Skip to content

BioMarkerImaging/CelerSightAI

Repository files navigation

Celer Sight AI

celer_sight_splash

Celer Sight AI is a free high-throughput imaging analysis solution that leverages AI to simplify microscopy image analysis through a community powered AI model segmentation catalogue that is constantly improving. Perfect for researchers and professionals who need powerful image analysis without coding expertise.


πŸš€ Key Features

  • πŸš€ High-Throughput Analysis: Process fluorescent images much faster than traditional methods
  • ⚑️ Adaptive AI Models: Community-driven model hub that continuously improves through collective usage
  • πŸ› οΈ Experiment Reproducibility: Save and load sessions with .bmics files
  • πŸ–₯️ Cross-Platform: Fully supported on macOS and Windows
  • πŸ€“ User-Friendly Interface: Pre-trained AI models and automated analysis - no coding required
  • πŸ’Ύ Optimized Performance: Efficient handling of large TIF files with smart disk caching
  • πŸ”¬ WSI Compatible: Support for pyramidal whole slide images
  • 🌐 Open Source Models: All models available through HuggingFace for community use
  • πŸ—‚οΈ Intuitive Data Import: Simple drag-and-drop interface for batch processing

Why Choose Celer Sight?

Traditional image analysis tools often require extensive training or programming knowledge. Celer Sight revolutionizes this workflow by providing:

  • Intuitive interface for immediate productivity
  • Community-driven model development
  • Automated analysis pipelines

Quick Start Guide

Please follow the quick start guide here

See it in action:

1) Sample Experiment

sample_worms

2) Easy importing

import patterns

3) Easily create new class / models && improve them

Create new model

4) High resolution / pyramidal image support

Untitled

πŸ”§ Installation for development

  1. Clone the repository:

    git clone https://github.com/BioMarkerImaging/SelerSightAI

    cd celer-sight-ai

  2. Create and activate a virtual environment:

    Windows:

    python -m venv .venv

    .venv\Scripts\activate

    macOS/Linux:

    python -m venv .venv

    source .venv/bin/activate

  3. Install dependencies:

    pip install -r requirements.txt pip install -v -e .

Model training / inference is expensive, how can this be free?

Initially, all users will have access to cloud compute, however, within the next year this will change. For free users Celer Sight will download a version of the requested model automatically and run inference on users machine in the background. If you still need cloud inference, there will be a paid option available for those who can support it.

Current Model (Beta)

  • Cloud inference available to all users
  • Zero data retention policy
  • No image or metadata storage
  • Daily limit to all users, with an increased limit to patreon members

Upcoming Tiers (2025)

Free Tier

  • Local inference by default
  • Potential cloud inference during low-traffic periods
  • Access to all community models
  • Community model contributions

Premium Tier

  • Priority cloud inference
  • Private model development possible
  • Dedicated processing resources
  • Premium support access

πŸ’ Support Celer Sight

Contributing

We welcome contributions from the community! Here are some ways you can help:

  • Submit bug reports and feature requests through GitHub Issues
  • Contribute code improvements via Pull Requests
  • Share annotated datasets to help improve our models through the app or by email
  • Help document and improve our wiki

Support Our Work

If you find Celer Sight useful, consider supporting its development:

  • β˜•οΈ Buy us a coffee on Ko-fi
  • πŸ’– Become a patron on Patreon
  • 🌟 Star us on GitHub

Your support helps us maintain and improve Celer Sight for everyone!

Data Collection

  • Account information for registered users (email only)
  • When a user contributes images we retain Image (without metadata) and annotation data
  • Zero retention policy during inference.
  • No data is shared with any other third parties.

License

Celer Sight AI is available under two licensing options:

1. Academic/Research License

  • Free for academic and research use under Creative Commons Attribution-NonCommercial 4.0 International License
  • Allows sharing and adaptation of the material for non-commercial purposes
  • Requires appropriate attribution
  • See LICENSE file for full terms

2. Commercial License

Note: The software includes third-party components under various open-source licenses (LGPL, BSD) - see the extra_libs directory for individual component licenses.