David F. Yankelevitz

The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI): A Completed Reference Database of Lung Nodules on CT Scans

PURPOSEnThe development of computer-aided diagnostic (CAD) methods for lung nodule detection, classification, and quantitative assessment can be facilitated through a well-characterized repository of computed tomography (CT) scans. The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI) completed such a database, establishing a publicly available reference for the medical imaging research community. Initiated by the National Cancer Institute (NCI), further advanced by the Foundation for the National Institutes of Health (FNIH), and accompanied by the Food and Drug Administration (FDA) through active participation, this public-private partnership demonstrates the success of a consortium founded on a consensus-based process.nnnMETHODSnSeven academic centers and eight medical imaging companies collaborated to identify, address, and resolve challenging organizational, technical, and clinical issues to provide a solid foundation for a robust database. The LIDC/IDRI Database contains 1018 cases, each of which includes images from a clinical thoracic CT scan and an associated XML file that records the results of a two-phase image annotation process performed by four experienced thoracic radiologists. In the initial blinded-read phase, each radiologist independently reviewed each CT scan and marked lesions belonging to one of three categories (nodule > or =3 mm, nodule <3 mm, and non-nodule > or =3 mm). In the subsequent unblinded-read phase, each radiologist independently reviewed their own marks along with the anonymized marks of the three other radiologists to render a final opinion. The goal of this process was to identify as completely as possible all lung nodules in each CT scan without requiring forced consensus.nnnRESULTSnThe Database contains 7371 lesions marked nodule by at least one radiologist. 2669 of these lesions were marked nodule > or =3 mm by at least one radiologist, of which 928 (34.7%) received such marks from all four radiologists. These 2669 lesions include nodule outlines and subjective nodule characteristic ratings.nnnCONCLUSIONSnThe LIDC/IDRI Database is expected to provide an essential medical imaging research resource to spur CAD development, validation, and dissemination in clinical practice.

The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI)

PURPOSEnThe development of computer-aided diagnostic (CAD) methods for lung nodule detection, classification, and quantitative assessment can be facilitated through a well-characterized repository of computed tomography (CT) scans. The Lung Image Database Consortium (LIDC) and Image Database Resource Initiative (IDRI) completed such a database, establishing a publicly available reference for the medical imaging research community. Initiated by the National Cancer Institute (NCI), further advanced by the Foundation for the National Institutes of Health (FNIH), and accompanied by the Food and Drug Administration (FDA) through active participation, this public-private partnership demonstrates the success of a consortium founded on a consensus-based process.nnnMETHODSnSeven academic centers and eight medical imaging companies collaborated to identify, address, and resolve challenging organizational, technical, and clinical issues to provide a solid foundation for a robust database. The LIDC/IDRI Database contains 1018 cases, each of which includes images from a clinical thoracic CT scan and an associated XML file that records the results of a two-phase image annotation process performed by four experienced thoracic radiologists. In the initial blinded-read phase, each radiologist independently reviewed each CT scan and marked lesions belonging to one of three categories (nodule > or =3 mm, nodule <3 mm, and non-nodule > or =3 mm). In the subsequent unblinded-read phase, each radiologist independently reviewed their own marks along with the anonymized marks of the three other radiologists to render a final opinion. The goal of this process was to identify as completely as possible all lung nodules in each CT scan without requiring forced consensus.nnnRESULTSnThe Database contains 7371 lesions marked nodule by at least one radiologist. 2669 of these lesions were marked nodule > or =3 mm by at least one radiologist, of which 928 (34.7%) received such marks from all four radiologists. These 2669 lesions include nodule outlines and subjective nodule characteristic ratings.nnnCONCLUSIONSnThe LIDC/IDRI Database is expected to provide an essential medical imaging research resource to spur CAD development, validation, and dissemination in clinical practice.

