Tashfeen Ekram MD, Presenter: Nothing to Disclose

Safwan Halabi MD, Abstract Co-Author: Nothing to Disclose

James Ciarelli MS, Abstract Co-Author: Nothing to Disclose

Michael James Flynn PhD, Abstract Co-Author: Nothing to Disclose

Increased use of IT solutions has led to the increased generation of unstructured data. The ability to extract useful information from this data to aid in clinical and business decision making becomes increasingly important. LEXIMER, a natural language processing software, was developed to quantify the rate of positivity of interpretations and determine the recommendation rate for further imaging.

NLP tools offer myriad benefits in business analytics and determination of imaging outcomes. Clinical validation of these solutions is critical to ensure reliability and consistency of the unstructured data analysis. Developers should consider defining and delineating varying degrees of positivity and recommendations being analyzed.

Increased use of IT solutions has led to the increased generation of unstructured data. The ability to extract useful information from this data to aid in clinical and business decision making becomes increasingly important. LEXIMER, a natural language processing software, was developed to quantify the rate of positivity of interpretations and determine the recommendation rate for further imaging.

Ekram, T, Halabi, S, Ciarelli, J, Flynn, M, Towards Validation of Natural Language Processing Tools for Business Analytics.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12034809.html