Stacked Multiscale Feature Learning for Domain Independent Medical Image Segmentation

Autor:	Martin Jagersand, Ryan Kiros, Dana Cobzas, Karteek Popuri
Rok vydání:	2014
Předmět:	Computer science Segmentation-based object categorization business.industry Feature (computer vision) Scale-space segmentation Pattern recognition Segmentation Image segmentation Artificial intelligence business Feature learning Energy (signal processing) Domain (software engineering)
Zdroj:	Machine Learning in Medical Imaging ISBN: 9783319105802 MLMI
Popis:	In this work we propose a feature-based segmentation approach that is domain independent. While most existing approaches are based on application-specific hand-crafted features, we propose a framework for learning features from data itself at multiple scales and depth. Our features can be easily integrated into classifiers or energy-based segmentation algorithms. We test the performance of our proposed method on two MICCAI grand challenges, obtaining the top score on VESSEL12 and competitive performance on BRATS2012.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::aff6cced9d886cbe5daaa748cf65f54b https://doi.org/10.1007/978-3-319-10581-9_4 Zobrazit plný text záznamu