Machine intelligence-based prioritization of non-emergent procedures and visits

1. A computer-implemented method for using machine-learning to generate a medical recommendation, comprising:

receiving, by one or more processors, historical patient data and one or more historical risk values associated with a plurality of patients;

modifying, by the one or more processors, the historical patient data by:

computing one or more characteristics missing from the historical patient data, and

adding the computed one or more characteristics to the historical patient data to obtain augmented historical patient data;

training a machine-learning model by:

inputting the augmented historical patient data into the machine-learning model to obtain one or more estimated risk values, and

updating the machine-learning model based on a comparison of the one or more estimated risk values to the one or more historical risk values;

receiving, by the one or more processors, first patient data associated with a first patient;

modifying, by the one or more processors, the retrieved first patient data by:

computing a first set of characteristics missing from the retrieved first patient data, and

adding the computed first set of characteristics to the retrieved first patient data to obtain augmented first patient data;

receiving, by the one or more processors, second patient data associated with a second patient;

modifying, by the one or more processors, the retrieved second patient data by:

computing a second set of characteristics missing form the retrieved second patient data, and

adding the computed second set of characteristics to the retrieved second patient data to obtain augmented second patient data;

inputting the augmented first patient data into the trained machine-learning model to determine a first set of one or more risk values for the first patient;

inputting the augmented second patient data into the trained machine-learning model to determine a second set of one or more risk values for the second patient;

comparing the first set of one or more risk values and the second set of one or more risk values to determine a priority for distributing care to the first patient and the second patient; and

in accordance with the determination that the first patient has priority, generating a medical recommendation based on the priority, wherein the medical recommendation comprises an identification of a treatment for the first patient.

2. The method of claim 1 , wherein a risk value of the first set of one or more risk values corresponds to: a probability of an outcome within a time frame, a number of events of the outcome within a time frame, a time to a first occurrence of the outcome, a time-series of outcome events as a function of time, or the time-series of outcome probabilities as a function of time.

3. The method of claim 1 , further comprising automatically scheduling an appointment for the first patient.

4. The method of claim 1 , further comprising sending a notification to a provider, wherein the notification recommends distributing care to the first patient based on the priority.

5. The method of claim 1 , further comprising displaying the medical recommendation to a provider.

6. The method of claim 1 , further comprising sending a notification to the first patient, wherein the notification recommends next steps for distributing care to the first patient.

7. The method of claim 1 , wherein generating the medical recommendation comprises identifying a first treatment for the first patient.

8. The method of claim 1 , further comprising sending the second set of one or more risk values to a provider, wherein generating the medical recommendation comprises identifying a second treatment for the second patient based on the priority.

9. The method of claim 1 , wherein electronic health records of the first patient are updated to include the priority and the first set of one or more risk values.

10. The method of claim 1 , further comprising selecting the trained machine-learning model from a plurality of trained machine-learning models based on one or more events that have occurred to the first patient and the second patient.

11. The method of claim 1 , further comprising determining an updated first set of one or more risk values and an updated second set of one or more risk values at regular intervals, and determining an updated priority based on the updated first set of one or more risk values and the updated second set of one or more risk values.

12. The method of claim 1 , wherein the first patient data comprises patient characteristics, care service records, or a combination thereof.

13. The method of claim 12 , wherein the patient characteristics comprise demographic information, acute and chronic health history, history and physical exam findings, medical history, surgical history, prescription history, family history, occupational history, social history, review of systems, social determinants of health, or a combination thereof.

14. The method of claim 12 , wherein the care service records comprise a medical professional claim, a medical facility claim, a pharmacy claim, electronic health records, self-reported outcomes, digital health records, social media records, data from laboratory test, or a combination thereof.

15. The method of claim 1 , wherein the trained machine-learning model is configured to receive a patient-specific dataset and output a third set of one or more risk values indicative of a health of a corresponding patient in the absence of care.

16. The method of claim 1 , wherein the historical patient data comprises a plurality of patient-specific datasets for the plurality of patients, each patient-specific dataset comprising one or more of: patient characteristics at diagnosis, patient outcomes in the absence of non-emergent care, and delays between the diagnosis and receiving the non-emergent care.

