Sridevi Sarma

Johns Hopkins University, Biomedical Engineering, Faculty Member

Followers

Following

122

Co-authors

Public Views

Interests

Uploads

Papers by Sridevi Sarma

Measuring the signal-to-noise ratio of a neuron

by Sridevi Sarma and Wendy Suzuki

Proceedings of the National Academy of Sciences of the United States of America, Jan 20, 2015

The signal-to-noise ratio (SNR), a commonly used measure of fidelity in physical systems, is defi... more The signal-to-noise ratio (SNR), a commonly used measure of fidelity in physical systems, is defined as the ratio of the squared amplitude or variance of a signal relative to the variance of the noise. This definition is not appropriate for neural systems in which spiking activity is more accurately represented as point processes. We show that the SNR estimates a ratio of expected prediction errors and extend the standard definition to one appropriate for single neurons by representing neural spiking activity using point process generalized linear models (PP-GLM). We estimate the prediction errors using the residual deviances from the PP-GLM fits. Because the deviance is an approximate χ(2) random variable, we compute a bias-corrected SNR estimate appropriate for single-neuron analysis and use the bootstrap to assess its uncertainty. In the analyses of four systems neuroscience experiments, we show that the SNRs are -10 dB to -3 dB for guinea pig auditory cortex neurons, -18 dB to -...

Download

A computationally efficient method for modeling circuits housed in non-rectangular packages

1994 IEEE MTT-S International Microwave Symposium Digest (Cat. No.94CH3389-4), 1994

Point process models have been shown to be useful in characterizing neural spiking activity (NSA)... more Point process models have been shown to be useful in characterizing neural spiking activity (NSA) as a function of extrinsic and intrinsic factors. Most point process models of NSA are parametric as they are often efficiently computable. However, if the actual point process does not lie in the assumed parametric class of functions, misleading inferences can arise. Nonparametric methods are attractive due to fewer assumptions, but computation grows with the size of the data. We propose a computationally efficient method for nonparametric maximum likelihood estimation when the conditional intensity function, which characterizes the point process in its entirety, is assumed to be a Lipschitz continuous function but otherwise arbitrary. We show that by exploiting much structure, the problem becomes efficiently solvable and we compare our nonparametric estimation method to the most commonly used parametric approaches on goldfish retinal ganglion neural data. In this example, our nonparametric method gives a superior absolute goodness-of-fit measure than all parametric approaches analyzed.

Download

A Computationally Efficient Method for Nonparametric Modeling of Neural Spiking Activity with Point Processes

Neural Computation, 2010

Download

Synthesis of simple feed-forward networks: a first-order example

Stability analysis of networks has been the focus of much research over the past decade. Presentl... more Stability analysis of networks has been the focus of much research over the past decade. Presently, researchers are investigating performance and synthesis of controllers and channel coding schemes in networks. Such design problems are difficult in general, as there is a strong interplay between control objectives and communication constraints, which forces the synthesis of controllers and channel encoders to be done simultaneously. Current approaches typically fix one, while the other is designed to meet some objective. In this paper, we consider a simple network, in which the plant and controller are local to each other, but are together driven by a remote reference signal that is transmitted through a noisy discrete channel. We first construct a model matching performance metric that captures the tradeoffs between coding the reference command to achieve more accuracy at the remote site and designing a controller to meet performance. We then simultaneously synthesize the controller and encoder block lengths that meet the specified objective for a first-order plant and model case. Finally, we illustrate performance sensitivity to the poles of the plant and model, and to the channel noise.

Download

On time-varying bit-allocation maintaining input-output stability: a convex parameterization

In [9], we constructed a parameterization of general time-varying quantizers. The construction is... more In [9], we constructed a parameterization of general time-varying quantizers. The construction is general in that it can have infinite memory and be time-varying in that the strategy it follows in allocating a total of R bits to its inputs, is a function of time. We derived sufficient conditions for input-output stability as functions of the quantizer's time-dependent bit-allocation strategy for bounded reference inputs. Our generalized construction of the quantizer also led to the result that the set of allocation strategies that maintains stability for bounded inputs is convex, allowing the search for the most efficient strategy to ensure stability to be formulated as a convex optimization problem. In this paper, we extend our stability analysis and derive sufficient conditions for decaying reference signals. We further show that the set of allocation strategies that maintains stability remains convex, as in the case for bounded inputs. We then compute optimal bit-allocation strategies for a class of finite-memory quantizers for various plant and controller pairs, and observe that the most efficient strategies are non-trivial and time-varying. Throughout, we consider a system in which the plant and feedback controller are separated by a noiseless finiterate communication channel.

