Classification of Internal Carotid Arterial Doppler Signals Using Wavelet-based Neural Networks

Doppler ultrasound is a noninvasive technique which is widely used in medicine for the assessment of blood flow in vessels. Therefore, Doppler ultrasonography is known as a reliable technique, which demonstrates the flow characteristics and resistance of internal carotid arteries in stenosis and occlusion conditions. In this study, internal carotid arterial Doppler signals recorded from 130 subjects that 45 of them had suffered from internal carotid artery stenosis, 44 of them had suffered from internal carotid artery occlusion and the rest of them had been healthy subjects were classified using waveletbased neural network. Spectral analysis of internal carotid arterial Doppler signals was performed using wavelet transform for determining the neural network inputs. Multilayer perceptron neural network employing quick propagation training algorithm was used to detect internal carotid artery stenosis and occlusion. The network was trained, cross validated and tested with subject’s internal carotid arterial Doppler signals. The correct classification rate was 95.45% for healthy subjects, 92.00% for subjects having internal carotid artery stenosis and 95.65% for subjects having internal carotid artery occlusion. The classification results showed that multilayer perceptron neural network employing quick propagation training algorithm was effective to detect internal carotid artery stenosis and occlusion.