Search

Computing the connected component (CC) of a graph is a basic graph computing problem, which has numerous applications like graph partitioning and pattern recognition. Existing methods for computing CC suffer from memory wall problems because of the frequent data transmission between CPU and memory. To overcome this challenge, in this article, we propose to accelerate CC computation with the emerging processing-in-memory (PIM) architecture through an algorithm–architecture co-design manner. The innovation lies in computing CC with bitwise logical operations (such as AND and OR), and the customized data flow management methods to accelerate computation and reduce energy consumption. As a proof of concept, experimental results with computational spin-transfer torque magnetic RAM (STT-MRAM) arrays demonstrate on average $19.8\times $ and $12.4\times $ speedups compared with the CPU and GPU implementations, and a $35.4 \times $ energy efficiency improvement over the CPU implementation. Moreover, we investigate the potential associations between graph computing and bitwise Boolean logic, which could help design more general in-memory graph computing accelerators in the future. [ABSTRACT FROM AUTHOR]

Objective Methods Results Conclusion In this study, we investigated the safety and practicability of ultra‐fast track anesthesia (UFTA) for endoscopic thoracic sympathectomy (ETS).A total of 72 patients with palmar hyperhidrosis undergoing ETS were randomly divided into three groups: the UFTA group (group I), the group undergoing single‐lumen tracheal intubation with local infiltration anesthesia technique (group II), and the group undergoing single‐lumen tracheal intubation with routine anesthesia (group III). Mean arterial pressure (MAP) and heart rate (HR) were recorded for all three groups at the following six time points: Before anesthetics administration (T0), the time of intubating or inserting laryngeal mask airway (T1), the time of incising skin (T2), the time of disconnecting of the right sympathetic nerve (T3), the time of disconnecting of the left sympathetic nerve (T4), the time of withdrawing the tracheal tube or laryngeal mask airway (T5), and the time of transferring the patient to a post‐anesthesia care unit (PACU) (T6). The three groups were compared from the following perspectives: surgery duration; anesthesia recovery duration, that is, the duration from discontinuation of anesthesia to extubating the tracheal tube; the dose of propofol and remifentanil per kilogram body mass per unit time interval (the time at the end of the procedure, which lasted from anesthesia induction to incision suturing); and the visual analog scale (VAS) in the resting state in the PACU.Based on pairwise comparisons, the average HR and average MAP values of the three groups differed significantly from T2 to T6 (p < 0.05). As demonstrated by the correlation analysis between remifentanil and propofol with HR and MAP, the doses of the total amount of remifentanil and propofol were lower, and group I used less remifentanil and propofol than group II. No patient in group I experienced throat discomfort following surgery. Patients in groups II and III experienced a range of postoperative discomfort. The VAS scores of groups I and II were significantly lower than those of group III, with group I lower than group II.When utilized in ETS, UFTA can provide effective anesthesia for minor traumas. It is safe, effective, and consistent with the enhanced recovery philosophy of fast‐track surgery departments. [ABSTRACT FROM AUTHOR]

Triangles are the basic substructure of networks and triangle counting (TC) has been a fundamental graph computing problem in numerous fields such as social network analysis. Nevertheless, like other graph computing problems, due to the high memory-computation ratio and random memory access pattern, TC involves a large amount of data transfers thus suffers from the bandwidth bottleneck in the traditional Von-Neumann architecture. To overcome this challenge, in this paper, we propose to accelerate TC with the emerging processing-in-memory (PIM) architecture through an algorithm-architecture co-optimization manner. To enable the efficient in-memory implementations, we come up to reformulate TC with bitwise logic operations (such as AND), and develop customized graph compression and mapping techniques for efficient data flow management. With the emerging computational Spin-Transfer Torque Magnetic RAM (STT-MRAM) array, which is one of the most promising PIM enabling techniques, the device-to-architecture co-simulation results demonstrate that the proposed TC in-memory accelerator outperforms the state-of-the-art GPU and FPGA accelerations by $12.2\times$ 12. 2 × and $31.8\times$ 31. 8 × , respectively, and achieves a $34\times$ 34 × energy efficiency improvement over the FPGA accelerator. [ABSTRACT FROM AUTHOR]

The study of music-generated dance is a novel and challenging image generation task. It aims to input a piece of music and seed motions, then generate natural dance movements for the subsequent music. Transformer-based methods face challenges in time-series prediction tasks related to human movements and music due to their struggle in capturing the nonlinear relationship and temporal aspects. This can lead to issues like joint deformation, role deviation, floating, and inconsistencies in dance movements generated in response to the music. In this paper, we propose a quaternion-enhanced attention network for visual dance synthesis from a quaternion perspective, which consists of a spin position embedding (SPE) module and a quaternion rotary attention (QRA) module. First, SPE embeds position information into self-attention in a rotational manner, leading to better learning of features of movement sequences and audio sequences and improved understanding of the connection between music and dance. Second, QRA represents and fuses 3D motion features and audio features in the form of a series of quaternions, enabling the model to better learn the temporal coordination of music and dance under the complex temporal cycle conditions of dance generation. Finally, we conducted experiments on the dataset AIST++, and the results show that our approach achieves better and more robust performance in generating accurate, high-quality dance movements. Our source code and dataset can be available from https://github.com/MarasyZZ/QEAN and https://google.github.io/aistplusplus_dataset, respectively. [ABSTRACT FROM AUTHOR]

The pressure‐induced‐flow (PIF) processing method is used to fabricate oriented, self‐toughening poly(lactic acid) (PIF‐PLA) with excellent mechanical properties. However, residual internal stress in PIF‐PLA may lead to their deformation and cracking. Hence, to eliminate internal stress, solid‐state supercritical carbon dioxide foaming (sc‐CO2 foaming) is applied to PIF‐PLA, with the resultant structure being called FOAM‐PLA. The microstructure and mechanical properties of the PLAs are characterized pre‐ and post‐foaming. Scanning electron microscopy results show PIF‐PLA to exhibit oriented texture structures composed of staggered microfibers. Following sc‐CO2 foaming, the PIF‐PLA is converted into a bimodal cellular structure consisting of micro and nano‐cells. According to the 2D wide‐angle X‐ray diffraction and Raman spectra results, this structure of the FOAM‐PLA can, through a marginal reduction of orientation, release significant amounts of internal stress. Moreover, the impact strength of FOAM‐PLA is found to be 32.7 kJ m−2, which is 10.2 times higher than that of crystalline PLA. This paper depicts the evolution of the PLA microstructure through the stages of PIF‐processing and the subsequent solid‐state sc‐CO2 foaming. Meanwhile, the mechanism of the internal stress reduction is promoted. [ABSTRACT FROM AUTHOR]