Knowledge-based topographic feature extraction in medical images

JianZhong Qian; Mohammad M. Khair

doi:10.1117/12.216785

18 August 1995 Knowledge-based topographic feature extraction in medical images

JianZhong Qian, Mohammad M. Khair

Proceedings Volume 2622, Optical Engineering Midwest '95; (1995) https://doi.org/10.1117/12.216785
Event: Optical Engineering Midwest '95, 1995, Chicago, IL, United States

Abstract

Diagnostic medical imaging often contains variations of patient anatomies, camera mispositioning, or other imperfect imaging condiitons. These variations contribute to uncertainty about shapes and boundaries of objects in images. As the results sometimes image features, such as traditional edges, may not be identified reliably and completely. We describe a knowledge based system that is able to reason about such uncertainties and use partial and locally ambiguous information to infer about shapes and lcoation of objects in an image. The system uses directional topographic features (DTFS), such as ridges and valleys, labeled from the underlying intensity surface to correlate to the intrinsic anatomical information. By using domain specific knowledge, the reasoning system can deduce significant anatomical landmarks based upon these DTFS, and can cope with uncertainties and fill in missing information. A succession of levels of representation for visual information and an active process of uncertain reasoning about this visual information are employed to realiably achieve the goal of image analysis. These landmarks can then be used in localization of anatomy of interest, image registration, or other clinical processing. The successful application of this system to a large set of planar cardiac images of nuclear medicine studies has demonstrated its efficiency and accuracy.

Citation Download Citation

JianZhong Qian and Mohammad M. Khair "Knowledge-based topographic feature extraction in medical images", Proc. SPIE 2622, Optical Engineering Midwest '95, (18 August 1995); https://doi.org/10.1117/12.216785

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available