Biomedical Engineering - Research Publications

Permanent URI for this collection

Search Results

Now showing 1 - 10 of 29
  • Item
    No Preview Available
    Model Parameter Estimation As Features to Predict the Duration of Epileptic Seizures From Onset.
    Liu, Y ; Xia, S ; Soto-Breceda, A ; Karoly, P ; Cook, MJ ; Grayden, DB ; Schmidt, D ; Kuhlmann, L (IEEE, 2023-07)
    The durations of epileptic seizures are linked to severity and risk for patients. It is unclear if the spatiotemporal evolution of a seizure has any relationship with its duration. Understanding such mechanisms may help reveal treatments for reducing the duration of a seizure. Here, we present a novel method to predict whether a seizure is going to be short or long at its onset using features that can be interpreted in the parameter space of a brain model. The parameters of a Jansen-Rit neural mass model were tracked given intracranial electroencephalography (iEEG) signals, and were processed as time series features using MINIROCKET. By analysing 2954 seizures from 10 patients, patient-specific classifiers were built to predict if a seizure would be short or long given 7 s of iEEG at seizure onset. The method achieved an area under the receiver operating characteristic curve (AUC) greater than 0.6 for five of 10 patients. The behaviour in the parameter space has shown different mechanisms are associated with short/long seizures.Clinical relevance-This shows that it is possible to classify whether a seizure will be short or long based on its early characteristics. Timely interventions and treatments can be applied if the duration of the seizures can be predicted.
  • Item
    No Preview Available
    Continuous synthesis of artificial speech sounds from human cortical surface recordings during silent speech production
    Meng, K ; Goodarzy, F ; Kim, E ; Park, YJ ; Kim, JS ; Cook, MJ ; Chung, CK ; Grayden, DB (IOP Publishing Ltd, 2023-08-01)
    Objective. Brain-computer interfaces can restore various forms of communication in paralyzed patients who have lost their ability to articulate intelligible speech. This study aimed to demonstrate the feasibility of closed-loop synthesis of artificial speech sounds from human cortical surface recordings during silent speech production.Approach. Ten participants with intractable epilepsy were temporarily implanted with intracranial electrode arrays over cortical surfaces. A decoding model that predicted audible outputs directly from patient-specific neural feature inputs was trained during overt word reading and immediately tested with overt, mimed and imagined word reading. Predicted outputs were later assessed objectively against corresponding voice recordings and subjectively through human perceptual judgments.Main results. Artificial speech sounds were successfully synthesized during overt and mimed utterances by two participants with some coverage of the precentral gyrus. About a third of these sounds were correctly identified by naïve listeners in two-alternative forced-choice tasks. A similar outcome could not be achieved during imagined utterances by any of the participants. However, neural feature contribution analyses suggested the presence of exploitable activation patterns during imagined speech in the postcentral gyrus and the superior temporal gyrus. In future work, a more comprehensive coverage of cortical surfaces, including posterior parts of the middle frontal gyrus and the inferior frontal gyrus, could improve synthesis performance during imagined speech.Significance.As the field of speech neuroprostheses is rapidly moving toward clinical trials, this study addressed important considerations about task instructions and brain coverage when conducting research on silent speech with non-target participants.
