Video inspired lesion segmentation algorithm for OCT retinal images

Optical Coherence Tomography (OCT) retinal lesion segmentation is critical for ophthalmic diagnosis and treatment. However, current OCT retinal image segmentation methods primarily focus on individual B-scans and neglect the continuity condition of lesions across OCT volume B-scans, which leads to suboptimal segmentation results. To address this issue, we propose an innovative 2.5D segmentation algorithm that treats OCT retinal images as video sequences and leverage advanced video segmentation models. We compared our method with state-of-the-art 2D image-based segmentation methods on a public dataset. Experimental results demonstrate that our method significantly improves both visual quality and quantitative metrics, particularly in terms of structural continuity and robustness of the segmentation results.