Abstract
ObjectiveTo evaluate performance of computer-aided detection (CAD) beyond double reading for pulmonary nodules on low-dose computed tomography (CT) by nodule volume.MethodsA total of 400 low-dose chest CT examinations were randomly selected from the NELSON lung cancer screening trial. CTs were evaluated by two independent readers and processed by CAD. A total of 1,667 findings marked by readers and/or CAD were evaluated by a consensus panel of expert chest radiologists. Performance was evaluated by calculating sensitivity of pulmonary nodule detection and number of false positives, by nodule characteristics and volume.ResultsAccording to the screening protocol, 90.9 % of the findings could be excluded from further evaluation, 49.2 % being small nodules (less than 50 mm3). Excluding small nodules reduced false-positive detections by CAD from 3.7 to 1.9 per examination. Of 151 findings that needed further evaluation, 33 (21.9 %) were detected by CAD only, one of them being diagnosed as lung cancer the following year. The sensitivity of nodule detection was 78.1 % for double reading and 96.7 % for CAD. A total of 69.7 % of nodules undetected by readers were attached nodules of which 78.3 % were vessel-attached.ConclusionsCAD is valuable in lung cancer screening to improve sensitivity of pulmonary nodule detection beyond double reading, at a low false-positive rate when excluding small nodules.Key Points • Computer-aided detection (CAD) has known advantages for computed tomography (CT). • Combined CAD/nodule size cut-off parameters assist CT lung cancer screening. • This combination improves the sensitivity of pulmonary nodule detection by CT. • It increases the positive predictive value for cancer detection.
Highlights
Materials and methodsThe rapid development of multi-detector computed tomography (CT) (MDCT) has increased the amount of data for radiologists to analyse
Computer-aided detection (CAD) is valuable in lung cancer screening to improve sensitivity of pulmonary nodule detection beyond double reading, at a low false-positive rate when excluding small nodules
The purpose of our study was to assess the performance of CAD for detection of pulmonary nodules as a complementary tool in a large-scale, low-dose CT lung cancer screening study compared to double reading, with stratification according to nodule volume
Summary
The mean age of the 400 participants was 59±6 years (range 51–76 years). On 332 of the 400 baseline CT examinations at least one finding was reported (Fig. 1). NA not applicable, NS not significantly different a For 5 non-/part-solid nodules, volume was not available detected by readers were attached nodules, and 78.3 % of these were vessel-attached (Fig. 2). Of the non-peripheral, vessel-attached nodules, 7 out of 11 were missed by readers but all were detected by CAD. Of 33 nodules missed by readers at baseline, 24 were detected at 3-month or 1-year follow-up CT examinations. Lung cancer was diagnosed in one solid intraparenchymal nodule, found to have grown at the second-year screening CT. A solid pleural-based nodule with volume 217.8 mm missed by CAD was diagnosed as lung cancer after it was found to be growing on the 3-month follow-up CT examination, with volume doubling time less than 400 days (Fig. 3). None of the benign-appearing pulmonary nodules presented with malignant behaviour during subsequent screening rounds
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
