Autonomous Stabilization of Retinal Videos for Streamlining Assessment of Spontaneous Venous Pulsations.

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.

Autonomous assessment of spontaneous retinal venous pulsations in fundus videos using a deep learning framework.

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.

RAGE and its ligand amyloid beta promote retinal ganglion cell loss following ischemia-reperfusion injury.

Introduction: 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 via internalizing Aß within the neurons. In this study, we investigated whether the RAGE-Aß axis plays a role in RGC loss in experimental glaucoma. Methods: 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. Results: Retinal ischemia resulted in 1.9-fold higher RGC loss in WT mice compared to RAGE-/- mice (36 ± 3% p < 0.0001 vs. 19 ± 2%, p = 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% p = 0.008 vs. 24 ± 2%, p = 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%). Discussion: 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.

Distribution of Copper, Iron, and Zinc in the Retina, Hippocampus, and Cortex of the Transgenic APP/PS1 Mouse Model of Alzheimer's Disease.

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-ß 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.

Cognitive Performance on the Montreal Cognitive Assessment Test and Retinal Structural and Functional Measures in Glaucoma.

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.

A combined convolutional and recurrent neural network for enhanced glaucoma detection.

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.

Objective Quantification of Spontaneous Retinal Venous Pulsations Using a Novel Tablet-Based Ophthalmoscope.

Purpose: Dynamic assessment of retinal vascular characteristics can aid in identifying glaucoma-specific biomarkers. More specifically, a loss of spontaneous retinal venous pulsations (SVPs) has been reported in glaucoma, but a lack of readily available tools has limited the ability to explore the full potential of SVP analysis in glaucoma assessment. Advancements in smart technology have paved the way for the development of portable, noninvasive, and inexpensive imaging modalities. By combining off-the-shelf optical elements and smart devices, the current study aims to determine whether SVPs can be detected and quantified using a novel tablet-based ophthalmoscope in glaucoma and glaucoma suspects. Methods: Thirty patients, including 21 with confirmed glaucoma (9 men; average age 75 ± 8 years) and 9 glaucoma suspects (5 men; average age 64 ± 9 years), were studied. All patients had intraocular pressure measurements, Humphrey visual field assessment, optical coherence tomography, and a 10-second videoscopy of the retinal circulation. The retinal vasculature recordings (46° field of view at 30 frames per second) were analyzed to extract SVP amplitudes. Results: SVPs were detected and quantified in 100% of patients with glaucoma and those with suspected glaucoma using the novel device. The average SVP amplitudes in glaucoma and glaucoma suspects were 42.6% ± 10.7% and 34% ± 6.7%, respectively. Conclusions: Our results suggest that a novel tablet-based ophthalmoscope can aid in documenting and objectively quantifying SVPs in all patients. Translational Relevance: Outcomes of this study provide an innovative, portable, noninvasive, and inexpensive solution for objective assessment of SVPs, which may have clinical relevance in glaucoma screening.

A longitudinal assessment of retinal function and structure in the APP/PS1 transgenic mouse model of Alzheimer's disease.

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 (Aß) 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.

A novel method for retinal vessel segmentation and diameter measurement using high speed video.

Studying dynamic characteristics of retinal vessels, as opposed to static measures, may provide additional insight into pathophysiological changes associated with local and systemic abnormalities such as glaucoma and hypertension. Various approaches have been developed to derive static biomarkers from retinal still images, but not many for dynamic analysis of video sequences. This study presents a novel method for the assessment of time-dependent diameter changes in high-speed videos (125 fps) from the rat retina. The proposed method is composed of a vessel segmentation and a diameter measurement module. The specificity and sensitivity of the segmentation method over 25 images were 95.1% and 97.3% respectively. The mean and standard deviation of the diameter measurement errors were -0.147±0.41 pixels over 100 measurements.

Retinal vascular and structural changes are associated with amyloid burden in the elderly: ophthalmic biomarkers of preclinical Alzheimer's disease.

