Making Better Sense of Medical Records
November 1, 2012
New research out of CSAIL’s Clinical Decision Making Group should make it easier and more efficient for computers to parse electronic medical records. In a new paper to be presented at the American Medical Informatics Association’s (AMIA) annual symposium next week, researchers will explain a new system they have developed for disambiguating the meaning of words used in clinical notes written by doctors and nurses.
At present, it is difficult for computer systems to analyze electronic medical records as many medical terms can have multiple meanings. The new system developed by Professor Peter Szolovits, CSAIL postdoc Anna Rumshisky, graduate student Rachel Chasin, and research affiliate Özlem Uzuner, is 75 percent accurate in disambiguating words with two senses, a marked improvement over previous methods.
For more information on the Clinical Decision Making Group’s work with medical records, please visit: http://web.mit.edu/newsoffice/2012/digital-medical-records-offer-insights-1031.html.