Method and system for segmenting choroidal neovascularization in optical coherence tomography image

The invention discloses a method and system for segmenting choroidal neovascularization in an optical coherence tomography image, and the method comprises the steps: constructing a UNet-based segmentation network, setting two paths during down-sampling, and respectively extracting multi-scale global features and multi-scale global features with local semantic information and long-distance dependence information; the features extracted on the two paths are fused, and the fused features of each scale are output, so that feature information can be fully extracted; in the up-sampling process, the fused features and the up-sampling features of the same scale are enhanced and then fused, generation of lesion features in the up-sampling process can be more focused, lesion boundary information in the up-sampling process is made to serve as supervision information to enhance expression of boundary information, fine segmentation can be carried out, and the segmentation effect is improved. 本发明公开了一种光学相干断层扫描图像中脉络膜新生血管分割方法