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.

Background A great body of evidence suggests that there are retinal functional and structural changes that occur in Alzheimer’s disease (AD). However, whether such changes are primary or secondary remains to be elucidated. We studied a range of retinal functional and structural parameters in association with AD- specific pathophysiological markers in the double transgenic APP/PS1 and control mice across age. Methods Electroretinogram (ERG) and optical coherence tomography (OCT) was performed in APP/PS1 and wild type (WT) control mice every 3 months from 3 to 12 months of age. For functional assessment, the a- and b-wave of the ERG, amplitude of oscillatory potentials (OP) and the positive scotopic threshold response (pSTR) were quantified at each time point. For structural assessment, the inner and outer retinal thickness was segmented and measured from OCT scans. Episodic memory was evaluated at 6, 9 and 12 months of age using the novel object recognition test. Amyloid beta (Aβ) distribution in the hippocampus and the retina were visualised at 3, 6 and 12 months of age. Inter- and intra- group analysis was performed to study rate of change for each parameter between the two groups. Results Inter-group analysis revealed a significant difference in b-wave and OPs of APP/PS1 compared to WT controls starting from 3 months ( p  < 0.001). There was also a significant difference in the amplitude of pSTR between the two groups starting from 6 months ( p  < 0.001). Furthermore, a significant difference in the inner retinal thickness, between the two groups, was observed starting from 9 months ( p  < 0.001). Conclusions We observed an age-related decline in retinal functional and structural parameters in both APP/PS1 and WT controls, however, inter-group analysis revealed that inner retinal functional and structural decline is exacerbated in APP/PS1 mice, and that retinal functional changes precede structural changes in this strain. Further studies are required to confirm whether such phenomenon occurs in humans and if studying retinal functional changes can aid-in early assessment of AD.

Glaucoma is a progressive neurodegenerative disease associated with age. Accumulation of amyloid-beta (Aß) proteins in the ganglion cell layer (GCL) and subsequent retinal ganglion cell (RGC) loss is an established pathological hallmark of the disease. The mechanism through which Aß provokes RGC loss remains unclear. The receptor for the advanced glycation end product (RAGE), and its ligand Aß, have been shown to mediate neuronal loss internalizing Aß within the neurons. In this study, we investigated whether the RAGE-Aß axis plays a role in RGC loss in experimental glaucoma. Retinal ischemia was induced by an acute elevation of intraocular pressure in RAGE and wild-type (WT) control mice. In a subset of animals, oligomeric Aß was injected directly into the vitreous of both strains. RGC loss was assessed using histology and biochemical assays. Baseline and terminal positive scotopic threshold (pSTR) were also recorded. Retinal ischemia resulted in 1.9-fold higher RGC loss in WT mice compared to RAGE mice (36 ± 3% < 0.0001 vs. 19 ± 2%, = 0.004). Intravitreal injection of oligomeric Aß resulted in 2.3-fold greater RGC loss in WT mice compared to RAGE mice, 7-days post-injection (55 ± 4% = 0.008 vs. 24 ± 2%, = 0.02). We also found a significant decline in the positive scotopic threshold response (pSTR) amplitude of WT mice compared to RAGE (36 ± 3% vs. 16 ± 6%). RAGE mice are protected against RGC loss following retinal ischemia. Intravitreal injection of oligomeric Aß accelerated RGC loss in WT mice but not RAGE . A co-localization of RAGE and Aß, suggests that RAGE-Aß binding may contribute to RGC loss.

