages; and
the emotional sentiment to the user is determined based on the overall sentiment identified by the neural network.
14. The system of claim 13 , wherein identifying the overall sentiment indicated by the one or more terms within the text messages comprises:
obtaining, by the neural network, data indicating (i) a collection of terms, and (ii) a predetermined sentiment corresponding to each term included within the collection of terms; and
determining that a particular set of terms within the text messages are included within the collection of terms; and
determining the overall sentiment based on predetermined sentiments of terms within the particular set of terms.
15. At least one non-transitory computer-readable storage device storing instructions that are executable by one or more processors and cause the one or more processors to perform operations comprising:
determining a biological state for a user during a particular time period, wherein:
the biological state is determined based on biomarker data collected for the user during the particular time period, the biomarker data comprising at least a measured glucose level and a measured ketone level, and
the biological state is associated with a nutritional plan that the user participates in during the particular time period;
identifying, based on the biological state, activity data representing (i) one or more terms included in messages submitted by the user on a computing device during the particular time period and (ii) one or more external actions performed by the user on the computing device during the particular time period;
classifying, using one or more machine learning models trained to process historical account data of the user, the activity data as corresponding to a particular conversational pattern from among multiple conversational patterns associated with the biological state, wherein:
the particular conversational pattern is classified based on the one or more external actions performed by the user on the computing device during the particular time period and a context associated with the user when performing the one or more external actions—an emotional sentiment of the user during the particular time period based on the particular conversational;
predicting a preference or an aversion to the nutritional plan by the user based on the one or more terms included in messages submitted by the user on the computing device during the particular time period;
determining a change to a user-specific requirement associated with the nutritional plan based at least on the preference or the aversion and the emotional sentiment of the user during the particular time period;
selecting content to provide to the user based on determining the change to the user-specific requirement associated with the nutritional plan; and
providing, for output to a computing device associated with the user, a communication that includes the selected content.
16. The non-transitory computer-readable storage device of claim 15 , wherein the operations further comprise:
determining that at least one of the measured level of glucose biomarker and the measured level of ketone biomarker exceeds a corresponding predetermined threshold value within the one or more samples of the user; and
in response to determining that at least one of the measured level of glucose biomarker or the measured level of ketone biomarker exceeds a corresponding predetermined threshold value within the one or more samples of the user, automatically determining that a carbohydrate intake of the user is above a metabolic tolerance for the user.
17. The non-transitory computer-readable storage device of claim 15 , wherein the biomarker data indicating the measured levels of the one or more metabolic biomarkers comprises at least a measured level of palmitoleic acid and a measured level of dihomo-γ-linolenic acid within the one or more samples of the user.
18. The non-transitory computer-readable storage device of claim 15 , wherein determining the change to implement in the nutritional plan comprises determining a change to at least one of a user-specific nutritional requirement to sustain a carbohydrate remission for the user, a weight loss goal specified for the user, or a communication provided to the user corresponding to the biological state for the user.
19. The non-transitory computer-readable storage device of claim 15 , wherein:
the one or more machine learning models comprise a neural network trained to identify an overall sentiment indicated by the one or more terms within the text messages; and
the emotional sentiment to the user is determined based on the overall sentiment identified by the neural network.