Virtual Library
Start Your Search
B. Van Ginneken
Author of
-
+
MINI 36 - Imaging and Diagnostic Workup (ID 163)
- Event: WCLC 2015
- Type: Mini Oral
- Track: Screening and Early Detection
- Presentations: 1
-
+
MINI36.04 - Automated Measurement of Malignancy Risk of Lung Nodule Detected by Screening Computed Tomography (ID 1737)
18:45 - 18:50 | Author(s): B. Van Ginneken
- Abstract
- Presentation
Background:
We have previously reported a practical predictive tool that accurately estimates the probability of malignancy for lung nodules detected at baseline screening LDCT (New Engl J Med. 2013;369:908-17). Manual measurement of nodule dimensions and generation of malignancy risk scores is time consuming and subjected to intra- and inter-observer variability. The goal of this study is to prepare a nodule malignancy risk prediction model based on automated computer generated nodule data and compare it to an established model based on radiologists’ generated data.
Methods:
Using the same published PanCan dataset (New Engl J Med. 2013;369:908-17) with the number of lung cancers updated, we prepared a logistic regression model predicting lung cancer using computer-generated imaging data from the CIRRUS Lung Screening software (Diagnostic Imaging Analysis Group, Nijmegen, The Netherlands). Ninety-one of the 2,537 baseline (first) scans were not available or could not be processed by CIRRUS. The remaining 2,446 scans were first annotated by the CIRRUS software. A human non-radiologist reader then accepted/rejected the annotated marks and manually searched the LDCT for nodules missed by CIRRUS or the study radiologist. New nodules found that were not recorded by the study radiologist were reviewed by a subspecialty trained chest radiologist with 14 years experience in lung cancer screening (JM). Nodule morphometric measurements (maximum and mean diameter, volume, mass, density) and total nodule count per scan irrespective of size were automatically generated by the CIRRUS software. The nodule type (nonsolid, part-solid, or solid), nodule description (lobulated, spiculated or well defined) and nodule location (upper versus middle or lower lobe) were manually entered. The variables were evaluated in models as untransformed and natural log transformed variables. Nonlinear relationships with lung cancer were also evaluated. Socio-demographic and clinical history predictors were not included in the model.
Results:
Radiologists evaluation identified 8,570 pulmonary nodules of any size in 2063 individuals - 124 nodules in 119 individuals were diagnosed as cancer in follow-up. Based on CIRRUS software annotated marks that were accepted by a human reader, computer analysis identified 11,520 pulmonary nodules in 2174 individuals - 121 nodules in 115 individuals were diagnosed as cancer in follow-up. Thirty-six percent of new nodules found by CIRRUS and/or second human reader were ≥4 mm (mean±SD, 5.9± 3.5 mm). Both the computer generated imaging data model (Model-CIRRUS) and the radiologist generated data model (Model-RAD) demonstrated excellent discrimination and calibration. Their predictive performances were also similar. Comparing Model-CAD to Model-RAD, the AUCs were 0.9537 versus 0.9541, the 90[th] percentile absolute errors were 0.0008 versus 0.0007, and the Brier scores were 0.0093 versus 0.0137. Mean nodule diameter is a better risk predictor than maximum nodule diameter, nodule density or mass.
Conclusion:
The predictive performances of computer and radiologist generated data models were similar. The model can be integrated to the CIRRUS Lung Screening software to automatically generate a nodule malignancy risk score to facilitate nodule management recommendation. Supported by the Terry Fox Research Institute, The Canadian Partnership Against Cancer and the BC Cancer Foundation on behalf of the Pan-Canadian Early Detection of Lung Cancer Study Group.
Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.
-
+
ORAL 09 - CT Screening - New Data and Risk Assessment (ID 95)
- Event: WCLC 2015
- Type: Oral Session
- Track: Screening and Early Detection
- Presentations: 1
- Moderators:J. Mulshine, J.K. Field
- Coordinates: 9/07/2015, 10:45 - 12:15, Mile High Ballroom 2a-3b
-
+
ORAL09.05 - Lung-RADS versus the McWilliams Nodule Malignancy Score for Risk Prediction: Evaluation on the Danish Lung Cancer Screening Trial (ID 356)
11:28 - 11:39 | Author(s): B. Van Ginneken
- Abstract
Background:
Lung-RADS published in 2014 by the American College of Radiology is based on literature review and expert opinion and uses nodule type, size, and growth to recommend nodule management adjusted to malignancy risk. The McWilliams model (N Engl J Med 2013;369:910-9) is a multivariate logistic regression model derived from the Pan-Canadian Early Detection of Lung Cancer Study and provides a nodule malignancy probability based on nodule size, type, morphology and subject characteristics. We compare the performance of both approaches on an independent data set.
Methods:
We selected 60 cancers from the Danish Lung Cancer Screening Trial as presented in the first scan they were visible, and randomly added 120 benign nodules from baseline scans, all from different participants. Data had been acquired using a low-dose (16x0.75mm, 120kVp, 40mAs) protocol, and 1mm section thickness reconstruction. For each nodule, the malignancy probability was calculated using McWilliams model 2b. Parameters were available from the screening database or scored by an expert radiologist. Completely calcified nodules and perifissural nodules were assigned a malignancy probability of 0, in accordance with model guidelines. All nodules were categorized into their Lung-RADS category based on nodule type and diameter. Perifissural nodules were treated as solid nodules, in accordance with Lung-RADS guidelines. For each Lung-RADS category cut-off sensitivity and specificity were calculated. Corresponding sensitivities and specificities using the McWilliams model were determined.
Results:
Defining Lung-RADS category 2/3/4A/4B and higher as a positive screening result, specificities to exclude lung malignancy were 21%/65%/86%/99% and vice versa sensitivities to predict malignancy were 100%/85%/58%/32%. At the same sensitivity levels as Lung-RADS, McWilliams model yielded overall higher specificities with 2%/86%/98%/100%, respectively (red arrows in Figure 1). Similarly, at the same specificities McWilliams’s model achieved higher sensitivities with 100%/95%/85%/48%, respectively (green arrows in Figure 1). Figure 1
Conclusion:
For every cut-off level of Lung-RADS, the McWilliams model yields superior specificity to reduce unnecessary work-up for benign nodules, and higher sensitivity to predict malignancy. The McWilliams model seems to be a better tool than Lung-RADS to provide a malignancy risk, thus reducing unnecessary work-up and helping radiologists determine which subgroup of nodules detected in a screening setting need more invasive work-up.