Download

An Optimal Control Approach to Seizure Detection in Drug-Resistant Epilepsy

A Systems Theoretic Approach to Systems and Synthetic Biology I: Models and System Characterizations, 2014

Physiology of functional and effective networks in epilepsy

by Pierre Mégevand, Ashesh Mehta, Sabato Santaniello, and Sridevi Sarma

Clinical Neurophysiology, 2014

The relationships between functional and structural networks provide insights into brain abnormal... more The relationships between functional and structural networks provide insights into brain abnormalities that are observed in epilepsy. Functional and effective connectivity methods have been used to identify the ictal onset zone as well as to characterize the onset, spread, and termination of seizures. Studies of the dynamics of epileptic networks suggest mechanisms that may explain the sudden onset and termination of seizures.

Download

Point process models show temporal dependencies of basal ganglia nuclei under deep brain stimulation

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2010

Deep Brain Stimulation (DBS) is an effective treatment for patients with Parkinsons disease, but ... more Deep Brain Stimulation (DBS) is an effective treatment for patients with Parkinsons disease, but its impact on basal ganglia nuclei is not fully understood. DBS applied to the subthalamic nucleus (STN) affects neurons in the Globus Pallidus pars interna (GPi) through direct projections, as well as indirectly through the Globus Pallidus pars externa (GPe). Since traditional statistical analyses of electrophysiological data provide too coarse a view of circuit dynamics, and mesoscopic biophysical dynamic models contain an intractable number of state variables for small populations of neurons, we apply a modular approach and treat each region in the STN-GPe-GPi circuit as a multi-input multi-output point process system. We use microelectrode recordings of a normal primate with DBS applied to STN at 100 and 130 Hz to estimate point process models (PPMs) for recorded regions in GPi. Our PPMs uncovered distinct dependencies between regions of GPe and GPi neurons, separated by the position...

Download

A Bayesian framework for analyzing iEEG data from a rat model of epilepsy

by Sridevi Sarma and Marek Mirski

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2011

The early detection of epileptic seizures requires computing relevant statistics from multivariat... more The early detection of epileptic seizures requires computing relevant statistics from multivariate data and defining a robust decision strategy as a function of these statistics that accurately detects the transition from the normal to the peri-ictal (problematic) state. We model the afflicted brain as a hidden Markov model (HMM) with two hidden clinical states (normal and peri-ictal). The output of the HMM is a statistic computed from multivariate neural measurements. A Bayesian framework is developed to analyze the a posteriori conditional probability of being in peri-ictal state given current and past output measurements. We apply this method to multichannel intracortical EEGs (iEEGs) from the thalamo-cortical ictal pathway in an epilepsy rat model. We first define the output statistic as the max singular value of a connectivity matrix computed on the EEG channels with spectral techniques Then, we estimate the HMM transition probabilities from this statistic and track the a poste...

Download

Restoring the basal ganglia in Parkinson's disease to normal via multi-input phase-shifted deep brain stimulation

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2010

Deep brain stimulation (DBS) injects a high frequency current that effectively disables the disea... more Deep brain stimulation (DBS) injects a high frequency current that effectively disables the diseased basal ganglia (BG) circuit in Parkinson's disease (PD) patients, leading to a reversal of motor symptoms. Though therapeutic, high frequency stimulation consumes significant power forcing frequent surgical battery replacements and causing widespread influence into other brain areas which may lead to adverse side effects. In this paper, we conducted a rigorous study to assess whether low frequency signals can restore behavior in PD patients by restoring neural activity in the BG to the normal state. We used a biophysical-based model of BG nuclei and motor thalamus whose parameters can be set to simulate the normal state and the PD state with and without DBS. We administered pulse train DBS waveforms to the subthalamic nucleus (STN) with frequencies ranging from 1-150Hz. For each DBS frequency, we computed statistics on the simulated neural activity to assess whether it is restored...

An analytical study of relay neuron's reliability: dependence on input and model parameters

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2011

Relay neurons are widely found in our nervous system, including the Thalamus, spinal cord and lat... more Relay neurons are widely found in our nervous system, including the Thalamus, spinal cord and lateral geniculate body. They receive a modulating input (background activity) and a reference input. The modulating input modulates relay of the reference input. This modulation is critical for correct functioning of relay neurons, but is poorly understood. In this paper, we use a biophysical-based model and systems theoretic tools to calculate how well a single relay neuron relays a reference input signal as a function of the neuron's electro physiological properties (i.e. model parameters), the modulating signal, and the reference signal parameters. Our analysis is more rigorous than previous related works and is generalizable to all relay cells in the body. Our analytical expression matches relay performance obtained in simulation and suggest that increasing the frequency of a sinusoidal modulating input or decreasing its DC offset increases the relay cell reliability.

