Search

Predictive analytics and event forecasting using deep learning techniques require processing long-term historical data which is infeasible in low-power wearable devices. Such devices are constrained in memory and computational power, and are pushed to their limits by resource hungry deep neural networks. Current techniques either ignore historical data, or convert temporal sequences to pattern sequences, eliminating valuable properties such as time and/or recency. The proposed model maps arbitrary-length multivariate discrete time series to a concise sequence, called mapped interval sequence (MIS). MIS retains original data properties such as time, recency, and scale, without being susceptible to missing values. Life model for time series (LMts) mapping, is capable of mapping billions of data elements with sampling rate of several kHz or higher into a sequence of 32 elements or fewer. Furthermore, a new loss function called as tolerance error is introduced to improve long-term forecasting events using LMts. In a smart health Internet of Things environment, LMts enables real-time health predictions depending on both recent and historical data. In addition, the LMts model can predict the approximate time of events, with granularity of seconds and up to years. Experimental results show that, compared to previous studies in fall prediction, LMts achieves the same 100% accuracy with a single long short-term memory layer, while covering $16{\boldsymbol \times }$ longer time period and using $80{\boldsymbol \times }$ less weight parameters. LMts is also used to forecast human fall up to 14 s in advance even with 50% missing values.

Point-of-care ultrasound (POCUS) is transforming healthcare worldwide as a diagnostic tool with the potential to significantly reduce the delay between symptom onset and initiation of therapy. Conventional POCUS systems are based on piezoelectric transducers and cable-connected electronics, which require a costly manufacturing process and usually come with an undesirably limited channel count. Such devices typically serve a specific subset of clinical applications, as imaging at different body parts calls for different ultrasound frequencies that are beyond the bandwidth of a single piezoelectric transducer. To enable whole-body imaging, multiple probes with different frequencies, apertures and beamforming (BF) methods are generally required. This further limits the affordability and accessibility of POCUS. Recent advances in micromachined ultrasound transducers (MUTs) have offered an alternative path to addressing these challenges. However, previous attempts to integrate MUTs with chips have been incomplete, neither solving the integration problem [1, 2] nor achieving full ultrasound processing capabilities [3, 4].

Low power/area cytomorphic chips may be interfaced and ultimately implanted in the human body for cell‐sensing and cell‐control applications of the future. In such electronic platforms, it is crucial to accurately mimic the biological timescales and operate in real time. This paper proposes a methodology where slow nonlinear dynamical systems describing the behaviour of naturally encountered biological systems can be efficiently realised in hardware. To this end, as a case study, a low power and efficient digital Application‐Specific Integrated Circuit capable of emulating slow intracellular calcium dynamics with timescales reaching to seconds has been fabricated in the commercially available AMS 0.35 μm technology and compared with its analog counterpart. The fabricated chip occupies an area of 1.5 mm2 (excluding the area of the pads) and consumes 18.93 nW for each calcium unit from a power supply of 3.3 V. The presented cytomimetic topology follows closely the behaviour of its biological counterpart, exhibiting similar time‐domain calcium ions dynamics. Results show that the implemented design has the potential to speed up large–scale simulations of slow intracellular dynamics by sharing cellular units in real time.

This brief presents a reconfigurable and efficient 2-D neuron model capable of extending to higher dimensions. The model is applied to the Izhikevich and FitzHugh-Nagumo neuron models as 2-D case studies and to the Hindmarsh-Rose model as a 3-D case study. Hardware synthesis and physical implementations show that the resulting circuits can reproduce neural dynamics with acceptable precision and considerably low hardware overhead compared to previously published piecewise linear models.

Calcium is a highly widespread and versatile intracellular ion that can control a wide range of temporal dynamics in the brain such as synaptic plasticity. This brief presents a novel and efficient digital circuit for implementing a calcium-based plasticity model aimed at reproducing relevant biological dynamics. Accordingly, we investigate the feasibility of the proposed model in a minimal neural network stressing on the effect of calcium oscillations on synaptic plasticity with various neuronal stimulation protocols. MATLAB simulations and physical implementations on field-programmable gate array confirm that the proposed model, with considerably low hardware overhead, can fairly mimic the relevant biological dynamics.

This paper presents a modified astrocyte model that allows a convenient digital implementation. This model is aimed at reproducing relevant biological astrocyte behaviors, which provide appropriate feedback control in regulating neuronal activities in the central nervous system. Accordingly, we investigate the feasibility of a digital implementation for a single astrocyte and a biological neuronal network model constructed by connecting two limit-cycle Hopf oscillators to an implementation of the proposed astrocyte model using oscillator-astrocyte interactions with weak coupling. Hardware synthesis, physical implementation on field-programmable gate array, and theoretical analysis confirm that the proposed astrocyte model, with considerably low hardware overhead, can mimic biological astrocyte model behaviors, resulting in desynchronization of the two coupled limit-cycle oscillators.

