Automatic detection of mental health condition and patient classification using machine learning
Methods and systems are provided for detecting a mental health condition. Structured and unstructured information is analyzed using natural language processing to extract information including clinical data values and medical concepts pertaining to a user. Reference medical information is evaluated using natural language processing to correlate medical data with mental health conditions. A classification for a mental health condition of the user is determined using a machine learning model and based on the extracted information and correlations, wherein the extracted information includes blood analysis for the user. The user is assigned to a segment of users based on the extracted information. A treatment for the mental health condition of the user is indicated based on the classification and the assigned segment of users.
1. A method of detecting a mental health condition comprising:
analyzing, via a processor, structured and unstructured information using natural language processing to extract information including clinical data values and medical concepts pertaining to a user, wherein the extracted information includes blood analysis for the user including biomarkers derived from genomic, proteomic, and/or immunoassay studies;
evaluating, via the processor, reference medical information using natural language processing to correlate medical data with mental health conditions;
determining, via the processor, a classification for a mental health condition of the user using a machine learning model and based on the extracted information and correlations, wherein determining the classification includes:
performing sentiment analysis to determine values for sentiment of textual terms of the extracted information associated with the mental health conditions;
assigning values to the biomarkers of the blood analysis associated with the textual terms for indicating at least one of the mental health conditions; and
summing the values for sentiment of the textual terms with the values of the associated biomarkers to produce a score used for determining the classification;
assigning, via the processor, the user to a segment of users based on the extracted information;
indicating, via the processor, a treatment for the mental health condition of the user based on the classification and the assigned segment of users; and
continually training the machine learning model, via the processor, based on new user data and verification of classifications as the new user data is analyzed and the classifications are verified.
2. The method of claim 1 , wherein indicating the treatment further comprises:
indicating, via the processor, one or more additional blood tests to perform based on the classification.
3. The method of claim 1 , wherein the machine learning model includes a support vector machine.
4. The method of claim 1 , wherein the unstructured information includes one or more from a group of: notes of a clinician, a patient journal, medical literature, and medical standards.
5. The method of claim 1 , wherein the structured information includes one or more from a group of: blood work profiles, patient case files, electronic health records, and patient attributes.
6. The method of claim 1 , wherein determining the classification for the mental health condition of the user further comprises:
determining, via the processor, a risk of injury to the user due to the mental health condition of the user; and
providing, via the processor, an alert of the mental health condition of the user in response to the risk satisfying a threshold.
7. The method of claim 1 , wherein the classification is based on medical mental disorder standards.
8. An apparatus for detecting a mental health condition, the apparatus comprising:
one or more processors;
one or more computer readable storage media;
program instructions stored on the one or more computer readable storage media for execution by at least one of the one or more processors, the program instructions comprising instructions to:
analyze structured and unstructured information using natural language processing to extract information including clinical data values and medical concepts pertaining to a user, wherein the extracted information includes blood analysis for the user including biomarkers derived from genomic, proteomic, and/or immunoassay studies;
evaluate reference medical information using natural language processing to correlate medical data with mental health conditions;
determine a classification for a mental health condition of the user using a machine learning model and based on the extracted information and correlations, wherein determining the classification includes:
performing sentiment analysis to determine values for sentiment of textual terms of the extracted information associated with the mental health conditions;
assigning values to the biomarkers of the blood analysis associated with the textual terms for indicating at least one of the mental health conditions; and
summing the values for sentiment of the textual terms with the values of the associated biomarkers to produce a score used for determining the classification;
assign the user to a segment of users based on the extracted information;
indicate a treatment for the mental health condition of the user based on the classification and the assigned segment of users; and
continually train the machine learning model based on new user data and verification of classifications as the new user data is analyzed and the classifications are verified.
9. The apparatus of claim 8 , wherein the program instructions further comprise instructions to:
indicate one or more additional blood tests to perform based on the classification.
10. The apparatus of claim 8 , wherein indicating the treatment further comprises:
indicating a score card including one or more mental health conditions that the user is at-risk of developing, a listing of users subdivided according to risk level, treatment plans, and outcomes.
11. The apparatus of claim 8 , wherein the unstructured information includes one or more from a group of: notes of a clinician, a patient journal, medical literature, and medical standards.
12. The apparatus of claim 8 , wherein the structured information includes one or more from a group of: blood work profiles, patient case files, electronic health records, and patient attributes.
13. The apparatus of claim 8 , wherein the program instructions further comprise instructions to:
determine a risk of injury to the user due to the mental health condition of the user; and
provide an alert of the mental health condition of the user in response to the risk satisfying a threshold.
14. The apparatus of claim 8 , wherein the classification is based on medical mental disorder standards.
15. A computer program product for detecting a mental health condition, the computer program product comprising one or more computer readable storage media collectively having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to:
analyze structured and unstructured information using natural language processing to extract information including clinical data values and medical concepts pertaining to a user, wherein the extracted information includes blood analysis for the user including biomarkers derived from genomic, proteomic, and/or immunoassay studies;
evaluate reference medical information using natural language processing to correlate medical data with mental health conditions;
determine a classification for a mental health condition of the user using a machine learning model and based on the extracted information and correlations, wherein determining the classification includes:
performing sentiment analysis to determine values for sentiment of textual terms of the extracted information associated with the mental health conditions;
assigning values to the biomarkers of the blood analysis associated with the textual terms for indicating at least one of the mental health conditions; and
summing the values for sentiment of the textual terms with the values of the associated biomarkers to produce a score used for determining the classification;
assign the user to a segment of users based on the extracted information;
indicate a treatment for the mental health condition of the user based on the classification and the assigned segment of users; and
continually train the machine learning model based on new user data and verification of classifications as the new user data is analyzed and the classifications are verified.
16. The computer program product of claim 15 , wherein the program instructions further cause the computer to indicate one or more additional blood tests to perform based on the classification.
17. The computer program product of claim 15 , wherein the machine learning model includes a support vector machine.
18. The computer program product of claim 15 , wherein the unstructured information includes one or more from a group of: notes of a clinician, a patient journal, medical literature, and medical standards.
19. The computer program product of claim 15 , wherein the structured information includes one or more from a group of: blood work profiles, patient case files, electronic health records, and patient attributes.
20. The computer program product of claim 15 , wherein the program instructions further cause the computer to:
determine a risk of injury to the user due to the mental health condition of the user; and
provide an alert of the mental health condition of the user in response to the risk satisfying a threshold.