Systems, methods, and apparatus for processing documents to identify structures

• US 9,031,977 B2
• Filed: 04/02/2013
• Issued: 05/12/2015
• Est. Priority Date: 05/03/2010
• Status: Active Grant

• Current Assignee: Revvity Signals Software Incorporated (Revvity, Inc.)
• Original Assignee: PerkinElmer Informatics, Inc. (Revvity, Inc.)
• Inventors: Smith, Robin Y., Ballard, William B., Flicker, Scott G., Greenhow, Sean G.
• Primary Examiner(s): Arjomandi, Noosha
• Application Number:
  US13/855,342
• Publication Number:
  US 20130218878A1
• Time in Patent Office: 770 Days
• Field of Search: 707/769, 707/791, 707/803
• US Class Current: 707/769
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06V 30/422 Technical drawings; Geographical maps
• CPC Class Codes: G16C 20/40 Searching chemical structures or physicochemical data
• CPC Class Codes: G16C 20/80 Data visualisation
• CPC Class Codes: G16C 20/90 Programming languages; Computing architectures; Database systems; Data warehousing

Abstract

In various embodiments, multiple heterogeneous documents are processed to identify structures, such as chemical structures, contained therein, including non-embedded structures. Also described is a graphical user interface that permits a user to search for a structure or substructure within a set of electronic documents, then displays the matching structures as well as the actual pages of the documents on which the matching structures are found. Display of the actual pages allows the user to verify the matches and provides helpful context for the user.

Claims

1. A system for automatically identifying chemical structures found in one or more electronic files, the system comprising:

a memory having a set of instructions stored thereon; and

a processor, wherein the instructions, when executed by the processor, cause the processor to;

(a) identify one or more candidate chemical structures in an electronic file, wherein

the electronic file comprises at least one non-embedded image of a chemical structure, and

identifying each candidate chemical structure of the one or more candidate chemical structures comprises identifying one or more graphical features common to chemical structures;

(b) for each candidate chemical structure of the one or more candidate chemical structures, derive a respective chemical structure object with an associated set of properties, wherein

one or more properties of the set of properties is derived from at least a portion of the one or more graphical features common to chemical structures,

a first property of the set of properties is a number of carbons, wherein the number of carbons is derived from the one or more graphical features common to chemical structures, and

a second property of the set of properties comprises one of the following;

(A) number of hetero atoms, (B) number of bonds, (C) number of bonds of a selected bond order, (D) number of rings, and (E) formula weight;

(c) for each chemical structure object, apply one or more filters to at least one property of the associated set of properties, wherein

the one or more filters includes a filter configured to eliminate chemical structure objects having a value of the first property of the set of properties less than a predetermined number of carbons; and

(d) provide, for storage in a searchable electronic compendium of identified chemical structure objects, chemical structure objects not eliminated by the one or more filters.

2. The system of claim 1, wherein the instructions cause the processor to, prior to identifying the one or more candidate chemical structures in the electronic file:

retrieve, from a document storage facility, the electronic file;

extract, from the electronic file, one or more embedded objects; and

identify, from the one or more embedded objects, a one or more candidate embedded objects,

wherein each embedded object of the one or more candidate embedded objects comprises a MIME type identified as being related to chemistry.

3. The system of claim 2, wherein the instructions cause the processor to:

identify that at least a first embedded object of the one or more candidate embedded objects lacks an associated molecule connection table; and

add the first embedded object to a data stream for processing according to steps (a) through (d).

4. The system of claim 1, wherein the one or more filters are configured to identify expected content within the electronic file.

5. The system of claim 4, wherein the expected content comprises at least one of a chemical structure, a chemical substructure, and a chemical reaction.

6. The system of claim 1, wherein the one or more filters includes a filter configured to eliminate chemical structure objects having fewer than a threshold number of at least one of the following:

double bonds, triple bonds, and wedge bonds.

7. The system of claim 1, wherein the one or more filters includes a filter configured to eliminate chemical structure objects having fewer than a threshold number of at least one of the following:

heteroatoms, tetra-substituted carbon atoms, and tri-substituted carbon atoms.

