Michael D. Furman

Non-parametric early seizure detection in an animal model of temporal lobe epilepsy

The performance of five non-parametric, univariate seizure detection schemes (embedding delay, Hurst scale, wavelet scale, nonlinear autocorrelation and variance energy) were evaluated as a function of the sampling rate of EEG recordings, the electrode types used for EEG acquisition, and the spatial location of the EEG electrodes in order to determine the applicability of the measures in real-time closed-loop seizure intervention. The criteria chosen for evaluating the performance were high statistical robustness (as determined through the sensitivity and the specificity of a given measure in detecting a seizure) and the lag in seizure detection with respect to the seizure onset time (as determined by visual inspection of the EEG signal by a trained epileptologist). An optimality index was designed to evaluate the overall performance of each measure. For the EEG data recorded with microwire electrode array at a sampling rate of 12 kHz, the wavelet scale measure exhibited better overall performance in terms of its ability to detect a seizure with high optimality index value and high statistics in terms of sensitivity and specificity.

Using recurrence quantification analysis determinism for noise removal in cardiac optical mapping

Selecting signal processing parameters in optical imaging by utilizing the change in Determinism, a measure introduced in Recurrence Quantification Analysis, provides a novel method using the change in residual noise Determinism for improving noise quantification and removal across signals exhibiting disparate underlying tissue pathologies. The method illustrates an improved process for selecting filtering parameters and how using measured signal-to-noise ratio alone can lead to improper parameter selection.

ANALYSIS OF HIGH-RESOLUTION MICROELECTRODE EEG RECORDINGS IN AN ANIMAL MODEL OF SPONTANEOUS LIMBIC SEIZURES

We perform a systematic data analysis on high resolution (0.5–12 kHz) multiarray microelectrode recordings from an animal model of spontaneous limbic epilepsy, to investigate the role of high frequency oscillations and the occurrence of early precursors for seizures. Results of spectral analysis confirm the importance of very high frequency oscillations (even greater than 600 Hz) in normal (healthy) and abnormal (epileptic) hippocampus. Furthermore, we show that the measures of Recurrence Quantification Analysis (RQA) and Recurrence Time Statistics (RTS) are successful in indicating, rather uniquely, the onset of ictal state and the occurrence of some warnings/precursors during the pre-ictal state, in contrast to the linear measures investigated.

Biologically-inspired object recognition system with features from complex wavelets

In this paper, a novel cortex-inspired feed-forward hierarchical object recognition system based on complex wavelets is proposed and tested. Complex wavelets contain three key properties for object representation: shift invariance, which enables the extraction of stable local features; good directional selectivity, which simplifies the determination of image orientations; and limited redundancy, which allows for efficient signal analysis using the multi-resolution decomposition offered by complex wavelets. In this paper, we propose a complete cortex-inspired object recognition system based on complex wavelets. We find that the implementation of the HMAX model for object recognition in [1, 2] is rather over-complete and includes too much redundant information and processing. We have optimized the structure of the model to make it more efficient. Specifically, we have used the Caltech 5 standard dataset to compare with Serres model in [2] (which employs Gabor filter bands). Results demonstrate that the complex wavelet model achieves a speed improvement of about 4 times over the Serre model and gives comparable recognition performance.

Nonlinear Synchronization Analysis of Spatiotemporal Heart Data

A high‐speed video camera and voltage‐sensitive dyes were used to acquire high resolution (80×80 pixels) and high‐speed (500 μs/frame) optical signals of ventricular fibrillation in a Langendorff‐perfused porcine heart. The resulting spatiotemporal dynamics were recorded before and after the application of a defibrillation shock in order to study the mechanism of defibrillation failure. We calculate nonlinear synchronization index measures to qualify the evolution of different types of activity on the heart surface (focal, reentry). We observe changes with time in the spatial distribution of the first Fourier mode, showing that two main types of activity compete on the heart surface during a failed defibrillation.

Nonlinear analysis of cardiac optical mapping data reveals ordered period in defibrillation failure

A high-speed video camera and voltage-sensitive dyes were used to acquire high resolution (80times80 pixels) and high-speed (500mus/frame) optical signals of ventricular fibrillation in a Langendorff-perfused porcine heart. The resulting spatiotemporal dynamics were recorded before and after the application of a defibrillation shock in order to study the mechanism of defibrillation failure. We used recurrence plots as a tool to qualify the evolution of ordered behavior on the heart surface before fibrillation was reestablished in defibrillation failure. Such ordered periods may point to robust periods in which the defibrillation attempt has had the most effect and may provide a window in which a smaller, corrective shock may be applied to achieve defibrillation