Computer-Aided Diagnosis of Lung Nodules on CT Scans: . ROC Study of Its Effect on Radiologists' Performance

Ted Way, Heang Ping Chan, Lubomir Hadjiiski, Berkman Sahiner, Aamer Chughtai, Thomas K. Song, Chad Poopat, Jadranka Stojanovska, Luba Frank, Anil Attili, Naama Bogot, Philip N. Cascade, Ella A. Kazerooni

Research output: Contribution to journalArticle

29 Citations (Scopus)

Abstract

Rationale and Objectives: The aim of this study was to evaluate the effect of computer-aided diagnosis (CAD) on radiologists' estimates of the likelihood of malignancy of lung nodules on computed tomographic (CT) imaging. Methods and Materials: A total of 256 lung nodules (124 malignant, 132 benign) were retrospectively collected from the thoracic CT scans of 152 patients. An automated CAD system was developed to characterize and provide malignancy ratings for lung nodules on CT volumetric images. An observer study was conducted using receiver-operating characteristic analysis to evaluate the effect of CAD on radiologists' characterization of lung nodules. Six fellowship-trained thoracic radiologists served as readers. The readers rated the likelihood of malignancy on a scale of 0% to 100% and recommended appropriate action first without CAD and then with CAD. The observer ratings were analyzed using the Dorfman-Berbaum-Metz multireader, multicase method. Results: The CAD system achieved a test area under the receiver-operating characteristic curve (Az) of 0.857 ± 0.023 using the perimeter, two nodule radii measures, two texture features, and two gradient field features. All six radiologists obtained improved performance with CAD. The average Az of the radiologists improved significantly (P <.01) from 0.833 (range, 0.817-0.847) to 0.853 (range, 0.834-0.887). Conclusion: CAD has the potential to increase radiologists' accuracy in assessing the likelihood of malignancy of lung nodules on CT imaging.

Original languageEnglish (US)
Pages (from-to)323-332
Number of pages10
JournalAcademic Radiology
Volume17
Issue number3
DOIs
StatePublished - Mar 2010
Externally publishedYes

Fingerprint

Lung
ROC Curve
Neoplasms
Thorax
Radiologists

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this

Way, T., Chan, H. P., Hadjiiski, L., Sahiner, B., Chughtai, A., Song, T. K., ... Kazerooni, E. A. (2010). Computer-Aided Diagnosis of Lung Nodules on CT Scans: . ROC Study of Its Effect on Radiologists' Performance. Academic Radiology, 17(3), 323-332. https://doi.org/10.1016/j.acra.2009.10.016

Computer-Aided Diagnosis of Lung Nodules on CT Scans : . ROC Study of Its Effect on Radiologists' Performance. / Way, Ted; Chan, Heang Ping; Hadjiiski, Lubomir; Sahiner, Berkman; Chughtai, Aamer; Song, Thomas K.; Poopat, Chad; Stojanovska, Jadranka; Frank, Luba; Attili, Anil; Bogot, Naama; Cascade, Philip N.; Kazerooni, Ella A.

In: Academic Radiology, Vol. 17, No. 3, 03.2010, p. 323-332.

Research output: Contribution to journalArticle

Way, T, Chan, HP, Hadjiiski, L, Sahiner, B, Chughtai, A, Song, TK, Poopat, C, Stojanovska, J, Frank, L, Attili, A, Bogot, N, Cascade, PN & Kazerooni, EA 2010, 'Computer-Aided Diagnosis of Lung Nodules on CT Scans: . ROC Study of Its Effect on Radiologists' Performance', Academic Radiology, vol. 17, no. 3, pp. 323-332. https://doi.org/10.1016/j.acra.2009.10.016
Way, Ted ; Chan, Heang Ping ; Hadjiiski, Lubomir ; Sahiner, Berkman ; Chughtai, Aamer ; Song, Thomas K. ; Poopat, Chad ; Stojanovska, Jadranka ; Frank, Luba ; Attili, Anil ; Bogot, Naama ; Cascade, Philip N. ; Kazerooni, Ella A. / Computer-Aided Diagnosis of Lung Nodules on CT Scans : . ROC Study of Its Effect on Radiologists' Performance. In: Academic Radiology. 2010 ; Vol. 17, No. 3. pp. 323-332.
@article{2530fe19c3494cbe9bad705053c13969,
title = "Computer-Aided Diagnosis of Lung Nodules on CT Scans: . ROC Study of Its Effect on Radiologists' Performance",
abstract = "Rationale and Objectives: The aim of this study was to evaluate the effect of computer-aided diagnosis (CAD) on radiologists' estimates of the likelihood of malignancy of lung nodules on computed tomographic (CT) imaging. Methods and Materials: A total of 256 lung nodules (124 malignant, 132 benign) were retrospectively collected from the thoracic CT scans of 152 patients. An automated CAD system was developed to characterize and provide malignancy ratings for lung nodules on CT volumetric images. An observer study was conducted using receiver-operating characteristic analysis to evaluate the effect of CAD on radiologists' characterization of lung nodules. Six fellowship-trained thoracic radiologists served as readers. The readers rated the likelihood of malignancy on a scale of 0{\%} to 100{\%} and recommended appropriate action first without CAD and then with CAD. The observer ratings were analyzed using the Dorfman-Berbaum-Metz multireader, multicase method. Results: The CAD system achieved a test area under the receiver-operating characteristic curve (Az) of 0.857 ± 0.023 using the perimeter, two nodule radii measures, two texture features, and two gradient field features. All six radiologists obtained improved performance with CAD. The average Az of the radiologists improved significantly (P <.01) from 0.833 (range, 0.817-0.847) to 0.853 (range, 0.834-0.887). Conclusion: CAD has the potential to increase radiologists' accuracy in assessing the likelihood of malignancy of lung nodules on CT imaging.",
author = "Ted Way and Chan, {Heang Ping} and Lubomir Hadjiiski and Berkman Sahiner and Aamer Chughtai and Song, {Thomas K.} and Chad Poopat and Jadranka Stojanovska and Luba Frank and Anil Attili and Naama Bogot and Cascade, {Philip N.} and Kazerooni, {Ella A.}",
year = "2010",
month = "3",
doi = "10.1016/j.acra.2009.10.016",
language = "English (US)",
volume = "17",
pages = "323--332",
journal = "Academic Radiology",
issn = "1076-6332",
publisher = "Elsevier USA",
number = "3",

