A neuromorphic architecture for object recognition and motion anticipation using burst-STDP

In this work we investigate the possibilities offered by a minimal framework of artificial spiking neurons to be deployed in silico. Here we introduce a hierarchical network architecture of spiking neurons which learns to recognize moving objects in a visual environment and determine the correct motor output for each object. These tasks are learned through both supervised and unsupervised spike timing dependent plasticity (STDP). STDP is responsible for the strengthening (or weakening) of synapses in relation to pre- and post-synaptic spike times and has been described as a Hebbian paradigm taking place both in vitro and in vivo. We utilize a variation of STDP learning, called burst-STDP, which is based on the notion that, since spikes are expensive in terms of energy consumption, then strong bursting activity carries more information than single (sparse) spikes. Furthermore, this learning algorithm takes advantage of homeostatic renormalization, which has been hypothesized to promote memory consolidation during NREM sleep. Using this learning rule, we design a spiking neural network architecture capable of object recognition, motion detection, attention towards important objects, and motor control outputs. We demonstrate the abilities of our design in a simple environment with distractor objects, multiple objects moving concurrently, and in the presence of noise. Most importantly, we show how this neural network is capable of performing these tasks using a simple leaky-integrate-and-fire (LIF) neuron model with binary synapses, making it fully compatible with state-of-the-art digital neuromorphic hardware designs. As such, the building blocks and learning rules presented in this paper appear promising for scalable fully neuromorphic systems to be implemented in hardware chips.     

A computational approach to motion perception

In this paper it is shown that the computation of the optical flow from a sequence of timevarying images is not, in general, an underconstrained problem. A local algorithm for the computation of the optical flow which uses second order derivatives of the image brightness pattern, and that avoids the aperture problem, is presented. The obtained optical flow is very similar to the true motion field — which is the vector field associated with moving features on the image plane — and can be used to recover 3D motion information. Experimental results on sequences of real images, together with estimates of relevant motion parameters, like time-to-crash for translation and angular velocity for rotation, are presented and discussed. Due to the remarkable accuracy which can be achieved in estimating motion parameters, the proposed method is likely to be very useful in a number of computer vision applications.     

A genetic push to understand motion detection

Two articles in this issue of Neuron (Eichner et?al. and Clark et?al.) attack the problem of explaining how neuronal hardware in Drosophila implements the Reichardt motion detector, one of the most famous computational models in neuroscience, which has proven intractable up to now.     

Moving from spatially segregated to transparent motion: A modelling approach

Motion transparency, in which patterns of moving elements group together to give the impression of lacy overlapping surfaces, provides an important challenge to models of motion perception. It has been suggested that we perceive transparent motion when the shape of the velocity histogram of the stimulus is bimodal. To investigate this further, random-dot kinematogram motion sequences were created to simulate segregated (perceptually spatially separated) and transparent (perceptually overlapping) motion. The motion sequences were analysed using the multi-channel gradient model (McGM) to obtain the speed and direction at every pixel of each frame of the motion sequences. The velocity histograms obtained were found to be quantitatively similar and all were bimodal. However, the spatial and temporal properties of the velocity field differed between segregated and transparent stimuli. Transparent stimuli produced patches of rightward and leftward motion that varied in location over time. This demonstrates that we can successfully differentiate between these two types of motion on the basis of the time varying local velocity field. However, the percept of motion transparency cannot be based simply on the presence of a bimodal velocity histogram.     

Non-Faradaic electrochemical detection of protein interactions by integrated neuromorphic CMOS sensors

Electronic detection of the binding event between biotinylated bovine serum albumen (BSA) and streptavidin is demonstrated with the chemoreceptive neuron MOS (CnuMOS) device. Differing from the ion-sensitive field-effect transistors (ISFET), CnuMOS, with the potential of the extended floating gate determined by both the sensing and control gates in a neuromorphic style, can provide protein detection without requiring analyte reference electrodes. In comparison with the microelectrode arrays, measurements are gathered through purely capacitive, non-Faradaic interactions across insulating interfaces. By using a (3-glycidoxypropyl)trimethoxysilane (3-GPS) self-assembled monolayer (SAM) as a simple covalent link for attaching proteins to a silicon dioxide sensing surface, a fully integrated, electrochemical detection platform is realized for protein interactions through monotone large-signal measurements or small-signal impedance spectroscopy. Calibration curves were created to coordinate the sensor response with ellipsometric measurements taken on witness samples. By monitoring the film thickness of streptavidin capture, a sensitivity of 25ng/cm2 or 2A of film thickness was demonstrated. With an improved noise floor the sensor can detect down to 2ng/(cm2mV) based on the calibration curve. AC measurements are shown to significantly reduce long-term sensor drift. Finally, a noise analysis of electrochemical data indicates 1/f(alpha) behavior with a noise floor beginning at approximately 1Hz.     

