TY - JOUR
T1 - Benefit of computer-aided detection analysis for the detection of subsolid and solid lung nodules on thin- and thick-section CT
AU - Godoy, Myrna C.B.
AU - Kim, Tae Jung
AU - White, Charles S.
AU - Bogoni, Luca
AU - De Groot, Patricia
AU - Florin, Charles
AU - Obuchowski, Nancy
AU - Babb, James S.
AU - Salganicoff, Marcos
AU - Naidich, David P.
AU - Anand, Vikram
AU - Park, Sangmin
AU - Vlahos, Ioannis
AU - Ko, Jane P.
N1 - Copyright:
Copyright 2013 Elsevier B.V., All rights reserved.
PY - 2013/1
Y1 - 2013/1
N2 - OBJECTIVE. The objective of our study was to evaluate the impact of computer-aided detection (CAD) on the identification of subsolid and solid lung nodules on thin- and thick-section CT. MATERIALS AND METHODS. For 46 chest CT examinations with ground-glass opacity (GGO) nodules, CAD marks computed using thin data were evaluated in two phases. First, four chest radiologists reviewed thin sections (reader thin) for nodules and subsequently CAD marks (reader thin + CADthin). After 4 months, the same cases were reviewed on thick sections (readerthick) and subsequently with CAD marks (readerthick + CADthick). Sensitivities were evaluated. Additionally, reader thick sensitivity with assessment of CAD marks on thin sections was estimated (readerthick + CADthin). RESULTS. For 155 nodules (mean, 5.5 mm; range, 4.0-27.5 mm) - 74 solid nodules, 22 part-solid (part-solid nodules), and 59 GGO nodules - CAD stand-alone sensitivity was 80%, 95%, and 71%, respectively, with three false-positives on average (0-12) per CT study. Readerthin + CADthin sensitivities were higher than readerthin for solid nodules (82% vs 57%, p < 0.001), part-solid nodules (97% vs 81%, p = 0.0027), and GGO nodules (82% vs 69%, p < 0.001) for all readers (p < 0.001). Respective sensitivities for readerthick, readerthick + CAD thick, readerthick + CADthin were 40%, 58% (p < 0.001), and 77% (p < 0.001) for solid nodules; 72%, 73% (p = 0.322), and 94% (p < 0.001) for part-solid nodules; and 53%, 58% (p = 0.008), and 79% (p < 0.001) for GGO nodules. For readerthin, false-positives increased from 0.64 per case to 0.90 with CADthin (p < 0.001) but not for readerthick; false-positive rates were 1.17, 1.19, and 1.26 per case for readerthick, readerthick + CAD thick, and readerthick + CADthin, respectively. CONCLUSION. Detection of GGO nodules and solid nodules is significantly improved with CAD. When interpretation is performed on thick sections, the benefit is greater when CAD marks are reviewed on thin rather than thick sections.
AB - OBJECTIVE. The objective of our study was to evaluate the impact of computer-aided detection (CAD) on the identification of subsolid and solid lung nodules on thin- and thick-section CT. MATERIALS AND METHODS. For 46 chest CT examinations with ground-glass opacity (GGO) nodules, CAD marks computed using thin data were evaluated in two phases. First, four chest radiologists reviewed thin sections (reader thin) for nodules and subsequently CAD marks (reader thin + CADthin). After 4 months, the same cases were reviewed on thick sections (readerthick) and subsequently with CAD marks (readerthick + CADthick). Sensitivities were evaluated. Additionally, reader thick sensitivity with assessment of CAD marks on thin sections was estimated (readerthick + CADthin). RESULTS. For 155 nodules (mean, 5.5 mm; range, 4.0-27.5 mm) - 74 solid nodules, 22 part-solid (part-solid nodules), and 59 GGO nodules - CAD stand-alone sensitivity was 80%, 95%, and 71%, respectively, with three false-positives on average (0-12) per CT study. Readerthin + CADthin sensitivities were higher than readerthin for solid nodules (82% vs 57%, p < 0.001), part-solid nodules (97% vs 81%, p = 0.0027), and GGO nodules (82% vs 69%, p < 0.001) for all readers (p < 0.001). Respective sensitivities for readerthick, readerthick + CAD thick, readerthick + CADthin were 40%, 58% (p < 0.001), and 77% (p < 0.001) for solid nodules; 72%, 73% (p = 0.322), and 94% (p < 0.001) for part-solid nodules; and 53%, 58% (p = 0.008), and 79% (p < 0.001) for GGO nodules. For readerthin, false-positives increased from 0.64 per case to 0.90 with CADthin (p < 0.001) but not for readerthick; false-positive rates were 1.17, 1.19, and 1.26 per case for readerthick, readerthick + CAD thick, and readerthick + CADthin, respectively. CONCLUSION. Detection of GGO nodules and solid nodules is significantly improved with CAD. When interpretation is performed on thick sections, the benefit is greater when CAD marks are reviewed on thin rather than thick sections.
KW - CT
KW - Computer-aided detection
KW - Ground-glass nodule
KW - Lung nodule
UR - http://www.scopus.com/inward/record.url?scp=84871863453&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84871863453&partnerID=8YFLogxK
U2 - 10.2214/AJR.11.7532
DO - 10.2214/AJR.11.7532
M3 - Article
C2 - 23255744
AN - SCOPUS:84871863453
SN - 0361-803X
VL - 200
SP - 74
EP - 83
JO - American Journal of Roentgenology
JF - American Journal of Roentgenology
IS - 1
ER -