Download

Generalizing performance limitations of relay neurons: Application to Parkinson's disease

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2014

Relay cells are prevalent throughout sensory systems and receive two types of inputs: driving and... more Relay cells are prevalent throughout sensory systems and receive two types of inputs: driving and modulating. The driving input contains receptive field properties that must be transmitted while the modulating input alters the specifics of transmission. Relay reliability of a relay cell is defined as the fraction of pulses in the driving input that generate action potentials at the neuron's output, and is in general a complicated function of the driving input, the modulating input and the cell's properties. In a recent study, we computed analytic bounds on the reliability of relay neurons for a class of Poisson driving inputs and sinusoidal modulating inputs. Here, we generalize our analysis and compute bounds on the relay reliability for any modulating input. Furthermore, we show that if the modulating input is generated by a colored Gaussian process, closed form expressions for bounds on relay reliability can be derived. We applied our analysis to investigate relay reliabi...

Using Convex Optimization for Nonparametric Statistical Analysis of Point Processes

2007 IEEE International Symposium on Information Theory, 2007

ABSTRACT Point process models have been shown to be useful in characterizing neural spiking activ... more ABSTRACT Point process models have been shown to be useful in characterizing neural spiking activity as a function of extrinsic and intrinsic factors. Most point process models of neural spiking are parametric as they are often efficiently computable. However, if the actual point process does not lie in the assumed parametric class of functions, misleading inferences can arise. Nonparametric methods are attractive due to fewer assumptions, but most methods require excessively complex algorithms. We propose a computationally efficient method for nonparametric maximum likelihood estimation when the conditional intensity function, which characterizes the point process in its entirety, is assumed to satisfy a Lipschitz continuity condition. We show that by exploiting the structure of the likelihood function of a point process, the problem becomes efficiently solvable via Lagrangian duality and we compare our nonparametric estimation method to the most commonly used parametric approaches on goldfish retinal ganglion neural data. In this example, our nonparametric method gives a superior absolute goodness-of-fit measure than all parametric approaches analyzed.

Computing network-based features from physiological time series: Application to sepsis detection

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2014

Sepsis is a systemic deleterious host response to infection. It is a major healthcare problem tha... more Sepsis is a systemic deleterious host response to infection. It is a major healthcare problem that affects millions of patients every year in the intensive care units (ICUs) worldwide. Despite the fact that ICU patients are heavily instrumented with physiological sensors, early sepsis detection remains challenging, perhaps because clinicians identify sepsis by using static scores derived from bed-side measurements individually, i.e., without systematically accounting for potential interactions between these signals and their dynamics. In this study, we apply network-based data analysis to take into account interactions between bed-side physiological time series (PTS) data collected in ICU patients, and we investigate features to distinguish between sepsis and non-sepsis conditions. We treated each PTS source as a node on a graph and we retrieved the graph connectivity matrix over time by tracking the correlation between each pair of sources' signals over consecutive time windows...

Instability detector of a fragile neural network: Application to seizure detection in epilepsy

Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual Conference, 2014

It has recently been proposed that the epileptic cortex is fragile in the sense that seizures man... more It has recently been proposed that the epileptic cortex is fragile in the sense that seizures manifest through small perturbations in the synaptic connections that render the entire cortical network unstable. Therefore, one method for detecting seizures is to detect when the neuronal network has gone unstable. This is important for implementing a closed-loop therapy to suppress seizures. In this paper, we consider a widely used nonlinear stochastic model of a neuronal network, and assume that spiking dynamics during non-seizure periods correspond to certain synaptic connections that render its fixed point stable. We then apply a minimum energy perturbation theory we recently developed for networks to determine the changes in the most fragile node's synaptic connections that make the same fixed point unstable (our model during seizure). Then a detector is designed as follows. First a 2-state HMM is constructed (stable=state 1 and unstable=state 2) with fixed state transition prob...

