Convolutional neural networks (CNNs) have shown remarkable progress in medical image segmentation. However, the lesion segmentation remains a challenge to state-of-the-art CNN-based algorithms due to the variance in scales and shapes. On the one hand, tiny lesions are hard to delineate precisely from the medical images which are often of low resolutions. On the other hand, segmenting large-size lesions requires large receptive fields, which exacerbates the first challenge. In this article, we present a scale-aware super-resolution (SR) network to adaptively segment lesions of various sizes from low-resolution (LR) medical images. Our proposed network contains dual branches to simultaneously conduct lesion mask SR (LMSR) and lesion image SR (LISR). Meanwhile, we introduce scale-aware dilated convolution (SDC) blocks into the multitask decoders to adaptively adjust the receptive fields of the convolutional kernels according to the lesion sizes. To guide the segmentation branch to learn from richer high-resolution (HR) features, we propose a feature affinity (FA) module and a scale affinity (SA) module to enhance the multitask learning of the dual branches. On multiple challenging lesion segmentation datasets, our proposed network achieved consistent improvements compared with other state-of-the-art methods. Code will be available at: https://github.com/poiuohke/SASR_Net.

The knowledge gradient (KG) algorithm is a popular and effective algorithm for the best arm identification (BAI) problem. Due to the complex calculation of KG, theoretical analysis of this algorithm is difficult, and existing results are mostly about the asymptotic performance of it, e.g., consistency, asymptotic sample allocation, etc. In this research, we present new theoretical results about the finite-time performance of the KG algorithm. Under independent and normally distributed rewards, we derive lower bounds and upper bounds for the probability of error and simple regret of the algorithm. With these bounds, existing asymptotic results become simple corollaries. We also show the performance of the algorithm for the multi-armed bandit (MAB) problem. These developments not only extend the existing analysis of the KG algorithm, but can also be used to analyze other improvement-based algorithms. Last, we use numerical experiments to further demonstrate the finite-time behavior of the KG algorithm.

Guaranteeing safety is always a prerequisite in the process of realizing various automotive applications. However, the automotive functional safety design has been challenged by multiple factors, such as the complexity of the new generation automotive electrical and electronic (E/E) architecture, the continuous release and update of automotive functional safety standard International Standardization Organization (ISO) 26262, the release of new AUTOSAR adaptive platform standard, and the increase in different types of costs. In this article, we summarize the recent advances of automotive functional safety design methodologies through analysis, design, optimization, and runtime phases, respectively: 1) functional safety analysis; 2) functional safety guarantee; 3) safety-aware cost optimization; and 4) safety-critical multifunctional scheduling. Then, we provide the future trends in functional safety design methodologies that will be directly oriented to autonomous vehicles and adapt to the next generation functional safety standard ISO 21448.

The functional safety research for automotive cyber-physical systems (ACPS) has been studied in recent years; however, these studies merely consider the change in the exposure of the functional safety classification and assume that the driver’s controllability in the functional safety classification is always fixed and uncontrollable. In fact, the driver’s controllability is variable during the runtime phase, such that the execution process of safety-critical automotive functions is a human-interaction-aware process between the driver and ACPS. To adapt to the changes in the driver’s controllability, this article studies the human-interaction-aware adaptive functional safety processing for multi-functional ACPS in two main phases. In the design phase, where the driver’s controllability is fixed at the highest level (i.e., C3), we obtain the approximate optimal priority sequence of safety-critical functions without exhausting all sequences by proposing the refined exploration method. In the runtime phase, where the driver’s controllability level is variable (i.e., C0, C1, C2, or C3), we propose the human-interaction-aware task remapping method to autonomously respond to the change of the driver’s controllability. Examples and experiments confirm that the proposed adaptive functional safety processing can reduce overall task redundancy of safety-critical automotive functions while meeting their functional safety requirements, shorten the overall response time of safety-critical automotive functions, and increase the slack time for non-safety-critical automotive functions.

A new multihop relaying scheme based on repetition transmission is developed. The performance in Rayleigh fading of the proposed system with amplify-and-forward (AF) and decode-and-forward (DF) relaying is studied. In AF relaying systems, the source re-transmits the same signal in every three time slots, each relay simply amplifies and forwards the received signal, and the destination employs maximal ratio combining (MRC). In DF relaying systems, the source re-transmits the same signal in subsequent time slots, each relay detects the received signal based on square-law selection combining and then forwards the estimated signal, and the destination employs MRC or square-law combining. Using simple closed-form bounds on the average error probability, it is shown that the proposed repetition-based multihop transmission systems with AF relaying and DF relaying achieve diversity order equal to the number of repetitions. Numerical results are provided to assess the tightness of the proposed bounds. It is shown that repetition-based DF relaying systems achieve better symbol error rate performance than repetition-based AF relaying systems in medium to large SNRs. It is also shown that DF relaying systems with balanced links employing square-law combining at the destination achieve almost the same performance as those employing MRC. In addition, the throughputs of various multihop transmission systems are obtained. It is seen that multihop transmission systems employing a DF relaying scheme achieve higher throughput than multihop transmission systems employing an AF relaying scheme.

A new repetition-based multi-hop transmission with non-coherent detection is proposed. The symbol error probability (SEP) of this system is derived for both amplify-and-forward (AF) and decode-and-forward (DF) relaying schemes. It is shown that the repetition-based non-coherent AF relaying achieves diversity order equal to the number of repetitions, while the repetition-based non-coherent DF relaying does not achieve diversity order. In addition, throughput of the proposed systems is evaluated. Repetition does not enhance the throughput of the DF relaying systems, whereas it improves the throughput of the AF relaying systems for small and moderate values of signal-to-noise ratio.

The singularity locus analysis of the Stewart mechanism is one of the main concerns in the analysis and design of the parallel manipulators. Based on the new singularity kinematics theory we propose a singularity-equivalent-mechanism, thus the complex singularity analysis of the parallel Stewart mechanism is transformed into a simpler position analysis of the singularity-equivalent-mechanism. Using this method, the singularity locus equation of an analytical polynomial expression of degree three is transformed into a simpler equation and has finally been resolved into a plane equation and a hyperbola for all of the possible orientations of the moving platform.