A general model for visual motion detection

We propose a general model for detection of both first-order motion and second-order motion. In this model an input stimulus is divided into a number of partially overlapping spatiotemporal local regions. Spatiotemporal frequency analysis is done for every local region using Gabor filters, then the input stimulus (original spatiotemporal signal) is replaced by the outputs of Gabor filters. Local motion is detected by applying Gabor motion detectors to each local spatiotemporal pattern depicted by each local feature value. Outputs of all the detectors are integrated to give the final output for global motion of the input stimulus. The model was simulated on a computer and was confirmed to correctly detect second-order motion as well as first-order motion.     

A biologically plausible acoustic motion detection neural network

In this paper we present an acoustic motion detection system to be used in a small mobile robot. While the first purpose of the system has been to be a reliable computational implementation, cheap enough to be built in hardware, effort has also been taken to construct a biologically plausible solution. The motion detector consists of a neural network composed of motion-direction sensitive neurons with a preferred direction and a preferred region of the azimuth. The system was designed to produce a higher response when stimulated by motion in the preferred direction than in the null direction and that is in fact what the system does, which means that, as desired, the system can detect motion and distinguish its direction.     

A generalized approach to detection of organophosphate resistance in mosquitoes

Insecticide bioassays and biochemical microtitre assays were compared for detection of resistance to the organophosphate insecticides malathion and fenitrothion, using inbred laboratory strains of malaria vectors Anopheles albimanus Wiedemann, An.arabiensis Patton and An.stephensi Liston. With susceptible mosquitoes, the LT100 values determined from bioassays corresponded closely with times taken to abolish the activity of acetylcholinesterase activity in biochemical assays: approximately 2 h for malathion and 3 h for fenitrothion. Resistant strains of all three anophelines showed longer survival correlated with prolonged acetylcholinesterase activity. An.albimanus strains with insensitive acetylcholinesterase survived bioassays with discriminating doses of 1 h exposure to 5% malathion or 1% fenitrothion and were judged as resistant. It is concluded that enzyme-specific microassays provide a reliable means of detecting resistant individuals, with practical advantages over bioassays which do not reveal the resistance mechanism and require large numbers of healthy mosquitoes.     

A novel approach to predicting human ingress motion using an artificial neural network

Due to the increased availability of digital human models, the need for knowing human movement is important in product design process. If the human motion is derived rapidly as design parameters change, a developer could determine the optimal parameters. For example, the optimal design of the door panel of an automobile can be obtained for a human operator to conduct the easiest ingress and egress motion. However, acquiring motion data from existing methods provides only unrealistic motion or requires a great amount of time. This not only leads to an increased time consumption for a product development, but also causes inefficiency of the overall design process. To solve such problems, this research proposes an algorithm to rapidly and accurately predict full-body human motion using an artificial neural network (ANN) and a motion database, as the design parameters are varied. To achieve this goal, this study refers to the processes behind human motor learning procedures. According to the previous research, human generate new motion based on past motion experience when they encounter new environments. Based on this principle, we constructed a motion capture database. To construct the database, motion capture experiments were performed in various environments using an optical motion capture system. To generate full-body human motion using this data, a generalized regression neural network (GRNN) was used. The proposed algorithm not only guarantees rapid and accurate results but also overcomes the ambiguity of the human motion objective function, which has been pointed out as a limitation of optimization-based research. Statistical criteria were utilized to confirm the similarity between the generated motion and actual human motion. Our research provides the basis for a rapid motion prediction algorithm that can include a variety of environmental variables. This research contributes to an increase in the usability of digital human models, and it can be applied to various research fields.     

A stochastic model for the detection of coherent motion

A computational model is presented for the detection of coherent motion based on template matching and hidden Markov models. The premise of this approach is that the growth in detection sensitivity is greater for coherent motion of structured forms than for random coherent motion. In this preliminary study, a recent experiment was simulated with the model and the results are shown to be in agreement with the above premise. This model can be extended to be part of a more complex and elaborate computational visual system.     

A novel approach to remote homology detection: jumping alignments.

We describe a new algorithm for protein classification and the detection of remote homologs. The rationale is to exploit both vertical and horizontal information of a multiple alignment in a well-balanced manner. This is in contrast to established methods such as profiles and profile hidden Markov models which focus on vertical information as they model the columns of the alignment independently and to family pairwise search which focuses on horizontal information as it treats given sequences separately. In our setting, we want to select from a given database of "candidate sequences" those proteins that belong to a given superfamily. In order to do so, each candidate sequence is separately tested against a multiple alignment of the known members of the superfamily by means of a new jumping alignment algorithm. This algorithm is an extension of the Smith-Waterman algorithm and computes a local alignment of a single sequence and a multiple alignment. In contrast to traditional methods, however, this alignment is not based on a summary of the individual columns of the multiple alignment. Rather, the candidate sequence is at each position aligned to one sequence of the multiple alignment, called the "reference sequence." In addition, the reference sequence may change within the alignment, while each such jump is penalized. To evaluate the discriminative quality of the jumping alignment algorithm, we compare it to profiles, profile hidden Markov models, and family pairwise search on a subset of the SCOP database of protein domains. The discriminative quality is assessed by median false positive counts (med-FP-counts). For moderate med-FP-counts, the number of successful searches with our method is considerably higher than with the competing methods.     

