Abstract
Current computer-aided diagnosis (CAD) models for determining pulmonary nodule malignancy characterize nodule shape, density, and border in computed tomography (CT) data. Analyzing the lung parenchyma surrounding the nodule has been minimally explored. We hypothesize that improved nodule classification is achievable by including features quantified from the surrounding lung tissue. To explore this hypothesis, we have developed expanded quantitative CT feature extraction techniques, including volumetric Laws texture energy measures for the parenchyma and nodule, border descriptors using ray-casting and rubber-band straightening, histogram features characterizing densities, and global lung measurements. Using stepwise forward selection and leave-one-case-out cross-validation, a neural network was used for classification. When applied to 50 nodules (22 malignant and 28 benign) from high-resolution CT scans, 52 features (8 nodule, 39 parenchymal, and 5 global) were statistically significant. Nodule-only features yielded an area under the ROC curve of 0.918 (including nodule size) and 0.872 (excluding nodule size). Performance was improved through inclusion of parenchymal (0.938) and global features (0.932). These results show a trend toward increased performance when the parenchyma is included, coupled with the large number of significant parenchymal features that support our hypothesis: the pulmonary parenchyma is influenced differentially by malignant versus benign nodules, assisting CAD-based nodule characterizations.
Highlights
Lung cancer is the leading cause of cancer-related death, killing 1.37 million people in the world in 2008 alone.[1]
For the National Lung Screening Trial (NLST) study, lung cancer diagnoses were tracked as a part of the primary study outcomes; COPDGene subjects were separately contacted and consented in order to collect details related to nodules detected on their computed tomography (CT) studies that were acquired to study chronic obstructive pulmonary disease (COPD)
We expect in the future, especially within the lung cancer screening cohort, that this size bias between the malignant and benign cohorts will not be significant, as we describe in Sec. 4, and in this study, computer-aided diagnosis (CAD) tools were cross compared both incorporating and excluding size features
Summary
Lung cancer is the leading cause of cancer-related death, killing 1.37 million people in the world in 2008 alone.[1] While the overall 5-year survival rate of lung cancer is 15.9%, this statistic greatly improves with early diagnosis, up to 52% for the earliest stage of lung cancer.[2] The significant improvement in survival with earlier diagnosis has led to the implementation of screening for high-risk asymptomatic individuals, such as smokers and past smokers. The National Lung Screening Trial (NLST) found that using CT to screen for lung cancer reduced lung cancer mortality by 20% compared to screening using projection radiograph.[3] 96.4% of the nodules marked as suspicious on CT were found to be benign upon further evaluation.[4] These false positives result in unnecessary and invasive follow-up procedures and costs while incurring additional emotional stress for the patient
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
