Share Email Print
cover

Proceedings Paper

Evaluating segmentation algorithms for diffusion-weighted MR images: a task-based approach
Author(s): Abhinav K. Jha; Matthew A. Kupinski; Jeffrey J. Rodríguez; Renu M. Stephen; Alison T. Stopeck
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Apparent Diffusion Coefficient (ADC) of lesions obtained from Diffusion Weighted Magnetic Resonance Imaging is an emerging biomarker for evaluating anti-cancer therapy response. To compute the lesion's ADC, accurate lesion segmentation must be performed. To quantitatively compare these lesion segmentation algorithms, standard methods are used currently. However, the end task from these images is accurate ADC estimation, and these standard methods don't evaluate the segmentation algorithms on this task-based measure. Moreover, standard methods rely on the highly unlikely scenario of there being perfectly manually segmented lesions. In this paper, we present two methods for quantitatively comparing segmentation algorithms on the above task-based measure; the first method compares them given good manual segmentations from a radiologist, the second compares them even in absence of good manual segmentations.

Paper Details

Date Published: 27 February 2010
PDF: 8 pages
Proc. SPIE 7627, Medical Imaging 2010: Image Perception, Observer Performance, and Technology Assessment, 76270L (27 February 2010); doi: 10.1117/12.845515
Show Author Affiliations
Abhinav K. Jha, College of Optical Sciences, The Univ. of Arizona (United States)
Matthew A. Kupinski, College of Optical Sciences, The Univ. of Arizona (United States)
Jeffrey J. Rodríguez, The Univ. of Arizona (United States)
Renu M. Stephen, The Univ. of Arizona (United States)
Alison T. Stopeck, The Univ. of Arizona (United States)


Published in SPIE Proceedings Vol. 7627:
Medical Imaging 2010: Image Perception, Observer Performance, and Technology Assessment
David J. Manning; Craig K. Abbey, Editor(s)

© SPIE. Terms of Use
Back to Top