Search or filter publications

Filter by type:

Filter by publication type

Filter by year:

to

Results

  • Showing results for:
  • Reset all filters

Search results

  • JOURNAL ARTICLE
    Charalambous CC, Bharath AA,

    A data augmentation methodology for training machine/deep learning gait recognition algorithms

    There are several confounding factors that can reduce the accuracy of gaitrecognition systems. These factors can reduce the distinctiveness, or alter thefeatures used to characterise gait, they include variations in clothing,lighting, pose and environment, such as the walking surface. Full invariance toall confounding factors is challenging in the absence of high-quality labelledtraining data. We introduce a simulation-based methodology and asubject-specific dataset which can be used for generating synthetic videoframes and sequences for data augmentation. With this methodology, we generateda multi-modal dataset. In addition, we supply simulation files that provide theability to simultaneously sample from several confounding variables. The basisof the data is real motion capture data of subjects walking and running on atreadmill at different speeds. Results from gait recognition experimentssuggest that information about the identity of subjects is retained withinsynthetically generated examples. The dataset and methodology allow studiesinto fully-invariant identity recognition spanning a far greater number ofobservation conditions than would otherwise be possible.

  • JOURNAL ARTICLE
    Creswell A, Bharath AA,

    Denoising Adversarial Autoencoders

    Unsupervised learning is of growing interest because it unlocks the potentialheld in vast amounts of unlabelled data to learn useful representations forinference. Autoencoders, a form of generative model, may be trained by learningto reconstruct unlabelled input data from a latent representation space. Morerobust representations may be produced by an autoencoder if it learns torecover clean input samples from corrupted ones. Representations may be furtherimproved by introducing regularisation during training to shape thedistribution of the encoded data in latent space. We suggest denoisingadversarial autoencoders, which combine denoising and regularisation, shapingthe distribution of latent space using adversarial training. We introduce anovel analysis that shows how denoising may be incorporated into the trainingand sampling of adversarial autoencoders. Experiments are performed to assessthe contributions that denoising makes to the learning of representations forclassification and sample synthesis. Our results suggest that autoencoderstrained using a denoising criterion achieve higher classification performance,and can synthesise samples that are more consistent with the input data thanthose trained without a corruption process.

  • CONFERENCE PAPER
    Dávila-Montero S, Barsakcioglu DY, Jackson A, Constandinou TG, Mason AJet al.,

    Real-time clustering algorithm that adapts to dynamic changes in neural recordings

    , IEEE International Symposium on Circuits & Systems (ISCAS), Publisher: IEEE, Pages: 690-693

    This work presents a computationally efficient real-time adaptive clustering algorithm that recognizes and adapts to dynamic changes observed in neural recordings. The algorithm consists of an off-line training phase that determines initial cluster positions, and an on-line operation phase that continuously tracks drifts in clusters and periodically verifies acute changes in cluster composition. Analysis of chronic recordings from non-human primates shows that adaptive clustering achieves an improvement of 14% in classification accuracy and demonstrates an ability to recognize acute changes with 78% accuracy, with up to 29% computational efficiency compared to the state-of-the-art. The presented algorithm is suitable for long-term chronic monitoring of neural activity in various applications such as neuroscience research and control of neural prosthetics and assistive devices.

  • CONFERENCE PAPER
    Gao C, Ghoreishizadeh S, Liu Y, Constandinou TGet al.,

    On-chip ID generation for multi-node implantable devices using SA-PUF

    , IEEE International Symposium on Circuits & Systems (ISCAS), Publisher: IEEE, Pages: 678-681

    This paper presents a 64-bit on-chip identification system featuring low power consumption and randomness compensation for multi-node bio-implantable devices. A sense amplifier based bit-cell is proposed to realize the silicon physical unclonable function, providing a unique value whose probability has a uniform distribution and minimized influence from the temperature and supply variation. The entire system is designed and implemented in a typical 0.35 m CMOS technology, including an array of 64 bit-cells, readout circuits, and digital controllers for data interfaces. Simulated results show that the proposed bit-cell design achieved a uniformity of 50.24% and a uniqueness of 50.03% for generated IDs. The system achieved an energy consumption of 6.0 pJ per bit with parallel outputs and 17.3 pJ per bit with serial outputs.

