Emboli classification using RBF neural networks

Teixeira, C. A.; Ruano, M. Graça; Ruano, Antonio

http://hdl.handle.net/10400.1/2264

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
chair_get_file_doc_id=54&ession_id=13.pdf		254.94 KB	Adobe PDF	Download

Send Feedback

Authors

Teixeira, C. A.

Ruano, M. Graça

Ruano, Antonio

Abstract(s)

The presence of emboli in the medium cerebral artery may cause severe brain damage or, in the worst case, death. Emboli classification is not a deterministic task for the lack of universally accepted classification rules. In this work the use of Radial Basis Functions Neural Networks (RBFNN) is proposed to distinguish among solid emboli, liquid emboli, or artifacts, using as indicators a ratio between signal powers and emboli duration. The result of this work demonstrates that, with RBFNN, it is possible to classify emboli with approximately 88% success.

Keywords

Classification Neural Networks Radial Basis Functions

URI

http://hdl.handle.net/10400.1/2264

Citation

Teixeira, C. A.; Ruano, M. G.; Ruano, A. E. Emboli Classification using RBF Neural Networks. Trabalho apresentado em 6th Portuguese Conference on Automatic Control (Controlo 2004), In 6th Portuguese Conference on Automatic Control (Controlo 2004), Faro, 2004.