Patch-Based Techniques in Medical Imaging
Springer International Publishing (Verlag)
978-3-319-28193-3 (ISBN)
This book constitutes the thoroughly refereedpost-workshop proceedings of the First International Workshop on Patch-based Techniquesin Medical Images, Patch-MI 2015, which was held in conjunction with MICCAI2015, in Munich, Germany, in October 2015.
The 25 full papers presented in this volume werecarefully reviewed and selected from 35 submissions. The topics covered are suchas image segmentation of anatomical structures or lesions; image enhancement;computer-aided prognostic and diagnostic; multi-modality fusion; mono and multimodal image synthesis; image retrieval; dynamic, functional physiologic andanatomic imaging; super-pixel/voxel in medical image analysis; sparsedictionary learning and sparse coding; analysis of 2D, 2D+t, 3D, 3D+t, 4D, and4D+t data.
A Multi-level Canonical Correlation Analysis Scheme for Standard-dosePET Image Estimation.- Image Super-Resolution by Supervised Adaption ofPatchwise Self-Similarity from High-Resolution Image.- Automatic HippocampusLabeling Using the Hierarchy of Sub-Region Random Forests.- Isointense InfantBrain Segmentation by Stacked Kernel Canonical Correlation Analysis.- ImprovingAccuracy of Automatic Hippocampus Segmentation in Routine MRI by FeaturesLearned from Ultra-high Field MRI.- Dual-Layer l1-Graph Embedding forSemi-Supervised Image Labeling.- Automatic Liver Tumor Segmentation inFollow-up CT Studies Using Convolutional Neural Network.- Block-basedStatistics for Robust Non-Parametric Morphometry.- Automatic CollimationDetection in Digital Radiographs with the Directed Hough Transform andLearning-based Edge Detection.- Efficient Lung Cancer Cell Detection with DeepConvolutional Neural Network.- An Effective Approach for Robust Lung CancerCell Detection.- Laplacian Shape Editing with Local Patch Based Force Field forInteractive Segmentation.- Hippocampus Segmentation through Distance FieldFusion.- Learning a Spatiotemporal Dictionary for Magnetic ResonanceFingerprinting with Compress Sensing.- Fast Regions-of-Interest Detection inWhole Slide Histopathology Images.- Reliability Guided Forward and BackwardPatch-based Method for Multi-atlas Segmentation.- Correlating Tumour Histologyand ex vivo MRI Using Dense Modality-Independent Patch-Based Descriptor.- Multi-AtlasSegmentation using Patch-Based Joint Label Fusion with Non-Negative LeastSquares Regression.- A Spatially Constrained Deep Learning Framework forDetection of Epithelial Tumor Nuclei in Cancer Histology Images.- 3D MRIDenoising using Rough Set Theory and Kernel Embedding Method.- A Novel CellOrientation Congruence Descriptor for Superpixel based Epithelium Segmentationin Endometrial Histology Images.- Patch-based Segmentation from MP2RAGE Images:Comparison to Conventional Techniques.-Multi-Atlas and Multi-Modal HippocampusSegmentation for Infant MR Brain Images by Propagating Anatomical Labels onHypergraph.- Prediction of Infant MRI Appearance and Anatomical StructureEvolution using Sparse Patch-based Metamorphosis Learning Framework.- EfficientMulti-Scale Patch-based Segmentation.
| Erscheinungsdatum | 08.10.2016 |
|---|---|
| Reihe/Serie | Image Processing, Computer Vision, Pattern Recognition, and Graphics | Lecture Notes in Computer Science |
| Zusatzinfo | IX, 216 p. 81 illus. in color. |
| Verlagsort | Cham |
| Sprache | englisch |
| Maße | 155 x 235 mm |
| Themenwelt | Informatik ► Grafik / Design ► Digitale Bildverarbeitung |
| Schlagworte | Cell orientation • Computer Science • Computer Science, general • convolutional neural networks • Dictionary Learning • Generative model • Histology image analysis • image denoising • image enhancement • Image Registration • image retrieve • Image Segmentation • Image Synthesis • linear regression • MRI • multi-modality fusion • neural network • Patch-based segmentation • random forest • Rough Set Theory • Sparse Representation • super pixel |
| ISBN-10 | 3-319-28193-3 / 3319281933 |
| ISBN-13 | 978-3-319-28193-3 / 9783319281933 |
| Zustand | Neuware |
| Haben Sie eine Frage zum Produkt? |
aus dem Bereich
