Progress in two-dimensional and three-dimensional ultrasonic tissue-type imaging of the prostate based on spectrum analysis and nonlinear classifiers

Citation
Ej. Feleppa et al., Progress in two-dimensional and three-dimensional ultrasonic tissue-type imaging of the prostate based on spectrum analysis and nonlinear classifiers, MOL UROL, 3(3), 1999, pp. 303-310
Citations number
23
Language
INGLESE
art.tipo
Article
Categorie Soggetti
Urology & Nephrology
Journal title
MOLECULAR UROLOGY
ISSN journal
1091-5362 → ACNP
Volume
3
Issue
3
Year of publication
1999
Pages
303 - 310
Database
ISI
SICI code
1091-5362(199923)3:3<303:PITATU>2.0.ZU;2-H
Abstract
Spectrum analysis of radiofrequency (RF) ultrasonic echo signals often can sense tissue differences that are not visible on conventional ultrasonic im ages. Spectrum-analysis parameter values combined with other variables, suc h as serum prostate specific antigen (PSA) concentration, can be classified by neural networks to distinguish effectively between cancerous and noncan cerous prostate tissues. Images based on neural network classification of s pectral parameters and clinical variables can be advantageous for biopsy gu idance, staging, and treatment planning and monitoring. A study based on 64 4 biopsies from 137 patients showed that these methods are significantly su perior to B-mode image interpretation for differentiating cancerous from no ncancerous prostate tissues. Using the histologic determination of tissue t ypes as the gold standard, the area under the receiver-operator characteris tic (ROC) curve for neural network classification based on spectrum analysi s and PSA value for the 644 biopsies was 0.87 +/- 0.04, and the ROC curve a re for a level-of-suspicion (LOS) assignment based on B-mode imaging was 0. 64 +/- 0.04, Color-encoded and gray-scale images derived from neural networ k assignment of suspicion for cancer at each pixel location showed remarkab le detail and suggested potential clinical value for biopsy guidance using real-time two-dimensional (2D) images and staging, treatment planning, and monitoring using three-dimensional (3D) images.