期刊
ELIFE
卷 10, 期 -, 页码 -出版社
eLIFE SCIENCES PUBL LTD
DOI: 10.7554/eLife.60321
关键词
-
类别
资金
- National Institutes of Health [R21DC015652, R01NS096581, R01GM088333, R01GM108962, P40OD010440]
- National Science Foundation [1707401, 1764406]
- Direct For Biological Sciences
- Div Of Biological Infrastructure [1707401] Funding Source: National Science Foundation
- Direct For Mathematical & Physical Scien
- Division Of Mathematical Sciences [1764406] Funding Source: National Science Foundation
This article introduces a probabilistic-graphical-model framework, CRF_ID, based on Conditional Random Fields, for unbiased and automated cell identification. CRF_ID achieves higher accuracy and better robustness against challenging noise conditions in experimental data compared to existing methods. Additionally, it can boost accuracy by efficiently building atlases from annotated data and easily adding new features.
Although identifying cell names in dense image stacks is critical in analyzing functional whole-brain data enabling comparison across experiments, unbiased identification is very difficult, and relies heavily on researchers' experiences. Here, we present a probabilistic-graphical-model framework, CRF_ID, based on Conditional Random Fields, for unbiased and automated cell identification. CRF_ID focuses on maximizing intrinsic similarity between shapes. Compared to existing methods, CRF_ID achieves higher accuracy on simulated and ground-truth experimental datasets, and better robustness against challenging noise conditions common in experimental data. CRF_ID can further boost accuracy by building atlases from annotated data in highly computationally efficient manner, and by easily adding new features (e.g. from new strains). We demonstrate cell annotation in Caenorhabditis elegans images across strains, animal orientations, and tasks including gene-expression localization, multi-cellular and whole-brain functional imaging experiments. Together, these successes demonstrate that unbiased cell annotation can facilitate biological discovery, and this approach may be valuable to annotation tasks for other systems.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据