What is it about?
In this manuscript the authors present a strategy to accelerate automated tracking of growing yeast cells in time-lapse phase contrast images using an algorithm designed for off-the-shelf GPU installed on the control workstation. The yeast provides an established model for systems biology and beyond and therefore tools to help its study are interesting for the community.
Featured Image
Why is it important?
Beside the challenge of reliable non-invasive registration of yeast cell boundaries in dense cell populations (e.g. in bright-field, DIC or phase contrast images) as studied using microfluidics or high-content screening technologies, the bottleneck of the related methods is the computation time required for image processing. GPU computing is a promising tool to overcome this limitation. Thus, efforts aiming at accelerating the time-lapsed analysis of yeast or screening data is usually appreciated in the community. In this respect, the presented algorithm significantly speeds up a limiting step in a previously published image processing pipeline.
Perspectives
I think this article shows an effective perspective to introduce significant speed-up when processing data obtained in laboratories during experiments. The exploitation of easy-to-install computational accelerators like GPUs is a convenient way to efficiently improve the time-to-solution needed in many scientific experimental contexts.
Diego Romano
Consiglio Nazionale delle Ricerche
Read the Original
This page is a summary of: A GPU algorithm for tracking yeast cells in phase-contrast microscopy images, The International Journal of High Performance Computing Applications, September 2018, SAGE Publications,
DOI: 10.1177/1094342018801482.
You can read the full text:
Resources
Contributors
The following have contributed to this page
