Fractional flow reserve score prediction method and system based on deep learning

The invention provides a fractional flow reserve prediction method and system based on deep learning, and relates to the technical field of image processing, and the specific scheme comprises the steps: decomposing an original CTA image into a single-blood-vessel probe image sequence according to an extracted blood vessel center line and a segment center point; extracting a blood vessel segment image feature containing blood vessel complete time sequence information from the single blood vessel probe image sequence, and predicting a blood flow reserve score of a blood vessel segment based on the blood vessel segment image feature; the fractional flow reserve of the blood vessel segment is in one-to-one correspondence with segment center points on the blood vessel center line, and a complete fractional flow reserve prediction result corresponding to the original CTA image is obtained; according to the method, on the basis of the 3DCNN and the Transform, the fragment features containing the time sequence are ex