Wyniki wyszukiwania - BazTech

1

Hyp-Net: Automated detection of hypertension using deep convolutional neural network and Gabor transform techniques with ballistocardiogram signals

Gupta Kapil, Bajaj Varun, Ansari Irshad Ahmad, Acharya U. Rajendra

Biocybernetics and Biomedical Engineering

|

2022

|

Vol. 42, no. 3

784--796

EN

Hypertension (HPT) is a physiological abnormality characterized by high blood pressure, headache, wooziness, and fainting that may lead to various heart, kidney, or brain diseases. Detection and continuous monitoring of HPT by sphygmometer is arduous and hectic. Nowadays, ballistocardiogram (BCG) signals are used to determine HPT as it indicates the vibration of the heart. Usual linear and nonlinear hand-crafted machine learning methods are subjective, involve decomposition of signal, features elicitation, selection, and classification steps. In this work, a completely automated HPT detection system is proposed using time–frequency (T-F) spectral images and a convolutional neural network (CNN) for the accurate detection of HPTusing BCG signals. The BCG signals are subjected to Gabor transform (GT), smoothed pseudo-Wigner Ville distribution (SPWVD), and short-time Fourier transform (STFT) techniques to obtain T-F spectral images. These T-F spectral images are fed to benchmarked Alex-Net, Res-Net50, and proposed CNN (Hyp-Net) to develop the automated HPT detection model with a 10-fold cross-validation scheme. The proposed Hyp-Net obtained the highest detection accuracy of 97.65% with GT-based spectral images. In comparison to Alex-Net and Res-Net50 pre-trained models, our developed Hyp-Net needs minimal learnable parameters, makes it computationally fast and more efficient. This shows that our proposed model has outperformed other two transfer learning methods. The experimental results with collected BCG signals from a public dataset are provided to show the effectiveness of the presented technique for automated detection of HPT.

2

Wykorzystanie czujników inercyjnych do zdalnego monitorowania aktywności i parametrów życiowych

Osiadała Natalia, Kołakowski Marcin

Przegląd Telekomunikacyjny + Wiadomości Telekomunikacyjne

|

2019

|

nr 6

469--472, CD

PL

W referacie opisano czujnik do zdalnego monitorowania aktywności i parametrów życiowych, który opracowano z myślą o zastosowaniu w systemach inteligentnego domu przeznaczonych dla osób starszych. Zaprezentowany układ wyposażony jest w dwa akcelerometry oraz układ radiowy Wi-Fi pozwalający na komunikację z otoczeniem. W referacie zamieszczono wyniki badań eksperymentalnych układu.

EN

The paper describes a sensor for remote monitoring of activity and vital parameters, which was developed for use in smart home system for the elderly. The presented sensor has two accelerometers and a Wi-Fi radio module used for communication with system server. The paper includes the results of experiments.

3

Metody bezkontaktowego monitorowania pracy serca

Szymanowski R., Sondej T., Różanowski K.

Elektronika : konstrukcje, technologie, zastosowania

|

2013

|

Vol. 54, nr 10

108-111

PL

W artykule przedstawiono metody monitorowania pracy serca bez potrzeby bezpośredniego kontaktu elektrycznego ze skórą człowieka. Zastosowane bezkontaktowe metody pomiarowe eliminują stosowanie żeli przewodzących, powodujących nadmierną reakcję alergiczną skóry oraz uciążliwość wykonywania badań. Ponadto wyznaczona pośrednio na podstawie pomiarów wartość tętna i charakter zmian mogą być stosowane do oceny stanu psychofizjologicznego.

EN

We present two methods of monitoring of heart rate without direct conductive skin-contact. The contactless measurement techniques are utilized to avoid the use of conductive gel and the skin irritation due to such a contact. Moreover, this technologies are used to estimate the psychophysiological condition. Two non-contact methods similar to the electrocardiography and ballistocardiography for heart rate detection through clothes are presented.

4

Światłowodowy czujnik czynności oddechowej oraz pracy serca do zastosowań w środowisku rezonansu magnetycznego

Dziuda Ł., Skibniewski F., Nowicki G., Lewandowski J.

Przegląd Elektrotechniczny

|

2012

|

R. 88, nr 10b

120-124

PL

Artykuł prezentuje projekt oraz początkową fazę testów eksperymentalnych czujnika optycznego umożliwiającego rejestrację aktywności oddechowej oraz pracy serca. Czujnik zbudowany na bazie światłowodowej siatki Bragga charakteryzuje się odpornością na zakłócenia elektromagnetyczne, co pozwala na realizację pomiarów w środowisku rezonansu magnetycznego. Analiza statystyczna otrzymanych wyników wykazała, że czujnik charakteryzuje się dobrymi właściwościami pomiarowymi, mogącymi sprostać podstawowym potrzebom monitorowania pacjenta.

EN

This article shows the design and the initial phase of experimental tests carried out on a fiber-optic sensor allowing breathing and heart rate to be registered. The sensor, based on the fiber Bragg grating, is immune to electromagnetic interference, which enables measurements to be carried out in the magnetic resonance imaging environment. A statistical analysis of the obtained results shows that the sensor has good measuring properties able to meet the basic requirements for patient monitoring.