  • Item
    No Preview Available
    Evidence of Onset and Sustained Neural Responses to Isolated Phonemes from Intracranial Recordings in a Voice-based Cursor Control Task
    Meng, K ; Lee, S-H ; Goodarzy, F ; Vogrin, S ; Cook, MJ ; Lee, S-W ; Grayden, DB (ISCA-INT SPEECH COMMUNICATION ASSOC, 2022)
  • Item
    No Preview Available
    Implementation of a closed-loop BCI system for real-time speech synthesis under clinical constraints
    Meng, K ; Kim, E ; Vogrin, S ; Cook, MJ ; Goodarzy, F ; Grayden, DB ; Chung, CK (IEEE, 2022)
  • Item
    No Preview Available
    Sleep and seizure risk in epilepsy: bed and wake times are more important than sleep duration
    Stirling, RE ; Hidajat, CM ; Grayden, DB ; D'Souza, WJ ; Naim-Feil, J ; Dell, KL ; Schneider, LD ; Nurse, E ; Freestone, D ; Cook, MJ ; Karoly, PJ (OXFORD UNIV PRESS, 2023-07-03)
    Sleep duration, sleep deprivation and the sleep-wake cycle are thought to play an important role in the generation of epileptic activity and may also influence seizure risk. Hence, people diagnosed with epilepsy are commonly asked to maintain consistent sleep routines. However, emerging evidence paints a more nuanced picture of the relationship between seizures and sleep, with bidirectional effects between changes in sleep and seizure risk in addition to modulation by sleep stages and transitions between stages. We conducted a longitudinal study investigating sleep parameters and self-reported seizure occurrence in an ambulatory at-home setting using mobile and wearable monitoring. Sixty subjects wore a Fitbit smartwatch for at least 28 days while reporting their seizure activity in a mobile app. Multiple sleep features were investigated, including duration, oversleep and undersleep, and sleep onset and offset times. Sleep features in participants with epilepsy were compared to a large (n = 37 921) representative population of Fitbit users, each with 28 days of data. For participants with at least 10 seizure days (n = 34), sleep features were analysed for significant changes prior to seizure days. A total of 4956 reported seizures (mean = 83, standard deviation = 130) and 30 485 recorded sleep nights (mean = 508, standard deviation = 445) were included in the study. There was a trend for participants with epilepsy to sleep longer than the general population, although this difference was not significant. Just 5 of 34 participants showed a significant difference in sleep duration the night before seizure days compared to seizure-free days. However, 14 of 34 subjects showed significant differences between their sleep onset (bed) and/or offset (wake) times before seizure occurrence. In contrast to previous studies, the current study found undersleeping was associated with a marginal 2% decrease in seizure risk in the following 48 h (P < 0.01). Nocturnal seizures were associated with both significantly longer sleep durations and increased risk of a seizure occurring in the following 48 h. Overall, the presented results demonstrated that day-to-day changes in sleep duration had a minimal effect on reported seizures, while patient-specific changes in bed and wake times were more important for identifying seizure risk the following day. Nocturnal seizures were the only factor that significantly increased the risk of seizures in the following 48 h on a group level. Wearables can be used to identify these sleep-seizure relationships and guide clinical recommendations or improve seizure forecasting algorithms.
  • Item
    No Preview Available
    Brain model state space reconstruction using an LSTM neural network
    Liu, Y ; Soto-Breceda, A ; Karoly, P ; Grayden, DB ; Zhao, Y ; Cook, MJ ; Schmidt, D ; Kuhlmann, L (IOP Publishing Ltd, 2023-06-01)
    Objective. Kalman filtering has previously been applied to track neural model states and parameters, particularly at the scale relevant to electroencephalography (EEG). However, this approach lacks a reliable method to determine the initial filter conditions and assumes that the distribution of states remains Gaussian. This study presents an alternative, data-driven method to track the states and parameters of neural mass models (NMMs) from EEG recordings using deep learning techniques, specifically a long short-term memory (LSTM) neural network.Approach. An LSTM filter was trained on simulated EEG data generated by a NMM using a wide range of parameters. With an appropriately customised loss function, the LSTM filter can learn the behaviour of NMMs. As a result, it can output the state vector and parameters of NMMs given observation data as the input.Main results. Test results using simulated data yielded correlations withRsquared of around 0.99 and verified that the method is robust to noise and can be more accurate than a nonlinear Kalman filter when the initial conditions of the Kalman filter are not accurate. As an example of real-world application, the LSTM filter was also applied to real EEG data that included epileptic seizures, and revealed changes in connectivity strength parameters at the beginnings of seizures.Significance. Tracking the state vector and parameters of mathematical brain models is of great importance in the area of brain modelling, monitoring, imaging and control. This approach has no need to specify the initial state vector and parameters, which is very difficult to do in practice because many of the variables being estimated cannot be measured directly in physiological experiments. This method may be applied using any NMM and, therefore, provides a general, novel, efficient approach to estimate brain model variables that are often difficult to measure.
