Vortrag 14

Bookmark(0)

 

Robust slide cartography in colon cancer histology

Benz, Michaela; Bruns, Volker; Kuritcyn, Petr; Wittenberg, Thomas; Dexl, Jakob; Hartmann, David; Baghdadlian, Serop; Perrin, Dominik; Geppert, Carol; Eckstein, Markus; Hartmann, Arndt
Fraunhofer IIS, Erlangen

Abstract

Robustness against variations in color and resolution of digitized whole-slide images (WSIs) is an essential requirement for any computer-aided analysis in digital pathology. One common approach to encounter a lack of heterogeneity in the training data is data augmentation. We investigate the impact of different augmentation techniques for whole-slide cartography in colon cancer histology using a newly created multi scanner database of 39 slides each digitized with six different scanners. A state of the art convolutional neural network (CNN) is trained to differentiate seven tissue classes. Applying a model trained on one scanner to WSIs acquired with a different scanner results in a significant decrease in classification accuracy. Our results show that especially color augmentation is essential for robust models. Moreover, fine-tuning a pre-trained network using a small amount of annotated data from new scanners benefits the performance for these particular scanners, but this effect does not generalize to other unseen scanners.

Schlüsselwörter: 

Paper:

Persönlicher Kontakt:

Abstract

Robustness against variations in color and resolution of digitized whole-slide images (WSIs) is an essential requirement for any computer-aided analysis in digital pathology. One common approach to encounter a lack of heterogeneity in the training data is data augmentation. We investigate the impact of different augmentation techniques for whole-slide cartography in colon cancer histology using a newly created multi scanner database of 39 slides each digitized with six different scanners. A state of the art convolutional neural network (CNN) is trained to differentiate seven tissue classes. Applying a model trained on one scanner to WSIs acquired with a different scanner results in a significant decrease in classification accuracy. Our results show that especially color augmentation is essential for robust models. Moreover, fine-tuning a pre-trained network using a small amount of annotated data from new scanners benefits the performance for these particular scanners, but this effect does not generalize to other unseen scanners.

Schlüsselwörter: 

Paper:

Persönlicher Kontakt: