Surface-rendering imaging of gastrointestinal lesions by three-dimensionalendoscopic ultrasonography

Citation
J. Yoshino et al., Surface-rendering imaging of gastrointestinal lesions by three-dimensionalendoscopic ultrasonography, ENDOSCOPY, 31(7), 1999, pp. 541-545
Citations number
17
Categorie Soggetti
Surgery,"Medical Research Diagnosis & Treatment
Journal title
ENDOSCOPY
ISSN journal
0013726X → ACNP
Volume
31
Issue
7
Year of publication
1999
Pages
541 - 545
Database
ISI
SICI code
0013-726X(199909)31:7<541:SIOGLB>2.0.ZU;2-H
Abstract
Background and Study Aims: In three-dimensional endoscopic ultrasonography (3D-EUS), a surface-rendering method can provide both a surface image and a cross-sectional ultrasonographic image. We evaluated the usefulness of thi s imaging method for digestive tract lesions. Patients and Methods: A total of 30 patients underwent 3D-EUS with surface- rendering using a 3D probe system which arranged individual radial scanning images into 128 points on a computer monitor to outline the surface of a l esion. A complete surface image of the lesion was displayed on the computer monitor using lines obtained from 40 radial scanning images. Results: Surface-rendering images of lesions were similar to endoscopic ima ges. The surface-rendering method permitted precise correlation of two-dime nsional images depicting a slice of a lesion with the corresponding surface of the lesion. Unlike conventional endoscopy, this approach permitted obse rvations of lesions at any desired angle. Complete images of lesions were a chieved in 14 patients and were half-completed in another five, but could n ot be obtained in 11 patients: in six the distance between the lesion and t he 3D probe was too short to avoid artifacts, while in five the lesion was larger than the longitudinal scanning length of 4 cm or greater than 90 deg rees in extent in radial scanning images. Artifacts caused by heartbeat led to irregular images in four patients, including three with esophageal canc er and one with gastric cancer. Conclusions: Despite some problems, surface-rendering imaging should prove useful for diagnosis, and the method will improve as software is perfected.