Researchers from the University of Southern California (USC) have archived and shared an open-source data set of brain MRI from stroke patients known as the Anatomical Tracings of Lesion After Stroke (ATLAS).
Published online Feb. 20 in Scientific Data, the information is available to researchers who want to develop algorithms to process MRI images from stroke patients.
“This large, diverse dataset can be used to train and test lesion segmentation algorithms and provides a standardized dataset for comparing the performance of different segmentation methods,” wrote first author Sook-Lei Liew, PhD, an assistant professor at USC, and colleagues. “We hope ATLAS release 1.1 will be a useful resource to assess and improve the accuracy of current lesion segmentation methods.”
Liew and associates hope to automate the segmentation process to allow for examination of more images.
“We can't do it by hand at the scale of thousands, so we are really interested in helping find better automated ways, using machine learning and computer vision, to identify the lesions and have machines draw those boundaries,” said Liew et al.
The full announcement is available for free.