Automated systems and methods for textual extraction of relevant data elements from an electronic clinical document

• US 10,789,461 B1
• Filed: 01/15/2020
• Issued: 09/29/2020
• Est. Priority Date: 10/24/2019
• Status: Active Grant

• Current Assignee: Innovaccer, Inc.
• Original Assignee: Innovaccer, Inc.
• Inventors: Agrawal, Vibhuti, Bhabesh, Gourav Sanjukta
• Primary Examiner(s): Colucci, Michael
• Application Number:
  US16/743,175
• Time in Patent Office: 258 Days
• Field of Search: 704 9, 704235, 704500, 707779, 707723, 707722, 707741, 707625, 707610, 705 4, 705 29, 715234, 715780, 700 96, 713189
• CPC Class Codes: G06F 40/14 Tree-structured documents parsing G06F40/205; validation G06F40/226
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/221 Parsing markup language streams streaming G06F40/149
• CPC Class Codes: G06F 40/279 Recognition of textual entities
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G06V 30/262 using context analysis, e.g. lexical, syntactic or semantic context
• CPC Class Codes: G06V 30/414 Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text
• CPC Class Codes: G06V 30/416 Extracting the logical structure, e.g. chapters, sections or page numbers; Identifying elements of the document, e.g. authors
• CPC Class Codes: G16H 10/60 for patient-specific data, e.g. for electronic patient records
• CPC Class Codes: G16H 15/00 ICT specially adapted for medical reports, e.g. generation or transmission thereof

Abstract

A system and method for extracting relevant data elements from a file for conversion to a tabular format includes a computing device receiving an XML format file having a loop with nested blocks. Each of the blocks has at least one data element. Features are extracted from each data element. These extracted features are processed using a machine learning algorithm to estimate a column header value for the data elements relative to a data schema. With the data element classified, a configuration file is generated to map the column header value to the data elements of the XML file. The configuration file is used to extract the data elements from the XML file to a tabular format. In the healthcare industry, the system and method may be used to extract relevant health information from a clinical document for conversion to a tabular format.

Claims

1. A computer-implemented method for extracting relevant data elements from an electronic file for conversion to tabular format, the method comprising:

receiving, in a computing device, an Extensible Markup Language (XML) format file, the XML file having at least one loop with nested blocks, wherein each of the nested blocks has at least one data element, the at least one data element having an unstructured or semi-structured format;

extracting features from the data elements;

processing, with a processor of the computing device, the extracted features using a machine learning algorithm to estimate a column header value for the data elements relative to a data schema;

classifying, by the processor, the data elements from the XML file using the extracted features;

generating, by the processor, a configuration file which maps the column header value to the data elements of the XML file;

parsing the XML file using the configuration file to extract unstructured or semi-structured alphanumeric data values of the data elements from the XML file and convert the data elements to a structured tabular format; and

ingesting the structured tabular format of the data elements into a data analytics processing system.

2. The method of claim 1, wherein the file having the XML format further comprises a Consolidated Clinical Document Architecture (C-CDA) file.

3. The method of claim 1, further comprising extracting features from at least one attribute describing the data element.

4. The method of claim 1, wherein the extracted features from the data elements further comprise at least one of:

a length of a string, a number of capital letters, a number of lowercase letters, a number of spaces, a number of digits, a number of hyphens, a number of symbols, an average length of each word in the string, a hash value of the string, a number of grammatical separators, or a symmetry of the string.

5. The method of claim 4, wherein the hash value of the string and the symmetry of the string are calculated using an American Standard Code for Information Interchange (ASCII) value of each character within the string.

6. The method of claim 1, wherein the configuration file has a key value pair, where the key is the column header value name from a data-lake schema and the value is an XPath to the data element of the XML file.

7. The method of claim 1, further comprising outputting the configuration file to a separate computing device prior to parsing the XML file using the configuration file.

8. The method of claim 1, wherein parsing XML file using the configuration file to convert the data elements from the XML file to the tabular format is implemented simultaneously in parallel computing processors.

