Konstantin Nikolic - Profile on Academia.edu (original) (raw)

Papers by Konstantin Nikolic

Proceedings of SPIE, Mar 9, 2016

Due to optical coherence tomography (OCT) high spatial and temporal resolution, this technique co... more Due to optical coherence tomography (OCT) high spatial and temporal resolution, this technique could be used to observe the quick changes in the refractive index that accompany action potential. In this study we explore the use of time domain Optical Coherence Tomography (TD-OCT) for real time action potential detection in ex vivo Xenopus Laevis sciatic nerve. TD-OCT is the easiest and less expensive OCT technique and, if successful in detecting real time action potential, it could be used for low cost monitoring devices. A theoretical investigation into the order of magnitude of the signals detected by a TD-OCT setup is provided by this work. A linear dependence between the refractive index and the intensity changes is observed and the minimum SNR for which the setup could work is found to be SN R = 2 × 10 4 .

Journal of open research software, Apr 14, 2013

This code contains detailed modelling of the single photon response, the quantum bump, of fly pho... more This code contains detailed modelling of the single photon response, the quantum bump, of fly photoreceptors. All known components participating in the primary phototransduction process are taken into account, and estimates are given for the both the physical and the chemical parameters. The same model can be used for multiphoton response, i.e. in the case of higher light intensity stimuli. The model successfully reproduces the experimental results for the statistical features of quantum bumps (average shape, peak current average value and variance, the latency distribution, etc), arrestin mutant behaviour, low extracellular Ca cases, etc. The TRP channel activity is modelled using the Monod-Wyman-Changeux (MWC) theory for allosteric interaction. The model can combine deterministic and stochastic approaches and allows for a detailed noise analysis. The computational model was coded in Matlab using the Parallel Computing Toolbox, which allows computations on multicore computers and computer clusters. An appropriate graphic user interface was developed which gives very convenient and instructive presentation of the parameters used in the modelling and could easily be expanded to other G-protein coupled cascade processes.

Proceedings of SPIE, Feb 10, 2011

A new technique pertaining to the optical contactless chip-to-board communication using a commerc... more A new technique pertaining to the optical contactless chip-to-board communication using a commercially available CMOS technology is under development. The main concept is to use mid-IR light from a small LED which will enter an IC from the underside, impinge upon one or more pn-junctions and be reflected by a metallic sheet so that it falls on an external, discrete photodetector. After propagation through the doped semiconductor, the light undergoes attenuation due to free carrier absorption. By varying the reverse bias across the pn-junction(s), the depletion region widths are changed and therefore modulation of light intensity can be achieved. Through this scheme, data readout can be realized optically, thus alleviating the need for galvanic contacts (most notably wirebonds).

bioRxiv (Cold Spring Harbor Laboratory), Dec 4, 2020

The longstanding theory of "parallel processing" predicts that, except for a sign reversal, ON an... more The longstanding theory of "parallel processing" predicts that, except for a sign reversal, ON and OFF cells are driven by a similar pre-synaptic circuit and have similar visual field coverage, direction/orientation selectivity, visual acuity and other functional properties. However, recent experimental data challenges this view. Here we present an information theory based receptive field (RF) estimation method -quadratic mutual information (QMI) -applied to multi-electrode array electrophysiological recordings from the mouse dorsal lateral geniculate nucleus (dLGN). This estimation method provides more accurate RF estimates than the commonly used Spike-Triggered Average (STA) method, particularly in the presence of spatially correlated inputs. This improved efficiency allowed a larger number of RFs (285 vs 189 cells) to be extracted from a previously published dataset. Fitting a spatial-temporal Difference-of-Gaussians (ST-DoG) model to the RFs revealed that while the structural RF properties of ON and OFF cells are largely symmetric, there were some asymmetries apparent in the functional properties of ON and OFF visual processing streams -with OFF cells preferring higher spatial and temporal frequencies on average, and showing a greater degree of orientation selectivity.

IEEE Photonics Technology Letters, Aug 1, 2011

Here we present a method and structures for mid-infrared, free-space optical communication using ... more Here we present a method and structures for mid-infrared, free-space optical communication using unmodified, commercially available complementary metal-oxidesemiconductor integrated circuits. The modulator is based on the free carrier absorption in parasitic PN junction structures under reverse bias. Measured results demonstrate the proofof-concept with speeds of 100bps (1.55µm wavelength), but at least two orders of magnitude improvement can be achieved. This technology will enable non-galvanic chip-to-chip and chipto-package communication as an alternative to wirebonding in applications that benefit from a planar top chip surface, such as chemical sensing lab-on-chip systems as well as general sensors and mid-infrared communication.

