Computing geodesic paths encoding a curvature prior for curvilinear structure tracking.

In this paper, we introduce an efficient method for computing curves minimizing a variant of the Euler-Mumford elastica energy, with fixed endpoints and tangents at these endpoints, where the bending energy is enhanced with a user-defined and data-driven scalar-valued term referred to as the curvature prior. In order to guarantee that the globally optimal curve is extracted, the proposed method involves the numerical computation of the viscosity solution to a specific static Hamilton-Jacobi-Bellman (HJB) partial differential equation (PDE). For that purpose, we derive the explicit Hamiltonian associated with this variant model equipped with a curvature prior, discretize the resulting HJB PDE using an adaptive finite difference scheme, and solve it in a single pass using a generalized fast-marching method. In addition, we also present a practical method for estimating the curvature prior values from image data, designed for the task of accurately tracking curvilinear structure centerlines. Numerical experiments on synthetic and real-image data illustrate the advantages of the considered variant of the elastica model with a prior curvature enhancement in complex scenarios where challenging geometric structures appear.

A denoising method of ECG signal based on variational autoencoder and masked convolution.

Wearable electrocardiogram (ECG) equipment can realize continuous monitoring of cardiovascular diseases, but these devices are more susceptible to interference from various noises, which will seriously reduce the diagnostic correctness. In this work, a novel noise reduction model for ECG signals is proposed based on variational autoencoder and masked convolution. The variational Bayesian inference is conducted to capture the global features of the ECG signals by encouraging the approximate posterior of the latent variables to fit the prior distribution, and we use the skip connection and feature concatenation to realize the information interaction across the channels. To strengthen the connection of local features of the ECG signals, the masked convolution module is used to extract local feature information, which supplement the global features and the noise reduction performance of whole model can be greatly improved. Experiments are carried out on the MIT-BIH arrythmia database, and the results display that the performance metrics of signal-to-noise ratio (SNR) and root mean square error (RMSE) are significantly improved compared with other approaches while causing less signal distortion.

Constant Force-Tracking Control Based on Deep Reinforcement Learning in Dynamic Auscultation Environment.

Intelligent medical robots can effectively help doctors carry out a series of medical diagnoses and auxiliary treatments and alleviate the current shortage of social personnel. Therefore, this paper investigates how to use deep reinforcement learning to solve dynamic medical auscultation tasks. We propose a constant force-tracking control method for dynamic environments and a modeling method that satisfies physical characteristics to simulate the dynamic breathing process and design an optimal reward function for the task of achieving efficient learning of the control strategy. We have carried out a large number of simulation experiments, and the error between the tracking of normal force and expected force is basically within ±0.5 N. The control strategy is tested in a real environment. The preliminary results show that the control strategy performs well in the constant force-tracking of medical auscultation tasks. The contact force is always within a safe and stable range, and the average contact force is about 5.2 N.

Sparse ECG Denoising with Generalized Minimax Concave Penalty.

The electrocardiogram (ECG) is an important diagnostic tool for cardiovascular diseases. However, ECG signals are susceptible to noise, which may degenerate waveform and cause misdiagnosis. In this paper, the ECG noise reduction techniques based on sparse recovery are investigated. A novel sparse ECG denoising framework combining low-pass filtering and sparsity recovery is proposed. Two sparsity recovery algorithms are developed based on the traditional ℓ 1 -norm penalty and the novel generalized minimax concave (GMC) penalty, respectively. Compared with the ℓ 1 -norm penalty, the non-differentiable non-convex GMC penalty has the potential to strongly promote sparsity while maintaining the convexity of the cost function. Moreover, the GMC punishes large values less severely than ℓ 1 -norm, which is utilized to overcome the drawback of underestimating the high-amplitude components for the ℓ 1 -norm penalty. The proposed methods are evaluated on ECG signals from the MIT-BIH Arrhythmia database. The results show that underestimating problem is overcome by the proposed GMC-based method. The GMC-based method shows significant improvement with respect to the average of output signal-to-noise ratio improvement ( S N R i m p ), the average of root mean square error (RMSE) and the percent root mean square difference (PRD) over almost any given SNR compared with the classical methods, thus providing promising approaches for ECG denoising.

Parameter Selection and Performance Comparison of Particle Swarm Optimization in Sensor Networks Localization.

