Knowledge infused cascade convolutional neural network for segmenting retinal vessels in volumetric optical coherence tomography

TL;DR

This paper introduces a cascade deep neural network for segmenting retinal vessels in volumetric OCT images by integrating histological and imaging knowledge, significantly improving accuracy over existing methods.

Contribution

The novel approach infuses prior biological and imaging knowledge into the network without altering its architecture, enhancing segmentation performance.

Findings

Outperforms state-of-the-art deep learning methods

Effective use of histology and imaging knowledge

Demonstrates the benefit of prior knowledge infusion

Abstract

We present a cascade deep neural network to segment retinal vessels in volumetric optical coherence tomography (OCT). Two types of knowledge are infused into the network for confining the searching regions. (1) Histology. The retinal vessels locate between the inner limiting membrane and the inner nuclear layer of human retina. (2) Imaging. The red blood cells inside the vessels scatter the OCT probe light forward and form projection shadows on the retinal pigment epithelium (RPE) layer, which is avascular thus perfect for localizing the retinal vessel in transverse plane. Qualitative and quantitative comparison results show that the proposed method outperforms the state-of-the-art deep learning and graph-based methods. This work demonstrates, instead of modifying the architectures of the deep networks, incorporating proper prior knowledge in the design of the image processing framework could be an efficient approach for handling such specific tasks.