ASAP focuses on fast and fluid image viewing with an easy-to-use interface for making annotations. It consists of two main components: an IO library for reading and writing multi-resolution images and a viewer component for visualizing such images.
The multi-resolution image IO and processing libraries are wrapped in Python for easy application of machine learning tools.
ASAP contains a fast and fluid viewing application for analysis of multi-resolution images including a mini-map and field of view tracking.
Several annotation tools are integrated, such as dots, polygons and measurements. Annotations are stored in human-readable XML.
ASAP can visualize images as overlays, which allows inspection of segmentation results or likelihood maps. LUTs can be customized.
The entire application is extensible via a plugin framework which allows developers to add new functionality.
Below you can find some example images highlighting some of the features of the viewing application.
ASAP currently supports several different types of annotation and allows easy correction of already made annotations. Currently we support the following types:
ASAP can overlay multi-resolution images with float and integer datatype on top of other multi-resolution images. Overlay colors can be determined manually with lookup tables and window leveling.
ASAP supports on-the-fly inspection of image analysis results via ImageFilters. Basic filters such as nuclei detection are implemented, but they can be extended by developers.
ASAP (Automated Slide Analysis Platform) was developed by the Computation Pathology Group, part of the Diagnostic Image Analysis Group, at the Radboud University Medical Center. It was started after frustration with the current freely available software for annotating multi-resolution digital pathology images. Furthermore, the currently available software did not allow visualization of image analysis or machine learning results in a intuitive and useful manner. We hope that ASAP can function as a way for other users to jumpstart their digital pathology research and move the field ahead as a whole.
The development of this application would not have been possible without the support of the Dutch Cancer Association and the StITPro Foundation, whom provided us with research funding.
If your interested in this software or in our research, please reach out to us.
Geert Grootteplein-Zuid 10, 6525GA, Nijmegen, The Netherlands