Supriya Gupta MBBS, Presenter: Nothing to Disclose

Thomas J. Schultz BS, Abstract Co-Author: Nothing to Disclose

Steven A. Graham MBA, Abstract Co-Author: Nothing to Disclose

Arun Krishnaraj MD, MPH, Abstract Co-Author: Research support, General Electric Company

Keith J. Dreyer DO, PhD, Abstract Co-Author: Medical Advisor, General Electric Company Medical Advisor, Siemens AG Medical Advisor, Nuance Communications, Inc Medical Advisor, Carestream Health, Inc Medical Advisor, Vital Images, Inc Medical Advisor, Amirsys, Inc Medical Advisor, Life Image Inc Medical Advisor, McKesson Corporation Medical Advisor, Merge Healthcare

Interactive data mining tools enable us to reap the benefits of investments in technology in health care systems. These can be extrapolated to enhance utilization of clinicians’ time and efficiency by reducing the time to browse through past radiology reports.

Medicine is an information-intensive platform with patient charts and histories, research literature and evidence. Tapping into the vast potential of associative data and text mining enable us to integrate patient information in a shell making it easy to browse. We propose the creation and implementation of graphic representation of chronological radiological history of a patient equipping the physician with the capacity to locate a finding at any given time.

The proposed software- RadAtlas- induces structuring of various pathology and radiology findings in the reports along with the exam dates for a given patient. RadAtlas communicates with Radiology Information System, PACS, Radiology Order Entry and employs RadLex terminology to locate radiology findings in text reports. Efficient queries (SQL/MDX) are designed against aggregated data and web services (.net, java script, flash) are used to display results in human graphical form. Each organ in RadAtlas display has a separate icon to view report findings represented chronologically such that user can scroll through an organ to detect changes in radiological findings over a period of time in few seconds. Evaluation would be performed for accurate description and correlation of radiology findings with their representation in RadAtlas and the time taken by a physician to browse through RadAtlas would be recorded.

RadAtlas would be deployed as a part of electronic medical record such that the physician can efficiently browse through entire radiological history of a patient with click of a button. The dawn of pay-per performance strategies in medical settings has spawned awareness regarding efficient time utilization among physicians. Advanced data mining tools such as RadAtlas can be utilized to recreate a graphic version of text reports such that physicians and radiologists can minimize delays incurred in browsing through past radiology reports of patients.

Gupta, S, Schultz, T, Graham, S, Krishnaraj, A, Dreyer, K, Data Designing Using Advanced Informatics: Sketching the Radiological History of a Patient through Graphic Representation of Report Findings.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11005906.html