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-ß (Aß) 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 18F-florbetaben (FBB)-positron emission tomographic amyloid imaging to measure neocortical Aß 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 Aß scores (p < 0.001), whereas the amplitude of retinal arterial pulsations correlated positively with neocortical Aß 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 Aß 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.

Diagnostic and Prognostic Potential of Retinal Biomarkers in Early On-Set Alzheimer's Disease.

OBJECTIVE: Accumulating evidence suggests that the eye can be used in the assessment of early on-set Alzheimer's disease (AD). The eye offers a natural window to the brain through the retina. The retina and brain share common developmental origins and patho-physiological origins and mechanisms, having been sequestered from it during early development, but retaining its connections with the brain via the optic nerve. Therefore, it is well understood that neurological abnormalities have a direct profound impact on the retina. Recent studies suggest an array of physiological and pathological changes in the retina in dementia and specifically in AD. There are also reports on imaging the two hallmark proteins of the disease, extracellular amyloid beta peptides and intracellular hyper phosphorylated tau protein, as a proxy to neuroimaging. RESULTS: In this review, we summarise retinal structural, functional and vascular changes reported to be associated with AD. We also review techniques employed to image these two major hall mark proteins of AD and their relevance for early detection of AD.

High speed in-vivo imaging of retinal hemodynamics in a rodent model of hypertension.

The eye is the only organ through which microcirculation can be visualized non-invasively. This unique feature makes the eye and specifically retinal vasculature an excellent target area to monitor and study micro-vascular damage in systemic diseases. Dynamic (real-time) changes of retinal vessels have been shown to be more specific to the disease in comparison with static measurements. In this study we utilize high speed imaging (i.e. 125 fps) to study and derive dynamic changes of retinal vessels in a rat model of hypertension. A Eulerian video magnification algorithm was used to extract retinal arterial and venous pulse amplitude from five Spontaneously Hypertensive Rats (SHR) and five Wister Kyoto (WKY) rats were used as the control group. Results showed that retinal arterial diameter and pulse amplitude are significantly lower in the SHRs compared with WKYs. Dynamic biomarkers of retinal micro-vasculature may be used as a diagnostic tool for systemic diseases.

Correlation of retinal nerve fibre layer thickness and spontaneous retinal venous pulsations in glaucoma and normal controls.

PURPOSE: 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. METHOD: 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). RESULTS: 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). CONCLUSION: 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.

Characterizing dynamic properties of retinal vessels in the rat eye using high speed imaging.

PURPOSE: The dynamic properties of retinal vessels including pulse wave propagation and pulsatility index provide new perspective in retinal hemodynamic analysis. In this study we utilize a high speed imaging system to capture these characteristics in the rat eye for the first time. METHODS: Retinal video images of 9 Wistar-Kyoto (WKY) rats were captured at a rate of 250 frames per second for 10s with a 50° field of view using a high speed camera (Optronis, Kehl, Germany). Two recordings were taken from each rat at the same sites for repeatability analysis. The electrocardiogram (ECG) was measured simultaneously with retinal images. Arterial retinal pulse wave velocity (rPWV) and arterial/venous pulse amplitude were calculated from recorded images. Arterial measurements were repeated in another normotensive strain of the same age (Sprague-Dawley, n=4). RESULTS: The average WKY rPWV was 11.4 ± 6.1 cm/s. The differences between repeated measures were not significant (-2.8 ± 2.9 cm/s, p=0.2). Sprague-Dawley animals had a similar rPWV (9.8 ± 2.2 cm/s, p=0.61). The average arterial and venous pulse amplitude was 7.1 ± 1.5 µm and 8.2 ± 2.0 µm respectively. There was a positive correlation between rPWV and heart rate in the WKY groups (r(2)=0.32). A positive correlation was also obtained between arterial and venous diameter and their pulse amplitude (r(2)=0.67 and r(2)=0.37 respectively). CONCLUSION: rPWV was associated with heart rate. Higher pulsation amplitude was also correlated with larger vessel diameter. High speed imaging of retinal vessels in the rat eye provides an accurate and robust method to study dynamic characteristics of these vessels and their relationship with ocular and systemic abnormalities.

