Jurgen Jacobs MSc, Presenter: Nothing to Disclose

Guy Jacques Marchal MD, PhD, Abstract Co-Author: Nothing to Disclose

Hilde Bosmans PhD, Abstract Co-Author: Research partner, MEVIC SA Research partner, CIMI Research partner, IBBT, Belgium

To propose automated methods to evaluate patient doses in pediatric CT.

We developed software tools to perform automated dose monitoring in CT, without the need for additional implemented protocols (Modality Performed Procedure Step; MPPS or Structured Report; SR). Complete CT series are sent to an analyzing DICOM server. Relevant DICOM headers are stored and dose information, available as overlay images, is extracted using optical character recognition (OCR) software (Able2Extract 6.0 Professional). Distributed over 7 different age groups (<6m, <1y, <3y, <6y, <10y, <14y and <=18y), we retrospectively collected dose information of 7089 studies from 6 different CT scanners (Siemens Sensation 64, Sensation Open, Somatom 16 and Somatom Plus 4; GE BrightSpeed 16; Philips Brilliance 64). In current study this information was used in a triple approach: (I) we used collected dose information of 4 examinations (abdomen, brain, chest, neck) to calculate diagnostic reference levels (DRL) for the CT dose index (CTDIvol) by taking the 75th percentile. (II) we checked whether used scan parameters were appropriate for the patient age. (III), two experienced radiologist (20+ years) compared the actual scanned length with the optimal scanned length of 100 randomly selected cases (chest, abdomen) using generated coronal projections.

We were able to extract dose information from all tested systems. Due to large data transfer volumes, the proposed methods are best run overnight. Calculated DRL values for the 4 evaluated examinations and per age group are: 2.38, 4.90, 4.67, 4.96, 5.95, 7.29 and 8.11 mGy (abdomen); 63.49, 83.63, 91.52, 107.30, 108.59, 112.19 and 120.96 mGy (brain); 3.79, 4.36, 4.89, 4.91, 5.48, 7.71 and 11.56 mGy (chest); 6.08, 6.33, 7.30, 8.63, 13.33, 22.68 and 28.38 mGy (neck). Besides fulfilling legal dose survey obligations, the first approach could be used to continuously compare dose data in a periodic way. It was found that in 78% (abdomen), 77% (brain), 51% (chest) and 21% (neck) of the cases, pediatric patients were scanned using adult scan parameters. On average, the total scanned volume was 15.89% (chest) and 4.9% (abdomen) larger than needed.

Proposed work enables the acquisition of large patient dose datasets and audit of scan protocols for CT systems which don’t have dose specific protocols implemented.

Dose audit opportunity for pediatric CT

Jacobs, J, Marchal, G, Bosmans, H, Automated Patient Dose Evaluation for Pediatric CT.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9014721.html