Magnetic Resonance in Medicine
SUSAN: Segment Unannotated image Structure using Adversarial Network Journal: Magnetic Resonance in Medicine Manuscript ID MRM-18-18874.R3 Wiley - Manuscript type: Full Paper Date Submitted by the 13-Nov-2018 Author: Complete List of Authors: Liu, Fang; University of Wisconsin - Madison, Radiology Image processing/Image analysis < Theoretical < Technical Research, Research Type: segmentation < Post-Processing < Technique Development < Technical Research Research Focus: Bone < Musculoskeletal, Cartilage < Musculoskeletal
Magnetic Resonance in Medicine
Page 1 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Magnetic Resonance in Medicine
SUSAN: Segment Unannotated image Structure using Adversarial Network by Fang Liu, Ph.D.1,2
Department of Radiology1 University of Wisconsin School of Medicine and Public Health 600 Highland Avenue Madison, Wisconsin 53705-2275
2Corresponding
Author:
Fang Liu, Ph.D. Department of Radiology Wisconsin Institutes for Medical Research 1111 Highland Avenue Madison, Wisconsin 53705-2275 Phone: 612-222-9728, FAX: 608-263-5112 E-Mail:
[email protected]
Full Paper Word Count: 5855
Magnetic Resonance in Medicine
Magnetic Resonance in Medicine 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 40
ABSTRACT Purpose: To describe and evaluate a segmentation method using joint adversarial and segmentation convolutional neural network (CNN) to achieve accurate segmentation using unannotated magnetic resonance (MR) image datasets. Methods: A segmentation pipeline was built using joint adversarial and segmentation network. A CNN technique called cycle-consistent generative adversarial network (CycleGAN) was applied as the core of the method to perform unpaired image-to-image translation between different MR image datasets. A joint segmentation network was incorporated into the adversarial network to obtain additional functionality for semantic segmentation. The fully-automated segmentation method termed as SUSAN was tested for segmenting bone and cartilage on two clinical knee MR image datasets using images and annotated segmentation masks from an online publicly available knee MR image dataset. The segmentation results were compared using quantitative segmentation metrics with the results from a supervised U-Net segmentation method and two registration methods. The Wilcoxon signed-rank test was used to evaluate the value difference of quantitative metrics between different methods. Results: The proposed method SUSAN provided high segmentation accuracy with results comparable to the supervised U-Net segmentation method (most quantitative metrics having p>0.05) and significantly better than a multi-atlas registration method (all quantitative metrics having p
