Spectral processing procedure on RadioFrequency (RF) echographic signals is proposed for detecting and characterizing mammary pathologies in order to improve echographic diagnosis on breast cancer that is the second leading cause of cancer death among women. The spectral content of each RF track of a frame is decomposed in N-subband obtained by a bank of filters derived from Morlet Wavelet. The proposed processing procedure works in a N-dimensional spectral hyperspace. Different biological structure can be differentiated by their position in the hyperspace. A Clustering technique is employed to detect the typical spatial distributions. The algorithm is developed in two phases: Training step and Classification step. In the first one, a set of patients are selected and only Regions Of Interest (ROI) are processed to define the suitable Clusters. The Classifications phase, which operates on entire frame, is applied over all patients. The method is amplitude independent and moreover it is capable to compensate for different frequency responses of ultrasonic transducers

Tissue Characterization in Echographic Spectral Hyperspace: Breast Pathologies Differentiation / E. Biagi; S.Granchi; E.Vannacci; L.Lucarini; L.Masotti. - ELETTRONICO. - (2010), pp. 1388-1391. (Intervento presentato al convegno IEEE Ultrasonic Symposium tenutosi a San Diego, California nel 11-14 Ottobre 2010).

Tissue Characterization in Echographic Spectral Hyperspace: Breast Pathologies Differentiation.

BIAGI, ELENA;GRANCHI, SIMONA;VANNACCI, ENRICO;LUCARINI, LEONARDO;MASOTTI, LEONARDO
2010

Abstract

Spectral processing procedure on RadioFrequency (RF) echographic signals is proposed for detecting and characterizing mammary pathologies in order to improve echographic diagnosis on breast cancer that is the second leading cause of cancer death among women. The spectral content of each RF track of a frame is decomposed in N-subband obtained by a bank of filters derived from Morlet Wavelet. The proposed processing procedure works in a N-dimensional spectral hyperspace. Different biological structure can be differentiated by their position in the hyperspace. A Clustering technique is employed to detect the typical spatial distributions. The algorithm is developed in two phases: Training step and Classification step. In the first one, a set of patients are selected and only Regions Of Interest (ROI) are processed to define the suitable Clusters. The Classifications phase, which operates on entire frame, is applied over all patients. The method is amplitude independent and moreover it is capable to compensate for different frequency responses of ultrasonic transducers
2010
Proceedings IEEE Ultrasonic Symposium
IEEE Ultrasonic Symposium
San Diego, California
11-14 Ottobre 2010
E. Biagi; S.Granchi; E.Vannacci; L.Lucarini; L.Masotti
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/406614
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact