Computerized Medical Imaging and Graphics最新文献

Cerebral X-ray digital subtraction angiography (DSA) is a widely used imaging technique in patients with neurovascular disease, allowing for vessel and flow visualization with high spatio-temporal resolution. Automatic artery–vein segmentation in DSA plays a fundamental role in vascular analysis with quantitative biomarker extraction, facilitating a wide range of clinical applications. The widely adopted U-Net applied on static DSA frames often struggles with disentangling vessels from subtraction artifacts. Further, it falls short in effectively separating arteries and veins as it disregards the temporal perspectives inherent in DSA. To address these limitations, we propose to simultaneously leverage spatial vasculature and temporal cerebral flow characteristics to segment arteries and veins in DSA. The proposed network, coined CAVE, encodes a 2D+time DSA series using spatial modules, aggregates all the features using temporal modules, and decodes it into 2D segmentation maps. On a large multi-center clinical dataset, CAVE achieves a vessel segmentation Dice of 0.84 ($\pm$0.04) and an artery–vein segmentation Dice of 0.79 ($\pm$0.06). CAVE surpasses traditional Frangi-based $k$-means clustering (P $<$ 0.001) and U-Net (P $<$ 0.001) by a significant margin, demonstrating the advantages of harvesting spatio-temporal features. This study represents the first investigation into automatic artery–vein segmentation in DSA using deep learning. The code is publicly available at https://github.com/RuishengSu/CAVE_DSA.

Accurate segmentation of cerebrovascular structures from Computed Tomography Angiography (CTA), Magnetic Resonance Angiography (MRA), and Digital Subtraction Angiography (DSA) is crucial for clinical diagnosis of cranial vascular diseases. Recent advancements in deep Convolution Neural Network (CNN) have significantly improved the segmentation process. However, training segmentation networks for all modalities requires extensive data labeling for each modality, which is often expensive and time-consuming. To circumvent this limitation, we introduce an approach to train cross-modality cerebrovascular segmentation network based on paired data from source and target domains. Our approach involves training a universal vessel segmentation network with manually labeled source domain data, which automatically produces initial labels for target domain training images. We improve the initial labels of target domain training images by fusing paired images, which are then used to refine the target domain segmentation network. A series of experimental arrangements is presented to assess the efficacy of our method in various practical application scenarios. The experiments conducted on an MRA-CTA dataset and a DSA-CTA dataset demonstrate that the proposed method is effective for cross-modality cerebrovascular segmentation and achieves state-of-the-art performance.

Accurate reconstruction of a high-resolution 3D volume of the heart is critical for comprehensive cardiac assessments. However, cardiac magnetic resonance (CMR) data is usually acquired as a stack of 2D short-axis (SAX) slices, which suffers from the inter-slice misalignment due to cardiac motion and data sparsity from large gaps between SAX slices. Therefore, we aim to propose an end-to-end deep learning (DL) model to address these two challenges simultaneously, employing specific model components for each challenge. The objective is to reconstruct a high-resolution 3D volume of the heart ($V_{\mathrm{HR}}$) from acquired CMR SAX slices ($V_{\mathrm{LR}}$). We define the transformation from $V_{\mathrm{LR}}$ to $V_{\mathrm{HR}}$ as a sequential process of motion correction and super-resolution. Accordingly, our DL model incorporates two distinct components. The first component conducts motion correction by predicting displacement vectors to re-position each SAX slice accurately. The second component takes the motion-corrected SAX slices from the first component and performs the super-resolution to fill the data gaps. These two components operate in a sequential way, and the entire model is trained end-to-end. Our model significantly reduced inter-slice misalignment from originally 3.33$\pm$0.74 mm to 1.36$\pm$0.63 mm and generated accurate high resolution 3D volumes with Dice of 0.974$\pm$0.010 for left ventricle (LV) and 0.938$\pm$0.017 for myocardium in a simulation dataset. When compared to the LAX contours in a real-world dataset, our model achieved Dice of 0.945$\pm$0.023 for LV and 0.786$\pm$0.060 for myocardium. In both datasets, our model with specific components for motion correction and super-resolution significantly enhance the performance compared to the model without such design considerations. The codes for our model are available at https://github.com/zhennongchen/CMR_MC_SR_End2End.