  • CONFERENCE PAPER
    Haci D, Liu Y, Constandinou TG,

    32-channel ultra-low-noise arbitrary signal generation platform for biopotential emulation

    , IEEE International Symposium on Circuits & Systems (ISCAS), Publisher: IEEE, Pages: 698-701

    This paper presents a multichannel, ultra-low-noise arbitrary signal generation platform for emulating a wide range of different biopotential signals (e.g. ECG, EEG, etc). This is intended for use in the test, measurement and demonstration of bioinstrumentation and medical devices that interface to electrode inputs. The system is organized in 3 key blocks for generating, processing and converting the digital data into a parallel high performance analogue output. These blocks consist of: (1) a Raspberry Pi 3 (RPi3) board; (2) a custom Field Programmable Gate Array (FPGA) board with low-power IGLOO Nano device; and (3) analogue board including the Digital-to-Analogue Converters (DACs) and output circuits. By implementing the system this way, good isolation can be achieved between the different power and signal domains. This mixed-signal architecture takes in a high bitrate SDIO (Secure Digital Input Output) stream, recodes and packetizes this to drive two multichannel DACs, with parallel analogue outputs that are then attenuated and filtered. The system achieves 32-parallel output channels each sampled at 48kS/s, with a 10kHz bandwidth, 110dB dynamic range and uV-level output noise.

  • CONFERENCE PAPER
    Maslik M, Liu Y, Lande TS, Constandinou TGet al.,

    A charge-based ultra-low power continuous-time ADC for data driven neural spike processing

    , IEEE International Symposium on Circuits & Systems (ISCAS), Publisher: IEEE, Pages: 1420-1423

    The paper presents a novel topology of a continuous-time analogue-to-digital converter (CT-ADC) featuring ultra-low static power consumption, activity-dependent dynamic consumption, and a compact footprint. This is achieved by utilising a novel charge-packet based threshold generation method, that alleviates the requirement for a conventional feedback DAC. The circuit has a static power consumption of 3.75uW, with dynamic energy of 1.39pJ/conversion level. This type of converter is thus particularly well-suited for biosignals that are generally sparse in nature. The circuit has been optimised for neural spike recording by capturing a 3kHz bandwidth with 8-bit resolution. For a typical extracellular neural recording the average power consumption is in the order of ~4uW. The circuit has been implemented in a commercially available 0.35um CMOS technology with core occupying a footprint of 0.12 sq.mm

  • CONFERENCE PAPER
    Troiani F, Nikolic K, Constandinou TG,

    Optical coherence tomography for compound action potential detection: a computational study

    , SPIE/OSA European Conferences on Biomedical Optics (ECBO)

    The feasibility of using time domain optical coherence tomography (TD-OCT)to detect compound action potential in a peripheral nerve and the setup characteristics, are studied through the use of finite-difference time-domain (FDTD) technique.

  • CONFERENCE PAPER
    Ghoreishizadeh S, Haci D, Liu Y, Constandinou Tet al., 2017,

    A 4-wire interface SoC for shared multi-implant power transfer and full-duplex communication

    , IEEE Latin American symposium on Circuits and Systems (LASCAS), Pages: 49-52

    This paper describes a novel system for recoveringpower and providing full-duplex communication over an AC-coupled 4-wire lead between active implantable devices. Thetarget application requires a singleChest Devicebe connectedto aBrain Implantconsisting of multiple identical optrodesthat record neural activity and provide closed loop opticalstimulation. The interface is integrated within each optrode SoCallowing full-duplex and fully-differential communication basedon Manchester encoding. The system features a head-to-chestuplink data rate (1.6 Mbps) that is higher than that of the chest-to-head downlink (100 kbps) superimposed on a power carrier.On-chip power management provides an unregulated 5 V DCsupply with up to 2.5 mA output current for stimulation, anda regulated 3.3 V with 60 dB PSRR for recording and logiccircuits. The circuit has been implemented in a 0.35μm CMOStechnology, occupying 1.4 mm2silicon area, and requiring a62.2μA average current consumption.

