Details

Autor(en) / Beteiligte

• Faryna, Khrystyna
• van der Laak, Jeroen
• Litjens, Geert

Titel

Automatic data augmentation to improve generalization of deep learning in H&E stained histopathology

Ist Teil von

• Computers in biology and medicine, 2024-03, Vol.170, p.108018-108018, Article 108018

Ort / Verlag

United States: Elsevier Ltd

Erscheinungsjahr

2024

Link zum Volltext

Quelle

MEDLINE

Beschreibungen/Notizen

• In histopathology practice, scanners, tissue processing, staining, and image acquisition protocols vary from center to center, resulting in subtle variations in images. Vanilla convolutional neural networks are sensitive to such domain shifts. Data augmentation is a popular way to improve domain generalization. Currently, state-of-the-art domain generalization in computational pathology is achieved using a manually curated set of augmentation transforms. However, manual tuning of augmentation parameters is time-consuming and can lead to sub-optimal generalization performance. Meta-learning frameworks can provide efficient ways to find optimal training hyper-parameters, including data augmentation. In this study, we hypothesize that an automated search of augmentation hyper-parameters can provide superior generalization performance and reduce experimental optimization time. We select four state-of-the-art automatic augmentation methods from general computer vision and investigate their capacity to improve domain generalization in histopathology. We analyze their performance on data from 25 centers across two different tasks: tumor metastasis detection in lymph nodes and breast cancer tissue type classification. On tumor metastasis detection, most automatic augmentation methods achieve comparable performance to state-of-the-art manual augmentation. On breast cancer tissue type classification, the leading automatic augmentation method significantly outperforms state-of-the-art manual data augmentation. •Automatic augmentation achieves state-of-the-art domain generalization performance.•Computational cost and ease of adaptation to new tasks should be taken into account.•Integrating domain-specific knowledge into automatic augmentation can be beneficial.•Randaugment is a simple way to get state-of-the-art performance data augmentation.•Automatic augmentation tuning is structured, reproducible, and cost-efficient.