Overcoming Dimensionality Constraints: A Gershgorin Circle Theorem-Based Feature Extraction for Weighted Laplacian Matrices in Computer Vision Applications.

In graph theory, the weighted Laplacian matrix is the most utilized technique to interpret the local and global properties of a complex graph structure within computer vision applications. However, with increasing graph nodes, the Laplacian matrix's dimensionality also increases accordingly. Therefore, there is always the "curse of dimensionality"; In response to this challenge, this paper introduces a new approach to reducing the dimensionality of the weighted Laplacian matrix by utilizing the Gershgorin circle theorem by transforming the weighted Laplacian matrix into a strictly diagonal domain and then estimating rough eigenvalue inclusion of a matrix. The estimated inclusions are represented as reduced features, termed GC features; The proposed Gershgorin circle feature extraction (GCFE) method was evaluated using three publicly accessible computer vision datasets, varying image patch sizes, and three different graph types. The GCFE method was compared with eight distinct studies. The GCFE demonstrated a notable positive Z-score compared to other feature extraction methods such as I-PCA, kernel PCA, and spectral embedding. Specifically, it achieved an average Z-score of 6.953 with the 2D grid graph type and 4.473 with the pairwise graph type, particularly on the E_Balanced dataset. Furthermore, it was observed that while the accuracy of most major feature extraction methods declined with smaller image patch sizes, the GCFE maintained consistent accuracy across all tested image patch sizes. When the GCFE method was applied to the E_MNSIT dataset using the K-NN graph type, the GCFE method confirmed its consistent accuracy performance, evidenced by a low standard deviation (SD) of 0.305. This performance was notably lower compared to other methods like Isomap, which had an SD of 1.665, and LLE, which had an SD of 1.325; The GCFE outperformed most feature extraction methods in terms of classification accuracy and computational efficiency. The GCFE method also requires fewer training parameters for deep-learning models than the traditional weighted Laplacian method, establishing its potential for more effective and efficient feature extraction in computer vision tasks.

Gershgorin circle theorem-based feature extraction for biomedical signal analysis.

Recently, graph theory has become a promising tool for biomedical signal analysis, wherein the signals are transformed into a graph network and represented as either adjacency or Laplacian matrices. However, as the size of the time series increases, the dimensions of transformed matrices also expand, leading to a significant rise in computational demand for analysis. Therefore, there is a critical need for efficient feature extraction methods demanding low computational time. This paper introduces a new feature extraction technique based on the Gershgorin Circle theorem applied to biomedical signals, termed Gershgorin Circle Feature Extraction (GCFE). The study makes use of two publicly available datasets: one including synthetic neural recordings, and the other consisting of EEG seizure data. In addition, the efficacy of GCFE is compared with two distinct visibility graphs and tested against seven other feature extraction methods. In the GCFE method, the features are extracted from a special modified weighted Laplacian matrix from the visibility graphs. This method was applied to classify three different types of neural spikes from one dataset, and to distinguish between seizure and non-seizure events in another. The application of GCFE resulted in superior performance when compared to seven other algorithms, achieving a positive average accuracy difference of 2.67% across all experimental datasets. This indicates that GCFE consistently outperformed the other methods in terms of accuracy. Furthermore, the GCFE method was more computationally-efficient than the other feature extraction techniques. The GCFE method can also be employed in real-time biomedical signal classification where the visibility graphs are utilized such as EKG signal classification.

Non-stationary neural signal to image conversion framework for image-based deep learning algorithms.

This paper presents a time-efficient preprocessing framework that converts any given 1D physiological signal recordings into a 2D image representation for training image-based deep learning models. The non-stationary signal is rasterized into the 2D image using Bresenham's line algorithm with time complexity O(n). The robustness of the proposed approach is evaluated based on two publicly available datasets. This study classified three different neural spikes (multi-class) and EEG epileptic seizure and non-seizure (binary class) based on shapes using a modified 2D Convolution Neural Network (2D CNN). The multi-class dataset consists of artificially simulated neural recordings with different Signal-to-Noise Ratios (SNR). The 2D CNN architecture showed significant performance for all individual SNRs scores with (SNR/ACC): 0.5/99.69, 0.75/99.69, 1.0/99.49, 1.25/98.85, 1.5/97.43, 1.75/95.20 and 2.0/91.98. Additionally, the binary class dataset also achieved 97.52% accuracy by outperforming several others proposed algorithms. Likewise, this approach could be employed on other biomedical signals such as Electrocardiograph (EKG) and Electromyography (EMG).

Roboneuron: a simple and robust real-time analog spike simulator and calibrator.

