Impact of pre-analytical variables on deep learning accuracy in histopathology.

Andrew D JonesJohn Paul GraffMorgan DarrowAlexander BorowskyKristin A OlsonRegina Gandour-EdwardsAnanya Datta MitraDongguang WeiGuofeng GaoBlythe Durbin-JohnsonHooman H Rashidi

Published in: Histopathology (2019)

Lossy JPG images were not inferior to lossless PNG images in our models. Large numbers of unique H&E-stained slides were not required for training optimal ML models. This reinforces the need for an evidence-based approach to best practices for histopathological ML.

Keyphrases

deep learning
convolutional neural network
artificial intelligence
primary care
machine learning
optical coherence tomography
virtual reality