On measuring the change in size of pulmonary nodules

The pulmonary nodule is the most common manifestation of lung cancer, the most deadly of all cancers. Most small pulmonary nodules are benign, however, and currently the growth rate of the nodule provides for one of the most accurate noninvasive methods of determining malignancy. In this paper, we present methods for measuring the change in nodule size from two computed tomography image scans recorded at different times; from this size change the growth rate may be established. The impact of partial voxels for small nodules is evaluated and isotropic resampling is shown to improve measurement accuracy. Methods for nodule location and sizing, pleural segmentation, adaptive thresholding, image registration, and knowledge-based shape matching are presented. The latter three techniques provide for a significant improvement in volume change measurement accuracy by considering both image scans simultaneously. Improvements in segmentation are evaluated by measuring volume changes in benign or slow growing nodules. In the analysis of 50 nodules, the variance in percent volume change was reduced from 11.54% to 9.35% (p=0.03) through the use of registration, adaptive thresholding, and knowledge-based shape matching.

Characterization of pulmonary nodules: effects of size and feature type on reported performance

Differences in the size distribution of malignant and benign pulmonary nodules in databases used for training and testing characterization systems have a significant impact on the measured performance. The magnitude of this effect and methods to provide more relevant performance results are explored in this paper. Two- and three-dimensional features, both including and excluding size, and two classifiers, logistic regression and distance-weighted nearest-neighbors (dwNN), were evaluated on a database of 178 pulmonary nodules. For the full database, the area under the ROC curve (AUC) of the logistic regression classifier for 2D features with and without size was 0.721 and 0.614 respectively, and for 3D features with and without size, 0.773 and 0.737 respectively. In comparison, the performance using a simple size-threshold classifier was 0.675. In the second part of the study, the performance was measured on a subset of 46 nodules from the entire subset selected to have a similar size-distribution of malignant and benign nodules. For this subset, performance of the size-threshold was 0.504. For logistic regression, the performance for 2D, with and without size, were 0.578 and 0.478, and for 3D, with and without size, 0.671 and 0.767. Over all the databases, logistic regression exhibited better performance using 3D features than 2D features. This study suggests that in systems for nodule classification, size is responsible for a large part of the reported performance. To address this, system performance should be reported with respect to the performance of a size-threshold classifier.

PULMONARY NODULE CLASSIFICATION: SIZE DISTRIBUTION ISSUES

Automated nodule classification systems determine a model based on features extracted from documented databases of nodules. These databases cover a large size range and have an unequal distribution of malignant and benign nodules, leading to a high correlation between malignancy and size. For two recent studies in the literature, much of the reported performance of the system may be derived from size based on analysis of their size distributions. We performed experiments to determine the effect of unequal size distribution on a nodule classification systems performance. Preliminary results indicate that the performance across the entire dataset (a sensitivity/specificity of 0.85/0.80) does not generalize to a subset of nodules (0.50/0.80), but performance can be improved by specifically training on that subset (0.60/0.80). Additional testing with larger datasets needs to be performed, but results reported in this area are overly optimistic.

Characterization of solid pulmonary nodules using three-dimensional features

With the development of high-resolution, multirow-detector CT scanners, the prospects for diagnosing and treating lung cancer at an early stage are much improved. However, it is often difficult to determine whether a nodule, especially a small nodule, is malignant from a single CT scan. We developed a computer-aided diagnostic algorithm to distinguish benign from malignant solid nodules based on features that can be extracted from a single CT scan. Our method uses 3D geometric and densitometric moment analysis of a segmented nodule image and surface curvature from a polygonal surface model of the nodule. After excluding features directly related to size, we computed a total of 28 features. Prior to classification, the number of features was reduced through stepwise feature selection. The features are used by two classifiers, k-nearest-neighbors (k-NN) and logistic regression. We used 48 malignant nodules whose status was determined by biopsy or resection, and 55 benign nodules determined to be clinically stable through two years of no change or biopsy. The k-NN classifier achieved a sensitivity of 0.81 with a specificity of 0.76, while the logistic regression classifier achieved a sensitivity of 0.85 and a specificity of 0.80.