Hemodynamic interactions in the eye: a review.

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.

Non-invasive estimation of cerebrospinal fluid pressure waveforms by means of retinal venous pulsatility and central aortic blood pressure.

Current techniques used for cerebrospinal fluid pressure (CSFp) measurements are invasive. They require a surgical procedure for placement of a pressure catheter in the brain ventricles or in the brain tissue. The human eye provides direct visualisation of its physiological structures and due to its anatomical connection with CSF via the retrolaminar optic nerve it may provide accessible information about CSFp. A total of 25 subjects were included in this study. 15 subjects were used to characterise the relationship between intraocular pressure (IOP), spontaneous retinal venous pulsatility (SRVP), and CSFp. IOP was manipulated and SRVP amplitudes recorded dynamically using the dynamic vessel analyzer (DVA). The relationship between IOP and SRVP amplitude was established to estimate CSFp. Additionally Doppler blood flow velocity of the middle cerebral artery and arterial blood pressure (ABP) were acquired for all subjects. This was to compare and validate our findings with an alternative approach (ICM+) which uses these values to estimate CSFp. A CSFp waveform was extracted from central blood pressure (CBP) waveform by removing its cardiac component frequency. Furthermore to calibrate the CSFp to CBP waveform ratio, invasive CSFp, and ABP was measured from 10 subjects with brain tumours who had a range of normal to elevated CSFp (i.e., 0-30 mmHg). Results show good agreement between the two methods (correlation r (2) = 0.55) Mean estimated CSFp for the two techniques did not show any significant difference (p > 0.05). A significant correlation between CBP pulse (CBPp) and invasive CSFp pulse (CSFpp) was observed (i.e., CSFpp = 0.0654CBBp + 3.91, p < 0.01). Estimated CSFpp was calibrated to CBPp according to this relation. In conclusion, the study demonstrated a good correlation between two different methods of estimating CSFp non-invasively and may provide a novel method to estimate CSF waveforms non-invasively.

Visualization of orbital flow by means of phase contrast MRI.

Magnetic Resonance Imaging (MRI) is a high resolution medical imaging technique to image internal anatomical structures. Phase contrast MRI (pcMRI) technique is an add-on specification of MRI devices in order to quantify flow. Although different attempts have been introduced to measure orbital flows, a relationship between different ophthalmic physiological structures including superior ophthalmic vein, ophthalmic artery and optic nerve sheath (containing cerebrospinal fluid) using phase contrast MRI has not been established. In this study we investigate orbital flow in 5 normal subjects using a 3 tesla MRI device. pcMRI technique has been applied to extract flow in the superior ophthalmic vein and optic nerve sheath. Electrocardiogram of each subject was monitored and gated to the MRI in order to extract flow waveforms. Results show multiple peaks when assessing orbital flow waveforms, suggesting possible reflection of flow from back of the eye. These peaks have been characterized and a possible explanation to this phenomenon has been provided. This study enhances understanding of interaction between physiological structures at the retrolaminar portion of the eye which may be responsible for different ophthalmic abnormalities.

Non-invasive cerebrospinal fluid pressure estimation using multi-layer perceptron neural networks.

