Skip to main content

A tool for image registration.

Project description

ARDENT

Affine and Regularized DEformative Numeric Transform (ARDENT) is a Python package for performing automated image registration using LDDMM.

ARDENT stands out for its ability to predict and correct for artifacts and image nonuniformity, perform registrations across image modalities, ease of use, and other features in development.

Overview

Experimental neuroscience produces a stunning amount of imaging data from light or electron microscopy, MRI, and other 3D modalities. To be of real use these datasets must be interpreted with respect to each other and to refined standards: well-characterized image datasets called atlases. To build these interpretations, dense spatial alignments must be computed. This process is known as image registration, in which one image is optimally deformed, or flowed, until it aligns with another. Accurate registration is challenged by the large scale of imaging data and the heterogeneity across species scales and modalities. Current tools can perform well on very standard images but perform poorly on data with various imperfections. This restricts our ability to analyze data from novel experiments performed in a majority of labs.

ARDENT is an accessible pure-python image registration package in development with these neuroimaging challenges in mind.

Documentation

The official documentation with usage is at https://ardent.neurodata.io/

Please visit the tutorial section in the official website for more in-depth usage.

System requirements

Hardware requirements

ARDENT package requires only a standard computer with enough RAM to support the in-memory operations.

Python Requirements

This package is written for Python3. Currently, it is supported for Python 3.6.

Python Dependencies

ARDENT mainly depends on the Python scientific stack with the notable addition of PyTorch. However, this is pending deprecation.

numpy
matplotlib
scipy
scikit-learn
simpleitk
nibabel
nilearn
pytorch

Installation Guide

Install from pip

pip install ardent

Install from Github

git clone https://github.com/neurodata/ardent
cd ardent
python3 setup.py install

License

This project is covered under the Apache 2.0 License.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

ardent-0.0.1.tar.gz (37.1 kB view hashes)

Uploaded Source

Built Distribution

ardent-0.0.1-py3-none-any.whl (47.0 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page