Real-time detection and clinical categorisation of ultrasound high intensity transient signal
Barcaro U., Di Bona S., Fontanelli R., La Manna S., Orlandi G., Salvetti O., Sartucci F.
Hierarchical Neural Networks Image Classification Eco-Doppler Imaging Emboli Classification
In this paper we address the problem of interpreting ultrasound images obtained from doppler devices in order to classify, and hence to distinguish, the dangerous cerebral microemboli from the innocuous ones. In order to obtain an automatic categorisation of the cerebral high intensity transient signal, a multilevel neural network based on a hierarchical architecture has been implemented for image processing and classification. The images, obtained by measuring the blood flow velocities in brain arteries and veins,have been acquired using the 'Multi Dop X4' ultrasound device by DWL. The approach proposed, applied to real clinical cases selected by expert neurologists for their peculiar characteristics, has shown to be a valid real-time support for the diagnosis of cerebral vascular diseases
Source: WSEAS transactions on systems 2 (2003): 921–926.
Publisher: WSEAS Press, Athens
Back to previous page
@article{oai:it.cnr:prodotti:68274,
title = {Real-time detection and clinical categorisation of ultrasound high intensity transient signal},
author = {Barcaro U. and Di Bona S. and Fontanelli R. and La Manna S. and Orlandi G. and Salvetti O. and Sartucci F.},
publisher = {WSEAS Press, Athens },
journal = {WSEAS transactions on systems },
volume = {2},
pages = {921–926},
year = {2003}
}
0000-0002-1300-3900