DOI10.5281/zenodo.8146426Zenodo8146426MaRDI QIDQ6700484FDOQ6700484
Dataset published at Zenodo repository.
Gerard McGuinness, Brendan Kelly, Aonghus Lawlor, Ronan Killeen, Henry Dillon, Prateek Mathur, Duncan Simpson
Publication date: 14 July 2023
Multiple Sclerosis (MS) is an idiopathic chronic inflammatory demyelinating disease of the central nervous system. Timely and accurate diagnosis via the McDonald criteria improves outcome but relies heavily on radiologist interpretation of MRI studies. A growing body of research is focused on identifying new MS lesions on MRI by comparing one time point with another but only limited longitudinal data are available. We present a dataset of which contains 496 scans of 172 patients with MS each with at least 2 time points including T1, Flair and T2 sequences. There are 110 stable scans and 214 instances of change. We also include demographic information. Data has been anonymised, processed and segmented with three stages of expert opinion contribute to ground truth. Future research using this dataset could include new lesion identification, radiomic characterisation of new lesions, relationship with disease activity and brain atrophy and time series applications including prediction of new lesions and active learning for efficient segmentation.
This page was built for dataset: Insight MS Longitudinal MRI dataset
