Learning Active Contour Models for Medical Image Segmentation

Image segmentation is an important step in medical image processing and has been widely studied and developed for refinement of clinical analysis and applications. New models based on deep learning have improved results but are restricted to pixel-wise fitting of the segmentation map. Our aim was to tackle this limitation by developing a new model based on deep learning which takes into account the area inside as well as outside the region of interest as well as the size of boundaries during learning. Specifically, we propose a new loss function which incorporates area and size information and integrates this into a dense deep learning model. We evaluated our approach on a dataset of more than 2,000 cardiac MRI scans. Our results show that the proposed loss function outperforms other mainstream loss function Cross-entropy on two common segmentation networks. Our loss function is robust while using different hyperparameter lambda.

[1]  Max A. Viergever,et al.  Automatic Segmentation and Disease Classification Using Cardiac Cine MR Images , 2017, STACOM@MICCAI.

[2]  Bram van Ginneken,et al.  A survey on deep learning in medical image analysis , 2017, Medical Image Anal..

[3]  Mert R. Sabuncu,et al.  Anatomical Priors in Convolutional Networks for Unsupervised Biomedical Segmentation , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[4]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[5]  Zhao Feng,et al.  Two-Dimensional Otsu's Curve Thresholding Segmentation Method for Gray-Level Images , 2007 .

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

[7]  Yoshua Bengio,et al.  The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentation , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[8]  D. Louis Collins,et al.  Patch-based segmentation using expert priors: Application to hippocampus and ventricle segmentation , 2011, NeuroImage.

[9]  Chi-Wing Fu,et al.  H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation From CT Volumes , 2018, IEEE Transactions on Medical Imaging.

[10]  Ke Chen,et al.  A general model for multiphase texture segmentation and its applications to retinal image analysis , 2013, Biomed. Signal Process. Control..

[11]  M. Mohammed Thaha,et al.  Brain Tumor Segmentation Using Convolutional Neural Networks in MRI Images , 2019, Journal of Medical Systems.

[12]  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).

[13]  Su-Lin Lee,et al.  Focal FCN: Towards Small Object Segmentation with Limited Training Data , 2017, ArXiv.

[14]  Yuri Boykov,et al.  Normalized Cut Loss for Weakly-Supervised CNN Segmentation , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[15]  F. Moldoveanu,et al.  Image segmentation based on active contours without edges , 2012, 2012 IEEE 8th International Conference on Intelligent Computer Communication and Processing.

[16]  D. Mumford,et al.  Optimal approximations by piecewise smooth functions and associated variational problems , 1989 .

[17]  Tony F. Chan,et al.  An Active Contour Model without Edges , 1999, Scale-Space.

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

[19]  Yeonggul Jang,et al.  Automatic Segmentation of LV and RV in Cardiac MRI , 2017, STACOM@MICCAI.

[20]  Klaus H. Maier-Hein,et al.  Statistical Atlases and Computational Models of the Heart. ACDC and MMWHS Challenges , 2017, Lecture Notes in Computer Science.

[21]  Marc Pollefeys,et al.  An Exploration of 2D and 3D Deep Learning Techniques for Cardiac MR Image Segmentation , 2017, STACOM@MICCAI.

[22]  Kaiming He,et al.  Focal Loss for Dense Object Detection , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

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

[24]  Xavier Bresson,et al.  Fast Global Minimization of the Active Contour/Snake Model , 2007, Journal of Mathematical Imaging and Vision.

[25]  Ross B. Girshick,et al.  Fast R-CNN , 2015, 1504.08083.

[26]  S. M. Masudur Rahman Al-Arif,et al.  Shape-Aware Deep Convolutional Neural Network for Vertebrae Segmentation , 2017, MSKI@MICCAI.

[27]  Edward J. Delp,et al.  Weighted Hausdorff Distance: A Loss Function For Object Localization , 2018, ArXiv.

[28]  Ke Chen,et al.  A Hierarchical Algorithm for Multiphase Texture Image Segmentation , 2012 .

[29]  Phi Vu Tran,et al.  A Fully Convolutional Neural Network for Cardiac Segmentation in Short-Axis MRI , 2016, ArXiv.

[30]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[31]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[32]  Jan Kautz,et al.  Learning Superpixels with Segmentation-Aware Affinity Loss , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[33]  Tony F. Chan,et al.  A Multiphase Level Set Framework for Image Segmentation Using the Mumford and Shah Model , 2002, International Journal of Computer Vision.