This study proposes a spiking neuro-fuzzy clustering system based on a novel spike encoding scheme and a compatible learning algorithm. In this system, we utilize an analog to binary encoding scheme that properly maps the concept of "distance" in multi-dimensional analog spaces to the concept of "dissimilarity" of binary bits in the equivalent binary spaces. When this scheme is combined with a novel binary to spike encoding scheme and a proper learning algorithm is applied, a powerful clustering algorithm is produced. This algorithm creates flexible fuzzy clusters in its analog input space and modifies their shapes to different convex shapes during the learning process. This system has plausible biological support due to its spike-based learning mechanism, its Quasi Spike Time Dependent Plasticity learning policy and its brain-like fuzzy clustering performance. Moreover, this neuro-fuzzy system is fully implementable on the hybrid memristor-crossbar/CMOS platform. The resultant circuit was simulated on one clustering task carried out in the binary input space on the Simon Lucas handwritten dataset and another clustering task carried out in the analog input space on Fisher's Iris standard dataset. The results show that it attained a higher clustering rate in comparison with other algorithms such as the Self Organizing Map, K-mean and the Spiking Radial Basis Function. The circuit was also successfully simulated on an image segmentation task and some clustering tasks performed in noisy spaces with various cluster sizes. Furthermore, the circuit variability analysis shows that device and signal variations up to 20% had no significant impact on the circuit's clustering performance, so the system is sufficiently immune to different variations due to its fuzzy nature.

This paper presents a digital hardware implementation of a frequency adaptive Hopf oscillator along with investigation on systematic behavior when they are coupled in a population. The mathematical models of the oscillator are introduced and compared in sense of dynamical behavior by using system-level simulations based on which a piecewise-linear model is developed. It is shown that the model is capable to be implemented digitally with high efficiency. Behavior of the oscillators in different network structures to be used for dynamic Fourier analysis is studied and a structure with more precise operation which is also more efficient for FPGA-based implementation is implemented. Conceptual block-diagram and a high level representation for this network structure are shown where design process and synthesis are explained based on which physical implementation is demonstrated and tested.

It is expensive to simulate large-scale neural networks on hardware while ensuring a high resemblance to the original neurons' behavior. This paper introduces a novel technique to facilitate digital implementation and computer simulation of neuron models that contain an exponential term. This technique is applied to a biologically realistic neuron model called Adaptive Exponential integrated and fire (AdEx). Hardware synthesis and physical implementations show that the resulting model can reproduce precise neural behavior with high performance and considerably lower implementation costs compared with the original AdEx model.

There has been a strong push recently to examine biological scale simulations of neuromorphic algorithms to achieve stronger inference capabilities. This paper presents a set of piecewise linear spiking neuron models, which can reproduce different behaviors, similar to the biological neuron, both for a single neuron as well as a network of neurons. The proposed models are investigated, in terms of digital implementation feasibility and costs, targeting large scale hardware implementation. Hardware synthesis and physical implementations on FPGA show that the proposed models can produce precise neural behaviors with higher performance and considerably lower implementation costs compared with the original model. Accordingly, a compact structure of the models which can be trained with supervised and unsupervised learning algorithms has been developed. Using this structure and based on a spike rate coding, a character recognition case study has been implemented and tested., Comment: 14 pages, 16 figures

This study firstly presents (i) a novel general cellular mapping scheme for two dimensional neuromorphic dynamical systems such as bio-inspired neuron models, and (ii) an efficient mixed analog-digital circuit, which can be conveniently implemented on a hybrid memristor-crossbar/CMOS platform, for hardware implementation of the scheme. This approach employs 4n memristors and no switch for implementing an n-cell system in comparison with 2n2 memristors and 2n switches of a Cellular Memristive Dynamical System (CMDS). Moreover, this approach allows for dynamical variables with both analog and one-hot digital values opening a wide range of choices for interconnections and networking schemes. Dynamical response analyses show that this circuit exhibits various responses based on the underlying bifurcation scenarios which determine the main characteristics of the neuromorphic dynamical systems. Due to high programmability of the circuit, it can be applied to a variety of learning systems, real-time applications, and analytically indescribable dynamical systems. We simulate the FitzHugh-Nagumo (FHN), Adaptive Exponential (AdEx) integrate and fire, and Izhikevich neuron models on our platform, and investigate the dynamical behaviors of these circuits as case studies. Moreover, error analysis shows that our approach is suitably accurate. We also develop a simple hardware prototype for experimental demonstration of our approach., This work was partially funded by EU HBP project under grant number FP7-ICT-2013-FET-F-604102, by EU H2020 ECOMODE project under grant agreement 604102, and by Spanish research grant (with support from the European Regional Development Fund) TEC2012-37868-C04-01 (BIOSENSE).

