Feng Li MD, PhD, Abstract Co-Author: License agreement, Hologic, Inc License agreement, General Electric Company License agreement, Toshiba Corporation License agreement, Deus Technologies, LLC License agreement, Riverain Medical License agreement, Median Technologies License agreement, Mitsubishi Corporation

Roger Engelmann MS, Abstract Co-Author: Grant, Riverain Medical Royalties, Riverain Medical License agreement, Hologic, Inc License agreement, General Electric Company License agreement, Toshiba Corporation License agreement, Deus Technologies, LLC License agreement, Riverain Medical License agreement, MEDIAN Technologies License agreement, Mitsubishi Corporation

Heber Macmahon MD, Presenter: Consultant, Riverain Medical Speaker, Konica Minolta Group Stockholder, Hologic, Inc

To evaluate radiologists’ ability to detect focal airspace opacity, such as pneumonia, by use of standard or portable chest radiographs (CXRs) alone compared to CXRs with bone suppression imaging (BSI).

Posteroanterior (PA) or anteroposterior (AP) standard CXRs in 30 patients and portable radiographs in 10 patients with subtle focal airspace opacities (7 patients with 2 focal opacities in different lungs) as well as 20 comparable patients without focal opacities were included in the observer study. All 47 focal airspace opacities in these cases were consistent with infection, and 38 of them were confirmed by CT. Others were confirmed by prior and/or subsequent radiographs, and expert consensus. The maximum diameter of the 47 focal opacities ranged from 10 to 60 mm. A bone suppression image processing system (SoftView, Version 2.2 Riverain Medical, Miamisburg Ohio), which can selectively suppress the opacity of bones in chest radiographs, was applied to the 60 CXRs to create corresponding BSI images. Two experienced radiologists jointly selected these focal pneumonias based on subtlety and available truth, before applying BSI processing. Among the 40 radiographs, nine lungs which had confusing or unverifiable findings, were judged indeterminate, and were not included for scoring purposes. Six observers, including four attending radiologists and two radiology residents, indicated their confidence level regarding the presence of a focal opacity compatible with pneumonia for each lung (total, 111 lungs), first by use of standard PA or AP CXRs, then with additional BSI images. Receiver operating characteristic (ROC) analysis was used for evaluation of the observers’ performance.

The mean value of the area under the ROC curve (AUC) for six observers was significantly improved from 0.845 with standard CXRs alone to 0.876 with the use of BSI images (P=0.004).

Bone Suppression images improved radiologists’ performance for detection of pneumonia on chest radiographs.

Radiologists' accuracy for detection of pneumonia and similar focal air space opacities on chest radiographs can be improved by use of bone suppression images.

Li, F, Engelmann, R, Macmahon, H, Improved Detection of Pneumonia by Chest Radiographs with Bone Suppression Imaging.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9004931.html