Texture analysis of digitized prostate pathologic cross section

David E. Pitts, Saganti B. Premkumar, A. G. Houston, Richard J. Babaian, Patricia Troncoso

Research output: Chapter in Book/Report/Conference proceedingConference contribution

11 Scopus citations

Abstract

Understanding the texture attributes of prostate cancer lesions and identifying corresponding features on the ultrasound images (transrectal ultrasound images of the prostate) has been the aim of present investigation. Prostate glands, surgically removed through prostatectomy, are provided as serial whole mount cross-sections at 4 mm intervals from base to apex of the gland. Digitized images of these whole mount cross-sections are obtained using a photographic capture procedure for the present study. Texture analyses of these digitized image planes have been conducted for classification of each cross-section into benign and cancer regions through a supervised classification scheme. Preliminary results of supervised maximum likelihood classification of these 512 × 512 images in a 9 × 9 pixel window size are presented.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
EditorsMurray H. Loew
PublisherPubl by Society of Photo-Optical Instrumentation Engineers
Pages465-470
Number of pages6
ISBN (Print)0819411310
StatePublished - 1993
Externally publishedYes
EventMedical Imaging 1993: Image Processing - Newport Beach, CA, USA
Duration: Feb 14 1992Feb 19 1992

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume1898
ISSN (Print)0277-786X

Other

OtherMedical Imaging 1993: Image Processing
CityNewport Beach, CA, USA
Period2/14/922/19/92

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Texture analysis of digitized prostate pathologic cross section'. Together they form a unique fingerprint.

Cite this