Wyniki wyszukiwania - BazTech

1

A novel multi-scale convolutional network with attention-based bidirectional gated recurrent unit for atrial fibrillation discrimination

Wang Tan, Qin Yan

Biocybernetics and Biomedical Engineering

|

2021

|

Vol. 41, no. 2

445--455

EN

Atrial fibrillation (AF) is a major cardiovascular disease that has affected thousands of individuals worldwide. The electrocardiogram (ECG) is the most extensively applied approach to detect AF at present, while the traditional detection strategy based on the visual observation of ECG data is often laborious and inefficient. In this work, we specially designed an intelligent recognition system based on a novel convolutional neural network that utilizes the multi-scale convolution kernel and bidirectional gated recurrent unit with attention mechanism for AF detection. Also, two standard control groups using 10-fold cross-validation were performed to assess the validity of the proposed model. The empirical results not only demonstrate the high efficiency of multi-scale convolution kernel, but also show that the model has a more superior classification performance to several state-of the-art methods with an accuracy of 98.3% and 97.7% on two public databases, respectively. Due to its high performance, we plan to develop the model into portable devices to benefit more individuals such as the elderly and athletes.

2

Automatic diagnosis of atrial fibrillation based on online RR interval analysis

Sokol Yevgen, Shapov Pavel, Shyshkin Mykhailo, Koval Sergiy, Tomashevskyi Roman

Zeszyty Naukowe Instytutu Pojazdów / Politechnika Warszawska

|

2019

|

z. 2/120

53--60

EN

The article presents the results of research of the tachogram of the cardiac signal with areas of atrial fibrillation. Using the wavelet transform to analyse non-stationary processes, it was shown that dispersion decomposition is an instantaneous correlation between the wavelet spectrum and has two implementations: frequency non-stationarity models - for frequency and time estimates, the ability to specify a parameter that should be relevant for the onset of atrial fibrillation. The calculation of this parameter can be used to detect fibrillation during the online recording of RR intervals.

PL

W artykule przedstawiono wyniki badań tachogramu sygnału sercowego z obszarami migotania przedsionków. Wykorzystanie transformacji falkowej do analizy procesów niestacjonarnych wykazało, że dekompozycja dyspersji jest chwilową korelacją między widmem falkowym i ma dwie implementacje: modele niestacjonarne częstotliwości - dla szacunków częstotliwości i czasu, możliwość określenia parametru, który powinien mieć znaczenie dla początku migotania przedsionków. Obliczenie tego parametru może być wykorzystane do wykrycia migotania podczas rejestracji online odstępów RR.

3

Automated diagnosis of atrial fibrillation ECG signals using entropy features extracted from flexible analytic wavelet transform

Kumar M., Pachori R. B., Rajendra Acharya U.

Biocybernetics and Biomedical Engineering

|

2018

|

Vol. 38, no. 3

564--573

EN

Atrial fibrillation (AF) is the most common type of sustained arrhythmia. The electrocardiogram (ECG) signals are widely used to diagnose the AF. Automated diagnosis of AF can aid the clinicians to make a more accurate diagnosis. Hence, in this work, we have proposed a decision support system for AF using a novel nonlinear approach based on flexible analytic wavelet transform (FAWT). First, we have extracted 1000 ECG samples from the long duration ECG signals. Then, log energy entropy (LEE), and permutation entropy (PEn) are computed from the sub-band signals obtained using FAWT. The LEE and PEn features are extracted from different frequency bands of FAWT.We have found that LEE features showed better classification results as compared to PEn. The LEE features obtained maximum accuracy, sensitivity, and specificity of 96.84%, 95.8%, and 97.6% respectively with random forest (RF) classifier. Our system can be deployed in hospitals to assist cardiac physicians in their diagnosis.

4

On ultrasound classification of stroke risk factors from randomly chosen respondents using non-invasive multispectral ultrasonic brain measurements and adaptive profiles

Wrobel M., Dabrowski A., Kolany A., Olak-Popko A., Olszewski R., Karlowicz P.

Biocybernetics and Biomedical Engineering

|

2016

|

Vol. 36, no. 1

19--28

EN

In this paper, we present a new brain diagnostic method based on a computer aided multispectral ultrasound diagnostics method (CAMUD). We explored the standard values of the relative time of flight (RIT), as well as the attenuation, ATN, of multispectral longitudinal ultrasound waves propagated non-invasively through the brains of a standard Caucasian volunteer population across different ages and genders. For the interpretation of the volunteers health questionnaire and ultrasound data we explored various clustering and classification algorithms, such as PCA and ANOVA. We showed that the RIT and ATN values provide very good estimators of possible physiological changes in the brain tissue and can differentiate the possible high-risk groups obtained by other groups and methods (Russo et al. [1]; Lloyd-Jones et al. [2]; Medscape [3]). Special attention should be given to the subgroup which included almost 39% of the volunteers. Respondents in this group have a significantly increased minimum ATN value (see Classification Trees). These values are strongly correlated with the identified risk of stroke factors being: age, increased alcohol consumption, cases of heart disease and stroke in the family as already shown by Rusco and as incorporated into Lloyd-Jones et al., ‘‘Heart Disease and Stroke Statistics – 2009 Update’’, by the American Heart Association (AHA) and American Stroke Association (ASA), as updated recently in the 2015 ‘‘Stroke Prevention Guidelines’’.