SNN deployment faces a dilemma: rate codes are power-hungry while temporal codes lack noise resilience. This paper proposes an SNN algorithm-hardware co-design, which uses sparse coding and a zero-skipping accelerator to alleviate the rate-temporal trade-off. The design reduces network spike count by 88% compared to rate coding while enhancing fault tolerance. Benchmarked against state-of-the-art rate-coding and temporal-coding accelerators, the prototype saves 88% and 89% energy, achieves $4.5 \times$ and $26.8 \times$ higher throughput, and uses 82% fewer LUTs, enabling efficient and robust edge inference.

Inspired by the heterogeneity of the brain, hybrid neural coding in SNN models has garnered increasing attention from researchers. However, most prior research relies on the ANN2SNN conversion method, which results in large time steps and decreased energy efficiency. To overcome these limitations, we propose a faster SNN model (Faster-SNN) based on hybrid neural coding and direct training. Faster-SNN assigns different coding schemes to the input layer, hidden layer, and output layer to achieve hybrid coding. The input layer uses temporal & spatial attention coding (TSAC), which incorporates a spatio-temporal attention mechanism to enhance spatio-temporal information processing. In addition, the hidden layer employs an optimized Burst-LIF neuron to implement burst coding, effectively leveraging the residual information in the membrane potential to improve information transfer efficiency. Finally, the output layer uses TTFS coding to ensure accurate and rapid decision-making. Experimental results demonstrate that our model achieves high-accuracy inference with extremely low latency through the use of hybrid neural coding and direct training methods.

Spiking neural networks (SNNs) have shown remarkable potential in processing spatio-temporal data by mimicking biological neuronal mechanisms and achieving low computational costs. However, previous SNNs often rely on neuron models with fixed and single threshold voltages and binary spikes across layers during training, which limits their capacity for accurate information representation and reduces their biological plausibility. Inspired by the diversity of neuronal behaviors in different brain regions, we propose a novel neuron called Stair-LIF, which introduces learnable incremental multi-threshold mechanisms to enhance neuronal representational capacity and utilizes multi-spike firing to improve the precision of information transmission. Furthermore, we propose a channel-wise parameterization method to expand representational capacity among Stair-LIF. Experimental results on static datasets (CIFAR-10, CIFAR-100) and neuromorphic dynamic datasets (CIFAR10-DVS and DVS128 Gesture) demonstrate that the Starir-LIF neuron achieves state-of-the-art performance.

Fault diagnosis in rotating machinery is crucial for ensuring the safety and efficient operation of industrial manufacturing processes. However, in practical applications, fault samples are typically noisy and limited in quantity, which reduces the performance of diagnostic models. To address this, this paper proposes a domain feature progressive fusion network. The network effectively integrates both time-domain and frequency-domain information, and through a progressive fusion approach, repeatedly combines time-domain and frequency-domain features, thereby significantly enhancing its capacity for feature extraction, learning, and transfer. Specifically, a multi-scale feature extraction module is proposed, aimed at thoroughly uncovering the multi-scale and multi-level latent features within both time-domain and frequency-domain samples. Additionally, a multi-scale cross-perception attention mechanism is proposed to enhance the representational power of key features within both time-domain and frequency-domain data. Moreover, a multi-classifier structure and a joint optimization strategy are proposed, further advancing the learning and transfer of key features. Experimental results demonstrate that, on two datasets, the proposed method achieved average diagnostic accuracies of 97.75% and 98.34%, respectively, outperforming the comparative methods.

Automated segmentation of coronary angiography images is highly significant for computer-aided diagnosis of coronary heart disease. However, existing segmentation methods suffer from the problem of poor segmentation results caused by insufficient extraction and fusion of the features of the complex topological structure of blood vessels. In view of this, this paper proposes an automated segmentation method for coronary angiography images based on SFAG-DeepLabv3+. This method utilizes the Swin Transformer network to screen coronary angiography images and proposes a Filtering Smoothing Equalization (FSE) image enhancement method to improve the quality of angiography images. Furthermore, this paper proposes an improved automatic segmentation network for coronary arteries based on the DeepLabv3+. In the encoder section, an Adaptive hybrid Dilated convolution and double Pooling (ADP) module is proposed to enhance the ability to extract topological features of coronary blood vessels. Between the encoder and decoder, a Gaussian Context Spatial Fusion (GCSF) module is proposed to reduce information loss during the compression and decompression of information from the encoder to the decoder. In the decoder section, bicubic interpolation upsampling is employed to improve the continuity of the segmented blood vessel topology. To validate the effectiveness of the proposed method, experiments were conducted using both the ARCADE public dataset and a self-constructed CSH dataset. Experimental results demonstrate that the method proposed in this paper can perform effective feature extraction, fusion and correction on coronary angiography images, achieving average Dice coefficients of 0.9249 on the CSH dataset and 0.9156 on the ARCADE dataset.

