What is MedLEE™?

MedLEE™ is a Natural Language Processing (NLP) application that extracts medical codes from typed or dictated free text narratives and converts the data into computerized clinical information.

Why is MedLEE™ called the "Gold Standard of Natural Language Processing"?

• MedLEE™ was developed at Columbia University during 20 years of research and field testing.
• MedLEE™ automatically extracts and codes medical reports fast, inexpensively and accurately.
• It covers the biomedical domain and utilizes national and international standards (UMLS, SNOMED and ICD-9).
• It is the most widely used natural language processor in the medical research community.
• In numerous evaluations carried out by independent users MedLEE™ has shown to be as accurate as
• physicians in interpreting narrative medical records.
• MedLEE™ has received US and international patents.

Track Record

• MedLEE™ has been sucessfully used by government agencies (DoD, NCI), evaluated and recommended by the VA, and successfully deployed at major medical centers in Pittsburgh, Chicago, Cincinnati, New York and others.
• In a biosurveillance project for the City of New York, MedLEE™ detected patient records identifying individuals at high risk for having active tuberculosis and reduced the rate of missed tuberculosis findings by almost one half at New York Presbyterian Hospital.
• MedLEE™ was used to code 889,921 radiographic reports on 251,186 patients. Its processor's sensitivity and specificity were comparable to those previously reported for expert coders.

MedLEE™ Benefits Sumamry

• Proven accuracy across all major medical domains.
• Most cost effective medical natural language processor for billing, reimbursement and decision support.
• MedLEE™'s multiple knowledge component structure allows users to select output formats (HL7, SML) to adjust pre- and post-processor parameters and to easily add or modify lexica.
• Real-time interface to speech recognition.