Data acquisition

All AIS patients who enrolled in the SALVO study underwent a research protocol MRI, including structural, diffusion and functional imaging, during their hospital admission at 2-5 days from the stroke symptom onset. A T1-weighted image was acquired with the following parameters: in-plane resolution, 0.430 mm; slice thickness, 6mm; matrix size, \(480 \times 512\); number of slices, 28. Gradient-echo echoplanar imaging (EPI) data depicting blood oxygen level-dependent contrast at rest were also acquired at 3.0T in Massachusetts General Hospital (Boston, USA). The resting-state fMRI (rfMRI) data consisted of 150 volumes with the following parameters: number of slices, 42 (interleaved); slice thickness, 3.51 mm; matrix size, \(64 \times 64\); flip angle, 90 degrees; repetition time (TR), 2400 ms; in-plane resolution, 3.437 mm.  In the majority of subjects, DWI was performed using a 3T (Siemens Skyra) scanner with the following parameters: numbers of slices, 160; slice thickness, 5mm; TR, 5500 ms; TE, 99ms; in-plane resolution, 1.375mm.

Image preprocessing

Both structural and functional images were preprocessed using the Configurable Pipeline for the Analysis of Connectomes (CPAC) \cite{Cameron_2013}. First, bias field correction of the anatomical images was performed \cite{Tustison_2010}, followed by brain extraction using an in-house tool that employs a convolutional neural network (CNN) and operates at a patch level. Subsequently, the images were registered to the MNI (Montreal Neurological Institute) anatomical template using non-linear registration (ANTs) \cite{Avants_2011}. Probability maps for grey matter, white matter and cerebrospinal fluid (CSF) were generated using FSL FAST \cite{Woolrich_2009}.