17. The method of claim 1 , wherein the historical patient data comprises one or more fiducial time points.

18. The method claim 1 , further comprising modifying, by the one or more processors, the received patient data to obtain respective augmented patient data, and wherein inputting the patient data into the first trained machine-learning model comprises inputting the augmented patient data into the first trained machine-learning model.

19. The method of claim 1 , wherein computing the first set of characteristics missing from the retrieved first patient data comprises computing one or more summary statistics.

20. The method of claim 1 , wherein computing the first set of characteristics missing from the retrieved first patient data comprises determining whether a predefined threshold is met.

21. The method of claim 1 , wherein computing the first set of characteristics missing from the retrieved first patient data comprises computing one or more values derived from a model.

22. The method of claim 21 , wherein the model comprises one or more of: a dimensionality reduction model, a clustering model, a graphical model, a classifier, a prediction model, and a regression model.

23. The method of claim 1 , wherein computing the first set of characteristics missing from the retrieved first patient data comprises retrieving one or more characteristics from a knowledge graph.

24. A system for generating a medical recommendation, comprising:

one or more processors;

a memory; and

one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for:

receiving, by one or more processors, historical patient data and one or more historical risk values associated with a plurality of patients;

modifying, by the one or more processors, the historical patient data by:

computing one or more characteristics missing from the historical patient data, and

adding the computed one or more characteristics to the historical patient data to obtain augmented historical patient data;

training a machine-learning model by:

inputting the augmented historical patient data into the machine-learning model to obtain one or more estimated risk values, and

updating the machine-learning model based on a comparison of the one or more estimated risk values to the one or more historical risk values;

receiving, by the one or more processors, first patient data associated with a first patient;

modifying, by the one or more processors, the retrieved first patient data by:

computing a first set of characteristics missing from the retrieved first patient data, and

adding the computed first set of characteristics to the retrieved first patient data to obtain augmented first patient data;

receiving, by the one or more processors, second patient data associated with a second patient;

modifying, by the one or more processors, the retrieved second patient data by:

computing a second set of characteristics missing form the retrieved second patient data, and

adding the computed second set of characteristics to the retrieved second patient data to obtain augmented second patient data;

inputting the augmented first patient data into the trained machine-learning model to determine a first set of one or more risk values for the first patient;

inputting the augmented second patient data into the trained machine-learning model to determine a second set of one or more risk values for the second patient;

comparing the first set of one or more risk values and the second set of one or more risk values to determine a priority for distributing care to the first patient and the second patient; and

in accordance with the determination that the first patient has priority, generating a medical recommendation based on the priority, wherein the medical recommendation comprises an identification of a treatment for the first patient.

25. A non-transitory computer-readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by one or more processors of one or more electronic devices, cause the electronic devices to:

receive, by one or more processors, historical patient data and one or more historical risk values associated with a plurality of patients;

modify, by the one or more processors, the historical patient data by:

computing one or more characteristics missing from the historical patient data, and

adding the computed one or more characteristics to the historical patient data to obtain augmented historical patient data;

train a machine-learning model by:

inputting the augmented historical patient data into the machine-learning model to obtain one or more estimated risk values, and

updating the machine-learning model based on a comparison of the one or more estimated risk values to the one or more historical risk values;

receive, by the one or more processors, first patient data associated with a first patient;

modify, by the one or more processors, the retrieved first patient data by:

computing a first set of characteristics missing from the retrieved first patient data, and

adding the computed first set of characteristics to the retrieved first patient data to obtain augmented first patient data;

receive, by the one or more processors, second patient data associated with a second patient;

modify, by the one or more processors, the retrieved second patient data by:

computing a second set of characteristics missing form the retrieved second patient data, and

adding the computed second set of characteristics to the retrieved second patient data to obtain augmented second patient data;

input the augmented first patient data into the trained machine-learning model to determine a first set of one or more risk values for the first patient;

input the augmented second patient data into the trained machine-learning model to determine a second set of one or more risk values for the second patient;

compare the first set of one or more risk values and the second set of one or more risk values to determine a priority for distributing care to the first patient and the second patient; and

in accordance with the determination that the first patient has priority, generate a medical recommendation based on the priority, wherein the medical recommendation comprises an identification of a treatment for the first patient.