No abstract

Loss of the NF1 tumor suppressor gene causes the autosomal dominant condition, neurofibromatosis type 1 (NF1). Children and adults with NF1 suffer from pathologies including benign and malignant tumors to cognitive deficits, seizures, growth abnormalities, and peripheral neuropathies. NF1 encodes neurofibromin, a Ras-GTPase activating protein, and NF1 mutations result in hyperactivated Ras signaling in patients. Existing NF1 mutant mice mimic individual aspects of NF1, but none comprehensively models the disease. We describe a potentially novel Yucatan miniswine model bearing a heterozygotic mutation in NF1 (exon 42 deletion) orthologous to a mutation found in NF1 patients. NF1+/ex42del miniswine phenocopy the wide range of manifestations seen in NF1 patients, including café au lait spots, neurofibromas, axillary freckling, and neurological defects in learning and memory. Molecular analyses verified reduced neurofibromin expression in swine NF1+/ex42del fibroblasts, as well as hyperactivation of Ras, as measured by increased expression of its downstream effectors, phosphorylated ERK1/2, SIAH, and the checkpoint regulators p53 and p21. Consistent with altered pain signaling in NF1, dysregulation of calcium and sodium channels was observed in dorsal root ganglia expressing mutant NF1. Thus, these NF1+/ex42del miniswine recapitulate the disease and provide a unique, much-needed tool to advance the study and treatment of NF1.

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 Quantitative computed tomography (CT) measures are increasingly being developed and used to characterize lung disease. With recent advances in CT technologies, we sought to evaluate the quantitative accuracy of lung imaging at low and ultra-low radiation doses with the use of iterative reconstruction (IR), tube current modulation (TCM), and spectral shaping. Methods We investigated the effect of five independent CT protocols reconstructed with IR on quantitative airway measures and global lung measures using an in-vivo large animal model as a human subject surrogate. A control protocol was chosen (NIH-SPIROMICS + TCM) and five independent protocols investigating TCM, low and ultra-low radiation dose, and spectral shaping. For all scans quantitative global parenchymal measurements (mean, median and standard deviation of the parenchymal HU, along with measures of emphysema) and global airway measurements (number of segmented airways and pi10) were generated. In addition, selected individual airway measurements (minor and major inner diameter, wall thickness, inner and outer area, inner and outer perimeter, wall area fraction, and inner equivalent circle diameter) were evaluated. Comparisons were made between control and target protocols using difference and repeatability measures. Results Estimated CT volume dose index (CTDIvol) across all protocols ranged from 7.32 mGy to 0.32 mGy. Low and ultra-low dose protocols required more manual editing and resolved fewer airway branches; yet, comparable pi10 whole lung measures were observed across all protocols. Similar trends in acquired parenchymal and airway measurements were observed across all protocols, with increased measurement differences using the ultra-low dose protocols. However, for small airways (1.9 ± 0.2mm) and medium airways (5.7 ± 0.4mm) the measurement differences across all protocols were comparable to the control protocol repeatability across breath-holds. Diameters, wall thickness, wall area fraction, and equivalent diameter had smaller measurement differences, than area and perimeter measurements. Conclusions In conclusion, the use of IR with low and ultra-low dose CT protocols with CT volume dose indices down to 0.32 mGy maintains selected quantitative parenchymal and airway measurements relevant to pulmonary disease characterization.