Improvement and evaluation of computerized method for detection of intracranial aneurysms for 3-D MR angiography

Hidetaka Arimura, Qiang Li, Yuknnori Korogi, Toshinori Hirai, Shigehiko Katsuragawa, Yasuyuki Yaniashita, Kazuhiro Tsuchiya, Kunio Doi

Research output: Contribution to journalArticle

Abstract

We have improved the computerized method for detection of intracranial aneurysms for three-dimensional (3-D) MRA, and evaluated the usefulness of the computer-aided diagnostic (CAD) system with three databases acquired from different institutions The improvement was achieved by a shape-based difference image (SBDI) technique, which was based on the shape-based difference between an original segmented vessel and a vessel with suppressed local change in thickness. We employed three databases obtained from three different institutions, i.e., Database A: 29 cases from a 1.0-T unit (Manufacturer S), Database B: 34 cases from a 1.5-T unit (T), and Database C: 53 aneurysm cases obtained from a 1.5-T unit (S). As a result of application to Database C, the number of false positives per patient decreased from 5.8 to 3.8, while a high sensitivity of 97% was maintained. The performances of our CAD scheme for Databases A and B indicated sensitivities of 97 and 94 % with 3.5 and 2.3 false positives per patient, respectively.

Original languageEnglish
Pages (from-to)385-386
Number of pages2
JournalInternational Journal of Computer Assisted Radiology and Surgery
Volume1
Issue numberSUPPL. 7
DOIs
Publication statusPublished - Jun 2006

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Surgery
  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging
  • Computer Vision and Pattern Recognition
  • Health Informatics
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design

Cite this