Radiomics is to provide quantitative descriptors of normal and abnormal tissues during classification and prediction tasks in radiology and oncology. Quantitative Imaging Network members are developing radiomic “feature” sets to characterize tumors, in general, the size, shape, texture, intensity, margin, and other aspects of the imaging features of nodules and lesions. Efforts are ongoing for developing an ontology to describe radiomic features for lung nodules, with the main classes consisting of size, local and global shape descriptors, margin, intensity, and texture-based features, which are based on wavelets, Laplacian of Gaussians, Law’s features, gray-level co-occurrence matrices, and run-length features. The purpose of this study is to investigate the sensitivity of quantitative descriptors of pulmonary nodules to segmentations and to illustrate comparisons across different feature types and features computed by different implementations of feature extraction algorithms. We calculated the concordance correlation coefficients of the features as a measure of their stability with the underlying segmentation; 68% of the 830 features in this study had a concordance CC of ≥0.75. Pairwise correlation coefficients between pairs of features were used to uncover associations between features, particularly as measured by different participants. A graphical model approach was used to enumerate the number of uncorrelated feature groups at given thresholds of correlation. At a threshold of 0.75 and 0.95, there were 75 and 246 subgroups, respectively, providing a measure for the features’ redundancy.

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.

Purpose This work aims to develop a new framework of image quality assessment using deep learning-based model observer (DL-MO) and to validate it in a low-contrast lesion detection task that involves CT images with patient anatomical background. Methods The DL-MO was developed using the transfer learning strategy to incorporate a pretrained deep convolutional neural network (CNN), a partial least square regression discriminant analysis (PLS-DA) model and an internal noise component. The CNN was previously trained to achieve the state-of-the-art classification accuracy over a natural image database. The earlier layers of the CNN were used as a deep feature extractor, with the assumption that similarity exists between the CNN and the human visual system. The PLSR model was used to further engineer the deep feature for the lesion detection task in CT images. The internal noise component was incorporated to model the inefficiency and variability of human observer (HO) performance, and to generate the ultimate DL-MO test statistics. Seven abdominal CT exams were retrospectively collected from the same type of CT scanners. To compare DL-MO with HO, 12 experimental conditions with varying lesion size, lesion contrast, radiation dose, and reconstruction types were generated, each condition with 154 trials. CT images of a real liver metastatic lesion were numerically modified to generate lesion models with four lesion sizes (5, 7, 9, and 11 mm) and three contrast levels (15, 20, and 25 HU). The lesions were inserted into patient liver images using a projection-based method. A validated noise insertion tool was used to synthesize CT exams with 50% and 25% of routine radiation dose level. CT images were reconstructed using the weighted filtered back projection algorithm and an iterative reconstruction algorithm. Four medical physicists performed a two-alternative forced choice (2AFC) detection task (with multislice scrolling viewing mode) on patient images across the 12 experimental conditions. DL-MO was operated on the same datasets. Statistical analyses were performed to evaluate the correlation and agreement between DL-MO and HO. Results A statistically significant positive correlation was observed between DL-MO and HO for the 2AFC low-contrast detection task that involves patient liver background. The corresponding Pearson product moment correlation coefficient was 0.986 [95% confidence interval (0.950, 0.996)]. Bland–Altman agreement analysis did not indicate statistically significant differences. Conclusions The proposed DL-MO is highly correlated with HO in a low-contrast detection task that involves realistic patient liver background. This study demonstrated the potential of the proposed DL-MO to assess image quality directly based on patient images in realistic, clinically relevant CT tasks.

The purpose of this study was to determine the correlation between human observer performance for localization of small low contrast lesions within uniform water background versus an anatomical liver background, under the conditions of varying dose, lesion size, and reconstruction algorithm. Liver lesions (5 mm, 7 mm, and 9 mm, contrast: −21 HU) were digitally inserted into CT projection data of ten normal patients in vessel-free liver regions. Noise was inserted into the projection data to create three image sets: full dose and simulated half and quarter doses. Images were reconstructed with a standard filtered back projection (FBP) and an iterative reconstruction (IR) algorithm. Lesion and noise insertion procedures were repeated with water phantom data. Two-dimensional regions of interest (66 lesion-present, 34 lesion-absent) were selected, randomized, and independently reviewed by three medical physicists to identify the most likely location of the lesion and provide a confidence score. Locations and confidence scores were assessed using the area under the localization receiver operating characteristic curve (AzLROC). We examined the correlation between human performance for the liver and uniform water backgrounds as dose, lesion size, and reconstruction algorithm varied. As lesion size or dose increased, reader localization performance improved. For full dose IR images, the AzLROC for 5, 7, and 9 mm lesions were 0.53, 0.91, and 0.97 (liver) and 0.51, 0.96, and 0.99 (uniform water), respectively. Similar trends were seen with other parameters. Performance values for liver and uniform backgrounds were highly correlated for both reconstruction algorithms, with a Spearman correlation of ρ = 0.97, and an average difference in AzLROC of 0.05 ± 0.04. For the task of localizing low contrast liver lesions, human observer performance was highly correlated between anatomical and uniform backgrounds, suggesting that lesion localization studies emulating a clinical test of liver lesion detection can be performed using a uniform background.

Determination of the effect of protocol modifications on diagnostic performance in CT with human observers is extremely time-consuming, limiting the applicability of such methods in routine clinical practice. In this work, we sought to determine whether a channelized Hotelling observer (CHO) could predict human observer performance for the task of liver lesion localization as background, reconstruction algorithm, dose, and lesion size were varied. Liver lesions (5 mm, 7 mm, and 9 mm) were digitally inserted into the CT projection data of patients with normal livers and water phantoms. The projection data were reconstructed with filtered back projection (FBP) and iterative reconstruction (IR) algorithms for three dose levels: full dose (liver CTDIvol = 10.5 ± 8.5 mGy, water phantom CTDIvol = 9.6 ± 0.1 mGy) and simulated half and quarter doses. For each of 36 datasets (3 dose levels x 2 reconstruction algorithms × 2 backgrounds × 3 sizes), 66 signal-present and 34 signal-absent 2D images were extracted from the reconstructed volumes. Three medical physicists independently reviewed each dataset and noted the lesion location and a confidence score for each image. A CHO with Gabor channels was calculated to estimate the performance for each of the 36 localization tasks. The CHO performances, quantified using localization receiver operating characteristic (LROC) analysis, were compared to the human observer performances. Performance values between human and model observers were highly correlated for equivalent parameters (same lesion size, dose, background, and reconstruction), with a Spearman’s correlation coefficient of 0.93 (95% CI: 0.82–0.98). CHO performance values for the uniform background were strongly correlated (ρ = 0.94, CI: 0.80–1.0) with the human observer performance values for the liver background. Performance values between human observers and CHO were highly correlated as dose, reconstruction type and object size were varied for the task of localization of patient liver lesions in both uniform and liver backgrounds.