• Mean Diffusivity (MD) (in 10-9 m2/s) [ Time Frame: Up to 72 hours ]
  Computation of Mean Diffusivity (MD) metrics (in 10-9 m2/s) including MDlow for b=1000s/mm2 and pseudo-MDhigh for maps calculated from the highest b-values within the rescued penumbra and within normal contralateral symmetric region of the brain
• Mean Diffusivity (MD) (in 10-9 m2/s) [ Time Frame: 3 month of follow-up. ]
  Computation of Mean Diffusivity (MD) metrics (in 10-9 m2/s) including MDlow for b=1000s/mm2 and pseudo-MDhigh for maps calculated from the highest b-values within the rescued penumbra and within normal contralateral symmetric region of the brain

• Mean Kurtosis (MK) [ Time Frame: Up to 72 hours ]
  Average of the diffusional kurtosis along all diffusion directions
• Mean Kurtosis (MK) [ Time Frame: 3 month of follow-up ]
  Average of the diffusional kurtosis along all diffusion directions
• Mean Diffusivity of the slow component (MDs) [ Time Frame: Up to 72 hours ]
  Mean Diffusivity of the slow component (MDs) within the rescued penumbra and within normal contralateral symmetric region
• Mean Diffusivity of the slow component (MDs) [ Time Frame: 3 month of follow-up ]
  Mean Diffusivity of the slow component (MDs) within the rescued penumbra and within normal contralateral symmetric region
• Magnetization Transfert Ratio [ Time Frame: Up to 72 hours ]
  Magnetization Transfert Ratio maps (MTR=(0-Ms)/M0 x100 where M0 and Ms represent the signal intensity with the saturation prepulse off and on, respectively) within the rescued penumbra and within normal contralateral symmetric region of the brain
• Magnetization Transfert Ratio [ Time Frame: 3 month of follow-up ]
  Magnetization Transfert Ratio maps (MTR=(0-Ms)/M0 x100 where M0 and Ms represent the signal intensity with the saturation prepulse off and on, respectively) within the rescued penumbra and within normal contralateral symmetric region of the brain

Thrombectomy has significantly improved the outcome of stroke patients. However, even after successful recanalization residual handicap including post-stroke cognitive and mood disorders impact the quality of life of patients. These symptoms correlate only moderately with the final stroke volume suggesting more widespread dysfunction than what is apparent on standard follow-up MRI. One hypothesis is that the rescued penumbra (i.e.; the tissue showing significant hypoperfusion at the acute stage but that appears normal on conventional imaging at follow up) could exhibit incomplete ischemic injury also known under the term of selective neuronal loss. The concept of selective neuronal loss in the rescued penumbra is admitted based on histological, animal and PET studies but the identification of such a graduation between pan-necrosis and normal tissue is very challenging to capture in vivo with MRI and is typically missed.

This study will prospectively include 100 stroke patients. Patients admitted with significant penumbra at the acute stage and successfully recanalized thanks to thrombectomy thrombolysis or spontaneously will be explored at 24h-to-72h and then at 3 months with the multi-b diffusion sequence. The investigators expect to be able to measure significant modifications within the rescued penumbra as compared to contralateral normal brain by using the advanced but also the simplified diffusion metrics. Investigators will test the impact of duration/severity of the initial hypoperfusion and they will explore the clinical relevance especially in terms of cognitive and mood disorders as measured at 3 months.

Inclusion Criteria:

• Patients older than 18 years old
• Cerebral infarct or cerebral hypoperfusion with neurological symptoms within the anterior circulation
• Due to occlusion or stenosis of intracranial carotid artery or its branches and/or middle cerebral artery (MCA) and/or anterior cerebral artery.
• With acute MRI performed within the first 24h and showing significant penumbra defined as ratio between the volumes of critically hypoperfused tissue (defined by Tmax>6s) and the ischemic core (defined by ADC<600 × 10-6 mm2/s) of 1.8 or more, with an absolute difference of 15 mL or more and ischemic core volume of less than 70 mL.
• Successful recanalization by thrombectomy (TICI 2b or 2c or 3) and/or by IV thrombolysis (if within the first 4.5h) or spontaneously.
• Patient/Legally Authorized Representative has signed the Informed Consent form.

Exclusion Criteria:

• History of symptomatic cerebral infarct with functional deficit (pre-stroke modified Rankin Scale score ≥1) to measure the impact of the infarct and selective neuronal loss on long-term outcome without being biased by pre-existing deficits.
• History of severe cognitive impairment (dementia) or DSMIV axis 1 psychiatric disorders that would confound the neurological evaluations.
• Pregnant or breast-feeding women.
• Contraindications to MRI.