To study the relationship between amplitude of spontaneous retinal venous pulsatility (SRVP) and retinal nerve fibre layer (RNFL) thickness in glaucomatous eyes, and to determine if this parameter may be a potential marker for glaucoma severity. 85 subjects including 50 glaucoma (21 males, 67±10 yrs) and 35 normals (16 males, 62±11 yrs) were studied. SRVP amplitude was measured using the Dynamic Vessel Analyser (DVA, Imedos, Germany) at four regions of the retina simultaneously within one disc diameter from the optic disc--temporal-superior (TS), nasal-superior (NS), temporal-inferior (TI) and nasal-inferior (NI)). This was followed by RNFL thickness measurement using spectral domain optical coherence tomography (Spectralis OCT). The correlation between SRVP amplitude and corresponding sectoral RNFL thickness was assessed by means of non-linear regression (i.e. logarithmic). Linear regression was also applied and slopes were compared using analysis of covariance (ANCOVA). Greater SRVP amplitude was associated with thicker RNFL. Global SRVP amplitude was significantly lower in glaucoma eyes compared with normals (p<0.0001). The correlation coefficient of the linear regression between RNFL and SRVP at TS, NS, TI and NI quadrants in the glaucoma group were r = 0.5, 0.5, 0.48, 0.62. Mean SRVP amplitude and RNFL thickness for TS, NS, TI and NI quadrants were 4.3±1.5, 3.5±1.3, 4.7±1.6, 3.1±1 μm and 96±30, 75±22, 89±35 and 88±30 μm, respectively. The ANCOVA test showed that the slope of linear regression between the four quadrants was not significant (p>0.05). Since the slopes are not significantly different, it is possible to calculate one slope for all the data. The pooled slope equals 10.8 (i.e. RNFL = 10.8SRVP+41). While SRVP was present and measurable in all individuals, the amplitude of SRVP is reduced in glaucoma with increasing RNFL loss. Our findings suggest the degree of SRVP may be an additional marker for glaucoma severity. Further studies are needed to determine the mechanism of reduction in SRVP, and whether changes can predict increased risk of progression.

A mis-metabolism of transition metals (i.e., copper, iron, and zinc) in the brain has been recognised as a precursor event for aggregation of Amyloid-β plaques, a pathological hallmark of Alzheimer’s disease (AD). However, imaging cerebral transition metals in vivo can be extremely challenging. As the retina is a known accessible extension of the central nervous system, we examined whether changes in the hippocampus and cortex metal load are also mirrored in the retina. Laser ablation inductively coupled plasma-mass spectrometry (LA-ICP-MS) was used to visualise and quantify the anatomical distribution and load of Cu, Fe, and Zn in the hippocampus, cortex, and retina of 9-month-old Amyloid Precursor Protein/Presenilin 1 (APP/PS1, n = 10) and Wild Type (WT, n = 10) mice. Our results show a similar metal load trend between the retina and the brain, with the WT mice displaying significantly higher concentrations of Cu, Fe, and Zn in the hippocampus (p < 0.05, p < 0.0001, p < 0.01), cortex (p < 0.05, p = 0.18, p < 0.0001) and the retina (p < 0.001, p = 0.01, p < 0.01) compared with the APP/PS1 mice. Our findings demonstrate that dysfunction of the cerebral transition metals in AD is also extended to the retina. This could lay the groundwork for future studies on the assessment of transition metal load in the retina in the context of early AD.

Purpose The retina provides a non-invasive window to monitor microvascular circulation. Carotid-femoral Pulse Wave Velocity (cfPWV) is an indicator of large artery stiffness and is associated with systemic cardiovascular diseases. This study investigates whether Retinal artery PWV (rPWV) reflects cfPWV changes in an elderly cohort. Methods A total of 37 elderly participants (28 female, age 79 ± 5 years) were studied. Participants with a history of diabetes, glaucoma, and any neurological or eye-related disorders were excluded. Twenty four subjects were taking antihypertensive medication. A 60-s recording of retinal arterial diameter changes were captured (25 Hz frame rate, Dynamic Vessel Analyzer). Systolic blood pressure, Diastolic Blood Pressure (DBP), cfPWV and Intraocular Pressure (IOP) were measured. A custom-written algorithm was used to extract pulse amplitudes from retinal arterial diameters within one- and three-disc diameters from the optic disc. rPWV was designated as the ratio of the distance between two sites along the artery to the time delay between pulses. Predictors of rPWV were assessed by linear regression and parameter selection techniques. Results There was a positive correlation between rPWV and cfPWV (Pearson’s correlation coefficient 0.82, p < 0.001). In parameter selection models, cfPWV, transmural pressure (DBP–IOP), male sex, and IOP were isolated as predictors of rPWV. Conclusion This study found a positive correlation between rPWV and cfPWV in an elderly cohort. This indicates that microvessel remodelling that parallels large vessel remodelling is able to be detected using rPWV measurement. Further longitudinal studies are required to elucidate the predictive value of screening rPWV in systemic cardiovascular abnormalities.

