Information Extraction From Medical Records

1
Information Extraction From Medical Records

•  
• by Alexander Barsky

2
Current Methodology

• Broad assessment of patient contained in
  beginning of chart with references to more
  specific areas. Specific divisions follow broad
  assessment. Records are listed in chronological
  order of activity.

3
Chart Example

• .

4
Problem 

•   
• A patient's medical chart is very detailed and
  very complex in nature. Any attempt to quickly
  locate specific information will be met with
  frustration.

5
Example

• .

6
Solution

• Create a system that properly extracts wanted
  information based on a predefined set of
  parameters.
•  
•  
• Example "Hormonal imbalance during puberty".
  Retrieve all references to hormonal imbalances
  but only between two specific time periods in
  medical chart.

7
Tool At our disposal

• JAPE  Java Annotation Patterns Engine.
•     Use pattern matching and semantic 
  extraction
•  
• GATE General Architecture for Text Engineering.
•     Use Information Extraction, document
  annotation, and 
•             XML output.
•  
• C      Visual C Winforms.
•     Use Medium for conversion between XML and
  .csv file                    formats.
•  
•          

8
Solution Methodology

• 1. Create corpus of documents in GATE.
• 2. Introduce rules for information extraction.
• 3. Annotate documents in corpus.
• 4. Output annotated documents in XML.
• 5. Strip file of unnecessary elements and convert
  to .csv.
•  

9
(No Transcript)
10
                        ANNIE

•         A-Nearly-New-Information-Extraction-System
  
•   
• -Tokeniser - splits sentence into simple tokens
• -Gazetter - identify entity names contained in
  lists
• -Sentence Splitter - splits text into sentences
  based on lists.
• -Parts of Speech Tagger - identifies text as
  different  POS.
• -Coreference Matcher- identifies relationships
  between previously defined entities.     

11
Success in Information Extraction is based on
integrating most if not all ANNIE components 

• -

12
        JAPE Key to Extraction

• -

13
                  JAPE Example

• -

14
XML Output

• -

15
Problem Too much unorganized information.
 Solution XLST to the rescue!!!

•  
• XLST - Extensible Stylesheet Language
  Transformations
•  
•  
• - Add specific rules to seperate needed from
  unnecessary information.

16
XLST Example

• -Find all the nodes within the ltLookupgt. Add
  string between the tags.

17
CSV File Type Comma  Seperated Value - Used to
present information in a tabular system. Useful
for analyzing large amount of data in an easy to
understand format. Most common program to use it
is Excel.  

• .

18
Potential Problem

• Regardless of how well all the ANNIE tools are
  utilized and how well the JAPE rules are defined,
  proper recall precentage won't ever be exact.

19
Solution Machine Learning

• Machine learning is our best chance to increase
  precision  of output results. Training a computer
  to recognize commonally used reporting
  phraseology will organize extraction better with
  more precise, concise outputs. Lucky for us, GATE
  include plugins to program machine learning.