BASIL perfusion

BASIL: Bayesian Inference for Arterial Spin Labeling MRI

Arterial Spin Labeling (ASL) MRI is a non-invasive method for the quantification of perfusion. Analysis of ASL data typically requires the inversion of a kinetic model of label inflow along with a separate calculation of the equilibrium magnetization of arterial blood. The BASIL toolbox provides a means to do this based on Bayeisan inference principles. The method was orginally developed for multi delay (inversion time) data where it can be used to greatest effect, but is also sufficiently fleixble to deal with the widely used single delay form of acquisition.

If you have resting ASL data with only a single inversion time then you can still use the tools in BASIL for perfusion quantification. If you want to perform analysis of a functional experiment with ASL data, i.e. one where you want to use a GLM, then you should consult the perfusion section of FEAT, or if you have dual-echo (combined BOLD and ASL) data then consult FABBER.

For single delay ASL data kinetic model inversion is realtively trivial and solutions to the standard model have been described int he literature. However, it is becoming increasingly common to aquire ASL data at multiple times post-inversion and fit the resultant data to a kinetic curve model. This permits problems in perfusion estimation associated with variable bolus arrival time to be avoided, since this becomes a paramter of the model whose value is determined from the data. Commonly the model fitting will be performed with a least squares technique providing parameter estimates, e.g. perfusion and bolus arrival time. In contrast to this BASIL uses a (fast) Bayesian inference method for the model inversion, this provides a number of advantages:

While the first two apply specfically to the case of mulitple delay data, the latter are also applicable to single delay ASL and are only available using the Bayesian technique employed by BASIL.

Latest version of BASIL now released!

The latest version of the BASIL tools are now included in FSL 6.0.1 (and higher) and we strongly recommend upgrading to this if you have been using an older version.

The new version includes an improved GUI, along with a range of other improvements.

BASIL perfusion


Fully updated documentation can be found via the link below.

Examples & Tutorials

An extensive set of examples of the use of BASIL (for pcASL) is available as part of the ASL primer. They are freely available via the website, you can link to them directly here.

Introduction to Perfusion Quantification using Arterial Spin Labelling, Oxford Neuroimaging Primers, Chappell, MacIntosh & Okell, Oxford University Press, 2017.

Further examples (for other tools within BASIL and more advanced command line usage) can be found via the documentation.

The BASIL toolset

BASIL is supplemented by a collection of tools that aid in the creation of quantitative CBF images from ASL data, you should either use one of the higher-level analysis packages or if you want a more specific analysis select the appropriate tool(s) for the data you have.

Higher-level packages, these provide a single means to quantify CBF from ASL data, including kinetic-model inversion, absolute quantification via a calibration image and registration of the data. This will generally be the first place to go for most people who want to do processing of ASL data.

The BASIL toolset:


If you employ BASIL in your research please reference the article below, plus any others that specifically relate to the analysis you have performed:

Chappell MA, Groves AR, Whitcher B, Woolrich MW. Variational Bayesian inference for a non-linear forward model. IEEE Transactions on Signal Processing 57(1):223-236, 2009.

If you employ spatial priors you should ideally reference this article too.

A.R. Groves, M.A. Chappell, M.W. Woolrich, Combined Spatial and Non-Spatial Prior for Inference on MRI Time-Series , NeuroImage 45(3) 795-809, 2009.

If you fit the macrovascular (arterial) contribution you should reference this article too.

Chappell MA, MacIntosh BJ, Donahue MJ, Gunther M, Jezzard P, Woolrich MW. Separation of Intravascular Signal in Multi-Inversion Time Arterial Spin Labelling MRI. Magn Reson Med 63(5):1357-1365, 2010.

If you employ the partial volume correction method then you should reference this article too.

Chappell MA, MacIntosh BJ, Donahue MJ,Jezzard P, Woolrich MW. Partial volume correction of multiple inversion time arterial spin labeling MRI data, Magn Reson Med, 65(4):1173-1183, 2011.

If you perform model-based analysis of QUASAR ASL data then you should reference this article too.

Chappell, M. A., Woolrich, M. W., Petersen, E. T., Golay, X., & Payne, S. J. (2012). Comparing model-based and model-free analysis methods for QUASAR arterial spin labeling perfusion quantification. doi:10.1002/mrm.24372

CategoryPerfusion CategoryBASIL


BASIL (last edited 15:45:03 12-03-2020 by MichaelChappell)