CPC G06F 16/2456 (2019.01) [G06F 16/248 (2019.01); G06F 16/2465 (2019.01); G06F 16/258 (2019.01); G16H 10/60 (2018.01)] | 15 Claims |
1. A method for selecting a treatment protocol for a patient, the method being implemented by a computer system comprising one or more processors executing computer program instructions that, when executed, perform the method, the method comprising:
receiving, at an integration server configured with a data integration specification and adapted to service requests from a client device, a query from the client device requesting analysis of treatment options for at least one disease;
mapping the query with respect to a data integration schema and integrity constraints, wherein the data integration schema identifies sources of information usable for answering the query;
storing, in a memory of the computer system, data source specific schema;
retrieving a first instance of biomedical data from at least one data source based on a first data source specific schema of the stored data source specific schema, wherein the first instance of biomedical data comprises patients treated for the at least one disease;
mapping the first instance of biomedical data from at least one data source to the data integration schema;
generating a first data source object by attaching the retrieved first instance of biomedical data from at least one data source to a first distributed data object;
conducting a first computation within the first data source object and generating a third data source object based on the first computation;
retrieving a second instance of biomedical data from the at least one data source based on a second data source specific schema of the stored data source specific schema, wherein the second instance of biomedical data comprises genetic profiles for the patients identified in the first instance;
mapping the second instance of biomedical data from the at least one data source to the data integration schema;
generating a second data source object by attaching the retrieved second instance of biomedical data from the at least one data source to a second distributed data object;
conducting a second computation within the second data source object and generating a fourth data source object based on the second computation;
combining the first data source object, the second data source object, the third data source object, and the fourth data source object to form a multidimensional data pool;
processing an operation corresponding to the query within the multidimensional data pool to generate a list of successful treatment protocols for the at least one disease; and
returning the list to the client device.
|