Skip to main content
. Author manuscript; available in PMC: 2018 Mar 27.
Published in final edited form as: Comput Intell Methods Bioinform Biostat. 2017 Oct 17;10477:42–58. doi: 10.1007/978-3-319-67834-4_4

Fig. 1.

Fig. 1

Proposed microscope-based saliency predictor pipeline workflow. The pathology session is recorded, the slide is scanned, the video frames are registered to scan patches. Lens change detection guides registration and viewing time is recorded for periods without motion. A convolutional neural net learns to classify patches as salient (long looks) or not.