Male pelvic CT multi-organ segmentation using synthetic MRI-aided dual pyramid networks

被引:10
作者
Lei, Yang [1 ,2 ]
Wang, Tonghe [1 ,2 ]
Tian, Sibo [1 ,2 ]
Fu, Yabo [1 ,2 ]
Patel, Pretesh [1 ,2 ]
Jani, Ashesh B. [1 ,2 ]
Curran, Walter J. [1 ,2 ]
Liu, Tian [1 ,2 ]
Yang, Xiaofeng [1 ,2 ]
机构
[1] Emory Univ, Dept Radiat Oncol, Atlanta, GA 30322 USA
[2] Emory Univ, Winship Canc Inst, Atlanta, GA 30322 USA
基金
美国国家卫生研究院;
关键词
deep learning; CT; segmentation; prostate; VOLUME DELINEATION; PROSTATE; RADIOTHERAPY; VARIABILITY; NORMALITY; IMAGES;
D O I
10.1088/1361-6560/abf2f9
中图分类号
R318 [生物医学工程];
学科分类号
0831 ;
摘要
The delineation of the prostate and organs-at-risk (OARs) is fundamental to prostate radiation treatment planning, but is currently labor-intensive and observer-dependent. We aimed to develop an automated computed tomography (CT)-based multi-organ (bladder, prostate, rectum, left and right femoral heads (RFHs)) segmentation method for prostate radiation therapy treatment planning. The proposed method uses synthetic MRIs (sMRIs) to offer superior soft-tissue information for male pelvic CT images. Cycle-consistent adversarial networks (CycleGAN) were used to generate CT-based sMRIs. Dual pyramid networks (DPNs) extracted features from both CTs and sMRIs. A deep attention strategy was integrated into the DPNs to select the most relevant features from both CTs and sMRIs to identify organ boundaries. The CT-based sMRI generated from our previously trained CycleGAN and its corresponding CT images were inputted to the proposed DPNs to provide complementary information for pelvic multi-organ segmentation. The proposed method was trained and evaluated using datasets from 140 patients with prostate cancer, and were then compared against state-of-art methods. The Dice similarity coefficients and mean surface distances between our results and ground truth were 0.95 +/- 0.05, 1.16 +/- 0.70 mm; 0.88 +/- 0.08, 1.64 +/- 1.26 mm; 0.90 +/- 0.04, 1.27 +/- 0.48 mm; 0.95 +/- 0.04, 1.08 +/- 1.29 mm; and 0.95 +/- 0.04, 1.11 +/- 1.49 mm for bladder, prostate, rectum, left and RFHs, respectively. Mean center of mass distances was within 3 mm for all organs. Our results performed significantly better than those of competing methods in most evaluation metrics. We demonstrated the feasibility of sMRI-aided DPNs for multi-organ segmentation on pelvic CT images, and its superiority over other networks. The proposed method could be used in routine prostate cancer radiotherapy treatment planning to rapidly segment the prostate and standard OARs.
引用
收藏
页数:22
相关论文
共 36 条
  • [1] Fully automated organ segmentation in male pelvic CT images
    Balagopal, Anjali
    Kazemifar, Samaneh
    Dan Nguyen
    Lin, Mu-Han
    Hannan, Raquibul
    Owrangi, Amir
    Jiang, Steve
    [J]. PHYSICS IN MEDICINE AND BIOLOGY, 2018, 63 (24)
  • [2] A System for Continual Quality Improvement of Normal Tissue Delineation for Radiation Therapy Treatment Planning
    Breunig, Jennifer
    Hernandez, Sophy
    Lin, Jeffrey
    Alsager, Stacy
    Dumstorf, Christine
    Price, Jennifer
    Steber, Jennifer
    Garza, Richard
    Nagda, Suneel
    Melian, Edward
    Emami, Bahman
    Roeske, John C.
    [J]. INTERNATIONAL JOURNAL OF RADIATION ONCOLOGY BIOLOGY PHYSICS, 2012, 83 (05): : E703 - E708
  • [3] Inter- and intra-observer variability in contouring of the prostate gland on planning computed tomography and cone beam computed tomography
    Choi, Hyuck Jae
    Kim, Young Seok
    Lee, Se Hyung
    Lee, Yu Sun
    Park, Geumju
    Jung, Jin Hong
    Cho, Byung Chul
    Park, Sung Ho
    Ahn, Hanjong
    Kim, Choung-Soo
    Yi, Seong Yoon
    Ahn, Seung Do
    [J]. ACTA ONCOLOGICA, 2011, 50 (04) : 539 - 546
  • [4] TESTS FOR DEPARTURE FROM NORMALITY - EMPIRICAL RESULTS FOR DISTRIBUTIONS OF B2 AND SQUARE ROOT B1
    DAGOSTIN.R
    PEARSON, ES
    [J]. BIOMETRIKA, 1973, 60 (03) : 613 - 622
  • [5] DAGOSTINO RB, 1971, BIOMETRIKA, V58, P341, DOI 10.1093/biomet/58.2.341
  • [6] Synthetic MRI-aided multi-organ segmentation on male pelvic CT using cycle consistent deep attention network
    Dong, Xue
    Lei, Yang
    Tian, Sibo
    Wang, Tonghe
    Patel, Pretesh
    Curran, Walter J.
    Jani, Ashesh B.
    Liu, Tian
    Yang, Xiaofeng
    [J]. RADIOTHERAPY AND ONCOLOGY, 2019, 141 : 192 - 199
  • [7] Automatic multiorgan segmentation in thorax CT images using U-net-GAN
    Dong, Xue
    Lei, Yang
    Wang, Tonghe
    Thomas, Matthew
    Tang, Leonardo
    Curran, Walter J.
    Liu, Tian
    Yang, Xiaofeng
    [J]. MEDICAL PHYSICS, 2019, 46 (05) : 2157 - 2168
  • [8] Automatic model-based segmentation of the heart in CT images
    Ecabert, Olivier
    Peters, Jochen
    Schramm, Hauke
    Lorenz, Cristian
    von Berg, Jens
    Walker, Matthew J.
    Vembar, Mani
    Olszewski, Mark E.
    Subramanyan, Krishna
    Lavi, Guy
    Weese, Juergen
    [J]. IEEE TRANSACTIONS ON MEDICAL IMAGING, 2008, 27 (09) : 1189 - 1201
  • [9] Multi-atlas segmentation of biomedical images: A survey
    Eugenio Iglesias, Juan
    Sabuncu, Mert R.
    [J]. MEDICAL IMAGE ANALYSIS, 2015, 24 (01) : 205 - 219
  • [10] Dosimetric evaluation of an atlas-based synthetic CT generation approach for MR-only radiotherapy of pelvis anatomy
    Farjam, Reza
    Tyag, Neelam
    Deasy, Joseph O.
    Hun, Margie A.
    [J]. JOURNAL OF APPLIED CLINICAL MEDICAL PHYSICS, 2019, 20 (01): : 101 - 109