Systems and methods for diagnosing a health condition based on patient time series data
Disclosed systems, methods, and computer readable media can diagnose a health condition based on patient time series data. For example, a method for diagnosing a health condition based on patient time series data includes identifying a training set of health records comprising a first set of patient time series data, training a neural network using the training set of health records, and executing the trained neural network model to diagnose a health condition based on a second set of patient time series data. In further examples, the first set of patient time series data and the second set of patient time series data can each comprise electrocardiogram data and the health condition can comprise pulmonary hypertension.
1. A method for diagnosing a health condition based on patient time series data, wherein the method comprises:
receiving, using one or more hardware processors, patient time series data, wherein the patient time series data comprises an electrocardiogram (ECG) waveform;
identifying, using the one or more hardware processors, a training set of health records, wherein:
the training set of health records comprises health records of patients who have been diagnosed with a health condition of interest and training data including ECG waveforms of the patients correlated to the health condition of interest; and
identifying the training set of health records comprises identifying one or more cohorts of the patients;
training, using the one or more hardware processors, a plurality of neural network models for each cohort of the patients using the training set of health records;
selecting, using the one or more hardware processors, one or more highest performing models from the plurality of trained neural network models;
executing, using the one or more hardware processors, the one or more highest performing models as a function of the patient time series data, wherein executing the one or more highest performing models comprises:
preprocessing the time series data, wherein preprocessing the time series data comprises:
extracting one or more discrete metrics as a function of the time series data, wherein the one or more discrete metrics comprises an interval of an ECG waveform; and
predicting, using the one or more hardware processors, a health condition as a function of the patient time series data, the interval of the ECG waveform, and the one or more highest performing models.
2. The method of claim 1 , wherein the interval of the ECG waveform comprises a QT interval.
3. The method of claim 2 , wherein the training set of health records further comprises ethnicity of patients.
4. The method of claim 3 , wherein the training set of health records further comprises QT intervals correlated with the patients who have been diagnosed with the health condition of interest.
5. The method of claim 4 , further comprising recommending, using the one or more hardware processors, an intervention as a function of the predicted heart condition.
6. The method of claim 1 , further comprising:
obtaining, using the one or more hardware processors, the training set of health records for each cohort of the one or more cohorts of the patients from a corpus of health records using a search query.
7. The method of claim 1 , wherein the training set of health records comprises exemplary pre-emptive time series data.
8. The method of claim 1 , further comprising:
receiving, using the one or more hardware processors, patient information as an input data for the one or more highest performing models, wherein the patient information comprises ethnicity of the patient.
9. The method of claim 8 , further comprising selecting, suing the one or more hardware processors, the one or more highest performing models from the plurality of trained neural network models as a function of the patient information.
10. The method of claim 1 , further comprising:
outputting, using the one or more hardware processors and the one or more highest performing models, a numerical score of a risk of having the health condition.
11. A system for diagnosing a health condition based on patient time series data, wherein the system comprises:
a non-transitory memory; and
one or more hardware processors configured to read instructions from the non-transitory that, when executed, cause the one or more hardware processors to perform operations comprising:
receive patient time series data, wherein the patient time series data comprises an electrocardiogram (ECG) waveform;
identify a training set of health records, wherein:
the training set of health records comprises health records of patients who have been diagnosed with a health condition of interest and training data including ECG waveforms of the patients correlated to the health condition of interest; and
identifying the training set of health records comprises identifying one or more cohorts of the patients;
train a plurality of neural network models for each cohort of the patients using the training set of health records;
select one or more highest performing models from the plurality of trained neural network models;
execute the one or more highest performing models as a function of the patient time series data, wherein executing the one or more highest performing models comprises:
preprocessing the time series data, wherein preprocessing the time series data comprises:
extracting one or more discrete metrics as a function of the time series data, wherein the one or more discrete metrics comprises an interval of an ECG waveform; and
predict a health condition as a function of the patient time series data, the interval o the ECG wave form, and the one or more highest performing models.
12. The system of claim 11 , wherein the interval of the ECG waveform comprises a QT interval.
13. The system of claim 12 , wherein the training set of health records further comprises ethnicity of patients.
14. The system of claim 13 , wherein the training set of health records further comprises QT intervals correlated with the patients who have been diagnosed with the health condition of interest.
15. The system of claim 14 , wherein the one or more hardware processors is further configured to recommend an intervention as a function of the predicted heart condition.
16. The system of claim 11 , wherein the one or more hardware processors is further configured to obtain the training set of health records for each cohort of the one or more cohorts of the patients from a corpus of health records using a search query.
17. The system of claim 11 , wherein the one or more hardware processors is further configured to output a numerical score of a risk of having the health condition using the one or more highest performing models.
18. The system of claim 11 , wherein the one or more hardware processors is further configured to convert the time series data to a spectrogram representation.
19. The system of claim 18 , wherein the one or more hardware processors is further configured to select the one or more highest performing models from the plurality of trained neural network models as a function of the patient information.
20. The system of claim 11 , wherein the one or more processors is further configured to output, using the one or more highest performing models, a numerical score of a risk of having the health condition.