  • JOURNAL ARTICLE
    Leene LB, Constandinou TG, 2017,

    Time Domain Processing Techniques Using Ring Oscillator-Based Filter Structures

    , IEEE Transactions on Circuits and Systems I: Regular Papers, Pages: 1-10, ISSN: 1549-8328
  • CONFERENCE PAPER
    Luan S, Williams I, De-Carvalho F, Grand L, Jackson A, Quian Quiroga R, Constandinou TGet al., 2017,

    Standalone headstage for neural recording with real-time spike sorting and data logging

    , BNA Festival of Neuroscience, Publisher: The British Neuroscience Association Ltd
  • JOURNAL ARTICLE
    Schultz SR, Copeland CS, Foust AJ, Quicke P, Schuck Ret al., 2017,

    Advances in Two-Photon Scanning and Scanless Microscopy Technologies for Functional Neural Circuit Imaging

    , PROCEEDINGS OF THE IEEE, Vol: 105, Pages: 139-157, ISSN: 0018-9219
  • CONFERENCE PAPER
    Arulkumaran K, Dilokthanakul N, Shanahan M, Bharath AAet al., 2016,

    Classifying Options for Deep Reinforcement Learning.

  • CONFERENCE PAPER
    Barsakcioglu DY, Constandinou TG, 2016,

    A 32-Channel MCU-based Feature Extraction and Classification for Scalable On-node Spike Sorting

    , IEEE International Symposium on Circuits and Systems (ISCAS), Publisher: IEEE, Pages: 1310-1313, ISSN: 0271-4302
  • JOURNAL ARTICLE
    Berditchevskaia A, Caze RD, Schultz SR, 2016,

    Performance in a GO/NOGO perceptual task reflects a balance between impulsive and instrumental components of behaviour

    , SCIENTIFIC REPORTS, Vol: 6, ISSN: 2045-2322
  • JOURNAL ARTICLE
    Chen S, Augustine GJ, Chadderton P, 2016,

    The cerebellum linearly encodes whisker position during voluntary movement

    , ELIFE, Vol: 5, ISSN: 2050-084X
  • JOURNAL ARTICLE
    Cheung K, Schultz SR, Luk W, 2016,

    NeuroFlow: A General Purpose Spiking Neural Network Simulation Platform using Customizable Processors

    , FRONTIERS IN NEUROSCIENCE, Vol: 9, ISSN: 1662-453X
  • CONFERENCE PAPER
    Elia M, Leene LB, Constandinou TG, 2016,

    Continuous-Time Micropower Interface for Neural Recording Applications

    , IEEE International Symposium on Circuits and Systems (ISCAS), Publisher: IEEE, Pages: 534-537, ISSN: 0271-4302
  • JOURNAL ARTICLE
    Evans BD, Jarvis S, Schultz SR, Nikolic Ket al., 2016,

    PyRhO: A Multiscale Optogenetics Simulation Platform

    , FRONTIERS IN NEUROINFORMATICS, Vol: 10, ISSN: 1662-5196
  • CONFERENCE PAPER
    Frehlick Z, Williams I, Constandinou TG, 2016,

    Improving Neural Spike Sorting Performance using Template Enhancement

    , 12th IEEE Biomedical Circuits and Systems Conference (BioCAS), Publisher: IEEE, Pages: 524-527, ISSN: 2163-4025
  • CONFERENCE PAPER
    Leene LB, Constandinou TG, 2016,

    A 2.7 mu W/MIPS, 0.88GOPS/mm(2) Distributed Processor for Implantable Brain Machine Interfaces