This paper investigates the dual-channel event-triggered quantitative load frequency control for discrete-time micro-grid system via a delay-square-dependent Lyapunov-Krasovskii functional approach. Firstly, a novel dual-channel dynamic event-triggered quantified control strategy has been proposed to enhance time-delay tolerance and the utilization of communication resources. Secondly, a discrete-time Lyapunov-Krasovskii functional incorporates the square of the delay is introduced to fully deploy the desired delay information about the system. Thirdly, improved stability criteria with strict dissipative performance indexes are established by employing the proposed delay-square-dependent Lyapunov-Krasovskii functional. Finally, two numerical examples are presented to illustrate the effectiveness and superiority of the proposed control strategy.Note to Practitioners—Micro-grid typically consist of distributed power resources, loads and energy storage devices, those are deployed spatially and remotely over some wireless and digital network mediums. The constrained resource issue, posed by finite bandwidth and processing capabilities of battery-powered system components. This paper aims at designing a reasonable control strategy to deal with issues such as band-limited channels and communication delays that undermine system stability in the process of network transmission. Specifically, we propose a dual-channel asynchronous dynamic event-triggered control strategy to save the limited channel bandwidth. To mitigate the impact of time delays, we comprehensively consider the state-dependent delays and control-related delays, further establishing a novel Lyapunov functional incorporating delay squared terms. The proposed control approach is validated to be capable of better reduce the consumption of bandwidth and achieve the desired stability through theoretical investigation and simulation cases.

Existing deep convolutional neural network-based survival analysis neither consider the modern attention mechanism nor use 3D tomographic medical images such as magnetic resonance images (MRI). This paper for the first time presents a 3D deep convolutional neural network using attention mechanism for survival prediction from multiparametric MRI in glioblastoma (GBM) patients. The attention module is incorporated into the residual network to enhance the representation power of meaningful features while suppress unimportant ones. The proposed model achieves an C-index of 0.71 in the training dataset and 0.68 in an independent test dataset, which outperforms both the traditional Cox model (0.60,0.54) and the non-attentive model (0.63,0.61). It indicates that the proposed 3D attention network has the potential of offering better performance in predicting survival using MRI than traditional survival analysis.

Brain tumors are the most common malignant neurologic tumors with the highest mortality and disability rate. Because of the delicate structure of the brain, the clinical use of several commonly used biopsy diagnosis is limited for brain tumors. Radiomics is an emerging technique for noninvasive diagnosis based on quantitative medical image analyses. However, current radiomics techniques are not standardized regarding feature extraction, feature selection, and decision making. In this paper, we propose a sparse representation-based radiomics (SRR) system for the diagnosis of brain tumors. First, we developed a dictionary learning- and sparse representation-based feature extraction method that exploits the statistical characteristics of the lesion area, leading to fine and more effective feature extraction compared with the traditional explicitly calculation-based methods. Then, we set up an iterative sparse representation method to solve the redundancy problem of the extracted features. Finally, we proposed a novel multi-feature collaborative sparse representation classification framework that introduces a new coefficient of regularization term to combine features from multi-modal images at the sparse representation coefficient level. Two clinical problems were used to validate the performance and usefulness of the proposed SRR system. One was the differential diagnosis between primary central nervous system lymphoma (PCNSL) and glioblastoma (GBM), and the other was isocitrate dehydrogenase 1 estimation for gliomas. The SRR system had superior PCNSL and GBM differentiation performance compared with some advanced imaging techniques and yielded 11% better performance for estimating IDH1 compared with the traditional radiomics methods.