BMC Neuroscience, Jul 1, 2013

This paper describes the design and modus of operation of a neuromorphic robotic system based on ... more This paper describes the design and modus of operation of a neuromorphic robotic system based on SpiNNaker platform, and its implementation on the goalkeeper task. The robotic system utilizes an Address Event Representation (AER) type of camera (DVS) to capture features of a moving ball, and a servo motor to position the goalkeeper to intercept the incoming ball. At the backbone of the system is a microprocessor (Arduino Due) which facilitates the communication between different robot parts. A spiking neural network, which is running on SpiNNaker, predicts the location of arrival of the moving ball and decides where to place the goalkeeper. In our setup, the maximum data transmission speed of the closed-loop system is approximately 3000 packets per second for both uplink and downlink, and the robot can intercept balls whose speed is up to 1 m/s starting from the distance of about 0.8 m. The interception accuracy is 85%, the response latency is 6.5 ms and the maximum power consumption is 7.15 W. A demo video of the robot goalie is available on YouTube: .

A bio-inspired ultrasensitive imaging chip — Phase one: Design paradigm

Recently we have completed a system level modelling of the G-protein coupled cascade in Drosophil... more Recently we have completed a system level modelling of the G-protein coupled cascade in Drosophila photoreceptors that converts single photons into transient electrical responses. Many interesting properties were revealed including the underlying mechanisms by which the system generates high quantum efficiency, single photon responses, huge signal amplification and fast recovery, as well as light adaptation to 11 orders of magnitude

This report examines a specific nanoscale computing technology, namely Magnetic Quantum Cellular ... more This report examines a specific nanoscale computing technology, namely Magnetic Quantum Cellular Automata (MQCA) systems. It provides estimates of the temperature stability and operating speed as a function of the device size, together with estimates of how such systems might behave if the device sizes could be scaled down to molecular dimensions. REPORT DOCUMENTATION PAGE Form Approved OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing the burden, to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.

IEEE Transactions on Biomedical Circuits and Systems, Oct 1, 2019

Raining fire upon modelling difficulties: PyRhO in the cloud

F1000Research, Sep 3, 2016

International Conference on Machine Learning, Jun 24, 2016

With biological systems it is often hard to adequately sample the entire input space. With sensor... more With biological systems it is often hard to adequately sample the entire input space. With sensory neural systems this can be a particularly acute problem, with very high dimensional natural inputs and typically sparse spiking outputs. Here we present an information theory based approach to analyse spiking data of an early sensory pathway, demonstrated on retinal ganglion cells (RGC) responding to natural visual scene stimuli . We used a non-parametric technique based on the concept of mutual information (MI), in particular, Quadratic Mutual Information (QMI). The QMI allowed us to very efficiently search the high dimensional space formed by the visual input for a much smaller dimensional subspace of Receptive Field Vectors (RFV). RFVs give the most information about the response of the cell to natural stimuli. This approach allows us to identify the RFVs far more efficiently using limited data as we can search the complete stimulus space for multiple vectors simultaneously. The RFVs were also used to predict the RGCs' responses to any natural stimuli. Another suitable area of application of this algorithm is in diagnostic inference. Currently we are adapting the method to be used for identifying the cancer markers in the volatile organic compounds present in exhaled breath. Once the maximally informative features are established they can be used for diagnostic predictions on new breath samples. Preliminary results of the breathomics analysis will be discussed at the conference. There are several other potential applications such as multiclass categorisation for bacterial strains using ISFET arrays for DNA sequencing. This algorithm can be part of a rapid point-of-care device for identifying the specific infectious agents and recommending appropriate antibiotics. Here we will focus on presenting the algorithm using the example of RFVs of RGCs.

Research features, Feb 10, 2019

omposed of billions of specialised nerve cells (called neurons) wired together in complex, intric... more omposed of billions of specialised nerve cells (called neurons) wired together in complex, intricate webs, the brain is inherently challenging to study. Neurons communicate with each other by transmitting electrical and chemical signals along neural circuits: signals that vary both in space and time. Until now, technical limitations in the available research methods to study the brain has meant that we are a long way from understanding how information is represented and processed by neural circuits responsible for perception, action and memory. But this is changing.

