TY - JOUR
T1 - Semi-Automated pulmonary nodule interval segmentation using the NLST data:
AU - Balagurunathan, Yoganand
AU - Beers, Andrew
AU - Kalpathy-Cramer, Jayashree
AU - McNitt-Gray, Michael
AU - Hadjiiski, Lubomir
AU - Zhao, Bensheng
AU - Zhu, Jiangguo
AU - Yang, Hao
AU - Yip, Stephen S.F.
AU - Aerts, Hugo J.W.L.
AU - Napel, Sandy
AU - Cherezov, Dmitrii
AU - Cha, Kenny
AU - Chan, Heang Ping
AU - Flores, Carlos
AU - Garcia, Alberto
AU - Gillies, Robert
AU - Goldgof, Dmitry
N1 - Funding Information:
The study teams are thankful to the National Cancer Institute’s Quantitative Imaging Network initiative, program managers, support staff, The Cancer Imaging Archive (TCIA), The NCIP-Hub, all of which provided the necessary collaborative platform for the study. The study teams are grateful to the funding support received through various agencies that funded the research time for the investigators. CUMC group acknowledge support received through: R01 CA149490, U01 CA140207, and U24 CA180927. SU group acknowledge support received through: U01 CA187947, R01 CA160251. MCC/ USF group acknowledge support received through: U01 CA 143062, 2KT01 State of Florida Department of Health and U24 CA180927. DFCC group acknowledge support received through: U24CA194354. UCLA group acknowledge support received through: U01CA181156. UMICH group acknowledges support received through: LH, KC and HPC are supported in part by NIH/NCI (U01CA179106). Large part of the general Michigan segmentation and feature extraction pipeline was developed under the support of NIH award number R01 CA93517 (PI: Heang-Ping Chan). MGH group acknowledge support received through: U24 CA180927.
Publisher Copyright:
© 2018 American Association of Physicists in Medicine.
PY - 2018/3
Y1 - 2018/3
N2 - Purpose: To study the variability in volume change estimates of pulmonary nodules due to segmentation approaches used across several algorithms and to evaluate these effects on the ability to predict nodule malignancy. Methods: We obtained 100 patient image datasets from the National Lung Screening Trial (NLST) that had a nodule detected on each of two consecutive low dose computed tomography (LDCT) scans, with an equal proportion of malignant and benign cases (50 malignant, 50 benign). Information about the nodule location for the cases was provided by a screen capture with a bounding box and its axial location was indicated. Five participating quantitative imaging network (QIN) institutions performed nodule segmentation using their preferred semi-Automated algorithms with no manual correction; teams were allowed to provide additional manually corrected segmentations (analyzed separately). The teams were asked to provide segmentation masks for each nodule at both time points. From these masks, the volume was estimated for the nodule at each time point; the change in volume (absolute and percent change) across time points was estimated as well. We used the concordance correlation coefficient (CCC) to compare the similarity of computed nodule volumes (absolute and percent change) across algorithms. We used Logistic regression model on the change in volume (absolute change and percent change) of the nodules to predict the malignancy status, the area under the receiver operating characteristic curve (AUROC) and confidence intervals were reported. Because the size of nodules was expected to have a substantial effect on segmentation variability, analysis of change in volumes was stratified by lesion size, where lesions were grouped into those with a longest diameter of <8 mm and those with longest diameter ≥ 8 mm. Results: We find that segmentation of the nodules shows substantial variability across algorithms, with the CCC ranging from 0.56 to 0.95 for change in volume (percent change in volume range was [0.15 to 0.86]) across the nodules. When examining nodules based on their longest diameter, we find the CCC had higher values for large nodules with a range of [0.54 to 0.93] among the algorithms, while percent change in volume was [0.3 to 0.95]. Compared to that of smaller nodules which had a range of [-0.0038 to 0.69] and percent change in volume was [-0.039 to 0.92]. The malignancy prediction results showed fairly consistent results across the institutions, the AUC using change in volume ranged from 0.65 to 0.89 (Percent change in volume was 0.64 to 0.86) for entire nodule range. Prediction improves for large nodule range (≥ 8 mm) with AUC range 0.75 to 0.90 (percent change in volume was 0.74 to 0.92). Compared to smaller nodule range (<8 mm) with AUC range 0.57 to 0.78 (percent change in volume was 0.59 to 0.77). Conclusions: We find there is a fairly high concordance in the size measurements for larger nodules (≥8 mm) than the lower sizes (<8 mm) across algorithms. We find the change in nodule volume (absolute and percent change) were consistent predictors of malignancy across institutions, despite using different segmentation algorithms. Using volume change estimates without corrections shows slightly lower predictability (for two teams).
