Roberto Lo Gullo MD, Presenter: Nothing to Disclose

Marie-Helene Levesque MD, Abstract Co-Author: Nothing to Disclose

Karl Sayegh MD, Abstract Co-Author: Nothing to Disclose

Sishir Rao MD, Abstract Co-Author: Nothing to Disclose

Scott Bruce Raymond MD, PhD, Abstract Co-Author: Nothing to Disclose

Subba Rao Digumarthy MD, Abstract Co-Author: Nothing to Disclose

Jo-Anne O. Shepard MD, Abstract Co-Author: Consultant, Agfa-Gevaert Group

Mannudeep K. S. Kalra MD, Abstract Co-Author: Nothing to Disclose

Atul Padole MD, Abstract Co-Author: Nothing to Disclose

Sarabjeet Singh MD, Abstract Co-Author: Research Grant, Siemens AG Research Grant, Toshiba Corporation Research Grant, General Electric Company Research Grant, Koninklijke Philips NV

Mark T. Madsen PhD, Abstract Co-Author: Nothing to Disclose

Alexi Otrakji MD, Abstract Co-Author: Nothing to Disclose

To assess the variation in lung nodule detection in low dose chest CT based on location, attenuation characteristics and reader experience.

In an IRB approved study, we selected and extracted 18 non-calcified lung nodules with solid (n=7 nodules), ground glass (7) and mixed (4) attenuation from 12 chest examinations from patients with known malignant lung disease. All nodules had similar size and shape (mean size 7mm, size range, 6-8 mm). These nodules were randomly inserted in to 34 normal low dose chest CT examinations belonging to 34 patients (mean age 57.5 years, 15M;19F) using a dedicated software (IDL virtual machine) at the following lung sites, apices, bases, peripheral 2 cm, close to heart and major vascular structures, branching points of bronchi and vessels and in the common regions of image artifacts such as behind the 1st rib. A total of 47 lung nodules (some nodules were inserted multiple times in each scan) were inserted. Two residents (Resident 1 with 2 weeks of CT training and resident 2 with at least one complete chest CT rotation) and two radiologists at the end of one year thoracic fellowship training, all blinded to the details of the study, interpreted the CT examinations in a routine fashion. Variation of nodule detection was assessed based on location, attenuation characteristics and reader experience; statistical significance was assessed by chi-square test.

Resident 2 and the two fellows identified significantly more nodules in all categories (p=0.01). There is significantly decreased detection of nodules at branching points, adjacent to heart and major vessels compared to those at apices, bases and at periphery of the lungs (p=0.004). Identification of mixed density nodules was significantly higher for fellows (p=0.008).

Dedicated chest CT training improves detection of lung nodules, particularly that of mixed attenuation nodules (which are more likely to be malignant). The detection of nodules adjacent to heart and major vessels and at branching points is difficult even for fellowship-trained radiologists.

Dedicated training is required for improved lung nodule detection. Nodules in certain locations may be difficult to detect, which may be improved with our dedicated training program. Review of MIP images and use of CAD program for lung nodule detection may also help.

Lo Gullo, R, Levesque, M, Sayegh, K, Rao, S, Raymond, S, Digumarthy, S, Shepard, J, Kalra, M, Padole, A, Singh, S, Madsen, M, Otrakji, A, Nodule Detection in Lung Cancer Screening: When, Where, and Why are Non-calcified Lung Nodules Missed?.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14008206.html