EXTRACTING RELATED MEDICAL INFORMATION FROM DIFFERENT DATA SOURCES FOR AUTOMATED GENERATION OF PROGNOSIS, DIAGNOSIS, AND PREDISPOSITION INFORMATION IN CASE SUMMARY
According to embodiments of the present invention, methods, systems and computer readable media are provided for extracting related medical information from various sources to produce a medical evaluation. Genomic information provided from a patient tumor sample is analyzed to determine the presence of one or more mutations in the tumor sample. Hierarchical matching is performed to match the one or more mutations from the patient sample to curated structured data derived from literature. One or more of a prognosis, diagnosis, or predisposition is evaluated based on the matching, wherein the one or more mutations is predictive of a prognosis for a type of tumor, and is a diagnostic marker of a type of tumor. When a pathogenic mutation is detected for a predisposition, a report is generated regarding whether the pathogenic mutation is associated with hereditary cancer.
1 . A method for extracting related medical information from various sources to produce a medical evaluation comprising:
analyzing, via a processor, genomic information provided from a patient tumor sample to determine the presence of one or more mutations in the tumor sample;
performing hierarchical matching, via the processor, to match the one or more mutations from the patient sample to curated structured data derived from literature; and
evaluating one or more of a prognosis, diagnosis, or predisposition based on the matching, wherein the one or more mutations is predictive of a prognosis for a type of tumor, and is a diagnostic marker of a type of tumor;
wherein when a pathogenic mutation is detected for a predisposition, reporting whether the pathogenic mutation is associated with hereditary cancer.
2 . The method of claim 1 , further comprising:
providing a cancer-specific ontology, which organizes diseases associated with abnormal cellular proliferation in a plurality of levels from specific categories to broad categories; and
applying the hierarchical matching at a level of the ontology, and when a match is not found, traversing the cancer-specific ontology and reapplying the hierarchical matching until a match is found or until the hierarchical matching has been applied to the entire cancer-specific ontology.
3 . The method of claim 2 , wherein the hierarchical matching comprises a first type of matching pertaining to a level of the cancer-specific ontology and a second type of matching pertaining to identifying cancer-specific mutations.
4 . The method of claim 1 , wherein the one or more mutations is a driver mutation.
5 . The method of claim 2 , wherein the cancer-specific ontology comprises at least a level comprising specific gene mutations, a level comprising organ-based cancers, and a level comprising solid and blood-borne cancers.
6 . The method of claim 1 , wherein the genomic information from the patient is translated into proteomic information for biomarker analysis.
7 . The method of claim 1 , wherein hierarchical matching to determine a mutation may include one or more of matching a fusion biomarker, matching based on cancer-specific codon transition bias, matching based on cancer-specific splicing isoforms, or matching based on copy number or gene expression levels.
8 . A system for extracting related medical information from various sources to produce a medical evaluation, wherein the system comprises at least one processor configured to:
analyze genomic information provided from a patient tumor sample to determine the presence of one or more mutations in the tumor sample;
perform hierarchical matching to match the one or more mutations from the patient sample to curated structured data derived from literature; and
evaluate one or more of a prognosis, diagnosis, or predisposition based on the matching, wherein the one or more mutations is predictive of a prognosis for a type of tumor, and is a diagnostic marker of a type of tumor;
wherein when a pathogenic mutation is detected for a predisposition, reporting whether the pathogenic mutation is associated with hereditary cancer.
9 . The system of claim 8 , wherein the at least one processor is configured to:
provide a cancer-specific ontology, which organizes diseases associated with abnormal cellular proliferation in a plurality of levels from specific categories to broad categories; and
apply the hierarchical matching at a level of the ontology, and when a match is not found, traversing the cancer-specific ontology and reapplying the hierarchical matching until a match is found or until the hierarchical matching has been applied to the entire cancer-specific ontology.
10 . The system of claim 9 , wherein the hierarchical matching comprises a first type of matching pertaining to a level of the cancer-specific ontology and a second type of matching pertaining to identifying cancer-specific mutations.
11 . The system of claim 8 , wherein the one or more mutations is a driver mutation.
12 . The system of claim 9 , wherein the cancer-specific ontology comprises at least a level comprising specific gene mutations, a level comprising organ-based cancers, and a level comprising solid and blood-borne cancers.
13 . The system of claim 8 , wherein the genomic information from the patient is translated into proteomic information for biomarker analysis.
14 . The system of claim 8 , wherein hierarchical matching to determine a mutation may include one or more of matching a fusion biomarker, matching based on cancer-specific codon transition bias, matching based on cancer-specific splicing isoforms, or matching based on copy number or gene expression levels.
15 . A computer program product for extracting related medical information from various sources to produce a medical evaluation, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to:
analyze genomic information via a processor provided from a patient tumor sample to determine the presence of one or more mutations in the tumor sample;
perform hierarchical matching via the processor, to match the one or more mutations from the patient sample to curated structured data derived from literature; and
evaluate one or more of a prognosis, diagnosis, or predisposition, wherein based on the matching, wherein the one or more mutations is predictive of a prognosis for a type of tumor, and is a diagnostic marker of a type of tumor;
wherein when a pathogenic mutation is detected for a predisposition, reporting whether the pathogenic mutation is associated with hereditary cancer.
16 . The computer program product of claim 15 , wherein the instructions are further executable by the computer to cause the computer to:
provide a cancer-specific ontology, which organizes diseases associated with abnormal cellular proliferation in a plurality of levels from specific categories to broad categories; and
apply the hierarchical matching at a level of the ontology, and when a match is not found, traversing the cancer-specific ontology and reapplying the hierarchical matching until a match is found or until the hierarchical matching has been applied to the entire cancer-specific ontology.
17 . The computer program product of claim 16 , wherein the hierarchical matching comprises a first type of matching pertaining to a level of the cancer-specific ontology and a second type of matching pertaining to identifying cancer-specific mutations.
18 . The computer program product of claim 15 , wherein the one or more mutations is a driver mutation.
19 . The computer program product of claim 15 , wherein the genomic information from the patient may be translated into proteomic information for biomarker analysis.
20 . The computer program product of claim 16 , wherein hierarchical matching to determine a mutation may include one or more of matching a fusion biomarker, matching a sequence comprising cancer-specific codon transition bias, matching cancer-specific splicing isoforms, or matching based on copy number or gene expression levels.