As the research on artificial intelligence booms, there is broad interest in brain‐inspired computing using novel neuromorphic devices. The potential of various emerging materials and devices for ...neuromorphic computing has attracted extensive research efforts, leading to a large number of publications. Going forward, in order to better emulate the brain's functions, its relevant fundamentals, working mechanisms, and resultant behaviors need to be re‐visited, better understood, and connected to electronics. A systematic overview of biological and artificial neural systems is given, along with their related critical mechanisms. Recent progress in neuromorphic devices is reviewed and, more importantly, the existing challenges are highlighted to hopefully shed light on future research directions.
A comprehensive overview of biological and artificial neural networks is presented, including their key computing elements and related important functions, such as synapses, neurons, plasticity, learning, and memory, along with their electronic demonstrations using emerging devices. As a perspective, the connections and gaps between them and the challenges for building more bio‐plausible artificial neural networks are discussed.
Abstract
Brain-machine interfaces are promising tools to restore lost motor functions and probe brain functional mechanisms. As the number of recording electrodes has been exponentially rising, the ...signal processing capability of brain–machine interfaces is falling behind. One of the key bottlenecks is that they adopt conventional von Neumann architecture with digital computation that is fundamentally different from the working principle of human brain. In this work, we present a memristor-based neural signal analysis system, where the bio-plausible characteristics of memristors are utilized to analyze signals in the analog domain with high efficiency. As a proof-of-concept demonstration, memristor arrays are used to implement the filtering and identification of epilepsy-related neural signals, achieving a high accuracy of 93.46%. Remarkably, our memristor-based system shows nearly 400× improvements in the power efficiency compared to state-of-the-art complementary metal-oxide-semiconductor systems. This work demonstrates the feasibility of using memristors for high-performance neural signal analysis in next-generation brain–machine interfaces.
Lithium–sulfur batteries are recognized as one of the most promising next-generation high-performance energy storage systems. However, obstacles like the irreversible capacity loss hinder its broad ...application. Herein, we fabricated an interconnected three-dimensional MoS
2
–MoN heterostructure (3D-MoS
2
–MoN) via a facile salt-template method, overcoming the intrinsic shortcomings such as poor conductivity and compact morphology of traditionally-synthesized transition metal sulfides (TMSs). Furthermore, excellent electrocatalysis ability and hierarchical pore structure effectively accelerate the sluggish lithium polysulfides conversions during cycling. As a result, 3D-MoS
2
–MoN showed a high initial specific capacity of 1466 mAh·g
−1
and excellent high-rate capability up to 4 °C. A stable cycling performance with a sulfur loading of 2 mg·cm
−2
was realized with a low decay rate of 0.069% per cycle. This work introduced a rational design route for the appliance of TMSs in the lithium-sulfur batteries.
Graphic abstract
Earthquake-induced landslides (EQILs) are an incredibly destructive geological disaster. Rapid landslide susceptibility assessments are indispensable and critical for risk analysis and emergency ...management. Previous studies mainly focus on the regional-scale assessment of EQIL susceptibility, while the global analyses of that are lacking. In this study, we constructed a global model for rapidly assessing earthquake-induced landslide susceptibility based on the random forest (RF) algorithm using globally available data. In total, 288,114 landslides from 16 high-quality EQIL inventories were utilized to develop the global landslide model. We split the data into 70% training dataset for model training and 30% testing data for model evaluation. We also used three blind test events to validate the model performance. The model showed excellent performance on the testing data (accuracy = 0.945, and AUC = 0.985). The RF model exhibited strong spatial generalizability and robustness, with an AUC exceeding 0.8 for each landslide inventory and showing good performance on the blind test events. The resulting landslide susceptibility maps also match relatively well with the actual landslide locations. Among the conditioning factors, modified Mercalli intensity (MMI), elevation and slope are the three most important conditioning factors. The susceptibility maps for each landslide event were produced. The developed RF model would be useful in studies of earthquake-induced landslide susceptibility and emergency response after an earthquake.
•A global random forest model assessing earthquake-induced landslide susceptibility.•Performances of the model are good in 16 landslide inventories and blind testing.•Modified Mercalli intensity, elevation and slope are the most important factors.