}

TY - JOUR

T1 - Computer-Aided Diagnosis of Lung Nodules on CT Scans

T2 - . ROC Study of Its Effect on Radiologists' Performance

AU - Way, Ted

AU - Chan, Heang Ping

AU - Hadjiiski, Lubomir

AU - Sahiner, Berkman

AU - Chughtai, Aamer

AU - Song, Thomas K.

AU - Poopat, Chad

AU - Stojanovska, Jadranka

AU - Frank, Luba

AU - Attili, Anil

AU - Bogot, Naama

AU - Cascade, Philip N.

AU - Kazerooni, Ella A.

PY - 2010/3

Y1 - 2010/3

N2 - Rationale and Objectives: The aim of this study was to evaluate the effect of computer-aided diagnosis (CAD) on radiologists' estimates of the likelihood of malignancy of lung nodules on computed tomographic (CT) imaging. Methods and Materials: A total of 256 lung nodules (124 malignant, 132 benign) were retrospectively collected from the thoracic CT scans of 152 patients. An automated CAD system was developed to characterize and provide malignancy ratings for lung nodules on CT volumetric images. An observer study was conducted using receiver-operating characteristic analysis to evaluate the effect of CAD on radiologists' characterization of lung nodules. Six fellowship-trained thoracic radiologists served as readers. The readers rated the likelihood of malignancy on a scale of 0% to 100% and recommended appropriate action first without CAD and then with CAD. The observer ratings were analyzed using the Dorfman-Berbaum-Metz multireader, multicase method. Results: The CAD system achieved a test area under the receiver-operating characteristic curve (Az) of 0.857 ± 0.023 using the perimeter, two nodule radii measures, two texture features, and two gradient field features. All six radiologists obtained improved performance with CAD. The average Az of the radiologists improved significantly (P <.01) from 0.833 (range, 0.817-0.847) to 0.853 (range, 0.834-0.887). Conclusion: CAD has the potential to increase radiologists' accuracy in assessing the likelihood of malignancy of lung nodules on CT imaging.

AB - Rationale and Objectives: The aim of this study was to evaluate the effect of computer-aided diagnosis (CAD) on radiologists' estimates of the likelihood of malignancy of lung nodules on computed tomographic (CT) imaging. Methods and Materials: A total of 256 lung nodules (124 malignant, 132 benign) were retrospectively collected from the thoracic CT scans of 152 patients. An automated CAD system was developed to characterize and provide malignancy ratings for lung nodules on CT volumetric images. An observer study was conducted using receiver-operating characteristic analysis to evaluate the effect of CAD on radiologists' characterization of lung nodules. Six fellowship-trained thoracic radiologists served as readers. The readers rated the likelihood of malignancy on a scale of 0% to 100% and recommended appropriate action first without CAD and then with CAD. The observer ratings were analyzed using the Dorfman-Berbaum-Metz multireader, multicase method. Results: The CAD system achieved a test area under the receiver-operating characteristic curve (Az) of 0.857 ± 0.023 using the perimeter, two nodule radii measures, two texture features, and two gradient field features. All six radiologists obtained improved performance with CAD. The average Az of the radiologists improved significantly (P <.01) from 0.833 (range, 0.817-0.847) to 0.853 (range, 0.834-0.887). Conclusion: CAD has the potential to increase radiologists' accuracy in assessing the likelihood of malignancy of lung nodules on CT imaging.

UR - http://www.scopus.com/inward/record.url?scp=75749101786&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=75749101786&partnerID=8YFLogxK

U2 - 10.1016/j.acra.2009.10.016

DO - 10.1016/j.acra.2009.10.016

M3 - Article

C2 - 20152726

AN - SCOPUS:75749101786

VL - 17

SP - 323

EP - 332

JO - Academic Radiology

JF - Academic Radiology

SN - 1076-6332

IS - 3

ER -