網膜インプラントと感覚代行のための低電力神経形態学的システム【Powered by NICT】

IEEE Conference Proceedings, 2016

Applied system innovation, Aug 17, 2022

The modern sedentary lifestyle is negatively influencing human health, and the current guidelines... more The modern sedentary lifestyle is negatively influencing human health, and the current guidelines recommend at least 150 min of moderate activity per week. However, the challenge is how to measure human activity in a practical way. While accelerometers are the most common tools to measure activity, current activity classification methods require calibration studies or labelled datasets-requirements that slow the research progress. Therefore, there is a pressing need to classify and quantify human activity efficiently. In this work, we propose an unsupervised approach to classify activities from accelerometer data using hidden semi-Markov models. We tune and infer the model parameters on accelerometer data from the UK Biobank and select the optimal model based on features used and informativeness of the prior. The best model achieves an average correlation of 0.4 between the inferred activities and the reference ones, with the overall physical activity obtaining a correlation of 0.8. Additionally, to prove the clinical significance of the method, we validate it by performing a linear regression between the inferred activities and anthropometric measures such as BMI and waist circumference. We show that for a sedentary behaviour and total physical activity, the proposed method achieves comparable regression coefficients to the reference labelled dataset. Moreover, the proposed method achieves a good agreement with a labelled dataset for daily time spent in a sedentary behaviour and total physical activity. The unsupervised nature of the method allows for a data-driven classification that does not require calibration studies or labelled datasets and can thus facilitate both clinical research as well as lifestyle recommendations.

Steady-states in a cooled p-Ge photoconductor via the Landsberg-Schöll-Shukla model

Solid-state Electronics, May 1, 1992

Abstract Landsberg, Scholl and Shukla have recently proposed a new model for generation-recombina... more Abstract Landsberg, Scholl and Shukla have recently proposed a new model for generation-recombination kinetics which can be used to explain the origin of chaos in semiconductors. In this paper, steady-states for the Landsberg-Scholl-Shukla (LSS) model are determined and compared with the steady-states arising from the classical (rate equation) model. In the case of cooled p -Ge, both models give a second-order nonequilibrium phase transition, from the nonconducting to the conducting state for the same value of control parameter A. This nonzero stable fixed point, in the case of the LSS model, can lose stability in a way which does not exist in the classical model. This leads to very significant differences between these two models. Furthermore, for the LSS model there is the possibility of extra fixed points appearing.

Design considerations for a CMOS Lab-on-Chip microheater array to facilitate the in vitro thermal stimulation of neurons

This paper identifies and addresses key design considerations and trade-offs in the implementatio... more This paper identifies and addresses key design considerations and trade-offs in the implementation of a CMOS high-resolution microheater array for Lab-on-Chip (LOC) applications. Specifically, this is investigated in the context of facilitating the in vitro thermal stimulation of single neurons. The paper analyses the electro-thermal response (by means of COMSOL simulations) and reliability issues (such as melting and electromigration) of different microheater designs. The analysis shows that a small-area heater is more efficient in terms of power, but it has more reliability problems essentially due to electromigration effects. For the proposed heater designs, the expected lifetime is a few days (in continuous operation) in the worst scenario, which is still generally acceptable for LOC applications.

The nonlinear transport regime of a T-shaped quantum interference transistor

Applied Physics Letters, Jun 17, 1996

We present the current–voltage characteristics of a T-shaped semiconductor electron waveguide str... more We present the current–voltage characteristics of a T-shaped semiconductor electron waveguide structure in the nonlinear regime. This system is interesting as a potential quantum interference transistor. Calculated zero-temperature I–V characteristics are with multiple peaks and exhibit regions of negative-differential resistance. The peak-to-valley ratio has maximum value of 3:1. The calculated values of the transconductance and the differential drain conductance are small, which suggests limited abilities for conventional applications of the transistor.

Chips, Feb 1, 2023

Neuromorphic computing is promising to become a future standard in low-power AI applications. The... more Neuromorphic computing is promising to become a future standard in low-power AI applications. The integration between new neuromorphic hardware and traditional microcontrollers is an open challenge. In this paper, we present an interface board and a communication protocol that allows communication between different devices, using a microcontroller unit (Arduino Due) in the middle. Our compact printed circuit board (PCB) links different devices as a whole system and provides a power supply for the entire system using batteries as the power supply. Concretely, we have connected a Dynamic Vision Sensor (DVS128), SpiNNaker board and a servo motor, creating a platform for a neuromorphic robotic system controlled by a Spiking Neural Network, which is demonstrated on the task of intercepting incoming objects. The data rate of the implemented interface board is 24.64 k symbols/s and the latency for generating commands is about 11ms. The complete system is run only by batteries, making it very suitable for robotic applications.