The methodological aspects implemented during this Ph.D. thesis consisted of the optimization of the acquisitions and data processing of ASL imaging for quantitative assessment of cerebral perfusion and vasoreactivity. First of all, an original technique called BoTuS (Bolus Turbo Sampling) was implemented and validated,with the aim to render the quantification of the pulsed ASL signal more robust. Cerebral blood flow measurements obtained using pulsed ASL were compared to gold standard techniques such as the first passage of gadolinium MRI and CT-scan perfusion in a population of patients with treated brain tumors. Secondly, a new processing technique was tested, taking into account the physiological state of the subject during the exam to model the ASL signal during the vasoreactivity paradigm, and thus to provide more reliable maps at the subject level. Finally, these methods were applied in various studies on healthy subjects and patients. A decrease in vasoreactivity was found in Alzheimer disease patients compared to elderly subjects. Studies on patients with severe stenosis were conducted to test our methods at the subject level. An increase in CBF and a decrease in vasoreactivity in subjects exposed to high altitude at 4365 m during 7 days was demonstrated and correlated to transcranial Doppler results.
