Background and Objectives: Lung cancer remains a leading cause of cancer mortality worldwide. Accurately classifying benign pulmonary nodules and malignant ones is crucial for early diagnosis and improved patient outcomes. The purpose of this study is to explore the deep-learning model of ResNet combined with a convolutional block attention module (CBAM) for the differentiation between benign and malignant lung cancer, based on computed tomography (CT) images, morphological features, and clinical information. Methods and materials: In this study, 8241 CT slices containing pulmonary nodules were retrospectively included. A random sample comprising 20% (n = 1647) of the images was used as the test set, and the remaining data were used as the training set. ResNet combined CBAM (ResNet-CBAM) was used to establish classifiers on the basis of images, morphological features, and clinical information. Nonsubsampled dual-tree complex contourlet transform (NSDTCT) combined with SVM classifier (NSDTCT-SVM) was used as a comparative model. Results: The AUC and the accuracy of the CBAM-ResNet model were 0.940 and 0.867, respectively, in test set when there were only images as inputs. By combining the morphological features and clinical information, CBAM-ResNet shows better performance (AUC: 0.957, accuracy: 0.898). In comparison, a radiomic analysis using NSDTCT-SVM achieved AUC and accuracy values of 0.807 and 0.779, respectively. Conclusions: Our findings demonstrate that deep-learning models, combined with additional information, can enhance the classification performance of pulmonary nodules. This model can assist clinicians in accurately diagnosing pulmonary nodules in clinical practice.

The basic emission process associated with PET imaging is Poisson in nature. Reconstructed images inherit some aspects of this—regional variability is typically proportional to the regional mean. Iterative reconstruction using expectationmaximization (EM), widely used in clinical imaging now, impose positivity constraints that impact noise properties. The present work is motivated by analysis of data from a physical phantom study of a PET/CT scanner in routine clinical use. Both traditional filtered back-projection (FBP) and EM reconstructions of the images are considered. FBP images are quite Gaussian but the EM reconstructions exhibit Gamma-like skewness. The Gamma structure has implications for how reconstructed PET images might be processed statistically. Post-reconstruction inference— model fitting and diagnostics for regions of interest are of particular interest. Although the relevant Gamma parameterization is not within the framework of generalized linear models (GLM), iteratively re-weighted least squares (IRLS) techniques, which are often used to find the maximum likelihood estimates of a GLM, can be adapted for analysis in this setting. Our work highlights the use of a Gamma-based probability transform in producing normalized residuals as model diagnostics. The approach is demonstrated for quality assurance analyses associated with physical phantom studies—recovering estimates of local bias and variance characteristics in an operational scanner. Numerical simulations show that when the Gamma assumption is reasonable, gains in efficiency are obtained. The work shows that the adaptation of standard analysis methods to accommodate the Gamma structure is straightforward and beneficial.

When a scanner is installed and begins to be used operationally, its actual performance may deviate somewhat from the predictions made at the design stage. Thus it is recommended that routine quality assurance (QA) measurements be used to provide an operational understanding of scanning properties. While QA data are primarily used to evaluate sensitivity and bias patterns, there is a possibility to also make use of such data sets for a more refined understanding of the 3-D scanning properties. Building on some recent work on analysis of the distributional characteristics of iteratively reconstructed PET data, we construct an auto-regression model for analysis of the 3-D spatial auto-covariance structure of iteratively reconstructed data, after normalization. Appropriate likelihood-based statistical techniques for estimation of the auto-regression model coefficients are described. The fitted model leads to a simple process for approximate simulation of scanner performance-one that is readily implemented in an R script. The analysis provides a practical mechanism for evaluating the operational error characteristics of iteratively reconstructed PET images. Simulation studies are used for validation. The approach is illustrated on QA data from an operational clinical scanner and numerical phantom data. We also demonstrate the potential for use of these techniques, as a form of model-based bootstrapping, to provide assessments of measurement uncertainties in variables derived from clinical FDG-PET scans. This is illustrated using data from a clinical scan in a lung cancer patient, after a 3-minute acquisition has been re-binned into three consecutive 1-minute time-frames. An uncertainty measure for the tumor SUVmax value is obtained. The methodology is seen to be practical and could be a useful support for quantitative decision making based on PET data.

Spatial transcriptomics (ST) has emerged as an efficient technology for mapping gene expression within tissue sections, offering informative spatial context for gene activities. However, most current ST techniques suffer from low spatial resolution, where each spatial location often contains cells of various types. Deconvolution methods are used to resolve the cell mixture within the spots, but conventional approaches rely on spot-by-spot analyses, which are limited by low gene expression levels and disregard spatial relationships between spots, ultimately reducing performance. Here, we introduce DECLUST, a cluster-based deconvolution method to accurately estimate the cell-type composition in ST data. The method identifies spatial clusters of spots using both gene expression and spatial coordinates, hence preserving the spatial structure of the tissue. Deconvolution is subsequently performed on the aggregated gene expression of individual clusters, mitigating the challenges associated with low expression levels in individual spots. We evaluate DECLUST on simulated ST datasets from a human breast cancer tissue and two real ST datasets from human ovarian cancer and mouse brain. We compare DECLUST to current methods including CARD, GraphST, Cell2location, and Tangram. The results indicate that DECLUST not only maintains the spatial integrity of tissues but also outperforms existing methods in terms of robustness and accuracy. In conclusion, DECLUST provides an effective and reliable approach for identifying cell-type compositions in ST data.