Skip to content

Latest commit

 

History

History
39 lines (25 loc) · 2.97 KB

README.md

File metadata and controls

39 lines (25 loc) · 2.97 KB
Raw

Localization3D module (luenn): A PyTorch-Based Package for 3D Single Molecule Localization Microscopy

Luenn is a powerful Python package built on PyTorch, designed for 3D Single Molecule Localization Microscopy (SMLM). It offers a comprehensive set of functionalities for data generation, sampling, model training, post-processing, 3D localization, and rendering. Leveraging deep learning techniques, Luenn achieves high accuracy across various imaging modalities and conditions.

Key Features

  • Data Generation: Luenn provides tools for generating synthetic data, enabling users to simulate diverse imaging scenarios for training and evaluation.

  • Sampling: The package includes sampling utilities to efficiently extract training and validation data from large datasets, optimizing the training process.

  • Model Training: Luenn utilizes a Deep Convolutional Neural Network to detect and localize emitters at sub-pixel resolution. Training is customizable, allowing users to adapt the model to specific experimental conditions.

  • Post-Processing Functions: Luenn offers post-processing functions to enhance and refine localization results, ensuring superior super-resolution reconstructions.

  • 3D Localization: Luenn specializes in 3D localization, enabling precise positioning of emitters in three-dimensional space, a crucial aspect in single molecule localization microscopy.

  • Rendering: The package facilitates the rendering of super-resolved images, providing visualization tools for the analyzed data. example of 3D reconstruction and luenn rendering tool for a live cell time-series image set

circle_LC13.mp4

Performance

Luenn has demonstrated exceptional accuracy across a broad spectrum of imaging conditions. Its ability to handle live-cell SMLM data with reduced light exposure in just 3 seconds makes it a valuable asset for dynamic imaging scenarios.

Getting Started

System Requirements

The software was tested on a Linux system with Ubuntu version 7.0, and a Windows system with Windows 10 Home. Training and evaluation were run on a standard workstation equipped with 32 GB of memory, an Intel(R) Core(TM) i7 − 8700, 3.20 GHz CPU, and a NVidia GeForce Titan Xp GPU with 12 GB of video memory.

Installation

  1. Download this repository as a zip file (or clone it using git).
  2. Go to the downloaded directory and unzip it.
  3. The conda environment for this project is given in environment_.yml where should be substituted with your operating system. For example, to replicate the environment on a linux system use the command: conda env create -f environment_linux.yml from within the downloaded directory. This should take a couple of minutes.
  4. After activation of the environment using: conda activate LUENN, you're set to go!

Contributers:

Armin Abdehkakha, Email: [email protected]
Craig Snoeyink, Email: [email protected]