Quickest detection of state-transition in point processes: Application to neuronal activity

2011 5th International IEEE/EMBS Conference on Neural Engineering, 2011

Quickest detection is the problem of detecting a change in the probability distribution of a sequ... more Quickest detection is the problem of detecting a change in the probability distribution of a sequence of random observations with as little delay as possible and with low probability of false alarm. To date, algorithms for quickest detection exist mainly for cases where the random observations are independent, and linear or exponential cost functions of the delay are used. We propose a dynamic programming-based algorithm to solve the quickest detection problem when dependencies exist among the observations, and for any nondecreasing cost function of the detection delay. We implement the algorithm for a Bayesian formulation (i.e., the change time in the probability distribution of the observations is a random variable with a priori fixed geometric distribution) when the observations distribute according to two distinct point processes. We apply the algorithm to spiking activity observations from neurons recorded in the subthalamic nucleus of Parkinson's disease patients during the execution of a motor task. The algorithm exploits the point-process characterization of the spike trains before and during the movement (two states), and optimally detects the state transition at movement onset. Performances significantly (i.e., with a p-value p<0.05) improve over a chance level predictor.

Download

Fragility in Networks: Application to the Epileptic Brain

Proceedings of the 19th IFAC World Congress, 2014

Fragility in Dynamic Networks: Application to Neural Networks in the Epileptic Cortex

Neural Computation, 2014

Epilepsy is a network phenomenon characterized by atypical activity at the neuronal and populatio... more Epilepsy is a network phenomenon characterized by atypical activity at the neuronal and population levels during seizures, including tonic spiking, increased heterogeneity in spiking rates, and synchronization. The etiology of epilepsy is unclear, but a common theme among proposed mechanisms is that structural connectivity between neurons is altered. It is hypothesized that epilepsy arises not from random changes in connectivity, but from specific structural changes to the most fragile nodes or neurons in the network. In this letter, the minimum energy perturbation on functional connectivity required to destabilize linear networks is derived. Perturbation results are then applied to a probabilistic nonlinear neural network model that operates at a stable fixed point. That is, if a small stimulus is applied to the network, the activation probabilities of each neuron respond transiently but eventually recover to their baseline values. When the perturbed network is destabilized, the activation probabilities shift to larger or smaller values or oscillate when a small stimulus is applied. Finally, the structural modifications to the neural network that achieve the functional perturbation are derived. Simulations of the unperturbed and perturbed networks qualitatively reflect neuronal activity observed in epilepsy patients, suggesting that the changes in network dynamics due to destabilizing perturbations, including the emergence of an unstable manifold or a stable limit cycle, may be indicative of neuronal or population dynamics during seizure. That is, the epileptic cortex is always on the brink of instability and minute changes in the synaptic weights associated with the most fragile node can suddenly destabilize the network to cause seizures. Finally, the theory developed here and its interpretation of epileptic networks enables the design of a straightforward feedback controller that first detects when the network has destabilized and then applies linear state feedback control to steer the network back to its stable state.

Download

Computing network-based features from intracranial EEG time series data: Application to seizure focus localization

2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2014

The surgical resection of the epileptogenic zone (EZ) is the only effective treatment for many dr... more The surgical resection of the epileptogenic zone (EZ) is the only effective treatment for many drug-resistant epilepsy (DRE) patients, but the pre-surgical identification of the EZ is challenging. This study investigates whether the EZ exhibits a computationally identifiable signature during seizures. In particular, we compute statistics of the brain network from intracranial EEG (iEEG) recordings and track the evolution of network connectivity before, during, and after seizures. We define each node in the network as an electrode and weight each edge connecting a pair of nodes by the gamma band cross power of the corresponding iEEG signals. The eigenvector centrality (EVC) of each node is tracked over two seizures per patient and the electrodes are ranked according to the corresponding EVC value. We hypothesize that electrodes covering the EZ have a signature EVC rank evolution during seizure that differs from electrodes outside the EZ. We tested this hypothesis on multi-channel iEEG recordings from 2 DRE patients who had successful surgery (i.e., seizures were under control with or without medications) and 1 patient who had unsuccessful surgery. In the successful cases, we assumed that the resected region contained the EZ and found that the EVC rank evolution of the electrodes within the resected region had a distinct &amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;quot;arc&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;quot; signature, i.e., the EZ ranks first rose together shortly after seizure onset and then fell later during seizure.

Reduction of a wave-variable biological arm control model

Proceedings of the 2000 American Control Conference. ACC (IEEE Cat. No.00CH36334), 2000

... feedback controller, or an indirect vs. ... match the gain and phase well at least up to the ... more

Sridevi Sarma

Uploads

Papers by Sridevi Sarma

Log In