Fully automated organ segmentation in male pelvic CT images

Accurate segmentation of prostate and surrounding organs at risk is important for prostate cancer radiotherapy treatment planning. We present a fully automated workflow for male pelvic CT image segmentation using deep learning. The architecture consists of a 2D organ volume localization network followed by a 3D segmentation network for volumetric segmentation of prostate, bladder, rectum, and femoral heads. We used a multi-channel 2D U-Net followed by a 3D U-Net with encoding arm modified with aggregated residual networks, known as ResNeXt. The models were trained and tested on a pelvic CT image dataset comprising 136 patients. Test results show that 3D U-Net based segmentation achieves mean (±SD) Dice coefficient values of 90 (±2.0)%, 96 (±3.0)%, 95 (±1.3)%, 95 (±1.5)%, and 84 (±3.7)% for prostate, left femoral head, right femoral head, bladder, and rectum, respectively, using the proposed fully automated segmentation method.

[1]  Zhiqiang Tian,et al.  PSNet: prostate segmentation on MRI based on a convolutional neural network , 2018, Journal of medical imaging.

[2]  Yaozong Gao,et al.  Accurate Segmentation of CT Male Pelvic Organs via Regression-Based Deformable Models and Multi-Task Random Forests , 2016, IEEE Transactions on Medical Imaging.

[3]  Rongrong Guo,et al.  Automatic segmentation of the prostate on CT images using deep learning and multi-atlas fusion , 2017, Medical Imaging.

[4]  Tian Liu,et al.  A new CT prostate segmentation for CT-based HDR brachytherapy , 2014, Medical Imaging.

[5]  Dinggang Shen,et al.  Segmenting CT prostate images using population and patient-specific statistics for radiotherapy , 2010, 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[6]  Pascal Haigron,et al.  Evaluation of multi-atlas-based segmentation of CT scans in prostate cancer radiotherapy , 2011, 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[7]  Yaozong Gao,et al.  CT Prostate Deformable Segmentation by Boundary Regression , 2014, MCV.

[8]  Hao Chen,et al.  Volumetric ConvNets with Mixed Residual Connections for Automated Prostate Segmentation from 3D MR Images , 2017, AAAI.

[9]  M. Herk.  Errors and margins in radiotherapy. , 2004 .

[10]  Wei Li,et al.  Learning Image Context for Segmentation of Prostate in CT-Guided Radiotherapy , 2011, MICCAI.

[11]  C. Fiorino,et al.  Intra- and inter-observer variability in contouring prostate and seminal vesicles: implications for conformal treatment planning. , 1998, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[12]  Nathan Lay,et al.  Rapid Multi-organ Segmentation Using Context Integration and Discriminative Models , 2013, IPMI.

[13]  Yinghuan Shi,et al.  Prostate Segmentation in CT Images via Spatial-Constrained Transductive Lasso , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[14]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[15]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[16]  K. Kihara,et al.  Computer‐aided diagnosis of prostate cancer on magnetic resonance imaging using a convolutional neural network algorithm , 2018, BJU international.

[17]  Marcel van Herk,et al.  Target definition in prostate, head, and neck. , 2005, Seminars in radiation oncology.

[18]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[19]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[20]  Steve B. Jiang,et al.  Segmentation of the prostate and organs at risk in male pelvic CT images using deep learning , 2018, Biomedical Physics & Engineering Express.

[21]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[23]  Thomas Brox,et al.  3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation , 2016, MICCAI.

[24]  Thomas Krauss,et al.  Conformal radiotherapy planning of cervix carcinoma: differences in the delineation of the clinical target volume. A comparison between gynaecologic and radiation oncologists. , 2003, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[25]  Fernando Arámbula Cosío,et al.  Use of simplex search in active shape models for improved boundary segmentation , 2010, Pattern Recognit. Lett..

[26]  Edward L. Chaney,et al.  Histogram Statistics of Local Model-Relative Image Regions , 2005, DSSCV.

