Visual Computing for Industry Biomedicine and Art最新文献

Low-dose computed tomography (LDCT) has gained increasing attention owing to its crucial role in reducing radiation exposure in patients. However, LDCT-reconstructed images often suffer from significant noise and artifacts, negatively impacting the radiologists' ability to accurately diagnose. To address this issue, many studies have focused on denoising LDCT images using deep learning (DL) methods. However, these DL-based denoising methods have been hindered by the highly variable feature distribution of LDCT data from different imaging sources, which adversely affects the performance of current denoising models. In this study, we propose a parallel processing model, the multi-encoder deep feature transformation network (MDFTN), which is designed to enhance the performance of LDCT imaging for multisource data. Unlike traditional network structures, which rely on continual learning to process multitask data, the approach can simultaneously handle LDCT images within a unified framework from various imaging sources. The proposed MDFTN consists of multiple encoders and decoders along with a deep feature transformation module (DFTM). During forward propagation in network training, each encoder extracts diverse features from its respective data source in parallel and the DFTM compresses these features into a shared feature space. Subsequently, each decoder performs an inverse operation for multisource loss estimation. Through collaborative training, the proposed MDFTN leverages the complementary advantages of multisource data distribution to enhance its adaptability and generalization. Numerous experiments were conducted on two public datasets and one local dataset, which demonstrated that the proposed network model can simultaneously process multisource data while effectively suppressing noise and preserving fine structures. The source code is available at https://github.com/123456789ey/MDFTN .

Early diagnosis and accurate prognosis of colorectal cancer is critical for determining optimal treatment plans and maximizing patient outcomes, especially as the disease progresses into liver metastases. Computed tomography (CT) is a frontline tool for this task; however, the preservation of predictive radiomic features is highly dependent on the scanning protocol and reconstruction algorithm. We hypothesized that image reconstruction with a high-frequency kernel could result in a better characterization of liver metastases features via deep neural networks. This kernel produces images that appear noisier but preserve more sinogram information. A simulation pipeline was developed to study the effects of imaging parameters on the ability to characterize the features of liver metastases. This pipeline utilizes a fractal approach to generate a diverse population of shapes representing virtual metastases, and then it superimposes them on a realistic CT liver region to perform a virtual CT scan using CatSim. Datasets of 10,000 liver metastases were generated, scanned, and reconstructed using either standard or high-frequency kernels. These data were used to train and validate deep neural networks to recover crafted metastases characteristics, such as internal heterogeneity, edge sharpness, and edge fractal dimension. In the absence of noise, models scored, on average, 12.2% ( $\alpha =0.012$ ) and 7.5% ( $\alpha =0.049)$ lower squared error for characterizing edge sharpness and fractal dimension, respectively, when using high-frequency reconstructions compared to standard. However, the differences in performance were statistically insignificant when a typical level of CT noise was simulated in the clinical scan. Our results suggest that high-frequency reconstruction kernels can better preserve information for downstream artificial intelligence-based radiomic characterization, provided that noise is limited. Future work should investigate the information-preserving kernels in datasets with clinical labels.

Speech is a highly coordinated process that requires precise control over vocal tract morphology/motion to produce intelligible sounds while simultaneously generating unique exhaled flow patterns. The schlieren imaging technique visualizes airflows with subtle density variations. It is hypothesized that speech flows captured by schlieren, when analyzed using a hybrid of convolutional neural network (CNN) and long short-term memory (LSTM) network, can recognize alphabet pronunciations, thus facilitating automatic speech recognition and speech disorder therapy. This study evaluates the feasibility of using a CNN-based video classification network to differentiate speech flows corresponding to the first four alphabets: /A/, /B/, /C/, and /D/. A schlieren optical system was developed, and the speech flows of alphabet pronunciations were recorded for two participants at an acquisition rate of 60 frames per second. A total of 640 video clips, each lasting 1 s, were utilized to train and test a hybrid CNN-LSTM network. Acoustic analyses of the recorded sounds were conducted to understand the phonetic differences among the four alphabets. The hybrid CNN-LSTM network was trained separately on four datasets of varying sizes (i.e., 20, 30, 40, 50 videos per alphabet), all achieving over 95% accuracy in classifying videos of the same participant. However, the network's performance declined when tested on speech flows from a different participant, with accuracy dropping to around 44%, indicating significant inter-participant variability in alphabet pronunciation. Retraining the network with videos from both participants improved accuracy to 93% on the second participant. Analysis of misclassified videos indicated that factors such as low video quality and disproportional head size affected accuracy. These results highlight the potential of CNN-assisted speech recognition and speech therapy using articulation flows, although challenges remain in expanding the alphabet set and participant cohort.

