Explore the vast range of titles available.

Glaucoma, a leading cause of blindness, is a multifaceted disease with several patho-physiological features manifesting in single fundus images (e.g., optic nerve cupping) as well as fundus videos (e.g., vascular pulsatility index). Current convolutional neural networks (CNNs) developed to detect glaucoma are all based on spatial features embedded in an image. We developed a combined CNN and recurrent neural network (RNN) that not only extracts the spatial features in a fundus image but also the temporal features embedded in a fundus video (i.e., sequential images). A total of 1810 fundus images and 295 fundus videos were used to train a CNN and a combined CNN and Long Short-Term Memory RNN. The combined CNN/RNN model reached an average F-measure of 96.2% in separating glaucoma from healthy eyes. In contrast, the base CNN model reached an average F-measure of only 79.2%. This proof-of-concept study demonstrates that extracting spatial and temporal features from fundus videos using a combined CNN and RNN, can markedly enhance the accuracy of glaucoma detection.

The presence or absence of spontaneous retinal venous pulsations (SVP) provides clinically significant insight into the hemodynamic status of the optic nerve head. Reduced SVP amplitudes have been linked to increased intracranial pressure and glaucoma progression. Currently, monitoring for the presence or absence of SVPs is performed subjectively and is highly dependent on trained clinicians. In this study, we developed a novel end-to-end deep model, called U3D-Net, to objectively classify SVPs as present or absent based on retinal fundus videos. The U3D-Net architecture consists of two distinct modules: an optic disc localizer and a classifier. First, a fast attention recurrent residual U-Net model is applied as the optic disc localizer. Then, the localized optic discs are passed on to a deep convolutional network for SVP classification. We trained and tested various time-series classifiers including 3D Inception, 3D Dense-ResNet, 3D ResNet, Long-term Recurrent Convolutional Network, and ConvLSTM. The optic disc localizer achieved a dice score of 95% for locating the optic disc in 30 milliseconds. Amongst the different tested models, the 3D Inception model achieved an accuracy, sensitivity, and F1-Score of 84 ± 5%, 90 ± 8%, and 81 ± 6% respectively, outperforming the other tested models in classifying SVPs. To the best of our knowledge, this research is the first study that utilizes a deep neural network for an autonomous and objective classification of SVPs using retinal fundus videos.

Spontaneous retinal venous pulsations (SVPs) are a dynamic vascular marker for glaucoma, which is a leading cause of irreversible blindness across the globe. Since the discovery of SVPs, their presence has been variably reported in patients with glaucoma, some even reporting absent SVPs. Recent objective quantification of SVPs has led to an increase in their detection. There is a need to explore the possibility of SVPs being used as a potential marker for glaucoma screening in clinical practice. Current devices that are used to assess SVPs pose limitations that deem current SVP detection unfeasible for screening outside metropolitan and in remote areas and for use in mobile clinics that service underprivileged and remote communities where there is likely to be many cases of undetected glaucoma.This thesis aims to assess the effectiveness of a novel tablet-based ophthalmoscope to detect and quantify SVPs in glaucoma with the aid of computer analysis. This device was used to perform fundus videoscopy in 170 participants recruited from three ophthalmic clinics in Sydney. All participants had a confirmed diagnosis of glaucoma or glaucoma suspect. SVP amplitudes were extracted from raw videos using a custom-written algorithm. Standard structural (RNFL thickness) and functional (VF loss) clinical markers for glaucoma, as well as intraocular pressure (IOP) and retinal ganglion cell (RGC) estimates were also recorded and documented. SVP distribution, and its association with the established clinical structural and functional measures were assessed.Using tablet-based ophthalmoscopy, SVPs were detected and quantified in all participants, regardless of glaucoma status. The largest SVP amplitudes were detected in normal tension glaucoma (NTG; 32.5%), followed by primary open-angle glaucoma (POAG; 28.7%) and glaucoma suspects (26.3%). A significant association was found between SVP amplitudes and clinical markers in NTG and POAG with the highest correlations being between SVP amplitude - RNFL thickness (p=0.1) and SVP amplitude - RGC count (p<0.001) in NTG and POAG respectively. When evaluating which clinical marker can distinguish between confirmed glaucoma and glaucoma suspects, SVP analysis was found to be comparable to standard clinical markers. More specifically, SVPs can separate POAG from glaucoma suspects as effectively as RNFL thickness. SVPs can also separate POAG from NTG as effectively as IOP measurements.The novel device used in this thesis overcomes many of the disadvantages of current commercial techniques, particularly portability and ease of use. The novel device and technique can be used to detect and quantify SVPs in all participants with glaucoma, regardless of glaucoma severity. The findings of this thesis indicate that SVPs are associated with clinical markers that are known to occur during the early glaucomatous changes. The potential benefits that this may offer in the early detection of glaucoma, consequent management and evaluation of progression are substantial. When combined with RGC count, SVP amplitude analysis may provide benefits to traditional glaucoma assessments where often structural and functional glaucomatous loss are only clinically detected once substantial RGC loss has already occurred. Further studies are required to determine if longitudinal SVP changes are associated with progressive glaucoma and whether SVPs can be used as a marker for monitoring disease progression.