5

Zastosowanie elektromechanicznej folii EMFi do pomiaru tętna

Sondej T., Różanowski K., Dziuda Ł.

Pomiary Automatyka Kontrola

|

2012

|

R. 58, nr 3

264-267

PL

W artykule przedstawiono sposób użycia elektromechanicznej folii EMFi do pomiaru częstości skurczów serca człowieka. Stanowisko pomiarowe składa się z dwóch sensorów EMFi, umieszczonych na typowym krześle biurowym, modułu referencyjnego do pomiaru EKG i SpO2 oraz karty DAQ. Wyznaczono wartość tętna dla sygnału EMFi i EKG. Błąd pomiaru tego parametru określony z serii pomiarów wykonanych w czasie 1 minuty i szacowany jako różnica między sygnałem pochodzącym z EMFi a sygnałem EKG wynosi < 1 %.

EN

The Electro-Mechanical Film (EMFi) sensor for measurement of the human heart rate is presented. EMFi material is used to non-invasive technique for the assessment of the cardiac function called ballistocardiography (BCG). Construction and sensor operation are shown in Fig. 2. There was built a measurement setup from a typical office chair with ECG and SpO2 reference module (Fig. 4) for obtaining the ballistocardiogram . The EMFi signals from a subject sitting on the chair are recorded by a specially-designed amplifiers, analog filters and dedicated data acquisition card. A Matlab PC software was used for recording, data processing and displaying the results. EMFi sensors were mounted on the chair seat and backrest. The example signals are shown in Figs. 5, 6. The authors' method was evaluated by comparing the beat-to-beat heart rate (HRBB) values obtained from the EMFi sensors with the accurate reference values calculated from the reference ECG signal. The experimental results (calculated heart rate) are presented in Figs. 9, 10. The measurement error for the HRBB parameter, estimated as a difference between the EMFi and reference signal, is below 1 %, during 60-second recording.

6

Modal analysis of physiological model of avian embryos – model parameter estimation. Part II

Iwaniec J., Iwaniec M., Pawlak K.

Vibrations in Physical Systems

|

2008

|

Vol. 23

171--176

EN

The research carried out by authors aimed at inventing the sensitive non-invasive method for ethical investigation into the influence of chemical substances and physical factors on the growth of organisms in the prenatal phase. The knowledge of effects of chemical substances on the course of embryogenesis provides the possibility of testing new medicines and assessing the teratogenous influence of chemical substances already present in the natural environment on alive organisms. The results of the research are presented in the form of the series of two papers under the same title. The first paper, denoted as the part I, concerns the invented noninvasive method dedicated for measurements of low-amplitude vibrations induced by the work of cardiac muscles. The current paper (part II) is dedicated to vibration analysis of measured ballistocardiograps carried out by means of the classical LSCE modal analysis method. Due to the rapid increase in the cardiac muscle mass, the parameters of the biomechanical model of developing avian embryo are nonlinear. Nevertheless in a single measurement session the changes in the object parameters are negligible, which enables application of the classical modal analysis methods.

7

Modal analysis of physiological model of avian embryos - identification experiment. Part I

Iwaniec J., Iwaniec M., Pawlak K.

Vibrations in Physical Systems

|

2008

|

Vol. 23

165--168

EN

Avian embryos are commonly used as ethically acceptable physiological models for research into prenatal live. Although easily available, avian embryos are difficult objects for investigation. The main difficulties result from the presence of shell separating the embryo from environmental influences and, what is more important, from the presence and interference of many biomechanical, chemical, electrical and other phenomena. The tiny changes in the object parameters of different nature are hardly measurable by means of the indirect methods. Investigation into the embryo prenatal life requires application of the non-invasive methods, which complicates the measurements. Taking into account the fact that as early as in the second day of incubation the avian embryo hart begins to work being by itself at the beginning of mutual conversions, it is easy to become conscious of the scale of the measurements and resulting identification problems. In the process of forming and growth, the heart of avian embryo undergoes constant biomechanical modifications. This aspects of embryogenesis are correlated to electrical and hydro-dynamical activity. Only mechanical vibrations and acoustical effects can be analysed on the basis of the indirect non-invasive measurements. The research carried out by authors is presented in the series of two papers. The current paper (part I) concerns the innovative noninvasive method dedicated for measurements of low-amplitude vibrations induced by the work of cardiac muscles while in the second paper (part II) there is presented vibration analysis carried out by means of the classical LSCE modal analysis method. Although the biomechanical model of developing heart is characterized by significant nonlinearity of observed parameters resulting from rapid increase in the cardiac muscle mass, during a given measurement session the changes in the object parameters are negligible and the assumptions of modal analysis are valid. The changes in the proportions and organ masses as well as in time histories and values of forces generated by beating heart are the source of changes not only in the estimated modal model parameters but also in the model order. Some natural frequencies and corresponding mode shapes evolve and change their values, decay or appear. This variability can be observed in the longer time scale (e.g. days). In a single identification experiment lasting up to tens of minutes per day, after elimination of disturbances, the measured vibroacoustical signal can be treated as quasi stationary. Modal model parameters and the model order were determined for each separate measurement session lasting 10 to 40 minutes. Parameter values estimated for the consecutive measurement sessions made it possible to track the changes in modal model in the consecutive days of incubation.