Department of Neurobiology and Anatomy, Kochi University Faculty of Medicine, Kochi, Japan.
Clinical & Medical Affairs, Ziemer Ophthalmic Systems AG, Port, Switzerland.
Curr Protoc. 2023 Jul;3(7):e819. doi: 10.1002/cpz1.819.
The explosive growth of Machine Learning provided scientists with insights into the data in the ways unattainable using established research techniques. It allowed the detection of biological features that were previously unrecognized and overlooked. Yet, since Machine Learning methodology originates from informatics, many cell biology laboratories experience difficulties with implementing it. In preparing this article, we targeted the rapidly expanding audience of cell and molecular biologists who perform analysis of microscopy images and seek to add Machine Learning models to their research workflow. We review the advantages of using Machine Learning in microscopy projects, describe the Machine Learning pipeline, and share practical guidelines for building the models. The latest developments in the rapidly expanding field are also given. The technical survey is concluded with an overview of the tools required for model creation and advice on their use. © 2023 Wiley Periodicals LLC.
机器学习的爆炸式增长为科学家提供了一种前所未有的洞察力，可以从数据中获取信息，而这些信息是使用既定的研究技术无法获得的。它使得以前未被识别和忽视的生物特征得以被检测到。然而，由于机器学习方法源于信息学，许多细胞生物学实验室在实施它方面遇到了困难。在撰写本文时，我们针对的是快速增长的细胞和分子生物学家群体，他们从事显微镜图像分析，并希望将机器学习模型添加到他们的研究工作流程中。我们回顾了在显微镜项目中使用机器学习的优势，描述了机器学习管道，并分享了构建模型的实用指南。还介绍了快速扩展领域的最新发展。技术调查以创建模型所需工具的概述以及关于其使用的建议结束。© 2023 威立出版社有限公司。