Rib cross-sectional shapes (characterized by the outer contour and cortical bone thickness) affect the rib mechanical response under impact loading, thereby influence the rib injury pattern and risk. A statistical description of the rib shapes or their correlations to anthropometrics is a prerequisite to the development of numerical human body models representing target demographics. Variational autoencoders (VAE) as anatomical shape generators remain to be explored in terms of utilizing the latent vectors to control or interpret the representativeness of the generated results. In this paper, we propose a pipeline for developing a multi-rib cross-sectional shape generative model from CT images, which consists of the achievement of rib cross-sectional shape data from CT images using an anatomical indexing system and regular grids, and a unified framework to fit shape distributions and associate shapes to anthropometrics for different rib categories. Specifically, we collected CT images including 3193 ribs, surface regular grid is generated for each rib based on anatomical coordinates, the rib cross-sectional shapes are characterized by nodal coordinates and cortical bone thickness. The tensor structure of shape data based on regular grids enable the implementation of CNNs in the conditional variational autoencoder (CVAE). The CVAE is trained against an auxiliary classifier to decouple the low-dimensional representations of the inter- and intra- variations and fit each intra-variation by a Gaussian distribution simultaneously. Random tree regressors are further leveraged to associate each continuous intra-class space with the corresponding anthropometrics of the subjects, i.e., age, height and weight. As a result, with the rib class labels and the latent vectors sampled from Gaussian distributions or predicted from anthropometrics as the inputs, the decoder can generate valid rib cross-sectional shapes of given class labels (male/female, 2nd to 11th ribs) for arbitrary populational percentiles or specific age, height and weight, which paves the road for future biomedical and biomechanical studies considering the diversity of rib shapes across the population.

Dual-energy computed tomography (CT) is an excellent substitute for identifying bone marrow edema in magnetic resonance imaging. However, it is rarely used in practice owing to its low contrast. To overcome this problem, we constructed a framework based on deep learning techniques to screen for diseases using axial bone images and to identify the local positions of bone lesions. To address the limited availability of labeled samples, we developed a new generative adversarial network (GAN) that extends expressions beyond conventional augmentation (CA) methods based on geometric transformations. We theoretically and experimentally determined that combining the concepts of data augmentation optimized for GAN training (DAG) and Wasserstein GAN yields a considerably stable generation of synthetic images and effectively aligns their distribution with that of real images, thereby achieving a high degree of similarity. The classification model was trained using real and synthetic samples. Consequently, the GAN technique used in the diagnostic test had an improved F1 score of approximately 7.8% compared with CA. The final F1 score was 80.24%, and the recall and precision were 84.3% and 88.7%, respectively. The results obtained using the augmented samples outperformed those obtained using pure real samples without augmentation. In addition, we adopted explainable AI techniques that leverage a class activation map (CAM) and principal component analysis to facilitate visual analysis of the network’s results. The framework was designed to suggest an attention map and scattering plot to visually explain the disease predictions of the network.

A late post-traumatic seizure (LPTS), a consequence of traumatic brain injury (TBI), can potentially evolve into a lifelong condition known as post-traumatic epilepsy (PTE). Presently, the mechanism that triggers epileptogenesis in TBI patients remains elusive, inspiring the epilepsy community to devise ways to predict which TBI patients will develop PTE and to identify potential biomarkers. In response to this need, our study collected comprehensive, longitudinal multimodal data from 48 TBI patients across multiple participating institutions. A supervised binary classification task was created, contrasting data from LPTS patients with those without LPTS. To accommodate missing modalities in some subjects, we took a two-pronged approach. Firstly, we extended a graphical model-based Bayesian estimator to directly classify subjects with incomplete modality. Secondly, we explored conventional imputation techniques. The imputed multimodal information was then combined, following several fusion and dimensionality reduction techniques found in the literature, and subsequently fitted to a kernel- or a tree-based classifier. For this fusion, we proposed two new algorithms: recursive elimination of correlated components (RECC) that filters information based on the correlation between the already selected features, and information decomposition and selective fusion (IDSF), which effectively recombines information from decomposed multimodal features. Our cross-validation findings showed that the proposed IDSF algorithm delivers superior performance based on the area under the curve (AUC) score. Ultimately, after rigorous statistical comparisons and interpretable machine learning examination using Shapley values of the most frequently selected features, we recommend the two following magnetic resonance imaging (MRI) abnormalities as potential biomarkers: the left anterior limb of internal capsule in diffusion MRI (dMRI), and the right middle temporal gyrus in functional MRI (fMRI).

Due to the high expenses involved, 4D-CT data for certain patients may only include five respiratory phases (0%, 20%, 40%, 60%, and 80%). This limitation can affect the subsequent planning of radiotherapy due to the absence of lung tumor information for the remaining five respiratory phases (10%, 30%, 50%, 70%, and 90%). This study aims to develop an interpolation method that can automatically derive tumor boundary contours for the five omitted phases using the available 5-phase 4D-CT data. The dynamic mode decomposition (DMD) method is a data-driven and model-free technique that can extract dynamic information from high-dimensional data. It enables the reconstruction of long-term dynamic patterns using only a limited number of time snapshots. The quasi-periodic motion of a deformable lung tumor caused by respiratory motion makes it suitable for treatment using DMD. The direct application of the DMD method to analyze the respiratory motion of the tumor is impractical because the tumor is three-dimensional and spans multiple CT slices. To predict the respiratory movement of lung tumors, a method called uniform angular interval (UAI) sampling was developed to generate snapshot vectors of equal length, which are suitable for DMD analysis. The effectiveness of this approach was confirmed by applying the UAI-DMD method to the 4D-CT data of ten patients with lung cancer. The results indicate that the UAI-DMD method effectively approximates the lung tumor’s deformable boundary surface and nonlinear motion trajectories. The estimated tumor centroid is within 2 mm of the manually delineated centroid, a smaller margin of error compared to the traditional BSpline interpolation method, which has a margin of 3 mm. This methodology has the potential to be extended to reconstruct the 20-phase respiratory movement of a lung tumor based on dynamic features from 10-phase 4D-CT data, thereby enabling more accurate estimation of the planned target volume (PTV).

