Wyniki wyszukiwania - Biblioteka Nauki

1

Prediction of fatigue and sleep onset using HRV analysis

100%

Szypulska M. , Piotrowski Z.

Elektronika : konstrukcje, technologie, zastosowania

|

2012

|

tom Vol. 53, nr 12

16-19

EN

In the paper objective syndromes associated with sleep onset and fatigue based on the analysis of heart rate variability (HRV) have been presented. Temporal and frequency parameters have been given particular attention. An algorithm for detection of the moment of sleep onset and fatigue has been described. It is based on the determination of the LF/HF ratio on the basis of an RR tachogram and assigning its value to three basic states: activity, drowsiness and sleep. Results of experiments conducted on people without dysfunctions in electrocardiogram (ECG) waveforms have been presented and discussed.

PL

W artykule przedstawiono obiektywne syndromy towarzyszące zasypianiu i zmęczeniu na podstawie analizy Zmienności Rytmu Serca (HRV). Wyszczególniono parametry analizy czasowej i częstotliwościowej. Opisano algorytm detekcji momentu zasypiania i zmęczenia bazujący na wyznaczeniu wartości współczynnika LF/HF na podstawie tachogramu RR i przypisaniu jego wartości do trzech podstawowych stanów: aktywności, senności oraz snu. Przedstawiono i omówiono wyniki eksperymentów przeprowadzonych na osobach bez dysfunkcji przebiegów elektrokardiogramu (EKG).

2

Classification of falling asleep states using HRV analysis

100%

Piotrowski Z. , Szypulska M.

Biocybernetics and Biomedical Engineering

|

2017

|

tom Vol. 37, no. 2

290--301

EN

The article presents the results of studies on drowsiness and drowsiness detection performed using heart rate variability analysis (HRV). The results of those studies indicate that the most significant parameters, from the standpoint of classification of drowsiness are the following parameters of the HRV analysis: the low and high frequency band the ratio of the tachogram power in the LF and HF bands, and the total power distribution. The best detection results were obtained for the following methods, in the following order: the nearest neighborhood with metrics: standardized Euclides and Mahalanobis, the square discriminant analysis, and the Bayesian classifier. In order to classify drowsiness periods, a neural network was also used; it consisted of four inputs, six neurons in the hidden layer, and three outputs, one of which was assigned to one of the accepted classes. In order to obtain the most effective learning, a linear feed forward network was designed using back propagation of errors and the RPROP algorithm. In the case of this type of networks, the achieved accuracy of the individual classes was on the level of 98.7%.

3

Modular microelectronic system for in vivo electrical stimulation and recording at up to 512 electrodes

41%

Szypulska M. , Wiacek P. , Skoczen A. , Fiutowski T. , Ahmed I. , Jedraczka M. , Dusik J. , Dwuznik M. , Kublik E. , Dabrowski W. , Hottowy P.

Acta Neurobiologiae Experimentalis

|

2017

|

tom 77

|

nr Suppl.1

EN

INTRODUCTION: Multielectrode silicon probes can record neuronal signals with combination of spatial and temporal resolution that other recording techniques cannot provide. Here we propose a novel microelectronic system that combines this functionality with advanced electrical stimulation. AIM(S): We designed a modular system for multielectrode electrical stimulation and recording in the brain of a living animal. It can be combined with any silicon probe used for brain research. It can generate complex sequences of stimulation pulses and simultaneously record at up to 512 electrodes. It can use up to 4 silicon probes in parallel, providing bidirectional communication with populations of neurons simultaneously in several brain areas. METHOD(S): The system is based on a dedicated multichannel CMOS chip. The chip includes 64 channels, digital circuitry for real-time communication with the control computer and a multiplexer that sends amplified signals from 64 electrodes into a single output line. The amplifier gain can be changed from 110 to 550. The low cut‑off frequency is set between 200 mHz and 3 Hz, the anti-aliasing filter is set at 7 kHz and the sampling rate is 40 kHz. The stimulation signal is controlled independently for each channel with 12-bit resolution and refresh rate of 40 kHz. Each amplifier can be disconnected from the electrode for the duration of the stimulation pulse for the artifact reduction. Up to 8 chips can be controlled in parallel with dedicated LabView software. RESULTS: Base version of the system was produced and tested with positive results. The final system is in the integration phase. We plan the first experiments to take place in the fall 2017 at the Nencki Institute for Experimental Biology. CONCLUSIONS: The reported system can generate complex sequences of stimulation pulses and record neuronal signals with very low artifacts at 512 electrodes, making it a powerful tool for mapping of the functional connections between brain circuits. FINANCIAL SUPPORT: Grant 2013/08/W/NZ4/00691, Polish National Science Centre.

4

Microelectronic system for low-artifact electrical stimulation and recording of brain activity at up to 512 electrodes

41%

Szypulska M. , Wiacek P. , Skoczen A. , Ahmed I. , Futowski T. , Jurgielewicz P. , Kolodziej K. , Czerwinski M. B. , Wojcik D. K. , Mindur B. , Kublik E. , Dabrowski W. , Hottowy P.

Acta Neurobiologiae Experimentalis

|

2019

|

tom 79

|

nr Suppl.1

EN

INTRODUCTION: We present a novel microelectronic system for in vivo stimulation and recording of neuronal activity. The system is intended for use with multielectrode silicon probes and is based on a dedicated 64‑channel CMOS chip. It can generate complex sequences of microstimulation pulses and simultaneously record (with low artifacts) neuronal responses at up to 512 electrodes. The system is compatible with most silicon probes used in the brain research and can use up to four probes in parallel, providing bidirectional communication with populations of neurons simultaneously in several brain areas. Each channel of the chip includes a recording amplifier and a stimulation circuit. The amplifier has adjustable gain (110‑550x), low cut‑off frequency (1.4‑7 Hz), and anti‑aliasing filter frequency (1.2‑14 kHz). The input‑referred noise is 6.8 µV. Signals from all the channels are digitized at 40 kHz. The stimulation signal is defined independently for each channel with 40 kHz refresh rate. The stimulation artifacts are reduced by temporally disconnecting the amplifiers from electrodes and optimization of the pulse waveform. METHOD(S): The system has been tested in experiments exploring somatosensory thalamo-cortical network in rodents. 2‑3 weeks before surgery, animals received injections of AAV‑hSyn‑ChR2‑EYFP viral vector. In anesthetized animals, multichannel probes were inserted into the barrel cortex and/or sensory thalamus for recording of LFPs and multi-unit responses to microstimulation delivered to various nodes of thalamo‑cortical network. Electrically evoked activity was compared with responses to natural whisker deflection and optical stimulation. RESULTS: The reported system can generate complex patterns of stimulation pulses and record neuronal signals with very low artifacts at up to 512 electrodes, making it a powerful tool for mapping of the functional connections between brain circuits. FINANCIAL SUPPORT: Supported by Polish National Science Centre grant 2013/08/W/NZ4/00691.