The ocular circulation provides readily visible information about the state of the systemic circulation, as well as being potentially of relevance to the pathogenesis of ocular disorders such as glaucoma. The interaction between intraocular pressure, retinal vessels and cerebrospinal fluid pressure located at the retrolaminar portion of the eye has been of great interest for both ophthalmic and neurological clinicians and researchers. Understanding the relationship between these physiological parameters can explain phenomena such as spontaneous retinal venous pulsatility, and characterize the effects of the translaminar pressure gradient. It may be feasible to use measurable changes in venous pulsatility to enhance clinical assessment in different diseases. In this article we review recent findings on ocular hemodynamics and the relevance of these parameters in the diagnosis of ophthalmic and neurological diseases.

Background Retinal imaging may serve as an alternative approach to monitor brain pathology in Alzheimer’s disease (AD). In this study, we investigated the association between retinal vascular and structural changes and cerebral amyloid-β (Aβ) plaque load in an elderly cohort. Methods We studied a total of 101 participants, including 73 elderly subjects (79 ± 5 years, 22 male) with no clinical diagnosis of AD but reporting some subjective memory change and an additional 28 subjects (70 ± 9 years, 16 male) with clinically established AD. Following a complete dilated ocular examination, the amplitude of retinal vascular pulsations and dynamic response, retinal nerve fibre layer thickness and retinal ganglion cell layer (RGCL) thickness were determined in all patients. Systemic blood pressure and carotid-to-femoral pulse wave velocity were measured. The elderly cohort also underwent magnetic resonance imaging and 18 F-florbetaben (FBB)-positron emission tomographic amyloid imaging to measure neocortical Aβ standardised uptake value ratio (SUVR), and this was used to characterise a ‘preclinical’ group (SUVR >1.4). Results The mean FBB neocortical SUVR was 1.35 ± 0.3. The amplitude of retinal venous pulsations correlated negatively with the neocortical Aβ scores ( p  < 0.001), whereas the amplitude of retinal arterial pulsations correlated positively with neocortical Aβ scores ( p  < 0.01). RGCL thickness was significantly lower in the clinical AD group ( p  < 0.05). Conclusions The correlation between retinal vascular changes and Aβ plaque load supports the possibility of a vascular component to AD. Dynamic retinal vascular parameters may provide an additional inexpensive tool to aid in the preclinical assessment of AD.

Spontaneous retinal venous pulsations (SRVP) are assessed as a clinical marker for patients with ophthalmic or neurological disorders. The pulsations are influenced by intraocular pressure (IOP), cerebrospinal fluid pressure (CSFp), and retinal venous pressure (RVP). However, little is known about the effect of cyanosis with polycythemia, a common finding in adults with complex congenital heart disease (CHD), on SRVP. This study investigated 11 subjects with long-standing cyanosis secondary to CHD and 11 control subjects to determine if there were measurable differences in resting pulsatility for a given IOP level. Intraocular pressure was measured using Goldman tonometry, and dynamic SRVP was recorded noninvasively using a retinal vessel imaging system. Peak amplitude of SRVP at each cardiac cycle was measured and compared with IOP. Heart rate was also monitored during the tests. Results show that for a similar baseline IOP, SRVP amplitudes are significantly lower in cyanotic patients compared with normal subjects ( P < 0.0001). This may be explained by an increased RVP or high CSFp in these patients. Mean venous diameter is also significantly higher in cyanotic patients ( P < 0.01), but no significant relationship was found between SRVP or diameter with blood parameters.