To date except for the larger striato-pallidal complex, there are no reliable imaging markers of small deep nuclei. Major improvement of the spatial resolution resulting from the use of ultra-high field MRI systems offers new perspectives of imaging of these deep structures.

We will use the new contrast mechanisms available in data acquired with ultra-high field MR systems (7T) as well as the most recent high angular diffusion imaging techniques in order to characterize the cytoarchitectonics of the deep brain structures and brainstem lesions in parkinsonian syndromes (with special interest in substantia nigra (SN) pathology and nigro-striatal fiber pathways; subthalamic nucleus (STN), red nucleus (RN), oculomotor structures (involved in PSP); pedonculo-pontine nucleus (involved in gait and posture control disorders) and the locus sub-coeruleus area (implicated in sleep disorders) The optimized MR sequences at 7T will be adapted and validated at 3T (on a more clinically oriented MRI approach).

The clinical goal of the project (via the characterization of deep brain structures) is the detection of new neuroimaging markers of neuronal lesions in PD. These biomarkers will be used to create a diagnostic tool at early stage of the disease that could be correlated to clinical signs such as gait disorders and help to identify predictive factors. In addition, this could contribute to establish an adequate therapeutic strategy (as for example with deep brain stimulation).

This project will be divided into several methodological steps. First, we will establish an imaging protocol using high field MR systems (7T) to deliver high resolution data. The new T1/T2-star contrasts mechanisms observed in high field data will make possible the segmentation of the central structures, more specifically those deeply located in the brainstem, the pons and the mesencephalon that are involved in Parkinsonian syndromes. Second, we will characterize the cytarchitectony of the deep structures and the anatomical connectivity involved in the Parkinsonian syndromes using high angular resolution diffusion imaging (HARDI). The local information about the organization of the tissue coming from the HARDI data will be used to subdivide the deep structures into smaller nuclei. We will also use the HARDI data to infer the connectivity of these structures, employing probabilistic tractography algorithms more efficient to detect the smallest bundles. Such techniques have never been used for Parkinsonian syndrome, and may highlight structural differences inside deep nuclei that could not be observed when performing statistics on the entire structure. Third, preclinical studies will be performed in order to tackle the best spatial and angular resolutions and to build an atlas of the brainstem that will help understanding its structural organization. Clinical acquisitions will be done to acquire two databases of healthy volunteers and patients suffering from Parkinsons disease and progressive supranuclear palsy. Fourth, we will develop a statistical tool for isolating the biomarkers of the Parkinsonian syndrome and to create a diagnosis tool of the pathology, relying on robust bootstrapping techniques. The markers will stem from comparisons between populations using the various informations available. Expected results In this project, we will develop ultra high field techniques that will allow us to investigate the characteristics of the brainstem structures and of the central deep nuclei in normal subjects and in parkinsonian syndromes. The findings obtained with these image innovation techniques will help to overcome the present technical difficulties and limits of 1.5 and 3T MRI as some of the methods will be transferred to the routine machines and results will be compared to the images obtained on 3T MRI, in order to make new diagnosis tools available, in clinical research or clinical practice. The ultimate goal is to improve the accuracy of diagnosis of various parkinsonian syndromes at a early stage, to detect at a pre-clinical stage abnormalities that may reflect the underlying anatomical lesions, to improve the accuracy of neurosurgical targeting for deep brain stimulation, to detect imaging markers that may be useful biomarkers of the progression of the disease in neuroprotective therapy.

Inclusion Criteria:

• PD according to the UK brain bank diagnosis criteria
• PSP according to the NINDS diagnosis criteria

Exclusion Criteria:

• contra indication to MRI

parkinson's disease (PD) supranuclear palsy (PSP) controls