Segmentation of Preretinal Space in Optical Coherence Tomography Images Using Deep Neural Networks

Abstract

This paper proposes an efficient segmentation of the preretinal area between the inner limiting membrane (ILM) and posterior cortical vitreous (PCV) of the human eye in an image obtained with the use of optical coherence tomography (OCT). The research was carried out using a database of three-dimensional OCT imaging scans obtained with the Optovue RTVue XR Avanti device. Various types of neural networks (UNet, Attention UNet, ReLayNet, LFUNet) were tested for semantic segmentation, their effectiveness was assessed using the Dice coefficient and compared to the graph theory techniques. Improvement in segmentation efficiency was achieved through the use of relative distance maps. We also show that selecting a larger kernel size for convolutional layers can improve segmentation quality depending on the neural network model. In the case of PVC, we obtain the effectiveness reaching up to 96.35%. The proposed solution can be widely used to diagnose vitreomacular traction changes, which is not yet available in scientific or commercial OCT imaging solutions.

Denoising of Optical Coherence Tomography Images in Ophthalmology Using Deep Learning: A Systematic Review

“…Since the studies included in this review solely reported either PSNR or SNR of the denoised images in comparison to multiple reference tests, it remains uncertain whether DL provides significant assistance in image denoising and speckle reduction. Ideally, the impact of DL for denoising OCT images in ophthalmology should be demonstrated in practice-based settings and validated by its ability to improve further objectives such as detection [60], classification [61], and segmentation [62], which the majority of included studies did not consider.…”

Biometric Technologies Based on Optical Coherence Tomography

“…Deep neural networks are also suitable for effective segmentation of the preretinal space [ 3 ]. In this case, effective segmentation of the posterior cortical vitreous (PVC) and inner limiting membrane (ILM) is required.…”

CA‐Unet++: An improved structure for medical CT scanning based on the Unet++ Architecture

“…In the image segmentation task, the Dice coefficient will be more sensitive to the segmentation internal data, while the Hausdorff distance is sensitive to the segmentation boundary data. The Hausdorff distance calculation formula is derived from Formula (10), Formula (11), Formula (12).…”

“…Recently, deep learning has achieved remarkable results in visual image recognition tasks. 12 The application of image segmentation in medicine can mark the lesions and extract features for diagnosis and treatment. The deep convolutional network has better applications in medical image segmentation.…”