Three-Dimensional NAND flash technology is one of the most competitive integrated solutions for the high-volume massive data storage. So far, there are few investigations on how to use 3-D NAND flash ...for in-memory computing in the neural network accelerator. In this brief, we propose using the 3-D vertical channel NAND array architecture to implement the vector-matrix multiplication (VMM) with for the first time. Based on the array-level SPICE simulation, the bias condition including the selector layer and the unselected layers is optimized to achieve high computation accuracy of VMM. Since the VMM can be performed layer by layer in a 3-D NAND array, the read-out latency is largely improved compared to the conventional single-cell read-out operation. The impact of device-to-device variation on the computation accuracy is also analyzed.
Blockchain and AI are promising techniques for next-generation wireless networks. Blockchain can establish a secure and decentralized resource sharing environment. AI can be explored to solve ...problems with uncertain, time-variant, and complex features. Both of these techniques have recently seen a surge in interest. The integration of these two techniques can further enhance the performance of wireless networks. In this article, we first propose a secure and intelligent architecture for next-generation wireless networks by integrating AI and blockchain into wireless networks to enable flexible and secure resource sharing. Then we propose a blockchain empowered content caching problem to maximize system utility, and develop a new caching scheme by utilizing deep reinforcement learning. Numerical results demonstrate the effectiveness of the proposed scheme.
LINKED CONTENT
This article is linked to Rivera–Andrade et al papers. To view these articles, visit https://doi.org/10.1111/apt.16948 and https://doi.org/10.1111/apt.17060
Background GALAD score, comprising five clinical parameters, is a predictive model developed for hepatocellular carcinoma (HCC) detection. Since its emergence, its diagnostic ability has been ...validated in different populations with a wide variation. Therefore, we conducted a meta-analysis to investigate its overall diagnostic performance in differentiating HCC in chronic liver diseases. Methods Eligible studies were searched in the
,
,
,
,
, and
databases by 29 May 2022. Pooled sensitivity, pooled specificity, and area under the receiver operating characteristic curve (AUC) with the corresponding 95% confidence intervals (CI) were estimated. Results Fifteen original studies (comprising 19,021 patients) were included. For detecting any-stage HCC, GALAD score yielded an excellent ability, with pooled sensitivity, specificity, and AUC of 0.82 (95%CI: 0.78-0.85), 0.89 (95%CI: 0.85-0.91), and 0.92 (95%CI: 0.89-0.94), respectively. Notably, further analyses demonstrated a good diagnostic accuracy of GALAD score for identifying Barcelona Clinic Liver Cancer staging (BCLC) 0/A HCC, with a moderate sensitivity (0.73 (95%CI: 0.66-0.79)) and a high specificity (0.87 (95%CI: 0.81-0.91)); by contrast, only 38% of early-stage patients can be identified by alpha-fetoprotein, with an AUC value of 0.70 (95%CI: 0.66-0.74). Following subgroup analyses based on different HCC etiologies, higher sensitivities and AUC values were observed in subgroups with hepatitis C or non-viral liver diseases. For detecting BCLC 0/A HCC in the cirrhotic population, GALAD score had a pooled sensitivity, specificity, and AUC of 0.78 (95%CI: 0.66-0.87), 0.80 (95%CI: 0.72-0.87), and 0.86 (95%CI: 0.83-0.89). Conclusions We highlighted the superior diagnostic accuracy of GALAD score for detecting any-stage HCC with a high sensitivity and specificity, especially for early-stage HCC, with a relatively stable diagnostic performance. The addition of GALAD score into ultrasound surveillance may identify more HCC patients. Our findings imply the robust power of the GALAD score as a HCC screening or diagnostic tool, and it should be further validated by more studies with high quality.
Target position and velocity estimation using a passive radar with multiple signals of opportunity and multiple receive stations is investigated. The maximum likelihood (ML) estimate of the unknown ...position and velocity vector of a target is presented. Formulas bounding the best possible mean square error are provided, via the Cramer-Rao lower bound, for any unbiased estimator of target position and velocity. The model assumes a single target, a single receive antenna at each receive station, spatially and temporally white Gaussian clutter-plus-noise, and uncorrelated reflection coefficients. To describe the best possible performance, it is assumed that the signals of opportunity are estimated perfectly from the direct path reception. For a specific example where the signals of opportunity come from the Global System for Mobile (GSM) communication system, the optimum possible estimation performance is presented using numerical examples. It is shown that it is possible to obtain large performance gains through using multiple signals of opportunity and multiple receive stations.