Share Email Print
cover

Proceedings Paper • new

Comparison of different deep learning approaches for parotid gland segmentation from CT images
Author(s): Annika Hänsch; Michael Schwier; Tobias Gass; Tomasz Morgas; Benjamin Haas; Jan Klein; Horst K. Hahn
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The segmentation of target structures and organs at risk is a crucial and very time-consuming step in radiotherapy planning. Good automatic methods can significantly reduce the time clinicians have to spend on this task. Due to its variability in shape and often low contrast to surrounding structures, segmentation of the parotid gland is especially challenging. Motivated by the recent success of deep learning, we study different deep learning approaches for parotid gland segmentation. Particularly, we compare 2D, 2D ensemble and 3D U-Net approaches and find that the 2D U-Net ensemble yields the best results with a mean Dice score of 0.817 on our test data. The ensemble approach reduces false positives without the need for an automatic region of interest detection. We also apply our trained 2D U-Net ensemble to segment the test data of the 2015 MICCAI head and neck auto-segmentation challenge. With a mean Dice score of 0.861, our classifier exceeds the highest mean score in the challenge. This shows that the method generalizes well onto data from independent sites. Since appropriate reference annotations are essential for training but often difficult and expensive to obtain, it is important to know how many samples are needed to properly train a neural network. We evaluate the classifier performance after training with differently sized training sets (50–450) and find that 250 cases (without using extensive data augmentation) are sufficient to obtain good results with the 2D ensemble. Adding more samples does not significantly improve the Dice score of the segmentations.

Paper Details

Date Published: 27 February 2018
PDF: 6 pages
Proc. SPIE 10575, Medical Imaging 2018: Computer-Aided Diagnosis, 1057519 (27 February 2018); doi: 10.1117/12.2292962
Show Author Affiliations
Annika Hänsch, Fraunhofer MEVIS (Germany)
Michael Schwier, Fraunhofer MEVIS (Germany)
Tobias Gass, Varian Medical Systems Imaging Lab. GmbH (Switzerland)
Tomasz Morgas, Varian Medical Systems, Inc. (United States)
Benjamin Haas, Varian Medical Systems Imaging Lab. GmbH (Switzerland)
Jan Klein, Fraunhofer MEVIS (Germany)
Horst K. Hahn, Fraunhofer MEVIS (Germany)


Published in SPIE Proceedings Vol. 10575:
Medical Imaging 2018: Computer-Aided Diagnosis
Nicholas Petrick; Kensaku Mori, Editor(s)

© SPIE. Terms of Use
Back to Top