According to the intracranial
aneurysm recognition and detection method, device and
system and the computer readable storage medium, the intracranial
aneurysm recognition and detection method comprises the steps of receiving and labeling original intracranial CTA images sequentially, and obtaining labeled image data; performing boundary pixel points of the
aneurysm in the three-dimensional space through the marked image data, distinguishing a
positive sample image block and a
negative sample image block of the image data according to the boundary pixel points, wherein the
positive sample image block is the area where the aneurysm is located, and the
negative sample image block is the area where the non-aneurysm is located; and detecting the intracranial aneurysm in the positive and
negative sample image blocks by using a
deep learning model, and outputting a detection result. According to the invention, positive and negative
sample image blocks are processed at the same time by using algorithms of an
encoder and a decoder in a
deep learning model, so that the small aneurysm
detection rate is increased; a long-range jump connection
algorithm introduced by a
deep learning model is taken as a main part, and a short-range jump connection
algorithm introduced by the long-range jump connection
algorithm is taken as an auxiliary part, so that the problem of poor continuity between different
layers is solved.