LABL tools and resources

Name Description Language Contact Last Mod Location Limitations
Patient database Semistructured patient database with structured and parsed data XML Stephen Johnson March 2003 lucid: /localprojects/pdb deidentified data for internal department use only
MEDLINE Database of citations to medical literature XML Stephen Johnson March 2003 lucid: /localprojects/medline Available only for research purposes by Columbia staff
Frequency analyzer for XML files Counts occurrences of elements in XML document Java or Perl? Hongfang Liu last modified? location? limitations on use?
Medical discourse analyzer Analysis of time, reference and other connections between events in MedLEE output Java Stephen Johnson not yet available lucid: /localprojects/tempo internal Columbia research
Lifelog surgical information on patients: vital signs, operative events, procedure, medications XML Greg Whalen last modified? location? confidential patient information requiring IRB permission
Op note parser shallow parser for operative notes Language? Greg Whalen last modification? location? limitations for use?
Textbook summarizer summarization of medical textbooks Language? Greg Whalen not yet available location? limitations for use?
Enhanced SAS XML LIBNAME Engine Tutorial for successfully importing an XML document to SAS SAS Jen-Hsiang Chuang 2001 http://www.sas.com/rnd/base/topics/sxle82/exp82/mapxmlexp.html needs SAS 8.2 program installed
UMLS lexical tools norm: normalizes a phrase (removes stop words, punctuation, etc, and alphabetizes resulting set of words. luiNorm: norm based on UMLS LUIs. wordInd: breaks up strings into a set of lowercased words Java Michael Cantor or Yves Lussier early 2002 /localprojects/ktlab/lvg2002/ no limitations
MMTx Implementation of UMLS Metamap, which analyzes phrases and maps them to UMLS terms Java Michael Cantor or Yves Lussier last modified? /localprojects/ktlab/mmtx no limitations
UMLS_2002 MRCON: contains cui, lui, sui and text of UMLS terms. MRXNS_ENG: contains lui, sui, text of normalized UMLS terms SQL Michael Cantor or Yves Lussier August 2002 /localprojects/ktlab/mysql/connect_mysql ktlab group
SNOMED tables of SNOMED concepts (normed), and concepts related to UMLS, QMR, OMIM SQL Michael Cantor or Yves Lussier August 2002 /localprojects/ktlab/mysql/connect_mysql ktlab group
HGDB tables from HGDB, a subset of OMIM published in Nature (1408 terms) SQL Michael Cantor or Yves Lussier September 2002 /localprojects/ktlab/mysql/connect_mysql ktlab group
OMIM tables from OMIM's genemap and morbid map, processed with various techniques sql Michael Cantor or Yves Lussier September 2002 /localprojects/ktlab/mysql/connect_mysql ktlab group
Visualizer for MedLEE data HTML views of MedLEE data generated with XSL and Protégé Java? Michael Krauthammer last modified? location? limitations for use?
Linearizer for MedLEE data creates in-line XML markup format for MedLEE data Java? Michael Krauthammer last modified? location? limitations for use?
GENIES information extraction from medical literature Prolog? Carol Friedman last modified? location? limitations for use?
Frequency analyzer for XML files Counts occurrences of elements in XML document Pure Java for small data and the combination of JAVA and Perl for large collection Hongfang Liu and Carol Friedman last modified September 2002 personal computer limitations on use (Available for research purpose for staff at Columbia, seek permissions from Carol)
The UMLS Normalizer Normalize UMLS including recognize and expand abbreviations (tag) Perl (requires the UMLS and Berkley DB installed) Hongfang Liu and Carol Friedman last modified May 2002 cat? Research purpose for staff at Columbia
Normalized report section headings A database for normalized report section headers (227 headers with a total of 586 header variants chosen using frequency) Perl Hashtable Hongfang Liu and Carol Friedman May 2001 leo limitations on use (available)
Supervised Machine Learning Package Naive Bayes Learning, Decision List Learning etc Perl Hongfang Liu and Carol Friedman May 2002 io limitations on use (available )
PhraseMap Map phrase to the UMLS Concept under following four modes: Exact, Specialist-Normalized, Stemmed, Synonym replacement (Last three modes include ignoring word order and stop words.) Perl (and Berkley DB) Hongfang Liu and Carol Friedman March 2002 Cat Not well-documented, but it is easy to use
PhraseGrouper Group textual variants with the same sense Perl (and Berkley DB) Hongfang Liu and Carol Friedman March 2002 Cat Not well-documented, but it is easy to use
MedLEE Linearization Tool Merges the split MedLEE XML representation (structured and tt section) into a linearized form where the text and the semantic tags are joined together. Java Michael Krauthammer July 2002 not currently available Some MedLEE documents won't parse yet, some linearization bugs.
MedLEE view generator Generates a 'physician view' (problem list) and a 'data mining view' out of a MedLEE parsed discharge summary Java Michael Krauthammer March 2001 not currently available Current version works with user interface. Distribution version may be working on a command line basis, but this is not implemented yet.