Premium
The use of atlas registration and graph cuts for prostate segmentation in magnetic resonance images
Author(s) -
Korsager Anne Sofie,
Fortunati Valerio,
Lijn Fedde,
Carl Jesper,
Niessen Wiro,
Østergaard Lasse Riis,
Walsum Theo
Publication year - 2015
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4914379
Subject(s) - segmentation , atlas (anatomy) , artificial intelligence , computer science , image registration , cut , magnetic resonance imaging , image segmentation , pattern recognition (psychology) , computer vision , ground truth , graph , nuclear medicine , medicine , radiology , image (mathematics) , theoretical computer science , anatomy
Purpose: An automatic method for 3D prostate segmentation in magnetic resonance (MR) images is presented for planning image‐guided radiotherapy treatment of prostate cancer. Methods: A spatial prior based on intersubject atlas registration is combined with organ‐specific intensity information in a graph cut segmentation framework. The segmentation is tested on 67 axial T 2 ‐weighted MR images in a leave‐one‐out cross validation experiment and compared with both manual reference segmentations and with multiatlas‐based segmentations using majority voting atlas fusion. The impact of atlas selection is investigated in both the traditional atlas‐based segmentation and the new graph cut method that combines atlas and intensity information in order to improve the segmentation accuracy. Best results were achieved using the method that combines intensity information, shape information, and atlas selection in the graph cut framework. Results: A mean Dice similarity coefficient (DSC) of 0.88 and a mean surface distance (MSD) of 1.45 mm with respect to the manual delineation were achieved. Conclusions: This approaches the interobserver DSC of 0.90 and interobserver MSD 0f 1.15 mm and is comparable to other studies performing prostate segmentation in MR.