Related Experiment Video
Updated: Mar 29, 2026
10:33
Expedited Radiation Biodosimetry by Automated Dicentric Chromosome Identification ADCI and Dose Estimation
Published on: September 4, 2017
16.8K
Open-Source Radiation Exposure Extraction Engine (RE3) with Patient-Specific Outlier Detection
Samuel J Weisenthal1,2, Les Folio1, William Kovacs1
1National Institutes of Health, Clinical Center, Radiology and Imaging Sciences, Clinical Image Processing Service (CIPS), 10 Center Drive, Bethesda, MD, 20892-1182, USA.
Journal of Digital Imaging
|December 9, 2015
Summary
We developed RE3, an open-source engine that integrates with PACS to automatically monitor computed tomography (CT) radiation exposure. RE3 accurately calculates dose length product (DLP) and identifies outliers using multivariable regression models.
Area of Science:
- Medical Imaging
- Radiology
- Health Informatics
Background:
- Automated monitoring of radiation exposure in computed tomography (CT) is crucial for patient safety and quality assurance.
- Existing systems may lack seamless integration with Picture Archiving and Communication Systems (PACS) and real-time data processing capabilities.
Purpose of the Study:
- To present RE3 (Radiation Exposure Extraction Engine), an open-source, PACS-integrated tool for automated, study-specific CT radiation dose monitoring.
- To validate RE3's accuracy in calculating dose length product (DLP) and its effectiveness in detecting radiation exposure outliers.
Main Methods:
- Developed RE3 using open-source components for seamless PACS integration.
- Calculated DLP from DICOM headers and compared it with vendor data.
- Constructed multivariable regression models incorporating patient demographics, scan parameters (scan length, water-equivalent diameter, scanned body volume), and pediatric status for outlier detection.
Main Results:
- RE3 demonstrated high agreement (R² = 0.99) in DLP calculations compared to vendor dose pages.
- Multivariable regression models effectively predicted DLP, with scan length (R² = 0.45), water-equivalent diameter (R² = 0.70), and scanned body volume (R² = 0.80) being significant continuous predictors.
- Gender and pediatric status were significant categorical predictors of DLP.
Conclusions:
- RE3 provides a robust, open-source solution for automated CT radiation exposure monitoring integrated with PACS.
- The system accurately quantifies radiation dose and effectively identifies study-specific outliers using predictive modeling.
- RE3 facilitates enhanced radiation safety and quality control in CT imaging practices.
