abstract |
The present invention relates to a system and a method for estimating an ASPECT score, which is a factor for identifying a stroke disease from a brain CT image, the method comprising: a pre-processing step of normalizing and standardizing features in an image data set; a separation step of separating each lesion from CT images classified into a supra-ganglion level and a ganglion level; and a determination step of independently constructing a neural network for learning positive/negative images by each lesion, so as to determine whether a lesion indicates a stroke, wherein an ASPECT score, which is an objective index for determining the state of a stroke patient, can be estimated by using a brain CT image of the patient. |