Premium
A novel MRI segmentation method using CNN ‐based correction network for MRI ‐guided adaptive radiotherapy
Author(s) -
Fu Yabo,
Mazur Thomas R.,
Wu Xue,
Liu Shi,
Chang Xiao,
Lu Yonggang,
Li H. Harold,
Kim Hyun,
Roach Michael C.,
Henke Lauren,
Yang Deshan
Publication year - 2018
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 - 10.1002/mp.13221
Subject(s) - softmax function , conditional random field , artificial intelligence , computer science , convolutional neural network , segmentation , sørensen–dice coefficient , contouring , pattern recognition (psychology) , hausdorff distance , voxel , deep learning , image segmentation , computer graphics (images)
Purpose The purpose of this study was to expedite the contouring process for MRI ‐guided adaptive radiotherapy ( MR ‐ IGART ), a convolutional neural network ( CNN ) deep‐learning ( DL ) model is proposed to accurately segment the liver, kidneys, stomach, bowel and duodenum in 3D MR images. Methods Images and structure contours for 120 patients were collected retrospectively. Treatment sites included pancreas, liver, stomach, adrenal gland, and prostate. The proposed DL model contains a voxel‐wise label prediction CNN and a correction network which consists of two sub‐networks. The prediction CNN and sub‐networks in the correction network each includes a dense block which consists of twelve densely connected convolutional layers. The correction network was designed to improve the voxel‐wise labeling accuracy of a CNN by learning and enforcing implicit anatomical constraints in the segmentation process. Its sub‐networks learn to fix the erroneous classification of its previous network by taking as input both the original images and the softmax probability maps generated from its previous sub‐network. The parameters of each sub‐network were trained independently using piecewise training. The model was trained on 100 datasets, validated on 10 datasets and tested on the remaining 10 datasets. Dice coefficient, Hausdorff distance ( HD ) were calculated to evaluate the segmentation accuracy. Results The proposed DL model was able to segment the organs with good accuracy. The correction network outperformed the conditional random field ( CRF ), a most comparable method that is usually applied as a post‐processing step. For the 10 testing patients, the average Dice coefficients were 95.3 ± 0.73, 93.1 ± 2.22, 85.0 ± 3.75, 86.6 ± 2.69, and 65.5 ± 8.90 for liver, kidneys, stomach, bowel, and duodenum, respectively. The mean Hausdorff Distance ( HD ) were 5.41 ± 2.34, 6.23 ± 4.59, 6.88 ± 4.89, 5.90 ± 4.05, and 7.99 ± 6.84 mm, respectively. Manual contouring, as to correct the automatic segmentation results, was four times as fast as manual contouring from scratch. Conclusion The proposed method can automatically segment the liver, kidneys, stomach, bowel, and duodenum in 3D MR images with good accuracy. It is useful to expedite the manual contouring for MR ‐ IGART .