Localization is a key technology in wireless sensor networks. Faced with the challenges of the sensors' memory, computational constraints, and limited energy, particle swarm optimization has been widely applied in the localization of wireless sensor networks, demonstrating better performance than other optimization methods. In particle swarm optimization-based localization algorithms, the variants and parameters should be chosen elaborately to achieve the best performance. However, there is a lack of guidance on how to choose these variants and parameters. Further, there is no comprehensive performance comparison among particle swarm optimization algorithms. The main contribution of this paper is three-fold. First, it surveys the popular particle swarm optimization variants and particle swarm optimization-based localization algorithms for wireless sensor networks. Secondly, it presents parameter selection of nine particle swarm optimization variants and six types of swarm topologies by extensive simulations. Thirdly, it comprehensively compares the performance of these algorithms. The results show that the particle swarm optimization with constriction coefficient using ring topology outperforms other variants and swarm topologies, and it performs better than the second-order cone programming algorithm.

Low Duty-Cycling MAC Protocol for Low Data-Rate Medical Wireless Body Area Networks.

Wireless body area networks (WBANs) are severely energy constrained, and how to improve the energy efficiency so as to prolong the network lifetime as long as possible is one of the most important goals of WBAN research. Low data-rate WBANs are promising to cut down the energy consumption and extend the network lifetime. Considering the characteristics and demands of low data-rate WBANs, a low duty-cycling medium access control (MAC) protocol is specially designed for this kind of WBAN in this paper. Longer superframes are exploited to cut down the energy consumed on the transmissions and receptions of redundant beacon frames. Insertion time slots are embedded into the inactive part of a superframe to deliver the frames and satisfy the quality of service (QoS) requirements. The number of the data subsections in an insertion time slot can be adaptively adjusted so as to accommodate low data-rate WBANs with different traffic. Simulation results show that the proposed MAC protocol performs well under the condition of low data-rate monitoring traffic.

An Energy-Efficient MAC Protocol for Medical Emergency Monitoring Body Sensor Networks.

Medical emergency monitoring body sensor networks (BSNs) monitor the occurrence of medical emergencies and are helpful for the daily care of the elderly and chronically ill people. Such BSNs are characterized by rare traffic when there is no emergency occurring, high real-time and reliable requirements of emergency data and demand for a fast wake-up mechanism for waking up all nodes when an emergency happens. A beacon-enabled MAC protocol is specially designed to meet the demands of medical emergency monitoring BSNs. The rarity of traffic is exploited to improve energy efficiency. By adopting a long superframe structure to avoid unnecessary beacons and allocating most of the superframe to be inactive periods, the duty cycle is reduced to an extremely low level to save energy. Short active time slots are interposed into the superframe and shared by all of the nodes to deliver the emergency data in a low-delay and reliable way to meet the real-time and reliable requirements. The interposition slots can also be used by the coordinator to broadcast network demands to wake-up all nodes in a low-delay and energy-efficient way. Experiments display that the proposed MAC protocol works well in BSNs with low emergency data traffic.

A MAC protocol for medical monitoring applications of wireless body area networks.

Targeting the medical monitoring applications of wireless body area networks (WBANs), a hybrid medium access control protocol using an interrupt mechanism (I-MAC) is proposed to improve the energy and time slot utilization efficiency and to meet the data delivery delay requirement at the same time. Unlike existing hybrid MAC protocols, a superframe structure with a longer length is adopted to avoid unnecessary beacons. The time slots are mostly allocated to nodes with periodic data sources. Short interruption slots are inserted into the superframe to convey the urgent data and to guarantee the real-time requirements of these data. During these interruption slots, the coordinator can break the running superframe and start a new superframe. A contention access period (CAP) is only activated when there are more data that need to be delivered. Experimental results show the effectiveness of the proposed MAC protocol in WBANs with low urgent traffic.

Grouping Boundary Proposals for Fast Interactive Image Segmentation.

Geodesic models are known as an efficient tool for solving various image segmentation problems. Most of existing approaches only exploit local pointwise image features to track geodesic paths for delineating the objective boundaries. However, such a segmentation strategy cannot take into account the connectivity of the image edge features, increasing the risk of shortcut problem, especially in the case of complicated scenario. In this work, we introduce a new image segmentation model based on the minimal geodesic framework in conjunction with an adaptive cut-based circular optimal path computation scheme and a graph-based boundary proposals grouping scheme. Specifically, the adaptive cut can disconnect the image domain such that the target contours are imposed to pass through this cut only once. The boundary proposals are comprised of precomputed image edge segments, providing the connectivity information for our segmentation model. These boundary proposals are then incorporated into the proposed image segmentation model, such that the target segmentation contours are made up of a set of selected boundary proposals and the corresponding geodesic paths linking them. Experimental results show that the proposed model indeed outperforms state-of-the-art minimal paths-based image segmentation approaches.

