The invention relates to the technical field of medical
image processing, and discloses a
liver cancer image
automatic segmentation identification method fusing a multi-
modal image. The method comprises the following steps: acquiring multi-
modal medical image data in a first
processing stage, and judging whether the data contains artifact interference characteristics or not; if artifact interference features are included, a first
processing flow is triggered,
time sequence registration
processing is carried out on the data, a registered first image
feature set is obtained in a second processing stage, and a first tumor boundary segmentation result is generated based on a first segmentation
network model and the first image
feature set; and if the artifact interference feature is not included, triggering a second processing flow, performing feature enhancement processing on the data, obtaining an enhanced second image
feature set in a third processing stage, and generating a second tumor boundary segmentation result based on the second segmentation
network model and the second image feature set. According to the method, multi-
modal image data with different qualities can be processed in a targeted manner, the segmentation adaptability and reliability are improved, and clinical application requirements are met.