============================ Advanced Normalization Tools

ANTs computes high-dimensional mappings to capture the statistics of brain structure and function. See the FAQ page.

ANTs allows one to organize, visualize and statistically explore large biomedical image sets.

ANTs integrates imaging modalities and related information in space and time.

ANTs works across species or organ systems with minimal customization.

ANTs and related tools have won several international and unbiased competitions.

ANTsR is the underlying statistical workhorse.

Questions: Discussion Site or new ANTsDoc or try this version ... also read our guide to evaluation strategies and addressing new problems with ANTs or other software. New ANTs handout, part of forthcoming ANTs tutorial material.

ANTsTalk - subject to change at any moment

ANTsRegistrationTalk - subject to change at any moment

Install ANTs via pre-built: Packages @ github older versions @ sourceforge ... also, Github Releases are here thanks to Arman Eshaghi.

Build ANTs from: Source-Code (recommended)

ANTs Dashboard thanks to Arman Eshaghi and Hans J. Johnson

ANTs extracts information from complex datasets that include imaging (Word Cloud). Paired with ANTsR (answer), ANTs is useful for managing, interpreting and visualizing multidimensional data. ANTs is popularly considered a state-of-the-art medical image registration and segmentation toolkit. ANTsR is an emerging tool supporting standardized multimodality image analysis. ANTs depends on the Insight ToolKit (ITK), a widely used medical image processing library to which ANTs developers contribute. A summary of some ANTs findings and tutorial material (most of which is on this page) is here.

Authors

Brian B. Avants - UPENN

Role: Creator, Algorithm Design, Implementation, more

Nicholas J. Tustison - UVA

Role: Compeller, Algorithm Design, Implementation Guru, more

Hans J. Johnson - UIowa

Role: Large-Scale Application, Testing, Software design

Team Members

Core: Gang Song (Originator), Philip A. Cook, Jeffrey T. Duda (DTI), Ben M. Kandel (Perfusion, multivariate analysis)

Image Registration

Diffeomorphisms: SyN, Independent Evaluation: Klein, Murphy, Template Construction (2004)(2010), Similarity Metrics, Multivariate registration, Multiple modality analysis and statistical bias

Image Segmentation

Atropos Multivar-EM Segmentation (link), Multi-atlas methods (link), Bias Correction (link), DiReCT cortical thickness (link), DiReCT in chimpanzees

Multivariate Analysis Eigenanatomy (1) (2)

Prior-Based Eigenanatomy (in prep), Sparse CCA (1), (2), Sparse Regression (link)

ImageMath Useful!

morphology, GetLargestComponent, CCA, FillHoles ... much more!

Application Domains

Frontotemporal degeneration PENN FTD center

Multimodality Neuroimaging

Structural MRI
Functional MRI
Network Analysis

Lung Imaging

Structure
Perfusion MRI
Branching

Multiple sclerosis (lesion filling) example

Background & Theory

The SyN and N4 bias correction papers and other relevant references in Pubmed
Visualization: e.g. a gource of ANTs development
DiReCT cortical thickness papers
A folder of relevant docs: segmentation, registration, usage(old), for clinical apps
ANTs redesigned for generality, automation, multi-core computation with ITKv4
Dev'd ITKv4 with Kitware, GE, Natl. Lib of Medicine & Academia

ANTs has won several unbiased & international competitions

ANTs finished in 1st rank in Klein 2009 intl. brain mapping competition
ANTs finished 1st overall in EMPIRE10 intl. lung mapping competition
ANTs is the standard registration for MICCAI-2013 segmentation competitions
Conducting ANTs-based R tutorial @ MICCAI-2013
ITK-focused Frontiers in Neuroinformatics research topic here
Won the BRATS 2013 challenge with ANTsR
Won the best paper award at the STACOM 2014 challenge

Learning about ANTs

antsRegistration bash example

Eigenanatomy for multivariate neuroimage analysis via PCA & CCA

ANTs and ITK paper

Pre-built ANTs templates with spatial priors download

The ANTs Cortical Thickness Pipeline example

"Cooking" tissue priors for templates example (after you build your template)

Basic Brain Mapping example

Large deformation example

Template construction example

Automobile example

Asymmetry example

Point-set mapping which includes the PSE metric and affine and deformable registration with (labeled) pointsets or iterative closest point

Feature matching example ... not up to date ...

Chimpanzee cortical thickness example

Global optimization example

Morphing example

fMRI or Motion Correction example

fMRI reproducibility example

fMRI prediction example ... WIP ...

Cardiac example

Brain extraction example

N4 bias correction <-> segmentation example

Cortical thickness example

Multi-atlas joint label/intensity fusion examples example 1 example 2 (thanks to @chsasank)

Lung and lobe estimation example

Bibliography bibtex of ANTs-related papers

ANTs google scholar page

Presentations: e.g. a Prezi about ANTs (WIP)

Reproducible science as a teaching tool: e.g. compilable ANTs tutorial (WIP)

References

Google Scholar

Pubmed

Boilerplate ANTs

Here is some boilerplate regarding ants image processing:

We will analyze multiple modality neuroimaging data with Advanced Normalization Tools (ANTs) version >= 2.1 [1] (http://stnava.github.io/ANTs/). ANTs has proven performance in lifespan analyses of brain morphology [1] and function [2] in both adult [1] and pediatric brain data [2,5,6] including infants [7]. ANTs employs both probabilistic tissue segmentation (via Atropos [3]) and machine learning methods based on expert labeled data (via joint label fusion [4]) in order to maximize reliability and consistency of multiple modality image segmentation. These methods allow detailed extraction of critical image-based biomarkers such as volumes (e.g. hippocampus and amygdala), cortical thickness and area and connectivity metrics derived from structural white matter [13] or functional connectivity [12]. Critically, all ANTs components are capable of leveraging multivariate image features as well as expert knowledge in order to learn the best segmentation strategy available for each individual image [3,4]. This flexibility in segmentation and the underlying high-performance normalization methods have been validated by winning several internationally recognized medical image processing challenges conducted within the premier conferences within the field and published in several accompanying articles [8][9][10][11].