AB - Purpose: To study the variability in volume change estimates of pulmonary nodules due to segmentation approaches used across several algorithms and to evaluate these effects on the ability to predict nodule malignancy. Methods: We obtained 100 patient image datasets from the National Lung Screening Trial (NLST) that had a nodule detected on each of two consecutive low dose computed tomography (LDCT) scans, with an equal proportion of malignant and benign cases (50 malignant, 50 benign). Information about the nodule location for the cases was provided by a screen capture with a bounding box and its axial location was indicated. Five participating quantitative imaging network (QIN) institutions performed nodule segmentation using their preferred semi-Automated algorithms with no manual correction; teams were allowed to provide additional manually corrected segmentations (analyzed separately). The teams were asked to provide segmentation masks for each nodule at both time points. From these masks, the volume was estimated for the nodule at each time point; the change in volume (absolute and percent change) across time points was estimated as well. We used the concordance correlation coefficient (CCC) to compare the similarity of computed nodule volumes (absolute and percent change) across algorithms. We used Logistic regression model on the change in volume (absolute change and percent change) of the nodules to predict the malignancy status, the area under the receiver operating characteristic curve (AUROC) and confidence intervals were reported. Because the size of nodules was expected to have a substantial effect on segmentation variability, analysis of change in volumes was stratified by lesion size, where lesions were grouped into those with a longest diameter of <8 mm and those with longest diameter ≥ 8 mm. Results: We find that segmentation of the nodules shows substantial variability across algorithms, with the CCC ranging from 0.56 to 0.95 for change in volume (percent change in volume range was [0.15 to 0.86]) across the nodules. When examining nodules based on their longest diameter, we find the CCC had higher values for large nodules with a range of [0.54 to 0.93] among the algorithms, while percent change in volume was [0.3 to 0.95]. Compared to that of smaller nodules which had a range of [-0.0038 to 0.69] and percent change in volume was [-0.039 to 0.92]. The malignancy prediction results showed fairly consistent results across the institutions, the AUC using change in volume ranged from 0.65 to 0.89 (Percent change in volume was 0.64 to 0.86) for entire nodule range. Prediction improves for large nodule range (≥ 8 mm) with AUC range 0.75 to 0.90 (percent change in volume was 0.74 to 0.92). Compared to smaller nodule range (<8 mm) with AUC range 0.57 to 0.78 (percent change in volume was 0.59 to 0.77). Conclusions: We find there is a fairly high concordance in the size measurements for larger nodules (≥8 mm) than the lower sizes (<8 mm) across algorithms. We find the change in nodule volume (absolute and percent change) were consistent predictors of malignancy across institutions, despite using different segmentation algorithms. Using volume change estimates without corrections shows slightly lower predictability (for two teams).
KW - CT lung
KW - change in volume segmentation
KW - lung nodule segmentation
KW - volume estimate
UR - http://www.scopus.com/inward/record.url?scp=85042145249&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85042145249&partnerID=8YFLogxK
U2 - 10.1002/mp.12766
DO - 10.1002/mp.12766
M3 - Article
C2 - 29363773
AN - SCOPUS:85042145249
VL - 45
SP - 1093
EP - 1107
JO - Medical Physics
JF - Medical Physics
SN - 0094-2405
IS - 3
ER -