Back

Accelerated and Reproducible Fiji for image processing using GPUs on the cloud

Hung, L.-H.; Straw, E.; Colburn, Z.; Yeung, K. Y.

2022-07-18 bioinformatics
10.1101/2022.07.15.500283 bioRxiv
Show abstract

SummaryGraphical processing units can greatly accelerate image processing but adoption has been hampered by the need for specialized hardware and software. The cloud offers inexpensive on-demand instances that can be pre-configured with the necessary software. Specifically, we use the Biodepot-workflow-builder (Bwb) to deploy a containerized version of Fiji that includes the CLIJ package to use GPUs on the cloud. In addition, we provide an Amazon Machine Image (AMI) with the correct drivers and Docker images pre-loaded. We demonstrate the portability and reproducibility of the platform by deploying an interactive Fiji/CLIJ workflow on both Amazon Web Services and IBM cloud. The workflows produce identical results while providing a 29-fold reduction in execution time.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.