ARPM‐net: A novel CNN‐based adversarial method with Markov random field enhancement for prostate and organs at risk segmentation in pelvic CT images

Autor: Zhuangzhuang Zhang, Tianyu Zhao, Hiram Gay, Weixiong Zhang, Baozhou Sun
Rok vydání: 2020
Předmět:
Male
Organs at Risk
FOS: Computer and information sciences
Computer science
Computer Vision and Pattern Recognition (cs.CV)
Computer Science - Computer Vision and Pattern Recognition
ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION
030218 nuclear medicine & medical imaging
03 medical and health sciences
0302 clinical medicine
Image Processing
Computer-Assisted
FOS: Electrical engineering
electronic engineering
information engineering
Humans
Segmentation
Image resolution
Retrospective Studies
Contouring
Markov random field
68T07(Primary)
68T45(Secondary)
business.industry
Deep learning
Image and Video Processing (eess.IV)
Prostate
Prostatic Neoplasms
Pattern recognition
General Medicine
Electrical Engineering and Systems Science - Image and Video Processing
Hausdorff distance
030220 oncology & carcinogenesis
Test set
Artificial intelligence
Tomography
X-Ray Computed
business
Zdroj: Medical Physics. 48:227-237
ISSN: 2473-4209
0094-2405
Popis: Purpose: The research is to develop a novel CNN-based adversarial deep learning method to improve and expedite the multi-organ semantic segmentation of CT images, and to generate accurate contours on pelvic CT images. Methods: Planning CT and structure datasets for 120 patients with intact prostate cancer were retrospectively selected and divided for 10-fold cross-validation. The proposed adversarial multi-residual multi-scale pooling Markov Random Field (MRF) enhanced network (ARPM-net) implements an adversarial training scheme. A segmentation network and a discriminator network were trained jointly, and only the segmentation network was used for prediction. The segmentation network integrates a newly designed MRF block into a variation of multi-residual U-net. The discriminator takes the product of the original CT and the prediction/ground-truth as input and classifies the input into fake/real. The segmentation network and discriminator network can be trained jointly as a whole, or the discriminator can be used for fine-tuning after the segmentation network is coarsely trained. Multi-scale pooling layers were introduced to preserve spatial resolution during pooling using less memory compared to atrous convolution layers. An adaptive loss function was proposed to enhance the training on small or low contrast organs. The accuracy of modeled contours was measured with the Dice similarity coefficient (DSC), Average Hausdorff Distance (AHD), Average Surface Hausdorff Distance (ASHD), and relative Volume Difference (VD) using clinical contours as references to the ground-truth. The proposed ARPM-net method was compared to several stateof-the-art deep learning methods.
Comment: 21 pages, 8 figures; accepted as a journal article at Medical Physics; abstract presented at AAPM 2020
Databáze: OpenAIRE
Externí odkaz: