System and method for phenotype vector manipulation of medical data
Cohort definition and selection system for a computer having a memory, a central processing unit and a display, the system including: a cohort definition module to configure the memory according to a phenotype vector. The phenotype vector includes a patient ID to uniquely associate the phenotype vector to a patient, a plurality of demographic dimension fields, each demographic dimension field to describe a respective demographic aspect of the patient, a calculated dimension field to describe a calculated information related to the patient, a plurality of phenotype-based dimension fields, each phenotype-based dimension field to indicate relevance of the respective phenotype-based dimension field to the patient, and a child phenotype vector to recursively define a phenotype-based dimension field, and a cohort selection module to select a set of phenotype vectors that are within a predetermined error from a cohort selection criteria.
1. A system to select a cohort, the system comprising:
a computer comprising a central processing unit (CPU), the computer configured to:
convert, patient data received from a plurality of databases into a phenotype vector structure, wherein the
phenotype vector structure is based on cohort definitions using a format and fields in a plurality of patient, medical, and claims databases;
prepare one or more phenotype vectors, by the CPU, using the phenotype vector structure, wherein the CPU combines the phenotype vector structure with time-based criteria, event-based criteria, and a unique patient key assigned for a respective patient for the observational tests wherein the time-based criteria is independent of the event-based criteria, wherein the event-based criteria includes a medical condition of the respective patient, wherein the unique patient key includes characteristics associated with the respective patient, wherein the phenotype vector structure includes:
a field to uniquely associate the phenotype vectors to the respective patient,
a plurality of demographic fields, each to describe a respective demographics aspect of the respective patient,
a plurality of phenotype-based fields to indicate relevance of a respective demographic field to the respective patient, and
a child phenotype vector to recursively define a phenotype-based field; and
select one or more of the prepared phenotype vectors associated with an error function to form the cohort.
2. The system of claim 1 , wherein the one or more phenotype vectors identify patients for a cohort study.
3. The system of claim 1 , wherein the phenotype-based fields comprise binary indicators.
4. The system of claim 1 , wherein the error function is determined using a Euclidean metric.
5. The system of claim 1 , wherein the error function is determined using respective weighting functions for at least one of the demographic dimension fields, the calculated field, and the phenotype-based fields.
6. The system of claim 1 , wherein the error function gives greater weight to characteristics relevant to the selection of the cohort.
7. The system of claim 1 , wherein at least one of the demographic fields comprises a statistic that represents a patient population as a whole.
8. The system of claim 1 , further comprising time relationships between the fields of the phenotype vectors.
9. The system of claim 1 , further comprising a time constraint on at least one field of the phenotype vectors.
10. The system of claim 1 , wherein a cohort definitions are determined based on a training data set of patient vectors.
11. The system of claim 1 , wherein the patients, medical and claims databases are expandable.
12. The system of claim 1 , wherein an additional phenotype vector structure is configured to be combined with the time-based criteria, event-based criteria, and the unique patient key when the one or more selected phenotype vectors are not within the predetermined error.
13. A method to select a cohort, the method comprising steps of:
configuring, on a computer, using a Central Processing Unit (CPU), a phenotype vector structure by converting received patient data to the phenotype vector structure from a plurality of databases, wherein the phenotype vector structure is based on cohort definitions using a common format and fields in a plurality of patient, medical, and claims databases, wherein the common format is used by a plurality of scientific groups and does not require knowledge of structures of the plurality of patient, medical, and claims databases;
preparing one or more phenotype vectors by the CPU, using the phenotype vector structure, wherein the CPU combines the phenotype vector structure with time-based criteria, event-based criteria, and the unique patient key for a respective patient for the observational studies, wherein the time-based criteria is independent of the event-based criteria, wherein the event-based criteria includes medical condition of the respective patient, wherein the unique patient key includes characteristics associated with the respective patient, wherein the configured phenotype vector structure includes:
a field to uniquely associate the phenotype vectors to the respective patient,
a plurality of demographic fields, each to describe a respective demographics aspect of the respective patient,
a plurality of phenotype-based fields to indicate relevance of a respective phenotype-based dimension to the respective patient, and
a child phenotype vector to recursively define a phenotype-based field; and
selecting, on the computer, one or more of the prepared phenotype vectors to form the cohort of members from the patient population in the plurality of patient, medical, and claims databases, wherein the two or more phenotype vectors are associated with an error distance to the configured cohort definitions.
14. The method of claim 13 , wherein the one or more selected phenotype vectors identify patients for a cohort study.
15. The method of claim 13 , wherein the phenotype-based fields comprise binary indicators.
16. The method of claim 13 , wherein the error distance is determined using a Euclidean metric.
17. The method of claim 13 , wherein the error distance is determined using respective weighting functions for at least one of the demographic dimension fields, the calculated field, and the phenotype-based dimension fields.
18. The method of claim 13 , wherein the error distance gives greater weight to characteristics relevant to selecting the cohort.
19. The method of claim 13 , wherein at least one of the demographic fields comprises a statistic that represents a patient population as a whole.
20. The method of claim 13 , further comprising time relationships between the fields of the phenotype vectors.
21. The method of claim 13 , further comprising a time constraint on at least one field of the phenotype vector.