FGSQA-Net: A Weakly Supervised Approach to Fine-Grained Electrocardiogram Signal Quality Assessment.

OBJECTIVE: Due to the lack of fine-grained labels, current research can only evaluate the signal quality at a coarse scale. This article proposes a weakly supervised fine-grained electrocardiogram (ECG) signal quality assessment method, which can produce continuous segment-level quality scores with only coarse labels. METHODS: A novel network architecture, i.e. FGSQA-Net, is developed for signal quality assessment, which consists of a feature shrinking module and a feature aggregation module. Multiple feature shrinking blocks, which combine residual CNN block and max pooling layer, are stacked to produce a feature map corresponding to continuous segments along the spatial dimension. Segment-level quality scores are obtained by feature aggregation along the channel dimension. RESULTS: The proposed method was evaluated on two real-world ECG databases and one synthetic dataset. Our method produced an average AUC value of 0.975, which outperforms the state-of-the-art beat-by-beat quality assessment method. The results are visualized for 12-lead and single-lead signals over a granularity from 0.64 to 1.7 seconds, demonstrating that high-quality and low-quality segments can be effectively distinguished at a fine scale. CONCLUSION: FGSQA-Net is flexible and effective for fine-grained quality assessment for various ECG recordings and is suitable for ECG monitoring using wearable devices. SIGNIFICANCE: This is the first study on fine-grained ECG quality assessment using weak labels and can be generalized to similar tasks for other physiological signals.

Geodesic Models With Convexity Shape Prior.

The minimal geodesic models established upon the eikonal equation framework are capable of finding suitable solutions in various image segmentation scenarios. Existing geodesic-based segmentation approaches usually exploit image features in conjunction with geometric regularization terms, such as euclidean curve length or curvature-penalized length, for computing geodesic curves. In this paper, we take into account a more complicated problem: finding curvature-penalized geodesic paths with a convexity shape prior. We establish new geodesic models relying on the strategy of orientation-lifting, by which a planar curve can be mapped to an high-dimensional orientation-dependent space. The convexity shape prior serves as a constraint for the construction of local geodesic metrics encoding a particular curvature constraint. Then the geodesic distances and the corresponding closed geodesic paths in the orientation-lifted space can be efficiently computed through state-of-the-art Hamiltonian fast marching method. In addition, we apply the proposed geodesic models to the active contours, leading to efficient interactive image segmentation algorithms that preserve the advantages of convexity shape prior and curvature penalization.

Active Learning for Deep Visual Tracking.

Convolutional neural networks (CNNs) have been successfully applied to the single target tracking task in recent years. Generally, training a deep CNN model requires numerous labeled training samples, and the number and quality of these samples directly affect the representational capability of the trained model. However, this approach is restrictive in practice, because manually labeling such a large number of training samples is time-consuming and prohibitively expensive. In this article, we propose an active learning method for deep visual tracking, which selects and annotates the unlabeled samples to train the deep CNN model. Under the guidance of active learning, the tracker based on the trained deep CNN model can achieve competitive tracking performance while reducing the labeling cost. More specifically, to ensure the diversity of selected samples, we propose an active learning method based on multiframe collaboration to select those training samples that should be and need to be annotated. Meanwhile, considering the representativeness of these selected samples, we adopt a nearest-neighbor discrimination method based on the average nearest-neighbor distance to screen isolated samples and low-quality samples. Therefore, the training samples' subset selected based on our method requires only a given budget to maintain the diversity and representativeness of the entire sample set. Furthermore, we adopt a Tversky loss to improve the bounding box estimation of our tracker, which can ensure that the tracker achieves more accurate target states. Extensive experimental results confirm that our active-learning-based tracker (ALT) achieves competitive tracking accuracy and speed compared with state-of-the-art trackers on the seven most challenging evaluation benchmarks. Project website: https://sites.google.com/view/altrack/.

An ECG Classification Method Based on Multi-Task Learning and CoT Attention Mechanism.

