Accelerating Convolutional Sparse Coding for Curvilinear Structures Segmentation by Refining SCIRD-TS Filter Banks

Deep learning has shown great potential for curvilinear structure (e.g., retinal blood vessels and neurites) segmentation as demonstrated by a recent auto-context regression architecture based on filter banks learned by convolutional sparse coding. However, learning such filter banks is very time-consuming, thus limiting the amount of filters employed and the adaptation to other data sets (i.e., slow re-training). We address this limitation by proposing a novel acceleration strategy to speed-up convolutional sparse coding filter learning for curvilinear structure segmentation. Our approach is based on a novel initialisation strategy (warm start), and therefore it is different from recent methods improving the optimisation itself. Our warm-start strategy is based on carefully designed hand-crafted filters (SCIRD-TS), modelling appearance properties of curvilinear structures which are then refined by convolutional sparse coding. Experiments on four diverse data sets, including retinal blood vessels and neurites, suggest that the proposed method reduces significantly the time taken to learn convolutional filter banks (i.e., up to -82%) compared to conventional initialisation strategies. Remarkably, this speed-up does not worsen performance; in fact, filters learned with the proposed strategy often achieve a much lower reconstruction error and match or exceed the segmentation performance of random and DCT-based initialisation, when used as input to a random forest classifier.

[1]  Max W. K. Law,et al.  Three Dimensional Curvilinear Structure Detection Using Optimally Oriented Flux , 2008, ECCV.

[2]  Shruti Aggarwal,et al.  Two-Dimensional Plane for Multi-Scale Quantification of Corneal Subbasal Nerve Tortuosity , 2016, Investigative ophthalmology & visual science.

[3]  Hui Ji,et al.  A Convergent Incoherent Dictionary Learning Algorithm for Sparse Coding , 2014, ECCV.

[4]  Anders P. Eriksson,et al.  Fast Convolutional Sparse Coding , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[5]  Yann LeCun,et al.  Structured sparse coding via lateral inhibition , 2011, NIPS.

[6]  Roberto Marcondes Cesar Junior,et al.  Retinal Vessel Segmentation Using the 2-D Morlet Wavelet and Supervised Classification , 2005, ArXiv.

[7]  Antonio Criminisi,et al.  Decision Forests: A Unified Framework for Classification, Regression, Density Estimation, Manifold Learning and Semi-Supervised Learning , 2012, Found. Trends Comput. Graph. Vis..

[8]  José Carlos Príncipe,et al.  A fast proximal method for convolutional sparse coding , 2013, The 2013 International Joint Conference on Neural Networks (IJCNN).

[9]  Rajat Raina,et al.  Efficient sparse coding algorithms , 2006, NIPS.

[10]  Vincent Lepetit,et al.  Learning Separable Filters , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[11]  Vincent Lepetit,et al.  Accurate and Efficient Linear Structure Segmentation by Leveraging Ad Hoc Features with Learned Filters , 2012, MICCAI.

[12]  Elisa Ricci,et al.  Retinal Blood Vessel Segmentation Using Line Operators and Support Vector Classification , 2007, IEEE Transactions on Medical Imaging.

[13]  Roberto Marcondes Cesar Junior,et al.  Retinal vessel segmentation using the 2-D Gabor wavelet and supervised classification , 2005, IEEE Transactions on Medical Imaging.

[14]  Emanuele Trucco,et al.  Leveraging Multiscale Hessian-Based Enhancement With a Novel Exudate Inpainting Technique for Retinal Vessel Segmentation , 2016, IEEE Journal of Biomedical and Health Informatics.

[15]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[16]  Tianfu Wang,et al.  A Cross-Modality Learning Approach for Vessel Segmentation in Retinal Images , 2016, IEEE Transactions on Medical Imaging.

[17]  A.D. Hoover,et al.  Locating blood vessels in retinal images by piecewise threshold probing of a matched filter response , 2000, IEEE Transactions on Medical Imaging.

[18]  Alejandro F. Frangi,et al.  Muliscale Vessel Enhancement Filtering , 1998, MICCAI.

[19]  Dinesh Kumar,et al.  Validating retinal fundus image analysis algorithms: issues and a proposal. , 2013, Investigative ophthalmology & visual science.

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

[21]  Ana Maria Mendonça,et al.  Segmentation of retinal blood vessels by combining the detection of centerlines and morphological reconstruction , 2006, IEEE Transactions on Medical Imaging.

[22]  Ke Chen,et al.  Automated Vessel Segmentation Using Infinite Perimeter Active Contour Model with Hybrid Region Information with Application to Retinal Images , 2015, IEEE Transactions on Medical Imaging.

[23]  Victor S. Lempitsky,et al.  N4-Fields: Neural Network Nearest Neighbor Fields for Image Transforms , 2014, ArXiv.

[24]  Sergei Vassilvitskii,et al.  k-means++: the advantages of careful seeding , 2007, SODA '07.

[25]  José Manuel Bravo,et al.  A New Supervised Method for Blood Vessel Segmentation in Retinal Images by Using Gray-Level and Moment Invariants-Based Features , 2011, IEEE Transactions on Medical Imaging.

[26]  Emanuele Trucco,et al.  Scale and Curvature Invariant Ridge Detector for Tortuous and Fragmented Structures , 2015, MICCAI.

[27]  Pascal Vincent,et al.  Representation Learning: A Review and New Perspectives , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[28]  Stephen P. Boyd,et al.  Proximal Algorithms , 2013, Found. Trends Optim..

[29]  Gordon Wetzstein,et al.  Fast and flexible convolutional sparse coding , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[30]  Peter Kontschieder,et al.  Deep Neural Decision Forests , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[31]  Andrew Y. Ng,et al.  Learning Feature Representations with K-Means , 2012, Neural Networks: Tricks of the Trade.

[32]  Marc Teboulle,et al.  A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems , 2009, SIAM J. Imaging Sci..

[33]  Julien Mairal,et al.  Convex optimization with sparsity-inducing norms , 2011 .

[34]  Emanuele Trucco,et al.  Boosting Hand-Crafted Features for Curvilinear Structure Segmentation by Learning Context Filters , 2015, MICCAI.

[35]  Max A. Viergever,et al.  Ridge-based vessel segmentation in color images of the retina , 2004, IEEE Transactions on Medical Imaging.

[36]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[37]  Alan Wee-Chung Liew,et al.  General Retinal Vessel Segmentation Using Regularization-Based Multiconcavity Modeling , 2010, IEEE Transactions on Medical Imaging.

[38]  Vincent Lepetit,et al.  Projection onto the Manifold of Elongated Structures for Accurate Extraction , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[39]  Li Cheng,et al.  Learning to Boost Filamentary Structure Segmentation , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[40]  Andrew Hunter,et al.  An Active Contour Model for Segmenting and Measuring Retinal Vessels , 2009, IEEE Transactions on Medical Imaging.

[41]  Quoc V. Le,et al.  On optimization methods for deep learning , 2011, ICML.

[42]  Ju Lu,et al.  The DIADEM Data Sets: Representative Light Microscopy Images of Neuronal Morphology to Advance Automation of Digital Reconstructions , 2011, Neuroinformatics.

[43]  Shruti Aggarwal,et al.  A fully automated tortuosity quantification system with application to corneal nerve fibres in confocal microscopy images , 2016, Medical Image Anal..

[44]  Y-Lan Boureau,et al.  Learning Convolutional Feature Hierarchies for Visual Recognition , 2010, NIPS.

[45]  Vincent Lepetit,et al.  Multiscale Centerline Detection , 2016, IEEE Transactions on Pattern Analysis and Machine Intelligence.