A calculation method for semi automatic follow up of multiple sclerosis by magnetic resonance eco planar perfusion imaging

Abstract

Multiple sclerosis (MS) is one of the most common chronic and disabling inflammatory and demyelinating disorders of the central nervous system (CNS). Magnetic Resonance Imaging (MRI) allows the observation of pathological changes in vivo. It has provided a number of important insights into the spatial-temporal evolution of MS pathology in vivo. Conventional MRI with T2-weighted images is useful in the assessment of oedema early in the inflammatory stage, tissue destruction with demyelination and axonal loss, and gliosis later in the chronic stage. Examination by conventional MRI usually requires more than one hour and it does not completely reveal rapid dynamic changes in blood flow. Recently introduced rapid MRI techniques, MR perfusion imaging and MR diffusion imaging, allow measurement of pathophysiological changes at the cellular level with a good temporal resolution. Nevertheless, until now there were no adequate analytical methods available within the clinical routine to differentiate between types of MS lesions using fast perfusion MRI. We present an analytical method capable of recognizing, and distinguishing, the status of MS lesions by calculating some numerical parameters from the MR perfusion images. The method has been tested on 14 patients affected by MS with different lesions and the results have been compared with those obtained with more expensive conventional MRI examinations. The proposed method made it possible to recognize the nature of the MS lesions in 100% of the examined cases without the need to perform long conventional MRI examinations, using instead perfusion MRI eco planar imaging which takes no more than two minutes. Moreover, the reduced time also allowed reduction of the quantity of contrast medium administered to the patient. Further studies may lead to the use of this technique for differential diagnoses in other white matter diseases.