Search

The success of deep learning in extending the frontiers of artificial intelligence has accelerated the application of AI-enabled systems in addressing various challenges in different fields. In healthcare, deep learning is deployed on edge computing platforms to address security and latency challenges, even though these platforms are often resource-constrained. Deep learning systems are based on conventional artificial neural networks, which are computationally complex, require high power, and have low energy efficiency, making them unsuitable for edge computing platforms. Since these systems are also used in critical applications such as bio-medicine, it is expedient that their reliability is considered when designing them. For biomedical applications, the spatio-temporal nature of information processing of spiking neural networks could be merged with a fault-tolerant 3-dimensional network on chip (3D-NoC) hardware to obtain an excellent multi-objective performance accuracy while maintaining low latency and low power consumption. In this work, we propose a reconfigurable 3D-NoC-based neuromorphic system for biomedical applications based on a fault-tolerant spike routing scheme. The performance evaluation results over X-ray images for pneumonia (i.e., COVID-19) detection show that the proposed system achieves 88.43% detection accuracy over the collected test data and could be accelerated to achieve 4.6% better inference latency than the ANN-based system while consuming 32% less power. Furthermore, the proposed system maintains high accuracy for up to 30% inter-neuron communication faults with increased latency. [ABSTRACT FROM AUTHOR]

The COVID-19 pandemic has increased the significance of remote collaboration. This study proposes a virtual reality (VR) whiteboard system that enables remote collaboration among multiple participants using natural handwriting. In total, three experiments were conducted to investigate, respectively, collaboration efficiency, user experience, and system delay. First, we compared the collaboration efficiency of the traditional whiteboard, the electronic whiteboard, and the proposed virtual reality whiteboard in a series of controlled experiments. It was discovered that the VR whiteboard significantly improves collaboration efficiency in comparison to the mouse-based electronic whiteboard and is comparable to the traditional whiteboard. Second, we assessed the user experience with a survey scale (questionnaires). The subsequent results demonstrate that the VR whiteboard provides a superior user experience in terms of efficiency, usability, etc., compared to the traditional whiteboard. We also measured an end-to-end latency of approximately 115 milliseconds, which is sufficient for remote collaboration. [ABSTRACT FROM AUTHOR]

To study the movement characteristics and separation mechanism of safflower petals and their impurities under the action of airflow and lower the impurity rate in the cleaning operation process, integration of computational fluid dynamics (CFD) and discrete element method (DEM) codes was performed to study the motion and sorting behavior of impurity particles and safflower petals under different airflow inclination angles, dust removal angles and inlet airflow velocities by establishing a true particle model. In this model, the discrete particle phase was applied by the DEM software, and the continuum gas phase was described by the ANSYS Fluent software. The Box-Behnken experimental design with three factors and three levels was performed, and parameters such as inlet airflow velocity, airflow inclined angle, and dust remover angle were selected as independent variables that would influence the cleaning impurity rate and the cleaning loss rate. A mathematical model was established, and then the effects of various parameters and their interactions were analyzed. The test results show that the cleaning effect is best when the inlet airflow velocity is 7 m/s, the airflow inclined angle is 0°, and the dust remover angle is 25°. Confirmatory tests showed that the average cleaning impurity rate and cleaning loss rate were 0.69% and 2.75%, respectively, which dropped significantly compared with those from previous optimization. An experimental device was designed and set up; the experimental results were consistent with the simulation results, indicating that studying the physical behavior of safflower petals-impurity separation in the airflow field by using the DEM-CFD coupling method is reliable. This result provides a basis for follow-up studies of separation and cleaning devices for lightweight materials such as safflower petals. [ABSTRACT FROM AUTHOR]