8. A method for automated identification of graphical representations of chemical structures comprising:

identifying, by a processor of a computing device, within an electronic file comprising one or more non-embedded images of chemical structures, at least one candidate chemical structure, wherein the at least one candidate chemical structure is identified based at least in part on graphical image processing, wherein the graphical image processing comprises identifying one or more graphical features common to chemical structures;

for each of the at least one candidate chemical structure, deriving, by the processor, a respective chemical structure object, wherein

the respective chemical structure object comprises a set of properties, wherein

one or more properties of the set of properties is derived from at least a portion of the one or more graphical features common to chemical structures,

a first property of the set of properties is a number of carbons, and

a second property of the set of properties comprises one of the following;

(A) number of hetero atoms, (B) number of bonds, (C) number of bonds of a selected bond order, (D) number of rings, and (E) formula weight;

for each chemical structure object, determining a respective confidence score based on two or more properties of the set of properties associated with the chemical structure object, wherein

each chemical structure of the at least one confirmed chemical structure is identified based at least in part upon respective confidence score associated with the respective chemical structure object derived therefrom; and

identifying, by the processor, at least one confirmed chemical structure, wherein each chemical structure of the at least one confirmed chemical structure is identified based at least in part upon the respective confidence score.

9. The method of claim 8, comprising, prior to identifying the at least one candidate chemical structure, generating, from the electronic file, at least one data stream, wherein identifying the at least one candidate chemical structure comprises identifying, from the at least one data stream, the at least one candidate chemical structure.

10. The method of claim 9, comprising generating a second data stream, and identifying, from the second data stream, at least a first candidate chemical structure of the at least one candidate chemical structure.

11. The method of claim 10, wherein a first data stream of the at least one data stream comprises a document-level data stream, and the second data stream comprises a page-level data stream.

12. The method of claim 10, wherein determining the respective confidence score for the first candidate chemical structure comprises determining the respective confidence score based in part upon a redundancy of identification of the first candidate chemical structure.

13. The method of claim 10 comprising, prior to deriving the respective chemical structure object, eliminating at least one redundant candidate chemical structure from the at least one candidate chemical structure.

14. A non-transitory computer readable medium having instructions stored thereon, wherein the instructions, when executed by a processor, cause the processor to:

identify, within an electronic file comprising one or more non-embedded images of chemical structures, at least one candidate chemical structure, wherein the at least one candidate chemical structure is identified based at least in part on graphical image processing, wherein

the graphical image processing comprises identifying one or more graphical features common to chemical structures;

for each of the at least one candidate chemical structure, derive a set of properties, wherein

one or more properties of the set of properties is derived from at least a portion of the one or more graphical features common to chemical structures, and

a first property of the set of properties comprises one of the following;

(A) number of hetero atoms, (B) number of bonds, (C) number of bonds of a selected bond order, (D) number of rings, (E) number of carbons, and (F) formula weight;

for each chemical structure object, determine a respective confidence factor value based on two or more properties of the set of properties associated with the chemical structure object; and

identify at least one confirmed chemical structure, wherein each chemical structure of the at least one confirmed chemical structure is identified based at least in part upon the respective confidence factor value.

15. The computer readable medium of claim 14, wherein, for each candidate chemical structure, the instructions cause the processor to apply one or more filters, wherein

each filter of the one or more filters is configured to compare one or more properties of the set of properties of a given candidate chemical structure to one or more predetermined values;

and

the at least one confirmed chemical structure is identified based at least in part upon the candidate chemical structure avoiding elimination by the one or more filters.

16. The computer readable medium of claim 14, wherein the respective set of properties associated with each confirmed chemical structure of the at least one confirmed chemical structure is collected within a respective chemical structure object.

17. The computer readable medium of claim 16, wherein the instructions cause the processor to provide, for storage in a document database, the electronic file and the respective chemical structure object of each confirmed chemical structure of the at least one confirmed chemical structure.

18. The computer readable medium of claim 17, wherein the instructions cause the processor to associate, with each confirmed chemical structure of the at least one confirmed chemical structure, an image file containing a visual representation of the confirmed chemical structure.

19. The computer readable medium of claim 18, wherein the image file comprises a thumbnail image of a portion of the electronic file containing the respective confirmed chemical structure.

20. The computer readable medium of claim 16, wherein the instructions cause the processor to add, to the respective chemical structure object of the at least one confirmed chemical structure, inventory information comprising one or more of a registration number, a serial number, a lot number, and a batch code associated with the respective chemical structure object.

21. The computer readable medium of claim 20, wherein the instructions cause the processor to, prior to adding the inventory information, access, via a network, at least one system comprising the inventory information.