How do I edit this website?

BiaPy

BiaPy logo

Overview

BiaPy is an open-source library to build deep learning workflows for bioimage analysis. This plugin brings it to Fiji: it installs BiaPy in its own Python environment and lets you create, run and follow BiaPy jobs from a single window, with no programming or command line.

Workflows: semantic segmentation, instance segmentation, object detection, denoising, super-resolution, self-supervised learning, image classification and image-to-image translation, in 2D and 3D.

  • Wizard: a few questions about your task, a check of your data (with a preview of the images next to their targets) and a configuration adapted to it, in the spirit of nnU-Net. It also looks for pretrained models of the BioImage Model Zoo that fit your workflow, such as Cellpose-SAM and Cellpose-DINO for cell and nucleus segmentation, which can be used as they are or fine-tuned with your data.
  • Run workflow: runs any BiaPy configuration on a GPU (NVIDIA, or Apple Silicon) or on the CPU, and shows the training curves, the predictions and the results as the job goes.
  • Job history: the jobs run, to open their results or run them again.

Installation

  1. Start Fiji (or download it first).
  2. Select Help › Update… from the menu bar.
  3. Click on Manage update sites.
  4. Tick the checkbox of the update site BiaPy and click Apply and Close.
  5. Click on Apply changes and restart Fiji.
  6. Open Plugins › BiaPy › Run BiaPy. In the HOME tab, click on Install BiaPy: it downloads BiaPy and PyTorch into the Fiji folder (several GB, only the first time).

Documentation and support

Citation

When you use BiaPy, please cite it (each job also writes the works to cite for its configuration in CITATIONS.md):

Franco-Barranco, D., et al. BiaPy: accessible deep learning on bioimages. Nature Methods (2025). https://doi.org/10.1038/s41592-025-02699-y