Polarization contrast and motion detection

Form and motion perception rely upon the visual system’s capacity to segment the visual scene based upon local differences in luminance or wavelength. It is not clear if polarization contrast is a sufficient basis for motion detection. Here we show that crayfish optomotor responses elicited by the motion of images derived from spatiotemporal variations in e-vector angles are comparable to contrast-elicited responses. Response magnitude increases with the difference in e-vector angles in adjacent segments of the scene and with the degree of polarization but the response is relatively insensitive to the absolute values of e-vector angles that compose the stimulus. The results indicate that polarization contrast can support visual motion detection.     

A control theory approach to the analysis and synthesis of the experimentally observed motion primitives

Recent experiments on frogs and rats, have led to the hypothesis that sensory-motor systems are organized into a finite number of linearly combinable modules; each module generates a motor command that drives the system to a predefined equilibrium. Surprisingly, in spite of the infiniteness of different movements that can be realized, there seems to be only a handful of these modules. The structure can be thought of as a vocabulary of "elementary control actions". Admissible controls, which in principle belong to an infinite dimensional space, are reduced to the linear vector space spanned by these elementary controls. In the present paper we address some theoretical questions that arise naturally once a similar structure is applied to the control of nonlinear kinematic chains. First of all, we show how to choose the modules so that the system does not loose its capability of generating a "complete" set of movements. Secondly, we realize a "complete" vocabulary with a minimal number of elementary control actions. Subsequently, we show how to modify the control scheme so as to compensate for parametric changes in the system to be controlled. Remarkably, we construct a set of modules with the property of being invariant with respect to the parameters that model the growth of an individual. Robustness against uncertainties is also considered showing how to optimally choose the modules equilibria so as to compensate for errors affecting the system. Finally, the motion primitive paradigm is extended to locomotion and a related formalization of internal (proprioceptive) and external (exteroceptive) variables is given.     

TomoAlign: A novel approach to correcting sample motion and 3D CTF in CryoET

TomoAlign is a software package that integrates tools to mitigate two important resolution limiting factors in cryoET, namely the beam-induced sample motion and the contrast transfer function (CTF) of the microscope. The package is especially focused on cryoET of thick specimens where fiducial markers are required for accurate tilt-series alignment and sample motion estimation. TomoAlign models the beam-induced sample motion undergone during the tilt-series acquisition. The motion models are used to produce motion-corrected subtilt-series centered on the particles of interest. In addition, the defocus of each particle at each tilt image is determined and can be corrected, resulting in motion-corrected and CTF-corrected subtilt-series from which the subtomograms can be computed. Alternatively, the CTF information can be passed on so that CTF correction can be carried out entirely within external packages like Relion. TomoAlign serves as a versatile tool that can streamline the cryoET workflow from initial alignment of tilt-series to final subtomogram averaging during in situ structure determination.     

MiST: A new approach to variant detection in deep sequencing datasets

MiST is a novel approach to variant calling from deep sequencing data, using the inverted mapping approach developed for Geoseq. Reads that can map to a targeted exonic region are identified using exact matches to tiles from the region. The reads are then aligned to the targets to discover variants. MiST carefully handles paralogous reads that map ambiguously to the genome and clonal reads arising from PCR bias, which are the two major sources of errors in variant calling. The reduced computational complexity of mapping selected reads to targeted regions of the genome improves speed, specificity and sensitivity of variant detection. Compared with variant calls from the GATK platform, MiST showed better concordance with SNPs from dbSNP and genotypes determined by an exonic-SNP array. Variant calls made only by MiST confirm at a high rate (>90%) by Sanger sequencing. Thus, MiST is a valuable alternative tool to analyse variants in deep sequencing data.     

A BAC clone fingerprinting approach to the detection of human genome rearrangements

We present a method, called fingerprint profiling (FPP), that uses restriction digest fingerprints of bacterial artificial chromosome clones to detect and classify rearrangements in the human genome. The approach uses alignment of experimental fingerprint patterns to in silico digests of the sequence assembly and is capable of detecting micro-deletions (1-5 kb) and balanced rearrangements. Our method has compelling potential for use as a whole-genome method for the identification and characterization of human genome rearrangements.