BACKGROUND: Modern computerized spike recording systems are increasingly powerful and sophisticated. However, this increases the importance of performing validation by recording signals from a system with a known input-output relationship. NEW METHOD: We present here a simple and robust analog circuit that uses a minimum number of commonly available components to simulate two independently spiking neurons. The two neurons generate asynchronous overlapping spikes. These can be independently set to spike at either a constant rate, or at a rate set by an external control voltage. RESULTS: The circuit is simple enough to easily assemble by hand, however, standard files for ordering commercial printed circuit boards are also supplied. Several units were built by different people, using both hand-assembly and commercially manufactured printed circuit boards: all worked well. The circuit is robust with respect to supply voltages and component values. COMPARISON WITH EXISTING METHODS: Existing analog circuits tend to be complex, hard to assemble, and use hard-to-find components. Digital simulators typically require specific development systems that have steep learning curves and are likely to change radically or become unavailable very quickly. This system has been optimized to be robust, simple, and use only commonly available components. CONCLUSIONS: When validating a system there could be an advantage to using a calibrator that is robust, whose input-output relationship is simple, and whose design is stable over time.

Effect of αlipoic acid and silymarin on bladder outlet obstruction.

The aim of the present study was to determine whether the treatment of obstructed rat bladders with αlipoic acid (ALA) and silymarin reverses the biochemical and physiological responses to bladder outlet obstruction (BOO). A total of 32 adult Sprague Dawley rats were divided into four groups (n=8 per group): sham (placebo surgery) animals with no treatment (group 1); control animals with surgically induced BOO (group 2); obstructed rats treated with ALA (group 3); and obstructed rats treated with silymarin (group 4). Histological evaluation, bladder weights, collagen structure, TdT-mediated biotin nick end-labeling (TUNEL), inducible nitric oxide sentase (iNOS) mRNA levels, malondialdehyde (MDA) levels and tumor necrosis factor (TNF) levels were investigated. The ALA-treated group had similar bladder weights, collagen levels and TUNEL positivity and decreased iNOS levels compared with the control group, while the silymarin group exhibited further differences. Serum MDA and TNF-α levels were both decreased in the ALA and silymarin groups. ALA treatment reduced the increased oxidative stress and bladder inflammation caused by BOO and may contribute to the protection of bladder function.

Short latency activation of cortex by clinically effective thalamic brain stimulation for tremor.

Deep brain stimulation (DBS) relieves disabling symptoms of neurologic and psychiatric diseases when medical treatments fail, yet its therapeutic mechanism is unknown. We hypothesized that ventral intermediate (VIM) nucleus stimulation for essential tremor activates the cortex at short latencies, and that this potential is related to the suppression of tremor in the contralateral arm. We measured cortical activity with electroencephalography in 5 subjects (seven brain hemispheres) across a range of stimulator settings, and reversal of the anode and cathode electrode contacts minimized the stimulus artifact, allowing visualization of brain activity. Regression quantified the relationship between stimulation parameters and both the peak of the short latency potential and tremor suppression. Stimulation generated a polyphasic event-related potential in the ipsilateral sensorimotor cortex, with peaks at discrete latencies beginning less than 1 ms after stimulus onset (mean latencies 0.9 ± 0.2, 5.6 ± 0.7, and 13.9 ± 1.4 ms, denoted R1, R2, and R3, respectively). R1 showed more fixed timing than the subsequent peaks in the response (P < 0.0001, Levene's test), and R1 amplitude and frequency were both closely associated with tremor suppression (P < 0.0001, respectively). These findings demonstrate that effective VIM thalamic stimulation for essential tremor activates the cerebral cortex at approximately 1 ms after the stimulus pulse. The association between this short latency potential and tremor suppression suggests that DBS may improve tremor by synchronizing the precise timing of discharges in nearby axons and, by extension, the distributed motor network to the stimulation frequency or one of its subharmonics.

Spectral cancellation of microstimulation artifact for simultaneous neural recording in situ.

A fundamental technical hurdle in systems neurophysiology has been to record the activity of individual neurons in situ while using microstimulation to activate inputs or outputs. Stimulation artifact at the recording electrode has largely limited the usefulness of combined stimulating and recording to using single stimulation pulses (e.g., orthodromic and antidromic activation) or to presenting brief trains of pulses to look for transient responses (e.g., paired-pulse stimulation). Using an adaptive filter, we have developed an on-line method that allows continuous extracellular isolation of individual neuron spikes during sustained experimental microstimulation. We show that the technique accurately and robustly recovers neural spikes from stimulation-corrupted records. Moreover, we demonstrate that the method should generalize to any recording situation where a stereotyped, triggered transient might obscure a neural event.

A new transparent multi-unit recording array system fabricated by in-house laboratory technology.

A new recording array system has been developed to record multi-unit activity in rabbit retina. The array consists of individually laid down layers of carbon fiber or tungsten microelectrodes whose center-center spacing can be made less than 100 microm. The array and associated electronics can be constructed by technology typically found in most electrophysiology laboratories. The array is mostly transparent, so that visual stimuli and microscopic examination can take place through it. The array can be manipulated much like a single electrode, and thus can be used to record from multiple tissue sites. Arrays as large as 32 elements have been used, with success rates of about 50% per electrode, with some electrodes picking up more than one cell. Stable recordings have been held for up to 6 h from groups of ganglion cells in an isolated eyecup preparation. These multi-electrode arrays have been used repeatedly in experiments for several months without any obvious degradation in recording quality. Although the arrays are hand-made, their layered method of assembly allows as many as 32 elements to be assembled.