• Ásgerdur Arna Pálsdóttir
  • Dagbjört Helga Eiriksdóttir
4. semester , Sundhedsteknologi, Kandidat (Kandidatuddannelse)
Cardiac disease is the leading cause of death globally. Most cardiovascular disease can be prevented or managed with early detection. Early detection could be increased by routine monitoring where a disease could potentially be found before any symptoms appear. Traditional methods for detecting heart disease include photoplethysmography, ECG, accelerometer etc. These methods are limited by the requirement of physical contact. A promising non-contact method for detecting cardiac activity utilizes the vibrations on the surface of the body to frequency modulate a carrier ultrasound wave that can be recorded by a microphone. The goal of this project is to examine this method further.
A cardiac signal was recorded from 13 healthy subjects using an ultrasound transducer and microphone, at four distances (10 - 40 cm) with T-shirt on or off. The ultrasound (US) signal was demodulated using arctan demodulation to get the displacement of the chest wall. Continuous wavelet transformation and different filters were used to analyze different frequency bands of the US signal. The waveform of an individual heart beat from the acceleration US signal was compared with a corresponding heart beat from an accelerometer. Lastly, physiological events associated with heart valve opening and closing were transposed from the acceleration US signal to the displacement US signal.
The linking of different frequency bands with movements caused by different physiological events such as breathing and heart beats corresponded to literature in terms of frequency and amplitude (mean peak-to-peak amplitude (ppAmp), 3.5 mm and 0.3 mm, respectively). Furthermore, mean ppAmp for low frequency seismocardiographic (SCG) signals and frequency band corresponding to valve sounds was calculated (0.08 mm and 0.007 mm, respectively).
The waveform correlated well with the accelerometer where the highest correlation was 0.969.
The displacement US signal contains a lot of information that has potential for being used in a clinical and out of hospital setting. Future research would include establishing physiological events on the displacement US signal along with improving the system setup.
Udgivelsesdato7 jun. 2018
Antal sider69
ID: 280486332