Computerized detection of lung nodules in thin-section CT images by use of selective enhancement filters and an automated rule-based classifier

Abstract

Rationale and objectives: We have been developing a computer-aided diagnostic (CAD) scheme for lung nodule detection in order to assist radiologists in the detection of lung cancer in thin-section computed tomography (CT) images.

Materials and methods: Our database consisted of 117 thin-section CT scans with 153 nodules, obtained from a lung cancer screening program at a Japanese university (85 scans, 91 nodules) and from clinical work at an American university (32 scans, 62 nodules). The database included nodules of different sizes (4-28 mm, mean 10.2 mm), shapes, and patterns (solid and ground-glass opacity (GGO)). Our CAD scheme consisted of modules for lung segmentation, selective nodule enhancement, initial nodule detection, feature extraction, and classification. The selective nodule enhancement filter was a key technique for significant enhancement of nodules and suppression of normal anatomic structures such as blood vessels, which are the main sources of false positives. Use of an automated rule-based classifier for reduction of false positives was another key technique; it resulted in a minimized overtraining effect and an improved classification performance. We used a case-based four-fold cross-validation testing method for evaluation of the performance levels of our computerized detection scheme.

Results: Our CAD scheme achieved an overall sensitivity of 86% (small: 76%, medium-sized: 94%, large: 95%; solid: 86%, mixed GGO: 89%, pure GGO: 81%) with 6.6 false positives per scan; an overall sensitivity of 81% (small: 69%, medium-sized: 91%, large: 91%; solid: 79%, mixed GGO: 88%, pure GGO: 81%) with 3.3 false positives per scan; and an overall sensitivity of 75% (small: 60%, medium-sized: 88%, large: 87%; solid: 70%, mixed GGO: 87%, pure GGO: 81%) with 1.6 false positives per scan.

Conclusion: The experimental results indicate that our CAD scheme with its two key techniques can achieve a relatively high performance for nodules presenting large variations in size, shape, and pattern.

Fig. 1

Distribution of nodule sizes in our database. The database contained nodules with a relatively wide range of sizes. There were 68 (44.4%) small (4 - 8 mm), 52 (34.0%) medium-sized (9 - 13 mm), and 33 (21.6%) large nodules (14 mm and above) in the database.

Fig. 2

Overall scheme of the computerized detection technique.

Fig. 3

Schemetic illustration for inclusion of a juxtapleural object.

Fig. 4

Maximum intensity projection of (a) two 3D original images with nodules identified by arrows and (b) nodule-enhanced images.

Fig. 4

Maximum intensity projection of (a) two 3D original images with nodules identified by arrows and (b) nodule-enhanced images.

Fig. 5

(a) Three low-contrast nodules with GGO that were successfully identified, and (b) the only two nodules that were missed by our initial nodule detection technique.

Fig. 5

(a) Three low-contrast nodules with GGO that were successfully identified, and (b) the only two nodules that were missed by our initial nodule detection technique.

Fig. 6

Mean FROC curves for training and testing of our CAD scheme.

Fig. 7

Mean FROC curves obtained from the testing of our CAD scheme for the nodules in the American dataset and the Japanese dataset.

Fig. 8

Mean FROC curves obtained from the testing of our CAD scheme for the small nodules (<9 mm), the medium-sized nodules (9-13 mm), the large nodules (>13 mm), and all nodules.

Fig. 9

Mean FROC curves obtained from the testing of our CAD scheme for the solid nodules, the mixed GGO nodules, the pure GGO nodules, and all nodules.