HistomicsML2.0: Fast interactive machine learning for whole slide imaging data
This addresses the problem of accessibility for investigators without image analysis expertise in histopathology, though it appears incremental as an update to existing software.
The authors tackled the challenge of enabling non-experts to extract quantitative phenotypic information from whole-slide images by developing HistomicsML2.0, a software that allows rapid learn-by-example training of machine learning classifiers for histologic pattern detection, resulting in a tool with a web-based interface and containerized deployment for ease of use.
Extracting quantitative phenotypic information from whole-slide images presents significant challenges for investigators who are not experienced in developing image analysis algorithms. We present new software that enables rapid learn-by-example training of machine learning classifiers for detection of histologic patterns in whole-slide imaging datasets. HistomicsML2.0 uses convolutional networks to be readily adaptable to a variety of applications, provides a web-based user interface, and is available as a software container to simplify deployment.