Development of financial markets and consequences of economic crises at international level caused effects on job environment and the companies’ future financial situation is a vital factor for different beneficiary groups, bankruptcy prediction can be used a mean to help them. Prediction methods are constantly evolving, and artificial neural networks have nowadays found a special position among these methods. Since learning constitutes a significant part of neural network models, learning methods of training these models are of particular importance. Therefore, finding a proper training method to reach the desired goals is necessary. Thus, this study seeks to find a better method of building and training artificial neural networks which leads to more accurate predictions of bankruptcy. Meanwhile, three neural networks of radial basis function type were built and trained separately by Altman model (1983), Zmijewski model (1984) and combinatory models’ variables. After evaluating the ability of these three models of bankruptcy prediction, their accuracy has been compared. Time span of 2004 to 2012 (eight years) has been used to select samples from the listed companies in Tehran Stock Exchange. Results show that all three models have the ability of predicting bankruptcy and the model trained with Altman Model’s variables is more accurate than the other two models in this regard.

Process variation has an increasingly dramatic effect on delay and power as process geometries shrink. Even if the amount of variation remains the same as in previous generations, it accounts for a greater percentage of process geometries as they get smaller. So an accurate prediction of path delay and power variability for real digital circuits in the current technologies is very important; however, its main drawback is the high runtime cost. In this paper, we present a new fast EDA tool which accelerates Monte Carlo based statistical static timing analysis (SSTA) for complex digital circuit. Parallel platforms like Message Passing Interface and POSIX® Threads and also the GPU-based CUDA platform suggests a natural fit for this analysis. So using these platforms, Monte Carlo based SSTA for complex digital circuits at 32, 45 and 65 nm has been performed. and of the pin-to-output delay and power distributions for all basic gates are extracted using a memory lookup from Hspice and then the results are extended to the complex digital circuit in a hierarchal manner on the parallel platforms. Results show that the GPU-based platform has the highest performance (speedup of 19×). The correctness of the Monte Carlo based SSTA implemented on a GPU has been verified by comparing its results with a CPU based implementation.

In this paper we demonstrate a dynamical system simulator that runs on a single GPU. The model (running on an NVIDIA GT325M with 1GB of memory) is up to 50 times faster than a CPU version when more than 10 million adaptive Hopf oscillators have been simulated. The simulation shows that the oscillators tune to the correct frequencies for both discrete and continuous spectra. Due to its dynamic nature the system is also capable to track non-stationary spectra. With the help of this model the frequency spectrum of an ECG signal (as a non-stationary signal) obtained and was showed that frequency domain representation of signal (i.e. FFT) is the same as one MATLAB generates.

Memristive devices have gained significant research attention lately because of their unique properties and wide application spectrum. In particular, memristor-based resistive random access memory (RRAM) offers the high density, low power, and low volatility required for next-generation nonvolatile memory. Nowadays, despite significant advances in hardware technology, in the case of massively parallel systems still new computational architectures are required. Simulation of large quantity of memristors in the crossbar structure is a known challenge encountering these barriers. Using graphic processing units (GPU) as a low-cost and high-performance computing platform is an efficient preferred approach to this problem. In this paper, we demonstrate an RRAM simulator that runs on a single GPU. The GPU-RRAM model (running on an NVIDIA GT325M with 1GB of memory) is up to 50 times faster than a CPU version. Besides a limitation on simulation of the memristor in the crossbar structure has been seen when more than 10 thousand of them are simulated but GPU can simulate more than one hundred million ones.

This study presents a cellular-based mapping for a special class of dynamical systems for embedding neuron models, by exploiting an efficient memristor crossbar-based circuit for its implementation. The resultant reconfigurable memristive dynamical circuit exhibits various bifurcation phenomena, and responses that are characteristic of dynamical systems. High programmability of the circuit enables it to be applied to real-time applications, learning systems, and analytically indescribable dynamical systems. Moreover, its efficient implementation platform makes it an appropriate choice for on-chip applications and prostheses. We apply this method to the Izhikevich, and FitzHugh–Nagumo neuron models as case studies, and investigate the dynamical behaviors of these circuits.