Cerebrospinal fluid pressure (CSFp) provides vital information in various neurological abnormalities including hydrocephalus, intracranial hypertension and brain tumors. Currently, CSFp is measured invasively through implanted catheters within the brain (ventricles and parenchyma) which is associated with a risk of infection and morbidity. In humans, the cerebrospinal fluid communicates indirectly with the ocular circulation across the lamina cribrosa via the optic nerve subarachnoid space. It has been shown that a relationship between retinal venous pulsation, intraocular pressure (IOP) and CSFp exists with the amplitude of retinal venous pulsation being associated with the trans-laminar pressure gradient (i.e. IOP-CSFp). In this study we use this characteristic to develop a non-invasive approach to estimate CSFp. 15 subjects were included in this study. Dynamic retinal venous diameter changes and IOP were measured and fitted into our model. Artificial neural networks (ANN) were applied to construct a relationship between retinal venous pulsation amplitude, IOP (input) and CSFp (output) and develop an algorithm to estimate CSFp based on these parameters. Results show a mean square error of 2.4 mmHg and 1.27 mmHg for train and test data respectively. There was no significant difference between experimental and ANN estimated CSFp values (p>0.01).This study suggests measurement of retinal venous pulsatility in conjunction with IOP may provide a novel approach to estimate CSFp non-invasively.

Minimising retinal vessel artefacts in optical coherence tomography images.

Optical coherence tomography (OCT) is commonly used to investigate the layers of the retina including retinal nerve fiber layer (RNFL) and retinal pigment epithelium (RPE). OCT images are altered by vessels on the retinal surface producing artefacts. We propose a new approach to compensate for these artefacts and enhance quality of OCT images. A total of 28 (20 normal and 8 glaucoma subjects) OCT images were obtained using Spectralis (Heidelberg, Germany). Shadows were detected along the image and compensated by the A-Scan intensity difference from surrounding non-affected areas. Images were then segmented and the area and thickness of RNFL and RPE were measured and compared. 10 subjects were tested twice to determine the effect of this on reproducibility of measurements. Shadow-suppressed images reflected the profile of the retinal layers more closely when assessed qualitatively, minimising distortion. The segmentation of RNFL and RPE thickness demonstrated a mean change of 2.4% ± 1 and 6% ± 1 from the original images. Much larger changes were observed in areas with vessels. Reproducibility of RNFL thickness was improved, specifically in the higher density vessel location, i.e. inferior and superior. Therefore, OCT images can be enhanced by an image processing procedure. Vessel artefacts may cause errors in assessment of RNFL thickness and are a source of variability, which has clinical implications for diseases such as glaucoma where subtle changes in RNFL need to be monitored accurately over time.

Dynamic association between intraocular pressure and spontaneous pulsations of retinal veins.

PURPOSE: The amplitude of spontaneous retinal venous pulsations (SRVP) is known to be affected by intraocular pressure (IOP), retinal venous pressure, and intracranial pressure (ICP). This study characterized SRVPs adjacent to the disc and quantified changes in the amplitude of these pulsations during IOP manipulation in normal subjects. METHODS: The study included 12 subjects (40 ± 15, 4 females, 8 males). Baseline IOP (range 10-25 mmHg) was measured and SRVP recorded using the dynamic retinal vessel analyzer (DVA). IOP was lowered using aproclonidine 0.5% and measured every 15 min, followed by dynamic recording of SRVP. Two subjects were also tested with timolol 0.5%, and three were treated with a placebo drop. Mean amplitude of SRVP was determined within each sample at the same site. Blood pressure and heart rate were tracked continuously. RESULTS: Amplitude of SRVP decreased in all subjects with reduction of IOP with aproclonidine and timolol. Mean SRVP amplitude was 8.5 ± 6 µm at baseline and reduced to 2.5 ± 1.8 µm after 45 min (p < 0.0001). IOP fell from 14.4 ± 2.6 mmHg to 10.2 ± 2.9 mmHg over the same period (p < 0.001). Venous diameter, blood pressure, and heart rate did not change significantly from the baseline. Analysis of waveforms showed a slight phase shift only (150 ± 78.5 ms, p = 0.93) between disc veins and adjacent retinal vein. CONCLUSION: SRVPs in the peripapillary retina have similar waveform characteristics to those at the disc. SRVP amplitudes are reduced by manipulation of IOP downwards with pharmacological intervention. The relationship was consistent in all individuals tested for two classes of drugs and was independent of BP or heart rate changes.