A hybrid human-computer approach for large-scale image-based measurements using web services and machine learning

摘要

We present a novel methodology combining web-based softwaredevelopment practices, machine learning, and spatial databases forcomputer-aided quantification of regions of interest (ROIs) inlarge-scale imaging data. We describe our main methodologicalchoices, and then illustrate the benefits of the approach (workloadreduction, improved precision, scalability, and traceability) onhundreds of whole-slide images of biological tissue slices in cancerresearch.