Advances in microscopy techniques have led to huge increases in the amount of imaging data being collected, prompting the use of deep-learning methods for image analysis. Deep-learning methods, known as convolutional neural networks (CNNs), have been used to automate segmentation and classification of phenotypes in developmental biology but have not been widely adopted by researchers without a computational background. In this Issue, Soeren Lienkamp and colleagues debunk some common misconceptions associated with CNNs, such as the requirements for expert computational knowledge, huge training datasets and specific image types, and describe the flexibility of U-Net (a Fiji plug-in) – their CNN architecture of choice. Using various models of human congenital diseases, the authors analyse renal, neural and craniofacial phenotypes in Xenopus embryos using trained networks. They demonstrate that their U-Net models can identify key morphological features and perform as well as expert annotators when trained on as few as five images of each phenotype. The authors also show that these networks can be applied to different imaging setups, developmental stages and staining protocols with minimal interventions. Together, these data demonstrate the power of CNNs in high-throughput analysis of embryonic development and disease, and increase the accessibility of deep-learning-based analysis to those without a computational background.
Deep learning for all
- Split-screen
- Views Icon Views
-
Article Versions Icon
Versions
- Version of Record 05 November 2021
- Share Icon Share
-
Tools Icon
Tools
- Search Site
Deep learning for all. Development 1 November 2021; 148 (21): e148_e2101. doi:
Download citation file:
Advertisement
Cited by
Call for papers: Uncovering Developmental Diversity
Development invites you to submit your latest research to our upcoming special issue: Uncovering Developmental Diversity. This issue will be coordinated by our academic Editor Cassandra Extavour (Harvard University, USA) alongside two Guest Editors: Liam Dolan (Gregor Mendel Institute of Molecular Plant Biology, Austria) and Karen Sears (University of California Los Angeles, USA).
Choose Development in 2024
In this Editorial, Development Editor-in-Chief James Briscoe and Executive Editor Katherine Brown explain how you support your community by publishing in Development and how the journal champions serious science, community connections and progressive publishing.
Journal Meeting: From Stem Cells to Human Development
Register now for the 2024 Development Journal Meeting From Stem Cells to Human Development. Early-bird registration deadline: 3 May. Abstract submission deadline: 21 June.
Pluripotency of a founding field: rebranding developmental biology
This collaborative Perspective, the result of a workshop held in 2023, proposes a set of community actions to increase the visibility of the developmental biology field. The authors make recommendations for new funding streams, frameworks for collaborations and mechanisms by which members of the community can promote themselves and their research.
Read & Publish Open Access publishing: what authors say
We have had great feedback from authors who have benefitted from our Read & Publish agreement with their institution and have been able to publish Open Access with us without paying an APC. Read what they had to say.