Automated detection system for acute ischemic stroke

An automated detection system for acute ischemic stroke comprising:

• a preprocessor configured to perform registration on a whole-brain image and a standard-brain spatial template to extract individual brain region masks from the whole-brain image, wherein the individual brain region masks form a three-dimensional (3D) brain region mask;
• a deep learning encoder coupled to the preprocessor and configured to perform feature extraction on the whole-brain image and the individual brain region masks, thereby converting the whole-brain image into a plurality of two-dimensional (2D) whole-brain slice images and identifying bounding boxes of brain regions of each of the individual brain masks;
• a first processor coupled to the deep learning encoder and configured to map the individual brain masks onto the whole-brain slice images for registration, wherein the first processor is configured to divide the plurality of two-dimensional whole-brain slice images into sets of brain region slice images based on the bounding boxes of the brain regions, and each of the sets of brain region slice images comprises slice images of a single brain region;
• a second processor coupled to the first processor and configured to compute stroke-related weight values of the slice images of each of the sets of brain region slice images, wherein the second processor is configured to sum the weight values to obtain a characteristic value of each of the brain regions;
• and
• a disparity-aware classifier coupled to the second processor and configured to determine whether any one of the brain regions has acute ischemic stroke according to the characteristic value of each of the brain regions.