Automatic creation of clinical study reports

• US 9,754,083 B2
• Filed: 10/24/2012
• Issued: 09/05/2017
• Est. Priority Date: 10/25/2011
• Status: Active Grant

• Current Assignee: Certara, L.P. (Certara, Inc.)
• Original Assignee: ClinGenuity, LLC (Synchrogenix Information Strategies, Inc.)
• Inventors: Kleeman, Keith M., Kowitz, Mickey W.
• Primary Examiner(s): Kanaan, Maroun
• Application Number:
  US14/353,513
• Publication Number:
  US 20140316822A1
• Time in Patent Office: 1,777 Days
• CPC Class Codes: G06F 16/2228 Indexing structures
• CPC Class Codes: G06F 16/2457 with adaptation to user needs
• CPC Class Codes: G06F 40/143 Markup, e.g. Standard Generalized Markup Language [SGML] or Document Type Definition [DTD]
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G16H 10/20 for electronic clinical trials or questionnaires
• CPC Class Codes: G16H 15/00 ICT specially adapted for medical reports, e.g. generation or transmission thereof

Abstract

An apparatus (102, 104) and method for generating a formatted clinical study report (206) by analyzing submitted clinical study data (150) based on semantics and maximum entropy analysis at a server (104). Sections of the formatted clinical study report are identified in the submitted data (111) using semantics analysis and maximum entropy analysis (204). The identified sections are formatted and output as the formatted clinical study report (206). The formatted clinical study report (206) may be reviewed and edited after generation by a user interfacing with the server (104).

Claims

1. A method of automatically generating a clinical study report from a plurality of data files, the method comprising:

receiving clinical study data including a plurality of data files at a server;

identifying to said server, a clinical study report, said clinical study report comprising a plurality of elements, each of said elements having an associated phrase and format; and

a processor of said server;

converting each of said plurality of data files to a common analysis format;

locating a plurality of sections within each of said plurality of data files from heading levels within said plurality of data files;

storing relationship information from said heading levels for said plurality of sections;

for each of said elements in said clinical study report;

indexing for each of said plurality of sections, occurrences of said phrase associated with said element within that section;

identifying any section with a positive index as a possible section for the clinical study report;

ranking the identified possible sections based on said indexing to provide a probability for each identified possible section to correspond to said element in said clinical study report; and

comparing each probability of each identified possible section to determine a best fit section from among the identified possible sections for said element; and

generating the formatted clinical study report by reproducing said best fit section for each said element as said element in said clinical study report in said format associated with said element.

2. The method of claim 1, further comprising displaying said formatted clinical study report at a user device in communication with said server.

3. The method of claim 2 further comprising allowing said user device to download said formatted clinical study report from said server only after said user device displayed said formatted clinical study report.

4. The method of claim 1 wherein said comparing uses maximum entropy analysis.

5. The method of claim 1 wherein:

said phrase comprises a keyword;

said identifying identifies any section including said keyword;

said ranking determines a number of occurrences of said keyword in each said identified section; and

said best fit section comprises said identified section with the most occurrences of said keyword.

6. The method of claim 1, wherein said server uses a placeholder to identify each element that does not yet have an associated best fit section.

7. The method of claim 6 wherein the indexing uses semantic similarity to score and identify phrases for building an overall confidence to which each section may have meaning to said placeholder.

8. The method of claim 6 wherein said comparing uses a maximum entropy technique to compare all sections having some meaning to a placeholder to see which section of all identified has the best meaning.

9. The method of claim 1 wherein said common analysis format is HTML.

10. The method of claim 1 wherein each said phrase can only be associated with one said element.

11. The method of claim 1 wherein said indexing is based at least in part on semantics of words included in text of said section.

12. The method of claim 1 wherein said relationship information indicates parent and child relationships between said sections so said indexing can be performed on each section independently or by treating multiple sections having a parent and child relationship with each other as a single section.