  • Item
    No Preview Available
    Ambient air pollution and epileptic seizures: A panel study in Australia
    Chen, Z ; Yu, W ; Xu, R ; Karoly, PJ ; Maturana, M ; Payne, DE ; Li, L ; Nurse, ES ; Freestone, DR ; Li, S ; Burkitt, AN ; Cook, MJ ; Guo, Y ; Grayden, DB (WILEY, 2022-07)
    OBJECTIVE: Emerging evidence has shown that ambient air pollution affects brain health, but little is known about its effect on epileptic seizures. This work aimed to assess the association between daily exposure to ambient air pollution and the risk of epileptic seizures. METHODS: This study used epileptic seizure data from two independent data sources (NeuroVista and Seer App seizure diary). In the NeuroVista data set, 3273 seizures were recorded using intracranial electroencephalography (iEEG) from 15 participants with refractory focal epilepsy in Australia in 2010-2012. In the seizure diary data set, 3419 self-reported seizures were collected through a mobile application from 34 participants with epilepsy in Australia in 2018-2021. Daily average concentrations of carbon monoxide (CO), nitrogen dioxide (NO2 ), ozone (O3 ), particulate matter ≤10 μm in diameter (PM10 ), and sulfur dioxide (SO2 ) were retrieved from the Environment Protection Authority (EPA) based on participants' postcodes. A patient-time-stratified case-crossover design with the conditional Poisson regression model was used to determine the associations between air pollutants and epileptic seizures. RESULTS: A significant association between CO concentrations and epileptic seizure risks was observed, with an increased seizure risk of 4% (relative risk [RR]: 1.04, 95% confidence interval [CI]: 1.01-1.07) for an interquartile range (IQR) increase of CO concentrations (0.13 parts per million), whereas no significant associations were found for the other four air pollutants in the whole study population. Female participants had a significantly increased risk of seizures when exposed to elevated CO and NO2 , with RRs of 1.05 (95% CI: 1.01-1.08) and 1.09 (95% CI: 1.01-1.16), respectively. In addition, a significant association was observed between CO and the risk of subclinical seizures (RR: 1.20, 95% CI: 1.12-1.28). SIGNIFICANCE: Daily exposure to elevated CO concentrations may be associated with an increased risk of epileptic seizures, especially for subclinical seizures.
  • Item
    Thumbnail Image
    A Neural Mass Model of Spontaneous Burst Suppression and Epileptic Seizures
    Freestone, DR ; Nesic, D ; Jafarian, A ; Cook, MJ ; Grayden, DB (IEEE, 2013)
    The paper presents a neural mass model that is capable of simulating the transition to and from various forms of paroxysmal activity such as burst suppression and epileptic seizure-like waveforms. These events occur without changing parameters in the model. The model is based on existing neural mass models, with the addition of feedback of fast dynamics to create slowly time varying parameters, or slow states. The goal of this research is to establish a link between system properties that modulate neural activity and the fast changing dynamics, such as membrane potentials and firing rates that can be manipulated using electrical stimulation. Establishing this link is likely to be a necessary component of a closed-loop system for feedback control of pathological neural activity.
  • Item
    Thumbnail Image
    INFERRING PATIENT-SPECIFIC PHYSIOLOGICAL PARAMETERS FROM INTRACRANIAL EEG: APPLICATION TO CLINICAL DATA
    Shmuely, S ; Freestone, DR ; Grayden, DB ; Nesic, D ; Cook, M (WILEY-BLACKWELL, 2012-09-01)
    Purpose: Intracranial EEG (iEEG) provides information regarding where and when seizures occur, whilst the underlying mechanisms are hidden. However physiologically plausible mechanisms for seizure generation and termination are explained by neural mass models, which describe the macroscopic neural dynamics. Fusion of models with patient-specific data allows estimation and tracking of the normally hidden physiological parameters. By monitoring changes in physiology, a new understanding of seizures can be achieved. This work addresses model-data fusion for iEEG for application in a clinical setting. Method: Data was recorded from three patients undergoing evaluation for epilepsy-related surgery at St. Vincent's Hospital, Melbourne. Using this data, we created patient-specific neural mass mathematical models based on the formulation of Jansen and Rit (1995). The parameters that were estimated include the synaptic gains, time constants, and the firing threshold. The estimation algorithm utilized the Unscented Kalman Filter (Julier and Uhlmann, 1997). Result: We demonstrate how parameters changed in relation to seizure initiation, evolution and termination. We also show within-patient (across different seizures) and between-patient specificity of the parameter estimates. Conclusion: The fusion of clinical data and mathematical models can be used to infer valuable information about the underlying mechanisms of epileptic seizure generation. This information could be used to develop novel therapeutic strategies
  • Item
    Thumbnail Image
    INFERRING PATIENT-SPECIFIC PHYSIOLOGICAL PARAMETERS FROM INTRACRANIAL EEG: THEORETICAL STUDIES
    Freestone, DR ; Grayden, DB ; Cook, M ; Nesic, D (WILEY-BLACKWELL, 2012-09)