Change Details

Currently the workbench does not support simple binary segs any more directly (only DICOM Seg and our own multi label nrrds are supported). They have to be converted manually. This could be tedious. How to make it easier? Ideas: - Command line tool to convert data into multi label seg format - python support module - offer a reader in case of binary itk images to load as seg - as a new reader: convinient, but slows the loading even more as the image has to be interpreted and completly processed on mime type level to decide if the reader should be applied - as a new mime type: faster, but not so convinient would enforce a pop up for all supported data formates in order to choose if it should be loaded as image or as segmentation)

Currently the workbench does not support simple binary segs any more directly (only DICOM Seg and our own multi label nrrds are supported). They have to be converted manually. This could be tedious. How to make it easier? Ideas: - Command line tool to convert data into multi label seg format - python support module - offer a reader in case of binary itk images to load as seg - as a new reader: convinient, but slows the loading even more as the image has to be interpreted and completly processed on mime type level to decide if the reader should be applied - as a new mime type: faster, but not so convinient would enforce a pop up for all supported data formates in order to choose if it should be loaded as image or as segmentation) - introduce a hook (at application level (here it could be more general ?!? or seg view level) that monitors the data storage and proposes conversion (or does auto convert as soon as a legacy segmentation is loaded).

Currently the workbench does not support simple binary segs any more directly (only DICOM Seg and our own multi label nrrds are supported). They have to be converted manually. This could be tedious. How to make it easier? Ideas: - Command line tool to convert data into multi label seg format - python support module - offer a reader in case of binary itk images to load as seg - as a new reader: convinient, but slows the loading even more as the image has to be interpreted and completly processed on mime type level to decide if the reader should be applied - as a new mime type: faster, but not so convinient would enforce a pop up for all supported data formates in order to choose if it should be loaded as image or as segmentation) - introduce a hook (at application level (here it could be more general ?!? or seg view level) that monitors the data storage and proposes conversion (or does auto convert as soon as a legacy segmentation is loaded).