    , 12th IEEE Biomedical Circuits and Systems Conference (BioCAS), Publisher: IEEE, Pages: 360-363, ISSN: 2163-4025
  • CONFERENCE PAPER
    Liu Y, Pereira JL, Constandinou TG, 2016,

    Clockless Continuous-Time Neural Spike Sorting: Method, Implementation and Evaluation

    , IEEE International Symposium on Circuits and Systems (ISCAS), Publisher: IEEE, Pages: 538-541, ISSN: 0271-4302
  • CONFERENCE PAPER
    Luan S, Liu Y, Williams I, Constandinou TGet al., 2016,

    An Event-Driven SoC for Neural Recording

    , 12th IEEE Biomedical Circuits and Systems Conference (BioCAS), Publisher: IEEE, Pages: 404-407, ISSN: 2163-4025
  • JOURNAL ARTICLE
    Ma Z-B, Yang Y, Liu Y-X, Bharath AAet al., 2016,

    Recurrently Decomposable 2-D Convolvers for FPGA-Based Digital Image Processing

    , IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II-EXPRESS BRIEFS, Vol: 63, Pages: 979-983, ISSN: 1549-7747
  • CONFERENCE PAPER
    Nicolaou N, Constandinou TG, 2016,

    Phase-Amplitude Coupling during propofol-induced sedation: an exploratory approach

    , FENS Forum of Neuroscience, Publisher: FENS
  • CONFERENCE PAPER
    Ramezani R, Dehkhoda F, Soltan A, Degenaar P, Liu Y, Constandinou Tet al., 2016,

    An Optrode with built-in self-diagnostic and fracture sensor for cortical brain stimulation

    , 12th IEEE Biomedical Circuits and Systems Conference (BioCAS), Publisher: IEEE, Pages: 392-395, ISSN: 2163-4025
  • JOURNAL ARTICLE
    Reichenbach CS, Braiman C, Schiff ND, Hudspeth AJ, Reichenbach Tet al., 2016,

    The Auditory-Brainstem Response to Continuous, Non-repetitive Speech Is Modulated by the Speech Envelope and Reflects Speech Processing

    , FRONTIERS IN COMPUTATIONAL NEUROSCIENCE, Vol: 10, ISSN: 1662-5188
  • CONFERENCE PAPER
    Reynolds S, Copeland CS, Schultz SR, Dragotti PLet al., 2016,

    AN EXTENSION OF THE FRI FRAMEWORK FOR CALCIUM TRANSIENT DETECTION

    , IEEE 13th International Symposium on Biomedical Imaging (ISBI), Publisher: IEEE, Pages: 676-679, ISSN: 1945-7928
  • JOURNAL ARTICLE
    Sollini J, Chadderton P, 2016,

    Comodulation Enhances Signal Detection via Priming of Auditory Cortical Circuits

    , JOURNAL OF NEUROSCIENCE, Vol: 36, Pages: 12299-12311, ISSN: 0270-6474
  • JOURNAL ARTICLE
    Tang J, Jimenez SCA, Chakraborty S, Schultz SRet al., 2016,

    Visual Receptive Field Properties of Neurons in the Mouse Lateral Geniculate Nucleus

    , PLOS ONE, Vol: 11, ISSN: 1932-6203
  • CONFERENCE PAPER
    Troiani F, Nikolic K, Constandinou TG, 2016,

    Optical coherence tomography for detection of compound action potential in Xenopus Laevis sciatic nerve

    , Conference on Clinical and Translational Neurophotonics; Neural Imaging and Sensing; and Optogenetics and Optical Manipulation, Publisher: SPIE-INT SOC OPTICAL ENGINEERING, ISSN: 0277-786X

This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.

Request URL: http://wlsprd.imperial.ac.uk:80/respub/WEB-INF/jsp/search-t4-html.jsp Request URI: /respub/WEB-INF/jsp/search-t4-html.jsp Query String: id=354&limit=30&respub-action=search.html Current Millis: 1501203052371 Current Time: Fri Jul 28 01:50:52 BST 2017