Dynamic MR images reconstruction

Description

A new and very interesting application of Magnetic Resonance Imaging (MRI) is the so called Dynamic Magnetic Resonance. In or volume (3D) of the human body are acquired at successive times and  they are used to examine the functionality of a particular organ with a contrast agent in preoperative contexts or in postprocessing studies of functional MRI (fMRI).

In both cases, it is necessary to acquire the data as fast as possible; for this reason, not all the MR impulses are registered for each image of the sequence. In the commercial systems a method, called keyhole, is usually implemented for the reconstruction of the images fromundersampled data, but it produces artifacts.
The project is concerned with the development of more efficient algorithms for the reconstruction of dynamic MR images from undersampled data. We provide software tested on real data obtained by our collaborators.

Main results

We have developed reconstruction algorithms based on the generalized series model proposed by Liang and Lauterbur (IEEE Trans. Med. Imag.,v.13,4). We use different basis functions for representing the image to be reconstructed, such as exponential functions (as in the original model) and B-spline functions, that are very efficient for this purpose.

We have produced a Matlab based software with graphical interface, MRITOOL v.1.0, for testing the methods on test problems and possibly on new data, if the raw data are available.

Funding

·MURST Project:"Analisi Numerica: Metodi e Software Matematico" (1997-1999) 

Publications

·E. Loli Piccolomini ,F. Zama, G. Zanghirati , A.R. Formiconi, S. Martini, MRITool: a Matlab tool for functional magnetic resonance imaging reconstruction, Technical report of the MURST national Project "Analisi Numerica: metodi e software matematico", 2000.

·A.R. Formiconi, E. Loli Piccolomini ,S. Martini,  F. Zama and G. Zanghirati , Numerical methods and software for functional magnetic resonance images reconstruction, Annali dell'Università di Ferrara, sez. VII Scienze Matematiche, suppl. vol. XLVI, Ferrara  , 2000

·A. R. Formiconi,.A.  Passeri, S.   Martini , A.  Pupi , E.  Loli Piccolomini, F.. Zama, G. Zanghirati ,  Regularization methods in dynamic MRI reconstruction, Eur. Radiology, suppl.1, pp. s121--s122, 1999.

·Formiconi A.R. Passeri A. Martini S. Pupi A. Loli Piccolomini E. Zama F. Zanghirati G., Regularization methods for the quantification of regional cerebral blood flow with MRI, European  Journal Nuclear Medicine , v. 25, pp. 1154, 1998.

Keywords