Initiative for Early Lung Cancer Research on Treatment: Development of Study Design and Pilot Implementation

Introduction: To maximize the benefits of computed tomographic screening for lung cancer, optimal treatment for small, early lung cancers is needed. Limiting the extent of surgery spares lung tissue, preserves pulmonary function, and decreases operative time, complications, and morbidities. It also increases the likelihood of resecting future new primary lung cancers. The goal is to assess alternative treatments in a timely manner. Methods: The focus sessions with patients and physicians separately highlighted the need to consider their perceptions. Literature reviews and analyses of treatment results using large databases were performed to formulate critical questions about long‐term treatment outcomes, recurrence, and quality of life of alternative treatments. Based on these analyses, the investigators developed a prospective multi‐institutional cohort study, the Initiative for Early Lung Cancer Research for Treatment, to compare treatments for stage I NSCLC. HIPAA compliant institutional review board approval was obtained and we performed a feasibility study of the first 206 surgical patients. Results: Lobectomy was performed in 89 (43.2%) patients, and sublobar resection was performed in 117 (56.7%) patients. Mediastinal lymph node resection was performed in 173 (84.0%) patients, 8 had N1 and 3 N2 lymph node metastases. Patients stated that both the surgeons opinion (93%) and the patients own opinion (93%) were extremely important, followed by the patients view that the chosen procedure would provide the best quality of life (90%). Conclusions: It was feasible to obtain pre‐ and postsurgical information from patients and surgeons. We anticipate statistically meaningful results about treatment alternatives in 3 to 5 years.

The Lung Image Database Consortium (LIDC) data collection process for nodule detection and annotation

The LIDC is developing a publicly available database of thoracic computed tomography (CT) scans as a medical imaging research resource. A unique multi-center data collection process and communication system were developed to share image data and to capture the location and spatial extent of lung nodules as marked by expert radiologists. A two-phase data collection process was designed to allow multiple radiologists at different centers to asynchronously review and annotate each CT image series. Four radiologists reviewed each case using this process. In the first or blinded phase, each radiologist reviewed the CT series independently. In the second or unblinded review phase, the results from all four blinded reviews are compiled and presented to each radiologist for a second review. This allows each radiologist to review their own annotations along with those of the other radiologists. The results from each radiologists unblinded review were compiled to form the final unblinded review. There is no forced consensus in this process. An XML-based message system was developed to communicate the results of each reading. This two-phase data collection process was designed, tested and implemented across the LIDC. It has been used for more than 130 CT cases that have been read and annotated by four expert readers and are publicly available at (http://ncia.nci.nih.gov). A data collection process was developed, tested and implemented that allowed multiple readers to review each case multiple times and that allowed each reader to observe the annotations of other readers.

Prediction of tumor volumes using an exponential model

Measurement of pulmonary nodule growth rate is important for the evaluation of lung cancer treatment. The change in nodule growth rate can be used as an indicator of the efficacy of a prescribed treatment. However, a change in growth rate may be due to actual physiological change, or it may be simply due to measurement error. To address this issue, we propose the use of an exponential model to predict the volume of a tumor based on two earlier scans. We examined 11 lung cancers presenting as solid pulmonary nodules that were not treated. Using 5 of these with optimal scan parameters, thin-slice (1.0mm or 1.25mm) with same axial resolution, we found an error ranging from 1.7% to 27.7%, with an average error of 14.9%. This indicates that we can estimate the growth of a lung cancer, as measured by CT, which includes the actual growth as well as the error due to the technique, by the amount indicated above. Using scans with non-optimal parameters, either thick-slice or different resolution thin-slice scans, resulted in errors ranging from 30% to 600%, suggesting that same resolution thin-slice CT scans are necessary for accurate measurement of nodule growth.