Semi-supervised learning has made significant progress in medical image segmentation. However, existing methods primarily utilize information from a single dimensionality, resulting in sub-optimal performance on challenging magnetic resonance imaging (MRI) data with multiple segmentation objects and anisotropic resolution. To address this issue, we present a Hybrid Dual Mean-Teacher (HD-Teacher) model with hybrid, semi-supervised, and multi-task learning to achieve effective semi-supervised segmentation. HD-Teacher employs a 2D and a 3D mean-teacher network to produce segmentation labels and signed distance fields from the hybrid information captured in both dimensionalities. This hybrid mechanism allows HD-Teacher to utilize features from 2D, 3D, or both dimensions as needed. Outputs from 2D and 3D teacher models are dynamically combined based on confidence scores, forming a single hybrid prediction with estimated uncertainty. We propose a hybrid regularization module to encourage both student models to produce results close to the uncertainty-weighted hybrid prediction to further improve their feature extraction capability. Extensive experiments of binary and multi-class segmentation conducted on three MRI datasets demonstrated that the proposed framework could (1) significantly outperform state-of-the-art semi-supervised methods (2) surpass a fully-supervised VNet trained on substantially more annotated data, and (3) perform on par with human raters on muscle and bone segmentation task. Code will be available at https://github.com/ThisGame42/Hybrid-Teacher.

Cardiovascular MRI (CMRI) is a non-invasive imaging technique adopted for assessing the blood circulatory system’s structure and function. Precise image segmentation is required to measure cardiac parameters and diagnose abnormalities through CMRI data. Because of anatomical heterogeneity and image variations, cardiac image segmentation is a challenging task. Quantification of cardiac parameters requires high-performance segmentation of the left ventricle (LV), right ventricle (RV), and left ventricle myocardium from the background. The first proposed solution here is to manually segment the regions, which is a time-consuming and error-prone procedure. In this context, many semi- or fully automatic solutions have been proposed recently, among which deep learning-based methods have revealed high performance in segmenting regions in CMRI data. In this study, a self-adaptive multi attention (SMA) module is introduced to adaptively leverage multiple attention mechanisms for better segmentation. The convolutional-based position and channel attention mechanisms with a patch tokenization-based vision transformer (ViT)-based attention mechanism in a hybrid and end-to-end manner are integrated into the SMA. The CNN- and ViT-based attentions mine the short- and long-range dependencies for more precise segmentation. The SMA module is applied in an encoder-decoder structure with a ResNet50 backbone named CardSegNet. Furthermore, a deep supervision method with multi-loss functions is introduced to the CardSegNet optimizer to reduce overfitting and enhance the model’s performance. The proposed model is validated on the ACDC2017 (n=100), M&Ms (n=321), and a local dataset (n=22) using the 10-fold cross-validation method with promising segmentation results, demonstrating its outperformance versus its counterparts.

Current methods of digital pathological images typically employ small image patches to learn local representative features to overcome the issues of computationally heavy and memory limitations. However, the global contextual features are not fully considered in whole-slide images (WSIs). Here, we designed a hybrid model that utilizes Graph Neural Network (GNN) module and Transformer module for the representation of global contextual features, called TransGNN. GNN module built a WSI-Graph for the foreground area of a WSI for explicitly capturing structural features, and the Transformer module through the self-attention mechanism implicitly learned the global context information. The prognostic markers of hepatocellular carcinoma (HCC) prognostic biomarkers were used to illustrate the importance of global contextual information in cancer histopathological analysis. Our model was validated using 362 WSIs from 355 HCC patients diagnosed from The Cancer Genome Atlas (TCGA). It showed impressive performance with a Concordance Index (C-Index) of 0.7308 (95% Confidence Interval (CI): (0.6283–0.8333)) for overall survival prediction and achieved the best performance among all models. Additionally, our model achieved an area under curve of 0.7904, 0.8087, and 0.8004 for 1-year, 3-year, and 5-year survival predictions, respectively. We further verified the superior performance of our model in HCC risk stratification and its clinical value through Kaplan–Meier curve and univariate and multivariate COX regression analysis. Our research demonstrated that TransGNN effectively utilized the context information of WSIs and contributed to the clinical prognostic evaluation of HCC.