Jurgen Jacobs MSc, Abstract Co-Author: Nothing to Disclose

Sam Shen, Abstract Co-Author: Nothing to Disclose

Gordon Mawdsley, Abstract Co-Author: Nothing to Disclose

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

Martin J. Yaffe PhD, Abstract Co-Author: Research collaboration, General Electric Company Scientific Advisory Board, ART Advanced Research Technologies Inc, Montreal, Canada Scientific Advisory Board, XCounter AB, Sweden

Hilde Bosmans PhD, Presenter: Nothing to Disclose

To report on the results of using a software network for remote quality control (QC) in digital mammographic breast cancer screening.

In breast cancer screening, it is important to have an effective QC program that closely monitors each X-ray system. For digital mammography, this could lead to large amounts of data arriving in a QC supervision center. We have developed a software package to simplify remote QC supervision. It connects previously developed software packages: MoniQA for monitor QC and Gladys for QC of digital mammography equipment. Technical and patient images from all local systems are automatically analyzed. Specific DICOM header tags (exposure factors, detector ID, etc., but no identifying information) are stored and transmitted by existing standards (FTP, E-mail) to a supervision site, where all data is reviewed by trained physicists. To simplify the reporting of problems, we developed a generic scoring mechanism and track system quality using color-coded time lines. This system has been used over 2 years to perform QC of 74 digital mammography units (15 different makes; 34 CR, 38 DR and 2 Slot Scan systems) and 148 monitors (8 different makes; 6 CRT and 142 LCD monitors). As a proof of concept, the same software QC network was installed in Canada for the Ontario Breast Cancer Screening Program. We report on the frequency and nature of the problems detected during this routine QC.

The implementation of the QC network was successful in both countries. With the help of the vendors, the QC procedures can be configured to be part of the normal radiological workflow, greatly reducing the effort required by the radiographer. At the supervision site, artifacts can typically be reported in less than 10 seconds. The most commonly seen problems were pixel defects, poor flat-field calibrations (DR) and dirty imaging plates (CR). Because artifacts were identified quickly, rapid follow up permitted collecting patient images to allow linking artifacts in QC and clinical data. The QC software was successfully extended to perform continuous patient dose monitoring (DR).

Using the software network results in efficient and responsive QC, reducing the workload and improving the quality in both the screening and the supervision center.

Remotely supervised QC is possible in digital mammography screening. Clinically relevant artifacts were detected and described.

Jacobs, J, Shen, S, Mawdsley, G, Marchal, G, Yaffe, M, Bosmans, H, Results of Remote Quality Control Supervision for Digital Mammography in Breast Cancer Screening.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8006822.html