WP1: Neuroimaging protocol
Scientific Background
In vivo qMRI imaging
Prof. Dr. Nikolaus Weiskopf, Director of the Department of Neurophysics at the MPI CBS, Leipzig, Germany, specializes in innovative non-invasive MRI-based in vivo histology . He collaborates with Dr Evgeniya Kirilina, who heads the MRI Biophysics Research Group developing advanced MRI techniques for measuring iron in the brain. Their approach integrates state-of-the-art MRI acquisition methods with biophysical models to extract information about tissue microstructure from macroscopic MR parameters. Weiskopf's group developed a method that captures intrinsically co-localized multiple quantitative parameter maps (MPM), including measures of tissue iron, myelin, and tissue water content and released an open-access toolbox for qMRI data analysis (hMRI). This method has been successfully applied in large aging cohorts and in multicenter clinical trials (NISCI) and in developing a framework for tracking neurodegenerative disease with cortical layer specificity. MPM at 7T (IronSleep) with sub-millimeter resolution (600 μm and 500 μm isotropic), combined with accelerated T2 mapping, diffusion-weighted imaging and neuromelanin-sensitive MRI, provides quantitative markers of the integrity of small nuclei (i.e. substantia nigra [SN] and locus coeruleus [LC]), and subtle changes in microstructure throughout the brain (iron, myelin, water content and neuronal fibers). The team has developed advanced biophysical models linking MRI parameters to cellular iron distribution in the SN, improving MRI specificity and proposing a unique quantitative biomarker for dopaminergic cell density in nigrosome 1 of the SN. By combining ultra-high resolution in vivo MPMs with multimodal 3D histology and biophysical modeling, they demonstrated together with Dr Anneke Alkemade’s team (WP2) that widely used radiological markers of nigrosome 1 are inaccurate and need to be reinterpreted.
Working plan and methodology
In vivo histology of brainstem nuclei using ultra-high resolution 7T qMRI
This WP will focus on the extension of the advanced neuroimaging methodology for ultra-high resolution histological MRI in early PD. It will build upon the previously established advanced neuroimaging protocol and advanced image analysis tools for histological qMRI developed within the IronSleep project. The protocol includes whole brain ultra-high resolution qMRI measurements at 3 and 7T combined with tailored quantitative histological MRI biomarkers of nigral neuronal integrity of nigrosomes in the SN. We will extend the developed methodology in two ways: (i) by a clinical translation of ultra-high resolution qMRI protocols, and (i) by a clinical translation of ultra-high resolution qMRI protocols, and (ii) by enriching it with quantitative histological MRI biomarkers of integrity of the LC and hypothalamic nuclei and by (iii) facilitating integration of an additional neuroimaging site and a second longitudinal timepoint, while ensuring highest level of data quality and comparability between dataset and time points.
The first clinical translation goal will be achieved by implementing advanced image acceleration and image reconstruction techniques to speed up the imaging protocol (~10x acceleration) and by translating the protocol to broadly clinically available 3T MRI. We will optimize the recently developed method for low-rank modeling of localk-space neighborhoods (LORAKS)50 for accelerated, high quality MRI and adapt it to the particular need of quantitative MRI by directly optimizing the quality of the quantitative maps and microstructural biomarkers rather than the weighted images as typically done. The performance of the imaging protocol will be tested in combination with advanced image analysis tools integrating advanced strategies for image reconstruction, image de-noising, and multimodal segmentation of small nuclei. The new developments can be applied to previously acquired datasets, since k-space raw data was stored as part of IronSleep. In addition, we will systematicallycompare the performance of optimized accelerated MPM protocol with recently developed variations of multiparametric qMRI including EPTI51, 3D segmented echo52 and universal RF pulse53 enabled acquisitions. The new protocols will be quantitatively compared and the difference in their sensitivity and specificity will be quantified in a traveling head study (as already initiated). We will leverage open API and software solutions (e.g. Siemens
open recon) for seamless integration on the scanner platforms. Importantly, these accelerated protocols will be deployed to translate the imaging capabilities and quantitative biomarkers tested in WP3, 4, and 5 towards clinical application. The additional microstructural biomarkers goal of WP1 will be achieved by developing a generative biophysical model of MRI contrast for the LC and small hypothalamic nuclei. Following the methodology developed for nigrosome imaging we will combine generative biophysical models of the MRI contrast with precise anatomical atlas. Informed by 3D histology acquired in WP2, which will include semi quantitative maps of iron and myelin, we will develop new empirical and generative models of MRI contrast in LC and hypothalamic nuclei. This model will be validated using quantitative MRI microscopy obtained on the same tissue blocks used and analyzed histologically in WP2 to link quantitative MRI parameters with tissue microstructure.
Finally for the third goal of WP1, we will extend the quality control procedure established for IronSleep to an additional 7T imaging site at ICM (WP5) and additional longitudinal time points, to assess reproducibility across sites, field strengths and time points. To this end, we will extend the traveling heads study by an additional time point and additional site (i.e. partner 5). We will compare the test-retest and inter-site comparability as well as establish a quantitative link to the protocol employed in the ICEBERG study. This will allow us to integrate the entire cohort dataset for computational modeling in WP6.
Team – MPI CBS
Professor Dr Nikolaus Weiskopf
Director