Mobile supplementation, extraction, and analysis of health records
A system, method, and mobile device application are configured to capture, with a mobile device, a document such as a next generation sequencing (NGS) report that includes NGS medical information about a genetically sequenced patient. The method includes receiving, from a mobile device, an image of a medical document comprising NGS medical information of the patient, extracting a first region from the image, extracting NGS medical information of the patient from the first region into a structured dataset, the extracted NGS medical information including at least one RNA expression, correlating a portion of the extracted NGS medical information that includes the at least one RNA expression with summarized medical information from a cohort of patients similar to the patient, and generating, for display on the mobile device, a clinical decision support report comprising the summarized medical information.
1. A method performed by a clinical laboratory, comprising:
a. receiving at least one specimen of, and a document including medical history information about, a cancer patient;
b. performing next generation sequencing on the at least one specimen to generate molecular information comprising at least five million reads;
c. extracting the medical history information of the patient from the document using an optical character recognition technique employed by a processor of a computer system in a first stage of a stateless pipeline, the optical character recognition technique selected from among a plurality of possible techniques available within the first stage of the stateless pipeline;
d. applying at least one natural language processing technique by the processor to the extracted medical history information to identify one or more categories of medical history information and values therefor in a second stage of the stateless pipeline;
e. arranging the identified one or more categories of medical history information of the patient and values therefor into structured medical history information;
f. storing the structured medical history information in a database system;
g. uploading the molecular information to a cloud service for processing by a variant calling bioinformatics pipeline;
h. generating one or more variant calls resulting from processing of the molecular information by the bioinformatics pipeline;
i. determining, for the cancer patient, a variant classification reflective of a degree of pathogenicity of each of the one or more variant calls;
j. storing in the database system the one or more variant calls and each variant call's corresponding pathogenic interpretation;
k. preparing a treatment plan comprising one or more of a therapy identification or a therapy efficacy, the preparing comprising evaluating, by the processor, the structured medical history information, the one or more variant calls, and the corresponding pathogenic interpretation stored in the database system in order to develop a customized treatment plan for the cancer patient; and
l. preparing a report including the one or more variant calls, the corresponding pathogenic interpretation, and the treatment plan.
2. The method of claim 1 , where the therapy identification is a therapy name.
3. The method of claim 1 , where the therapy identification is a therapy class.
4. The method of claim 1 , where the therapy identification is a clinical trial.
5. The method of claim 3 , where the therapy class is a radiation therapy.
6. The method of claim 3 , where the therapy class is a chemotherapy.
7. The method of claim 3 , where the therapy class is an immunotherapy.
8. The method of claim 3 , where the therapy class is a PARP inhibitor.
9. The method of claim 1 , where the therapy identification is a course of therapy.
10. The method of claim 1 , where the therapy efficacy is an indication of therapy resistance.
11. The method of claim 1 , where the therapy efficacy is an indication of therapy response.
12. The method of claim 1 , where the treatment plan is prepared at least in part using a therapy matching engine.
13. The method of claim 1 , where the treatment plan ranks therapies based on the one or more variant calls.
14. The method of claim 1 , where the treatment plan ranks therapies based on the pathogenic interpretation corresponding to the one or more variant calls.
15. The method of claim 1 , where the treatment plan ranks therapy identification based on whether the patient's cancer type matches a cancer type associated with an evidence level associated with the therapy identification.
16. The method of claim 1 , where the treatment plan is prepared using a therapy matching engine.
17. The method of claim 16 , where the therapy matching engine automatically determines that the one or more variant calls are associated with an evidence level associated with the therapy identification.
18. The method of claim 1 , where the report is prepared as a Portable Document Format document.
19. The method of claim 1 , further comprising displaying the report on a mobile device.
20. The method of claim 1 , where the at least one specimen comprises a FFPE slide and a normal specimen.
21. The method of claim 1 , where the at least one specimen comprises a normal sample and a tumor sample, where the processing of the molecular information by the bioinformatics pipeline comprises subtracting variants detected in the normal sample from variants detected in the tumor sample.
22. The method of claim 1 , where the molecular information generated by next generation sequencing comprises ribonucleic acid sequences.
23. The method of claim 1 , where the molecular information generated by next generation sequencing comprises ribonucleic acid sequences indicating fusion mRNAs.
24. The method of claim 1 , further comprising determining whether the generated molecular information meets a quality threshold level of a number of read pairs of ribonucleic acid sequences.
25. The method of claim 1 , where the variant calls resulting from the processing of the molecular information by the bioinformatics pipeline comprise single nucleotide polymorphisms, indels, copy number variants, and gene rearrangements.
26. The method of claim 1 , where the step of arranging the medical history information of the patient into structured medical history information comprises using an optical character recognition microservice to consume clinical records.
27. The method of claim 1 , further comprising preparing the treatment plan based on results of applying a cancer treatment to an organoid developed from a tumor specimen of the patient.
28. The method of claim 1 , further comprising preparing the treatment plan based on one or more features extracted from a radiology image associated with the patient.
29. The method of claim 1 , further comprising preparing the treatment plan based on one or more features extracted from a slide image associated with the patient.
30. The method of claim 1 , further comprising:
m. storing information from the treatment plan and the report in structured form in a searchable database comprising at least ten thousand records, each record comprising information stored in structured form from another treatment plan and another report associated with a patient with cancer; and
n. providing an interface for a recipient of the report to query the searchable database.