31 October 1997, Volume 26, Issue 6, Pages 332 – 336

Paper

Ultrasonographic texture characterization of salivary and neck masses using two-dimensional gray-scale clustering K Yoshiura1,4, K Miwa1, K Yuasa1, K Tokumori1, S Kanda1, Y Higuchi2 & M Shinohara3 1Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Kyushu University, Maidashi 3-1-1, Higashiku, Fukuoka 812, Japan     2Department of Dentistry and Oral Surgery, National Kyushu Medical Center Hospital, Japan     3Second Department of Oral and Maxillofacial Surgery, Faculty of Dentistry, Kyushu University, Japan     4Author for correspondence

Keywords

ultrasonic diagnosis;   image processing;   computer-assisted;   image interpretation, computer-assisted;   salivary gland diseases

Abstract

Objectives: To evaluate the clinical usefulness of texture analysis in ultrasonography (US) of head and neck masses. Methods: Sixty-one cases with salivary or neck masses were examined by US with a linear 7.5MHz array. Images were digitized and quantitatively evaluated by two-dimensional gray-scale clustering using two texture indices, M and SD, as measures of coarseness and homogeneity respectively. Results: Coarseness of Warthin’s tumors was significantly higher than that of the other salivary tumors. Malignant lymphomas had finer texture values than other lymph node lesions. Conclusions: Quantitative texture analysis appears to be a useful complement in diagnostic US.

Received 15 April, 97; Accepted 30 June, 97

^© Macmillan Publishers Ltd 1997