Electrocardiogram (ECG) is an efficient and simple method for the diagnosis of cardiovascular diseases and has been widely used in clinical practice. Because of the shortage of professional cardiologists and the popularity of electrocardiograms, accurate and efficient arrhythmia detection has become a hot research topic. In this paper, we propose a new multi-task deep neural network, which includes a shared low-level feature extraction module (i.e., SE-ResNet) and a task-specific classification module. Contextual Transformer (CoT) block is introduced in the classification module to dynamically model the local and global information of ECG feature sequence. The proposed method was evaluated on public CPSC2018 and PTB-XL datasets and achieved an average F1 score of 0.827 on the CPSC2018 dataset and an average F1 score of 0.833 on the PTB-XL dataset.

Autoimmune uveitis in Behçet's disease and Vogt-Koyanagi-Harada disease differ in tissue immune infiltration and T cell clonality.

Objectives: Non-infectious uveitis is often secondary to systemic autoimmune diseases, with Behçet's disease (BD) and Vogt-Koyanagi-Harada disease (VKHD) as the two most common causes. Uveitis in BD and VKHD can show similar clinical manifestations, but the underlying immunopathogenesis remains unclear. Methods: To understand immune landscapes in inflammatory eye tissues, we performed single-cell RNA paired with T cell receptor (TCR) sequencing of immune cell infiltrates in aqueous humour from six patients with BD (N = 3) and VKHD (N = 3) uveitis patients. Results: Although T cells strongly infiltrated in both types of autoimmune uveitis, myeloid cells only significantly presented in BD uveitis but not in VKHD uveitis. Conversely, VKHD uveitis but not BD uveitis showed an overwhelming dominance by CD4+ T cells (> 80%) within the T cell population due to expansion of CD4+ T cell clusters with effector memory (Tem) phenotypes. Correspondingly, VKHD uveitis demonstrated a selective expansion of CD4+ T cell clones which were enriched in pro-inflammatory Granzyme H+ CD4+ Tem cluster and showed TCR and Th1 pathway activation. In contrast, BD uveitis showed a preferential expansion of CD8+ T cell clones in pro-inflammatory Granzyme H+ CD8+ Tem cluster, and pathway activation for cytoskeleton remodelling, cellular adhesion and cytotoxicity. Conclusion: Single-cell analyses of ocular tissues reveal distinct landscapes of immune cell infiltration and T-cell clonal expansions between VKHD and BD uveitis. Preferential involvements of pro-inflammatory CD4+ Th1 cells in VKHD and cytotoxic CD8+ T cells in BD suggest a difference in disease immunopathogenesis and can guide precision disease management.

Interpatient ECG Arrhythmia Detection by Residual Attention CNN.

The precise identification of arrhythmia is critical in electrocardiogram (ECG) research. Many automatic classification methods have been suggested so far. However, efficient and accurate classification is still a challenge due to the limited feature extraction and model generalization ability. We integrate attention mechanism and residual skip connection into the U-Net (RA-UNET); besides, a skip connection between the RA-UNET and a residual block is executed as a residual attention convolutional neural network (RA-CNN) for accurate classification. The model was evaluated using the MIT-BIH arrhythmia database and achieved an accuracy of 98.5% and F 1 scores for the classes S and V of 82.8% and 91.7%, respectively, which is far superior to other approaches.

A large-scale multi-label 12-lead electrocardiogram database with standardized diagnostic statements.

Deep learning approaches have exhibited a great ability on automatic interpretation of the electrocardiogram (ECG). However, large-scale public 12-lead ECG data are still limited, and the diagnostic labels are not uniform, which increases the semantic gap between clinical practice. In this study, we present a large-scale multi-label 12-lead ECG database with standardized diagnostic statements. The dataset contains 25770 ECG records from 24666 patients, which were acquired from Shandong Provincial Hospital (SPH) between 2019/08 and 2020/08. The record length is between 10 and 60 seconds. The diagnostic statements of all ECG records are in full compliance with the AHA/ACC/HRS recommendations, which aims for the standardization and interpretation of the electrocardiogram, and consist of 44 primary statements and 15 modifiers as per the standard. 46.04% records in the dataset contain ECG abnormalities, and 14.45% records have multiple diagnostic statements. The dataset also contains additional patient demographics.

Trajectory Grouping With Curvature Regularization for Tubular Structure Tracking.

