Lung cancer screening aims to detect small pulmonary nodules and decrease the mortality rate of those affected. However, studies from large-scale clinical trials of lung cancer screening have shown that the false-positive rate is high and positive predictive value is low. To address these problems, a technical approach is greatly needed for accurate malignancy differentiation among these early-detected nodules. We studied the clinical feasibility of an additional protocol of localized thin-section CT for further assessment on recalled patients from lung cancer screening tests. Our approach of localized thin-section CT was integrated with radiomics features extraction and machine learning classification which was supervised by pathological diagnosis. Localized thin-section CT images of 122 nodules were retrospectively reviewed and 374 radiomics features were extracted. In this study, 48 nodules were benign and 74 malignant. There were nine patients with multiple nodules and four with synchronous multiple malignant nodules. Different machine learning classifiers with a stratified ten-fold cross-validation were used and repeated 100 times to evaluate classification accuracy. Of the image features extracted from the thin-section CT images, 238 (64%) were useful in differentiating between benign and malignant nodules. These useful features include CT density ( p = 0.002 518), sigma ( p = 0.002 781), uniformity ( p = 0.032 41), and entropy ( p = 0.006 685). The highest classification accuracy was 79% by the logistic classifier. The performance metrics of this logistic classification model was 0.80 for the positive predictive value, 0.36 for the false-positive rate, and 0.80 for the area under the receiver operating characteristic curve. Our approach of direct risk classification supervised by the pathological diagnosis with localized thin-section CT and radiomics feature extraction may support clinical physicians in determining truly malignant nodules and therefore reduce problems in lung cancer screening.
Purpose: A local damage model (LDM) was developed to estimate the biological efficiency of Auger-electron-emitting radionuclides. Materials and methods: The LDM required information on the local dose distribution, local energy spectrum, and clustered DNA damage yields in the cell nucleus. To apply the model, the nucleus was divided into concentric shells where each shell contributed its own local dose, energy spectrum, and damage yield. The local doses and energy spectra were computed using the PENELOPE (PENetration and Energy LOss of Positrons and Electrons) code. The DNA damage yields were estimated using the MCDS (Monte Carlo damage simulation) code. Results: For a typical 4-μm radius mammalian cell nucleus, the absorbed doses to the cell nucleus per unit cumulated activity, equal to 0.0065, 0.00418, 0.0028, 0.0027 and 0.0015 Gy Bq−1 s−1 for 125I, 119Sb, 123I, 111In and 99mTc, were within 6% difference with the MIRD (Medical Internal Radiation Dose) published data. The simulated total (simple and complex) single-strand break (SSB) and double-strand break (DSB) yields were in the same order, i.e., 125I > 119Sb > 123I > 111In > 99mTc. The agreement between present results and literature data for the DNA damage yields was generally good. More than 75% of the total SSB and DSB yields were contributed from regions within 2.5 μm of the nucleus center. Conclusions: The proposed methodology was computationally efficient and could be applied to other irradiation geometries such as cell clusters.
Previous studies have shown that the geometric development of femoral trabecular bone is affected by insufficient dietary intake of magnesium. However, it is not clear whether the development of femoral cortical bone can be quantitatively evaluated according to a diet with inadequate magnesium supplementation. Therefore, we used a micro computed tomography (CT) imaging approach with a laboratory mouse model to explore the potential application of texture analysis for the quantitative assessment of femoral cortical bones. C57BL/6J male mice were divided into two groups, where one group was fed a normal diet and the other group was fed a low-magnesium diet. We used a micro CT scanner for image acquisition, and the subsequent development of cortical bone was examined by texture analysis based on the statistical distribution of gray-scale intensities in which seven essential parameters were extracted from the micro CT images. Our calculations showed that the mean intensity increased by 7.20% (p = 0.000134), sigma decreased by 29.18% (p = 1.98E-12), skewness decreased by 19.52% (p = 0.0000205), kurtosis increased by 9.62% (p = 0.0877), energy increased by 24.19% (p = 3.32E-09), entropy decreased by 6.14% (p = 3.00E-10), and the Nakagami parameter increased by 104.32% (p = 4.13E-12) in the low-magnesium group when compared to the normal group. We found that the statistical parameters extracted from the gray-scale intensity distribution were able to differentiate between femoral cortical bone developments in the two different diet groups.