Related Experiment Videos
PathBot: A Radiology-Pathology Correlation Dashboard
Linda C Kelahan1, Amit D Kalaria2, Ross W Filice2,3
1Department of Radiology, MedStar Georgetown University Hospital, 3800 Reservoir Road NW, Washington, DC, 20007, USA. Lkelahan2@gmail.com.
Journal of Digital Imaging
|April 5, 2017
Summary
Radiologists can now automatically correlate pathology reports with their dictations using a new dashboard. This tool enhances self-education and peer review by linking diagnostic imaging with pathology findings.
Area of Science:
- Medical Imaging and Diagnostics
- Computational Pathology
Background:
- Pathology is the gold standard in diagnostic medicine.
- Radiology-pathology correlation is crucial but often hindered by practical constraints like time and electronic medical record complexity.
Purpose of the Study:
- To develop an automated system for correlating radiology and pathology reports.
- To create a user-friendly dashboard for presenting these correlations to radiologists.
Main Methods:
- Utilized the RadLex ontology and NCBO Annotator to identify and map anatomic concepts.
- Developed an algorithm to match pathology reports to corresponding radiology dictations for diagnostic imaging and image-guided procedures.
- Presented matched reports via a web-based dashboard.
Main Results:
- The developed algorithm demonstrated high specificity in detecting radiology-pathology matches.
- Sensitivity was lower than anticipated, potentially due to limitations in the RadLex ontology and mapping algorithms.
Conclusions:
- Automated radiology-pathology correlation via a dashboard can encourage pathology follow-up for self-education and peer review.
- This tool can facilitate the creation of educational materials like teaching files, lectures, and publications.
- Integrating pathology findings enhances the educational value of diagnostic images.