Synthetic brain MRI dataset for testing of cortical thickness estimation methods

This dataset is suitable for the evaluation of cortical thickness estimation methods. The synthetic dataset consists of 400 T1-weighted (T1-w) brain MRI scans, corresponding Partial Volume (PV) maps of three tissue classes, grey matter (GM), white matter (WM) and cerebrospinal fluid (CSF), as well as surface meshes - pial and WM. For the dataset synthesis, 20 healthy control (HC) baseline subjects from the ADNI dataset, ten females and ten males, were used for the generation of 20 different atrophy levels per subject uniformly. The atrophy levels fall into two ranges, [0.01,0.1] mm, with the atrophy step rate of 0.01 mm [0.1,1] mm (ten levels), with the atrophy step rate of 0.1mm (ten levels). The baseline subjects were also reconstructed using the same method to minimise the measurement bias when computing cortical thickness.