Premium
P4‐068: Cortical thickness estimation accuracy with three automated algorithms
Author(s) -
Redolfi Alberto,
Redolfi Alberto,
Borsci Genoveffa,
Revillard Jerome,
Manset David,
Grenier Baptiste,
Frisoni Giovanni
Publication year - 2011
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2011.05.2088
Subject(s) - computer science , artificial intelligence , pattern recognition (psychology) , computer vision , algorithm
Background: Beta-amyloid plaques and tau-containing neurofibrillary tangles are recognized biomarkers of Alzheimer’s disease (AD) onset and progression, yet hemodynamic and/or metabolic modulations that may precede such changes are currently debated. Importantly, detecting and evaluating early-stage pathogenesis is critical to characterizing disease etiology and guiding treatments intended to prevent irreversible tissue damage. Magnetic resonance imaging (MRI) poses potential for investigating pathological biomarkers owing to its variety of noninvasive contrast mechanisms. Therefore, we have initiated a collaborative effort between imaging physicists, neuropsychologists, and geneticists with the aim of identifying hemodynamic, neurochemical, and metabolic biomarkers in preclinical AD using novel MRI approaches at intermediate (3T) and high (7T) magnetic field. Methods: This study has been divided into three stages with distinct benchmarks for success: (i) to optimize high-field protocols for at-risk (APOE-e4 carrier or family history) AD populations, (ii) to measure neurochemical and hemodynamic parameters in preclinical at-risk populations (APOE-e4 carrier or family history), and (iii) to evaluate imaging biomarkers in patients with clinical AD. In addition to structural scans, novel approaches with AD-relevant contrasts have been implemented: (1) arterial spin labeling (cerebral blood flow), (2) vascular-space-occupancy (cerebral blood volume), (3) T1-? (beta-amyloid), (4) chemical-exchange-saturation-transfer (neurochemicals), (5) spontaneous blood-oxygenation-level-dependent (functional-connectivity), (6) susceptibility weighting (microbleeds) and (7) T2-relaxation-under-spin-tagging (oxygen extraction fraction). The imaging choices were motivated by (i) uniqueness, i.e. approaches that have not been robustly tested in AD and (ii) approaches that generate specific contrasts for parameters hypothesized to be implicated in early-stage AD. Results: This study represents an active trial with a target enrollment of 10 patients per month. We have implemented the above approaches, performed quality control measurements in healthy populations and are actively enrolling at-risk volunteers with mild cognitive impairment. Fig. 1 shows representative images from the AD protocol; pilot data from at-risk subjects, in the context of healthy volunteers, will be discussed. Conclusions: A novel high-field MRI protocol has been implemented for early detection of hemodynamic, neurochemical and metabolic changes that may precede structural or clinical changes in AD. This ongoing work is a multi-departmental effort and is expected to provide mechanistic clues regarding early-stage functional changes in AD.