[27]  Bo Du,et al.  Deeply-supervised CNN for prostate segmentation , 2017, 2017 International Joint Conference on Neural Networks (IJCNN).

[28]  Ali Kamen,et al.  Constrained Surface Evolutions for Prostate and Bladder Segmentation in CT Images , 2005, CVBIA.

[29]  Seyed-Ahmad Ahmadi,et al.  V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation , 2016, 2016 Fourth International Conference on 3D Vision (3DV).

[30]  Pascal Haigron,et al.  Multi-Atlas-Based Segmentation of Pelvic Structures from CT Scans for Planning in Prostate Cancer Radiotherapy , 2014 .

[31]  Lei Xing,et al.  SU‐E‐J‐92: Multi‐Atlas Based Prostate Segmentation Guided by Partial Active Shape Model for CT Image with Gold Markers , 2013 .

[32]  Xiaoli Tang,et al.  Geometric-model-based segmentation of the prostate and surrounding structures for image-guided radiotherapy , 2004, IS&T/SPIE Electronic Imaging.

[33]  Shu Liao,et al.  Prostate Segmentation by Sparse Representation Based Classification , 2012, MICCAI.

[34]  Siqi Chen,et al.  Level set segmentation with both shape and intensity priors , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[35]  Olivier Salvado,et al.  Atlas Based Segmentation and Mapping of Organs at Risk from Planning CT for the Development of Voxel-Wise Predictive Models of Toxicity in Prostate Radiotherapy , 2010, Prostate Cancer Imaging.

[36]  Michael Bremer,et al.  The delineation of target volumes for radiotherapy of lung cancer patients. , 2009, Radiotherapy and Oncology.

[37]  Zhuowen Tu,et al.  Aggregated Residual Transformations for Deep Neural Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[38]  Dimitris N. Metaxas,et al.  3D Meshless Prostate Segmentation and Registration in Image Guided Radiotherapy , 2009, MICCAI.

[39]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[40]  Yaozong Gao.  Accurate Segmentation of CT Pelvic Organs via Incremental Cascade Learning and Regression-based Deformable Models , 2016 .

[41]  C. Hess,et al.  The Impact of Gross Tumor Volume (GTV) and Clinical Target Volume (CTV) Definition on the Total Accuracy in Radiotherapy , 2003, Strahlentherapie und Onkologie.

[42]  Pascal Haigron,et al.  Segmentation of pelvic structures for planning CT using a geometrical shape model tuned by a multi-scale edge detector , 2014, Physics in medicine and biology.

[43]  Yoshua. Bengio,et al.  Learning Deep Architectures for AI , 2007, Found. Trends Mach. Learn..

[44]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[45]  Oscar Déniz-Suárez,et al.  A geodesic deformable model for automatic segmentation of image sequences applied to radiation therapy , 2011, International Journal of Computer Assisted Radiology and Surgery.

[46]  Corine van Vliet-Vroegindeweij,et al.  Multiinstitutional study on target volume delineation variation in breast radiotherapy in the presence of guidelines. , 2010, Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology.

[47]  Hervé Delingette,et al.  Automatic Segmentation of Bladder and Prostate Using Coupled 3D Deformable Models , 2007, MICCAI.

[48]  Rongrong Guo,et al.  Combining population and patient-specific characteristics for prostate segmentation on 3D CT images , 2016, SPIE Medical Imaging.

[49]  Lei Xing,et al.  Automatic multiorgan segmentation in CT images of the male pelvis using region-specific hierarchical appearance cluster models. , 2016, Medical physics.

[50]  Andrew Thall,et al.  A method and software for segmentation of anatomic object ensembles by deformable m-reps. , 2005, Medical physics.

[51]  Payel Ghosh,et al.  Prostate segmentation on pelvic CT images using a genetic algorithm , 2008, SPIE Medical Imaging.