A Practical Guide to Supervised Deep Learning for Bioimage Analysis: Challenges and good practices
Virginie Uhlmann, Laurène Donati, Daniel Sage · IEEE Signal Processing Magazine · 2022
The variety of bioimage data and their quality have dramatically increased over the last decade. In parallel, the number of proposed deep learning (DL) models for their analysis grows by the day. Yet, the adequate reuse of published tools by practitioners without DL expertise still raises many practical questions. In this article, we explore four categories of challenges faced by researchers when using supervised DL models in bioimaging applications. We provide examples in which each challenge arises and review the consequences that inadequate decisions may have. We then outline good practices that can be implemented to address the challenges of each category in a scientifically sound way. We provide pointers to the resources that are already available or in active development to help in this endeavor and advocate for the development of further community-driven standards. While primarily intended as a practical tutorial for life scientists, this article also aims at fostering discussions among method developers around the formulation of guidelines for the adequate deployment of DL, with the ultimate goal of accelerating the adoption of novel DL technologies in the biology community.
Cited by 15 works. doi:10.1109/msp.2021.3123589
Authors
References
- Deep Residual Learning for Image Recognition
- Densely Connected Convolutional Networks
- A Comprehensive Survey on Transfer Learning
- Cellpose: a generalist algorithm for cellular segmentation
- U-Net: deep learning for cell counting, detection, and morphometry
- Content-aware image restoration: pushing the limits of fluorescence microscopy
- A unifying view on dataset shift in classification
- Dataset Shift in Machine Learning
- Cell Detection with Star-Convex Polygons
- Nucleus segmentation across imaging experiments: the 2018 Data Science Bowl