Title :
Dependence of tissue characterization features on region of interest (ROI) size: studies on phantoms and simulations
Author :
He, Z. ; Huston, D.R. ; Grimm, S. ; Jannicky, E. ; Garra, B.S. ; Wagner, R.F. ; Wear, K.A.
Author_Institution :
Dept. of Mech. Eng., Vermont Univ., Burlington, VT, USA
Abstract :
In our previous work on prostate tissue, we noted an apparent dependence of the computed tissue characterization features and the size of the ROI chosen. We are studying this effect in a systematic fashion by computing feature values for ROI of different size in both phantoms and simulations. As expected, most features exhibited larger variance when small ROI were used. Most features exhibited a dependence on ROI size although some features were strongly dependent on ROI size only for small ROI sizes. The nature of the ROI size dependence varied widely for different features.
Keywords :
biological tissues; biomedical ultrasonics; cancer; phantoms; ultrasonic imaging; ROI size; phantoms; prostate tissue; region of interest; simulations; tissue characterization features; Attenuation; Backscatter; Cepstrum; Computational modeling; Data acquisition; Frequency; Imaging phantoms; Rayleigh scattering; Ultrasonic imaging; Ultrasonic transducers;
Conference_Titel :
Ultrasonics Symposium, 2004 IEEE
Print_ISBN :
0-7803-8412-1
DOI :
10.1109/ULTSYM.2004.1418246