Weakly Supervised Learning of Metric Aggregations for Deformable Image Registration
| Autor: | Rafael Marini Silva, Puneet K. Dokania, Nikos Paragios, Enzo Ferrante |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
FOS: Computer and information sciences
Similarity (geometry) Support Vector Machine Matching (graph theory) Computer science Computer Vision and Pattern Recognition (cs.CV) Physics::Medical Physics Computer Science - Computer Vision and Pattern Recognition ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Image registration Health Informatics 030218 nuclear medicine & medical imaging 03 medical and health sciences 0302 clinical medicine Health Information Management DEFORMABLE IMAGE REGISTRATION Abdomen Image Interpretation Computer-Assisted Image Processing Computer-Assisted Humans Graphical model Electrical and Electronic Engineering DISCRETE GRAPHICAL MODELS business.industry WEAKLY SUPERVISED LEARNING Supervised learning Brain Pattern recognition LATENT STRUCTURED SUPPORT VECTOR MACHINE (LSSVM) Magnetic Resonance Imaging Computer Science Applications Support vector machine Transformation (function) Ciencias de la Computación e Información Metric (mathematics) Artificial intelligence Supervised Machine Learning business Tomography X-Ray Computed Ciencias de la Información y Bioinformática 030217 neurology & neurosurgery CIENCIAS NATURALES Y EXACTAS Algorithms |
| Zdroj: | IEEE journal of biomedical and health informatics. 23(4) |
| ISSN: | 2168-2208 |
| Popis: | Deformable registration has been one of the pillars of biomedical image computing. Conventional approaches refer to the definition of a similarity criterion that, once endowed with a deformation model and a smoothness constraint, determines the optimal transformation to align two given images. The definition of this metric function is among the most critical aspects of the registration process. We argue that incorporating semantic information (in the form of anatomical segmentation maps) into the registration process will further improve the accuracy of the results. In this paper, we propose a novel weakly supervised approach to learn domain specific aggregations of conventional metrics using anatomical segmentations. This combination is learned using latent structured support vector machines (LSSVM). The learned matching criterion is integrated within a metric free optimization framework based on graphical models, resulting in a multi-metric algorithm endowed with a spatially varying similarity metric function conditioned on the anatomical structures. We provide extensive evaluation on three different datasets of CT and MRI images, showing that learned multi-metric registration outperforms single-metric approaches based on conventional similarity measures. Comment: Accepted for publication in IEEE Journal of Biomedical and Health Informatics, 2018 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|