Limits...
Automated lung nodule detection at low-dose CT: preliminary experience.

Goo JM, Lee JW, Lee HJ, Kim S, Kim JH, Im JG - Korean J Radiol (2003 Oct-Dec)

Bottom Line: Three of these, identified only at CAD, formed part of the data used to derive the gold standard.There were 8.0+/-5.2 false-positive CAD results per CT study.These preliminary results indicate that a CAD system may improve the detection of pulmonary nodules at low-dose CT.

View Article: PubMed Central - PubMed

Affiliation: Department of Radiology, Seoul National University College of Medicine, and the Institute of Radiation Medicine, SNUMRC, Seoul, Korea. jmgoo@plaza.snu.ac.kr

ABSTRACT

Objective: To determine the usefulness of a computer-aided diagnosis (CAD) system for the automated detection of lung nodules at low-dose CT.

Materials and methods: A CAD system developed for detecting lung nodules was used to process the data provided by 50 consecutive low-dose CT scans. The results of an initial report, a second look review by two chest radiologists, and those obtained by the CAD system were compared, and by reviewing all of these, a gold standard was established.

Results: By applying the gold standard, a total of 52 nodules were identified (26 with a diameter < or = 5 mm; 26 with a diameter >5 mm). Compared to an initial report, four additional nodules were detected by the CAD system. Three of these, identified only at CAD, formed part of the data used to derive the gold standard. For the detection of nodules >5 mm in diameter, sensitivity was 77% for the initial report, 88% for the second look review, and 65% for the CAD system. There were 8.0+/-5.2 false-positive CAD results per CT study.

Conclusion: These preliminary results indicate that a CAD system may improve the detection of pulmonary nodules at low-dose CT.

Show MeSH

Related in: MedlinePlus

A 6-mm nodule in the left upper lobe (arrow) was detected only during the second look review.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2698098&req=5

Figure 3: A 6-mm nodule in the left upper lobe (arrow) was detected only during the second look review.

Mentions: For the detection of nodules greater than 5 mm, the sensitivity rate was 77% (20 of 26) initially, and 88% (23 of 26) at the second look review. In the initial report, four of six nodules were missed because of a detection error (Fig. 3) and the other two because of an interpretation (Fig. 2). During the second look review, one was missed due to a detection error (Fig. 1), and two because of misinterpretation. The CAD system, without any user interaction, achieved a sensitivity of 65% (17 of 26) for nodules >5 mm in diameter. False-negative findings arose for the following reasons: six nodules were seen on only one slice (Fig. 3); one was elongated; one was less compact than the others (Fig. 4); and in one, attached to the pleura, a segmentation error was involved. One nodule, rejected the by CAD system becaused it was less compact, but accepted as a nodule at further review, was no longer visible at follow-up low-dose CT performed six months later, and was thought to be focal pneumonia (Fig. 4). There were 8.0±5.2 false-positive CAD results per CT study (total, 422 false-positive findings). The processing time of the CAD system, including the loading of images, image segmentation, preparation of 3D volume data, and analysis of 3D nodule candidates, was approximately 60 seconds.


Automated lung nodule detection at low-dose CT: preliminary experience.

Goo JM, Lee JW, Lee HJ, Kim S, Kim JH, Im JG - Korean J Radiol (2003 Oct-Dec)

A 6-mm nodule in the left upper lobe (arrow) was detected only during the second look review.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2698098&req=5

Figure 3: A 6-mm nodule in the left upper lobe (arrow) was detected only during the second look review.
Mentions: For the detection of nodules greater than 5 mm, the sensitivity rate was 77% (20 of 26) initially, and 88% (23 of 26) at the second look review. In the initial report, four of six nodules were missed because of a detection error (Fig. 3) and the other two because of an interpretation (Fig. 2). During the second look review, one was missed due to a detection error (Fig. 1), and two because of misinterpretation. The CAD system, without any user interaction, achieved a sensitivity of 65% (17 of 26) for nodules >5 mm in diameter. False-negative findings arose for the following reasons: six nodules were seen on only one slice (Fig. 3); one was elongated; one was less compact than the others (Fig. 4); and in one, attached to the pleura, a segmentation error was involved. One nodule, rejected the by CAD system becaused it was less compact, but accepted as a nodule at further review, was no longer visible at follow-up low-dose CT performed six months later, and was thought to be focal pneumonia (Fig. 4). There were 8.0±5.2 false-positive CAD results per CT study (total, 422 false-positive findings). The processing time of the CAD system, including the loading of images, image segmentation, preparation of 3D volume data, and analysis of 3D nodule candidates, was approximately 60 seconds.

Bottom Line: Three of these, identified only at CAD, formed part of the data used to derive the gold standard.There were 8.0+/-5.2 false-positive CAD results per CT study.These preliminary results indicate that a CAD system may improve the detection of pulmonary nodules at low-dose CT.

View Article: PubMed Central - PubMed

Affiliation: Department of Radiology, Seoul National University College of Medicine, and the Institute of Radiation Medicine, SNUMRC, Seoul, Korea. jmgoo@plaza.snu.ac.kr

ABSTRACT

Objective: To determine the usefulness of a computer-aided diagnosis (CAD) system for the automated detection of lung nodules at low-dose CT.

Materials and methods: A CAD system developed for detecting lung nodules was used to process the data provided by 50 consecutive low-dose CT scans. The results of an initial report, a second look review by two chest radiologists, and those obtained by the CAD system were compared, and by reviewing all of these, a gold standard was established.

Results: By applying the gold standard, a total of 52 nodules were identified (26 with a diameter < or = 5 mm; 26 with a diameter >5 mm). Compared to an initial report, four additional nodules were detected by the CAD system. Three of these, identified only at CAD, formed part of the data used to derive the gold standard. For the detection of nodules >5 mm in diameter, sensitivity was 77% for the initial report, 88% for the second look review, and 65% for the CAD system. There were 8.0+/-5.2 false-positive CAD results per CT study.

Conclusion: These preliminary results indicate that a CAD system may improve the detection of pulmonary nodules at low-dose CT.

Show MeSH
Related in: MedlinePlus