Mobile supplementation, extraction, and analysis of health records

Patent metadata

Field	Value
Publication number	US-10902952-B2
Application number	US-201916531005-A
Country	US
Kind code	B2
Filing date	Aug 2, 2019
Priority date	Oct 17, 2018
Publication date	Jan 26, 2021
Grant date	Jan 26, 2021

Abstract

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. At least some of the information is extracted from the document using an entity linking engine, and the extracted information is provided into a structured data repository where it is accessible to provide information regarding the patient specifically as well as collectively as part of a cohort of patients with similar genetic variants, medical histories, or other commonalities. In one aspect, the document is matched to a template model, and the document is processed using one or more masks segregating the template model, and therefore the document, into a series of distinct subregions.

First claim

The invention claimed is:

1. A method for providing a physician with clinical decision support information about a patient whose DNA has been sequenced with a next generation sequencing (NGS) system, comprising:
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 gene identifier;
correlating a portion of the extracted NGS medical information that includes the at least one gene identifier 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.

2. The method of claim 1 , wherein extracting the first region from the image is based at least in part from a template of a predefined model matched to the image, wherein the template encodes one or more rules.

3. The method of claim 2 , wherein the one or more rules comprise applying one or more masks to identify the first region.

4. The method of claim 2 , wherein the one or more rules are directed to document extraction features and comprise: a first rule for regular expressions; a second rule for natural language processing; and a third rule for column-row pairing to text identified in the one or more regions.

5. The method of claim 2 , wherein the one or more rules are directed to document format features and comprise one or more rules for validating the extracted NGS medical information based at least in part from a second region extracted from the medical document, wherein the second region is distinct from the first region.

6. The method of claim 1 , wherein the summarized medical information comprises treatment incidence information.

7. The method of claim 1 , wherein the summarized medical information comprises treatment response information.

8. The method of claim 1 , wherein correlating a portion of the extracted NGS medical information with summarized medical information from a cohort of patients similar to the patient comprises:
selecting a gene identifier in the extracted NGS medical information;
querying a structured data repository of medical information for patients with a gene identifier that is the same as the selected gene identifier;
assembling the cohort of patients similar to the patient from the results of the query.

9. The method of claim 1 , wherein the summarized medical information comprises clinical decision support information not found in the medical document.

10. The method of claim 7 , wherein the treatment response information comprises a treatment and a clinical response level of the cohort of patients prescribed the treatment.

11. The method of claim 1 , wherein the summarized medical information comprises a treatment name, the method further comprising:
providing the treatment name to an insurer of the patient; and
receiving a pre-authorization for the treatment from the insurer.

12. The method of claim 1 , wherein extracting NGS medical information from the first region further comprises:
identifying a first type of NGS medical information from a template;
identifying a first extraction rule for extracting the first type of NGS medical information from the template;
extracting the first type of NGS medical information from the first region based at least in part on the first extraction rule; and
classifying the extracted first type of NGS medical information.

13. The method of claim 12 , wherein extracting the NGS medical information based at least in part on the first extraction rule further comprises:
determining a first concept from the NGS medical information; and
identifying a match to the first concept in a first list of concepts.

14. The method of claim 13 , further comprising:
identifying a first classification from the template;
generating a comparison of a classification of the first concept to the first classification; and
storing the match to the first concept as the extracted first type of NGS medical information based on a positive generated comparison.

15. The method of claim 14 , wherein classifying the extracted first type of NGS medical information further comprises:
identifying an additional NGS medical information in the first region;
extracting the additional NGS medical information; and
generating structured data from the extracted first type of NGS medical information and the extracted additional NGS medical information.

16. The method of claim 14 , wherein identifying a first classification from the template further comprises selecting the first classification from an enumerated list of classifications within the template.

17. The method of claim 12 , wherein extracting the first type of NGS medical information from the first region further comprises:
extracting the first type of NGS medical information from a second region of the medical document;
validating the extracted first type of NGS medical information from the second region with a verified set of NGS medical information previously extracted from the second region; and
validating the extracted first type of NGS medical information from the first region with the validated extracted first type of NGS medical information from the second region.

18. The method of claim 12 , wherein extracting the first type of NGS medical information from the first region further comprises:
determining a first concept from the NGS medical information;
identifying a plurality of matches to the first concept in a first list of concepts;
identifying a first classification from the template;
generating, for each match of the plurality of matches, a comparison of a classification of the first concept to the first classification; and
storing a first match to the first concept, of the plurality of matches, as the extracted first type of NGS medical information based on a positive generated comparison.

19. The method of claim 12 , wherein extracting the first type of NGS medical information from the first region further comprises:
determining a first concept from the NGS medical information;
identifying a plurality of matches to the first concept in a first list of concepts;
identifying a first classification from the template;
generating, for each match of the plurality of matches, a classification of the match;
generating, for each match of the plurality of matches, an estimated confidence reliability rating of the match; and
storing the match with the highest estimated confidence, of the plurality of matches, as the extracted first type of NGS medical information.

20. The method of claim 12 , wherein extracting the NGS medical information of the patient from the first region further comprises:
determining a first concept from the NGS medical information;
identifying a match to the first concept in a first list of concepts, wherein the first list of concepts is not a preferred authority;
referencing the first concept to an entity in a database of related concepts; and
identifying a match to a second concept in a second list of concepts, the second list of concepts not directly linked to the first list of concepts except by a relationship between the entity and the first concept and the entity and the second concept, wherein the second list of concepts is a preferred authority.

21. Th

Assignees

• Tempus Labs
• Tempus Labs Inc

Inventors

• Lucas Michael Ryan
• Ozeran Jonathan
• Taylor Jason L
• Fernandes Louis E
• Neems Daniel
• Lane Hunter
• Lefkofsky Eric

Classifications

• G06V30/19013

  Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching (specially adapted for image segmentation G06T7/10; specially adapted for the analysis of motion G06T7/20; specially adapted for image alignment G06T7/30; specially adapted for the calculation of depth from stereo images G06T7/50; specially adapted for position determination G06T7/70) · CPC title

• G06Q40/08 Primary

  Insurance · CPC title

• G16H50/20 Primary

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• G06F18/217

  Validation; Performance evaluation; Active pattern learning techniques · CPC title

• G06F18/2431

  Multiple classes · CPC title