Pragya Ahuja Dang MBBS, Presenter: Nothing to Disclose

Mannudeep Karanvirsingh Kalra MD, Abstract Co-Author: Nothing to Disclose

Markus Stout, Abstract Co-Author: Nothing to Disclose

Paul Lemay, Abstract Co-Author: Nothing to Disclose

Thomas Schultz, Abstract Co-Author: Nothing to Disclose

Keith J. Dreyer DO, PhD, Abstract Co-Author: Clinical Trials Reader, Perceptics, LLC Medical Advisor, Agfa-Gevaert Group Medical Advisor, Fuji Photo Film Co, Ltd Medical Advisor, General Electric Company Medical Advisor, McKesson Corporation Medical Advisor, AuntMinnie.com Medical Advisor, AMICAS, Inc Medical Advisor, Dynamic Imaging, LLC Medical Advisor, Ascom Holding AG Medical Advisor, Bracco Group Medical Advisor, Merge Healthcare Medical Advisor, Emageon Inc Medical Advisor, RCG HealthCare Consulting Consultant, The Elizabeth Wende Breast Clinic Consultant, Valley Radiology Medical Associates, Inc Medical Advisor, Information Systems Consultants, Inc Medical Advisor, Siemens AG Medical Advisor, Barco nv Medical Advisor, Hue AS Medical Advisor, Planar Systems, Inc Medical Advisor, Vital Images, Inc Medical Advisor, Commissure, Inc Medical Advisor, TeraRecon, Inc Medical Advisor, Mercury Computer Systems, Inc Medical Advisor, IBM Corporation Medical Advisor, Hewlett-Packard Company Medical Advisor, EMC Corp Medical Advisor, Phase Forward Incorporated Medical Advisor, Winchester Systems, Inc Medical Advisor, Dell Inc Medical Advisor, Eastman Kodak Company Medical Advisor, Amirsys, Inc Medical Advisor, Reed Elsevier Editorial Board, Diagnostic Imaging Editorial Board, AuntMinnie.com Editorial Board, Imaging Economics Author, Springer Science+Business Media Deutschland GmbH

The purpose of our study was to compare between neuroradiologist and musculoskeletal radiologists, the rates of Relevant Radiology Findings (RRF) and recommendations in spine MR reports. In addition, we compared RRF and recommendations in spine MR reports read by different radiologists within the two radiology sub-specialties.

Leximer (a natural language processing engine) was used with HyperCube (an OnLine Analytic Processing engine) to analyze spine MRI reports obtained from a radiology report database of 1995 to 2006, reported by neuroradiologists (n=17), reports (n=10817), and musculoskeletal radiologists (n=10) reports (n= 2348) for presence of RRF and recommendations for further action. Reports were said to have RRF if they contained a significant finding and were classified to have recommendations if further imaging test was requested or suggested. The rates of RRF and recommendations for further action between the two groups of radiologists were compared. Subsequently, the RRF and recommendations rates for individual radiologists within the two groups were recorded and the variation of rates between the different radiologists was analyzed.

RRF rates for spine MR for neuroradiologists and musculoskeletal radiologists were 83.3% (9010/10817) and 86.2% (2023/2348), respectively. Recommendations for subsequent action in these reports were 12% (1301/10817) for neuroradiologists and 13% (297/2348) for musculoskeletal radiologists (p>0.05). However, when comparison for RRF and recommendations in spine MRI reports was made between different radiologists within the two groups, the rates varied significantly (p<0.05). Neuroradiologists showed a variation in RRF rates between 77% and 100%, and in the recommendation rates between 1.9%-44.4%. For the musculoskeletal radiologists, RRF rates varied between 75.8 and 93.9% and recommendation rates varied between 2.9 and 18.5%.

There are substantial differences in the recommendations and findings rates amongst radiologists of the same subspecialty.

The NLP allows assessment of patterns of findings and recommendations between different radiology sub-specialties and radiologists.

Dang, P, Kalra, M, Stout, M, Lemay, P, Schultz, T, Dreyer, K, Comparison of Findings and Recommendations in Spine MR Reports of Neuroradiologists and Musculoskeletal Radiologists.  Radiological Society of North America 2007 Scientific Assembly and Annual Meeting, November 25 - November 30, 2007 ,Chicago IL. http://archive.rsna.org/2007/5011790.html