9. A computer-implemented system for the extraction of relevant data elements from an electronic file for conversion to a tabular format, the system comprising:

a computing device receiving an Extensible Markup Language (XML) format file, the XML file having at least one loop with nested blocks, wherein each of the nested blocks has at least one data element, the at least one data element having an unstructured or semi-structured format;

a processor of the computing device executing instructions for;

extracting features from the data elements of an XML file;

processing the extracted features using a machine learning algorithm to estimate a column header value for the data elements relative to a data schema;

segregating the data elements from the XML file using the extracted features; and

generating a configuration file which maps the column header value to the data elements of the XML file;

parsing the XML file using the configuration file to extract unstructured or semi-structured alphanumeric data values of the data elements from the XML file and convert the data elements to a structured tabular format; and

ingesting the structured tabular format of the data elements into a data analytics processing system.

10. The system of claim 9, wherein the file having the XML format further comprises a Consolidated Clinical Document Architecture (C-CDA) file.

11. The system of claim 9, further comprising extracting features from at least one attribute describing the data element.

12. The system of claim 9, wherein the extracted features from the data elements further comprise at least one of:

a length of a string, a number of capital letters, a number of lowercase letters, a number of spaces, a number of digits, a number of hyphens, a number of symbols, an average length of each word in the string, a hash value of the string, a number of grammatical separators, or a symmetry of the string.

13. The system of claim 12, wherein the hash value of the string and the symmetry of the string are calculated using an American Standard Code for Information Interchange (ASCII) value of each character within the string.

14. The system of claim 9, wherein the configuration file has a key value pair, where the key is the column header value name from a data-lake schema and the value is an XPath to the data element of the XML file.

15. The system of claim 9, further comprising outputting the configuration file to at least one separate computing device prior to parsing the XML file using the configuration file.

16. The system of claim 9, wherein parsing the XML file using the configuration file to convert the data elements from the XML file to the tabular format is implemented simultaneously in parallel computing processors.

17. A computer-implemented system for extracting relevant health information from a clinical document in an Extensible Markup Language (XML) format for conversion to a tabular format, the system comprising:

a first computing device receiving the clinical document, wherein the clinical document has a plurality of loops with a plurality of attributes describing data elements, the plurality of data elements having an unstructured or semi-structured format, wherein the plurality of data elements correspond to health information within the clinical document;

a processor of the first computing device executing instructions for;

extracting features from at least one of the plurality of attributes or the data elements of a clinical document using textual analysis;

processing the extracted features using a machine learning algorithm to estimate a column header value for the data elements relative to a predefined data schema;

segregating the data elements from the clinical document using the extracted features; and

generating a configuration file which maps the column header value to the data elements of the clinical document using a key-value pair, where a key of the key-value pair provides a column header value name from a data-lake schema and a value from the key-value pair provides an XPath of the clinical document;

at least one second computing device in communication with the first computing device, wherein, at the at least one second computing device, the configuration file is used to parse the data elements from the clinical document to a tabular format by extracting unstructured or semi-structured alphanumeric data values of the data elements from the clinical document and converting the data elements to a structured tabular format; and

a data analytics processing system ingesting the structured tabular format of the data elements.

18. The system of claim 17, wherein the at least one second computing device further comprises a plurality of second computing devices operating simultaneously, wherein using the plurality of second computing devices in parallel, the configuration file is used to extract the data elements from the clinical document to the tabular format.

19. The system of claim 17, wherein the extracted features from the data elements further comprise at least one of:

a length of a string, a number of capital letters, a number of lowercase letters, a number of spaces, a number of digits, a number of hyphens, a number of symbols, an average length of each word in the string, a hash value of the string, a number of grammatical separators, or a symmetry of the string.

20. The system of claim 19, wherein the hash value of the string and the symmetry of the string are calculated using an American Standard Code for Information Interchange (ASCII) value of each character within the string.