Search

[Absctract]: OCT is a non-invasive imaging technique commonly used to obtain 3D volumes of the ocular structure. These volumes allow the monitoring of ocular and systemic diseases through the observation of subtle changes in the different structures present in the eye. In order to observe these changes it is essential that the OCT volumes have a high resolution in all axes, but unfortunately there is an inverse relationship between the quality of the OCT images and the number of slices of the cube. This results in routine clinical examinations using cubes that generally contain high-resolution images with few slices. This lack of slices complicates the monitoring of changes in the retina hindering the diagnostic process and reducing the effectiveness of 3D visualizations. Therefore, increasing the cross-sectional resolution of OCT cubes would improve the visualization of these changes aiding the clinician in the diagnostic process. In this work we present a novel fully automatic methodology to perform the synthesis of intermediate slices of OCT image volumes in an unsupervised manner. To perform this synthesis, we propose a fully convolutional neural network architecture that uses information from two adjacent slices to generate the intermediate synthetic slice. We also propose a training methodology, where we use three adjacent slices to train the network by contrastive learning and image reconstruction. We test our methodology with three different types of OCT volumes commonly used in the clinical setting and validate the quality of the synthetic slices created with several medical experts and using an expert system.

[Abstract] Optical Coherence Tomography (OCT) is a non-invasive imaging technique with a crucial role in the monitoring of a wide range of diseases. In order to make a good diagnosis it is essential that clinicians can observe any subtle changes that appear in the multiple ocular structures, so it is imperative that the 3D OCT volumes have good resolution in each axis. Unfortunately, there is a trade-off between image quality and the number of volume slices. In this work, we use a convolutional neural network to generate the intermediate synthetic slices of the OTC volumes and we propose a few variants of a 3D reconstruction algorithm to create visualizations that emphasize the changes present in multiple retinal structures to aid clinicians in the diagnostic process

[Abstract]: Optical coherence tomography (OCT) is a non-invasive diagnostic technique that can image ocular structures. Recently, this imaging technique has been used to diagnose and monitor patients with multiple sclerosis (MS), as several clinical studies have linked the development of MS to various changes in the eye. Among the different structures, one of the relevant biomarkers for MS analysis is the choroid. Systems such as Enhanced Depth Imaging (EDI) provide detailed images of the choroid region. However, OCT images are not routinely captured using this technology unless the study is specifically focused on choroidal analysis. In this work we propose a robust approach, based on convolutional neural networks to segment the choroid in non-EDI OCT images. The results obtained show that the proposed network manages to delimit the inferior contour of the choroid in a similar way to the experts.

[Abstract]: Optical Coherence Tomography (OCT) is a non-invasive imaging technique that provides high-resolution cross-sectional images of biological tissues. Biomarkers such as the thickness of retinal layers can be used for the evaluation of several diseases such as glaucoma or multiple sclerosis. In this work, we create the thickness profiles of healthy and pathological eyes using a statistical model and developed a methodology to compare the thickness profiles of OCT images with the models and visually inspect the abnormal areas. This approach allows for a quick assessment of the retinal layers health, assisting clinicians and easing the diagnostic burden.

[Abstract]: Multiple Sclerosis (MS) is a chronic neurological disease, in which immune-mediated mechanisms lead to pathological processes of neurodegeneration. Optical coherence tomography (OCT) has recently begun to be used to diagnose and monitor patients with MS. Morphological changes in the choroid have been linked to the onset of MS, so an accurate segmentation of this layer is critical. Conventional OCT has several limitations in obtaining accurate images of the choroid, which has been improved through the use of systems such as Enhanced Depth Imaging (EDI) OCT. Unfortunately, many longitudinal studies that have collected samples over the years in the past have been performed using highly variable settings and without the use of the EDI protocol (or similar variants). For these reasons, in this work we propose a series of fully automatic approaches, based on convolutional neural networks, capable of robustly segmenting the choroid in OCT images without using the EDI protocol. To test the robustness and efficiency of our method, we performed experiments on a public dataset and a collected one. The Dice score obtained by the best proposed architecture is 89.7 for the public dataset, and 93.7 for the collected dataset.

[Abstract]: Age-related macular degeneration is the leading cause of vision loss in developed countries, and wet-type AMD requires urgent treatment and rapid diagnosis because it causes rapid irreversible vision loss. Currently, AMD diagnosis is mainly carried out using images obtained by optical coherence tomography. This diagnostic process is performed by human clinicians, so human error may occur in some cases. Therefore, fully automatic methodologies are highly desirable adding a layer of robustness to the diagnosis. In this work, a novel computer-aided diagnosis and visualization methodology is proposed for the rapid identification and visualization of wet AMD. We adapted a convolutional neural network for segmentation of a similar domain of medical images to the problem of wet AMD segmentation, taking advantage of transfer learning, which allows us to work with and exploit a reduced number of samples. We generate a 3D intuitive visualization where the existence, position and severity of the fluid were represented in a clear and intuitive way to facilitate the analysis of the clinicians. The 3D visualization is robust and accurate, obtaining satisfactory 0.949 and 0.960 Dice coefficients in the different evaluated OCT cube configurations, allowing to quickly assess the presence and extension of the fluid associated to wet AMD.