4 results about "Optic disk" patented technology

This invention presents a passive domain adaptive fundus segmentation method based on evidence representation and marginal screening. Under the condition that the source domain training images and their pixel-level annotations are inaccessible, a Student network and a Teacher network are constructed to learn segmentation of the target domain fundus image. The network outputs positive foreground evidence and negative background evidence for each segmentation channel, and constructs a pixel-level Beta distribution. Based on the Beta distribution, foreground probability, evidence margin, and evidence strength are calculated. Reliable positive pseudo-labels are obtained through joint screening, and a weighted segmentation loss is constructed by combining pseudo-label weights. Simultaneously, regularization loss and evidence strength constraint loss are introduced to optimize the Student network. Finally, the exponential moving average coefficient is dynamically adjusted based on the amount of evidence to update the Teacher network parameters. This method can effectively suppress the propagation of pseudo-label noise and improve the accuracy, stability, and cross-domain adaptability of optic cup and optic disc segmentation in the target domain fundus image.

An apparatus includes an obtaining unit configured to obtain, using an optical coherence tomography, an analysis map for a disk and an analysis map for a macular area, which are analysis results of three-dimensional data obtained by capturing an image of a region of a fundus of a subject's eye that includes the disk and the macular area, and a control unit configured to cause a display unit to display side by side a two-dimensional image of the region including the disk and the macular area, the analysis map for the optic disk, and the analysis map for the macular area.

The invention discloses a lightweight optic disc segmentation method and system, and relates to the technical field of medical image segmentation. Comprising the following steps: carrying out data enhancement processing on a public data set iChallenge-PM, an iChallenge-AMD and a Drishiti-GS; a lightweight segmentation model is constructed, and the model comprises a C2f module used for extracting and fusing eye fundus image features; a GlobalEdgeInformationTransfer module, which is used for enhancing the edge feature extraction and the multi-scale fusion of the eye fundus image; the Wavelet Pooling module is used for realizing eye fundus image resolution conversion by using a fixed wavelet filter and reducing trainable parameters; and the C2fgConv module introduces a gating mechanism to optimize eye fundus image feature extraction and reduce parameter quantity. Through multi-scale feature extraction and fusion, WaveletPooling application and C2fgConv design, while the optic disk segmentation precision is ensured, model parameters are effectively reduced, light weight is realized, the training efficiency and generalization ability are improved, and the method is suitable for optic disk segmentation tasks under different resource conditions.