Background: Glaucoma, the leading cause of irreversible blindness, is classified as a neurodegenerative disease, and its incidence increases with age. Pathophysiological changes, such as the deposition of amyloid-beta plaques in the retinal ganglion cell layer, as well as neuropsychological changes, including cognitive decline, have been reported in glaucoma. However, the association between cognitive ability and retinal functional and structural measures in glaucoma, particularly glaucoma subtypes, has not been studied. We studied the association between cognitive ability and the visual field reliability indices as well as the retinal ganglion cell (RGC) count estimates in a cohort of glaucoma patients. Methods: A total of 95 eyes from 61 glaucoma patients were included. From these, 20 were normal-tension glaucoma (NTG), 25 were primary open-angle glaucoma (POAG), and 16 were glaucoma suspects. All the participants had a computerised Humphrey visual field (HVF) assessment and optical coherence tomography (OCT) scan and were administered the written Montreal Cognitive Assessment (MoCA) test. RGC count estimates were derived based on established formulas using the HVF and OCT results. A MoCA cut-off score of 25 and less was designated as cognitive impairment. Student’s t-test was used to assess differences between the groups. The Pearson correlation coefficient was used to assess the association between MoCA scores and retinal structural and functional measures. Results: Significant associations were found between MoCA scores and the false-negative and pattern standard deviation indices recorded on the HVF (r = −0.19, r = −0.22, p < 0.05). The mean IOP was significantly lower in the cognitively impaired group (i.e., MOCA ≤ 25) (13.7 ± 3.6 vs. 15.7 ± 4.5, p < 0.05). No significant association was found between RGC count estimates and MoCA scores. Analysis of these parameters in individual glaucoma subtypes did not reveal any group-specific significant associations either.

Spontaneous retinal Venous Pulsations (SVP) are rhythmic changes in the caliber of the central retinal vein and are observed in the optic disc region (ODR) of the retina. Its absence is a critical indicator of various ocular or neurological abnormalities. Recent advances in imaging technology have enabled the development of portable smartphone-based devices for observing the retina and assessment of SVPs. However, the quality of smartphone-based retinal videos is often poor due to noise and image jitting, which in return, can severely obstruct the observation of SVPs. In this work, we developed a fully automated retinal video stabilization method that enables the examination of SVPs captured by various mobile devices. Specifically, we first propose an ODR Spatio-Temporal Localization (ODR-STL) module to localize visible ODR and remove noisy and jittering frames. Then, we introduce a Noise-Aware Template Matching (NATM) module to stabilize high-quality video segments at a fixed position in the field of view. After the processing, the SVPs can be easily observed in the stabilized videos, significantly facilitating user observations. Furthermore, our method is cost-effective and has been tested in both subjective and objective evaluations. Both of the evaluations support its effectiveness in facilitating the observation of SVPs. This can improve the timely diagnosis and treatment of associated diseases, making it a valuable tool for eye health professionals.

Retinal vessel segmentation is generally grounded in image-based datasets collected with bench-top devices. The static images naturally lose the dynamic characteristics of retina fluctuation, resulting in diminished dataset richness, and the usage of bench-top devices further restricts dataset scalability due to its limited accessibility. Considering these limitations, we introduce the first video-based retinal dataset by employing handheld devices for data acquisition. The dataset comprises 635 smartphone-based fundus videos collected from four different clinics, involving 415 patients from 50 to 75 years old. It delivers comprehensive and precise annotations of retinal structures in both spatial and temporal dimensions, aiming to advance the landscape of vasculature segmentation. Specifically, the dataset provides three levels of spatial annotations: binary vessel masks for overall retinal structure delineation, general vein-artery masks for distinguishing the vein and artery, and fine-grained vein-artery masks for further characterizing the granularities of each artery and vein. In addition, the dataset offers temporal annotations that capture the vessel pulsation characteristics, assisting in detecting ocular diseases that require fine-grained recognition of hemodynamic fluctuation. In application, our dataset exhibits a significant domain shift with respect to data captured by bench-top devices, thus posing great challenges to existing methods. In the experiments, we provide evaluation metrics and benchmark results on our dataset, reflecting both the potential and challenges it offers for vessel segmentation tasks. We hope this challenging dataset would significantly contribute to the development of eye disease diagnosis and early prevention.