The regional industry network (RIN) is a type of financial network derived from industry networks that possess the capability to describe the connections between specific industries within a particular region. For most investors and financial analysts lacking extensive experience, the decision-support information provided by industry networks may be too vague. Conversely, RINs express more detailed and specific industry connections both within and outside the region. As RIN analysis is domain-specific and current financial network analysis tools are designed for generalized analytical tasks and cannot be directly applied to RINs, new visual analysis approaches are needed to enhance information exploration efficiency. In this study, we collaborated with domain experts and proposed V4RIN, an interactive visualization analysis system that integrates predefined domain knowledge and data processing methods to support users in uploading custom data. Through multiple views in the system panel, users can comprehensively explore the structure, geographical distribution, and spatiotemporal variations of the RIN. Two case studies were conducted and a set of expert interviews with five domain experts to validate the usability and reliability of our system.

Learning with noisy labels aims to train neural networks with noisy labels. Current models handle instance-independent label noise (IIN) well; however, they fall short with real-world noise. In medical image classification, atypical samples frequently receive incorrect labels, rendering instance-dependent label noise (IDN) an accurate representation of real-world scenarios. However, the current IDN approaches fail to consider the typicality of samples, which hampers their ability to address real-world label noise effectively. To alleviate the issues, we introduce typicality- and instance-dependent label noise (TIDN) to simulate real-world noise and establish a TIDN-combating framework to combat label noise. Specifically, we use the sample's distance to decision boundaries in the feature space to represent typicality. The TIDN is then generated according to typicality. We establish a TIDN-attention module to combat label noise and learn the transition matrix from latent ground truth to the observed noisy labels. A recursive algorithm that enables the network to make correct predictions with corrections from the learned transition matrix is proposed. Our experiments demonstrate that the TIDN simulates real-world noise more closely than the existing IIN and IDN. Furthermore, the TIDN-combating framework demonstrates superior classification performance when training with simulated TIDN and actual real-world noise.

This study proposes an image-based three-dimensional (3D) vector reconstruction of industrial parts that can generate non-uniform rational B-splines (NURBS) surfaces with high fidelity and flexibility. The contributions of this study include three parts: first, a dataset of two-dimensional images is constructed for typical industrial parts, including hexagonal head bolts, cylindrical gears, shoulder rings, hexagonal nuts, and cylindrical roller bearings; second, a deep learning algorithm is developed for parameter extraction of 3D industrial parts, which can determine the final 3D parameters and pose information of the reconstructed model using two new nets, CAD-ClassNet and CAD-ReconNet; and finally, a 3D vector shape reconstruction of mechanical parts is presented to generate NURBS from the obtained shape parameters. The final reconstructed models show that the proposed approach is highly accurate, efficient, and practical.

Cardiovascular disease, primarily caused by atherosclerotic plaque formation, is a significant health concern. The early detection of these plaques is crucial for targeted therapies and reducing the risk of cardiovascular diseases. This study presents PlaqueNet, a solution for segmenting coronary artery plaques from coronary computed tomography angiography (CCTA) images. For feature extraction, the advanced residual net module was utilized, which integrates a deepwise residual optimization module into network branches, enhances feature extraction capabilities, avoiding information loss, and addresses gradient issues during training. To improve segmentation accuracy, a depthwise atrous spatial pyramid pooling based on bicubic efficient channel attention (DASPP-BICECA) module is introduced. The BICECA component amplifies the local feature sensitivity, whereas the DASPP component expands the network's information-gathering scope, resulting in elevated segmentation accuracy. Additionally, BINet, a module for joint network loss evaluation, is proposed. It optimizes the segmentation model without affecting the segmentation results. When combined with the DASPP-BICECA module, BINet enhances overall efficiency. The CCTA segmentation algorithm proposed in this study outperformed the other three comparative algorithms, achieving an intersection over Union of 87.37%, Dice of 93.26%, accuracy of 93.12%, mean intersection over Union of 93.68%, mean Dice of 96.63%, and mean pixel accuracy value of 96.55%.