AAPM Banner
  • Volume/Page
  • Keyword
  • DOI
  • Citation
  • Advanced
   
 
 
 

You are not logged in Access to this article requires a subscription. Log In

Medical Physics / Volume 39 / Issue 1 / RADIATION IMAGING PHYSICS

Med. Phys. 39, 514 (2012); http://dx.doi.org/10.1118/1.3668058 (19 pages)

A new multi-object image thresholding method based on correlation between object class uncertainty and intensity gradient

Yinxiao Liu, Guoyuan Liang, and Punam K. Saha

Departments of Electrical and Computer Engineering and Radiology, University of Iowa, Iowa City, Iowa 52242

View MapView Map

(Received 24 May 2011; accepted 19 November 2011; revised 10 November 2011; published online 3 January 2012)

Full Text: Read Online (HTML) | Download PDF | Buy PDF (US$30) | View Cart
Purpose: Image thresholding and gradient analysis have remained popular image preprocessing tools for several decades due to the simplicity and straight-forwardness of their definitions. Also, optimum selection of threshold and gradient strength values are hidden steps in many advanced medical imaging algorithms. A reliable method for threshold optimization may be a crucial step toward automation of several medical image based applications. Most automatic thresholding and gradient selection methods reported in literature primarily focus on image histograms ignoring a significant amount of information embedded in the spatial distribution of intensity values forming visible features in an image. Here, we present a new method that simultaneously optimizes both threshold and gradient values for different object interfaces in an image that is based on unification of information from both the histogram and spatial image features; also, the method works for unknown number of object regions.
Methods: A new energy function is formulated by combining the object class uncertainty measure, a histogram-based feature, of each pixel with its image gradient measure, a spatial contextual feature in an image. The energy function is designed to measure the overall compliance of the theoretical premise that, in a probabilistic sense, image intensities with high class uncertainty are associated with high image gradients. Finally, it is expressed as a function of threshold and gradient parameters and optimum combinations of these parameters are sought by locating pits and valleys on the energy surface. A major strength of the algorithm lies in the fact that it does not require the number of object regions in an image to be predefined.
Results: The method has been applied on several medical image datasets and it has successfully determined both threshold and gradient parameters for different object interfaces even when some of the thresholds are almost impossible to locate in the histogram. Both accuracy and reproducibility of the method have been examined on several medical image datasets including repeat scan 3D multidetector computed tomography (CT) images of cadaveric ankles specimens. Also, the new method has been qualitatively and quantitatively compared with Otsu’s method along with three other algorithms based on minimum error thresholding, maximum segmented image information and minimization of homogeneity- and uncertainty-based energy and the results have demonstrated superiority of the new method.
Conclusions: We have developed a new automatic threshold and gradient strength selection algorithm by combining class uncertainty and spatial image gradient features. The performance of the method has been examined in terms of accuracy and reproducibility and the results found are better as compared to several popular automatic threshold selection methods.

© 2012 American Association of Physicists in Medicine

ACKNOWLEDGEMENT

This work was partially supported by the NIH Grant No. R01 AR054439.

Article Outline

  1. INTRODUCTION
  2. THEORY
    1. Principle of class uncertainty
    2. Energy surface and threshold/gradient optimization
  3. METHODS AND EXPERIMENTAL PLANS
    1. Optimization of threshold and gradient parameters
    2. Experimental plans
  4. RESULTS AND DISCUSSION
    1. Qualitative results
    2. Accuracy analysis
    3. Reproducibility analysis
    4. Conclusion and discussion
Digital Object Identifier
http://dx.doi.org/10.1118/1.3668058

KEYWORDS and PACS

Keywords

computerised tomography, image resolution, image segmentation, medical image processing, thresholding, gradient, probability, entropy, class uncertainty, energy surface, optimization

PACS

PUBLICATION DATA

ISSN

0094-2405 (print)  

Publisher

$abstract.society.value is a Member of CrossRef American Association of Physicists in Medicine

For access to fully linked references, you need to log in.

Figures (14)

Access to article objects (figures, tables, multimedia) requires a subscription; log in to view available files.
(Access to supplementary files, where available, is free for this journal.)



Close

Medical Physics is an official science journal of the AAPM and of the COMP/CCPM/IOMP
William R. Hendee, Editor
Departments of Radiology, Radiation Oncology, Biophysics, Community and Public Health, Medical College of Wisconsin
One Physics Ellipse
College Park, Maryland 20740
301-209-3352

© 2012, American Association of Physicists in Medicine. Individual readers of this journal, and nonprofit libraries acting for them, are freely permited to make fair use of the material in it, such as to copy an article for use in teaching or research. (For other kinds of copying see "Copying Fees.") Permission is granted to quote from this journal in scientific works with the customary acknowledgment of the source. To reprint a figure, table, or other excerpt requires, in addition, AAPM may require that permission also be obtained from one of the authors. Address inquiries and notices to Penny Slattery, Journal Manager, Medical Physics Journal, AAPM, One Physics Ellipse, College Park, MD 20740-3846; email: journal@aapm.org.

close