Tubular structure tracking is a crucial task in the fields of computer vision and medical image analysis. The minimal paths-based approaches have exhibited their strong ability in tracing tubular structures, by which a tubular structure can be naturally modeled as a minimal geodesic path computed with a suitable geodesic metric. However, existing minimal paths-based tracing approaches still suffer from difficulties such as the shortcuts and short branches combination problems, especially when dealing with the images involving complicated tubular tree structures or background. In this paper, we introduce a new minimal paths-based model for minimally interactive tubular structure centerline extraction in conjunction with a perceptual grouping scheme. Basically, we take into account the prescribed tubular trajectories and curvature-penalized geodesic paths to seek suitable shortest paths. The proposed approach can benefit from the local smoothness prior on tubular structures and the global optimality of the used graph-based path searching scheme. Experimental results on both synthetic and real images prove that the proposed model indeed obtains outperformance comparing with the state-of-the-art minimal paths-based tubular structure tracing algorithms.

Two-dimensional ECG-based cardiac arrhythmia classification using DSE-ResNet.

Electrocardiogram (ECG) is mostly used for the clinical diagnosis of cardiac arrhythmia due to its simplicity, non-invasiveness, and reliability. Recently, many models based on the deep neural networks have been applied to the automatic classification of cardiac arrhythmia with great success. However, most models independently extract the internal features of each lead in the 12-lead ECG during the training phase, resulting in a lack of inter-lead features. Here, we propose a general model based on the two-dimensional ECG and ResNet with detached squeeze-and-excitation modules (DSE-ResNet) to realize the automatic classification of normal rhythm and 8 cardiac arrhythmias. The original 12-lead ECG is spliced into a two-dimensional plane like a grayscale picture. DSE-ResNet is used to simultaneously extract the internal and inter-lead features of the two-dimensional ECG. Furthermore, an orthogonal experiment method is used to optimize the hyper-parameters of DSE-ResNet and a multi-model voting strategy is used to improve classification performance. Experimental results based on the test set of China Physiological Signal Challenge 2018 (CPSC2018) show that our model has average [Formula: see text] for classifying normal rhythm and 8 cardiac arrhythmias. Meanwhile, compared with the state-of-art model in CPSC2018, our model achieved the best [Formula: see text] in 2 sub-abnormal types. This shows that the model based on the two-dimensional ECG and DSE-ResNet has advantage in detecting some cardiac arrhythmias and has the potential to be used as an auxiliary tool to help doctors perform cardiac arrhythmias analysis.

Geodesic Paths for Image Segmentation With Implicit Region-Based Homogeneity Enhancement.

Minimal paths are regarded as a powerful and efficient tool for boundary detection and image segmentation due to its global optimality and the well-established numerical solutions such as fast marching method. In this paper, we introduce a flexible interactive image segmentation model based on the Eikonal partial differential equation (PDE) framework in conjunction with region-based homogeneity enhancement. A key ingredient in the introduced model is the construction of local geodesic metrics, which are capable of integrating anisotropic and asymmetric edge features, implicit region-based homogeneity features and/or curvature regularization. The incorporation of the region-based homogeneity features into the metrics considered relies on an implicit representation of these features, which is one of the contributions of this work. Moreover, we also introduce a way to build simple closed contours as the concatenation of two disjoint open curves. Experimental results prove that the proposed model indeed outperforms state-of-the-art minimal paths-based image segmentation approaches.

Automated Atrial Fibrillation Detection Based on Feature Fusion Using Discriminant Canonical Correlation Analysis.

Atrial fibrillation (AF) is one of the most common cardiovascular diseases, with a high disability rate and mortality rate. The early detection and treatment of atrial fibrillation have great clinical significance. In this paper, a multiple feature fusion is proposed to screen out AF recordings from single lead short electrocardiogram (ECG) recordings. The proposed method uses discriminant canonical correlation analysis (DCCA) feature fusion. It fully takes intraclass correlation and interclass correlation into consideration and solves the problem of computation and information redundancy with simple series or parallel feature fusion. The DCCA integrates traditional features extracted by expert knowledge and deep learning features extracted by the residual network and gated recurrent unit network to improve the low accuracy of a single feature. Based on the Cardiology Challenge 2017 dataset, the experiments are designed to verify the effectiveness of the proposed algorithm. In the experiments, the F1 index can reach 88%. The accuracy, sensitivity, and specificity are 91.7%, 90.4%, and 93.2%, respectively.