Choroidalyzer: An Open-Source, End-to-End Pipeline for Choroidal Analysis in Optical Coherence Tomography.

Purpose: To develop Choroidalyzer, an open-source, end-to-end pipeline for segmenting the choroid region, vessels, and fovea, and deriving choroidal thickness, area, and vascular index. Methods: We used 5600 OCT B-scans (233 subjects, six systemic disease cohorts, three device types, two manufacturers). To generate region and vessel ground-truths, we used state-of-the-art automatic methods following manual correction of inaccurate segmentations, with foveal positions manually annotated. We trained a U-Net deep learning model to detect the region, vessels, and fovea to calculate choroid thickness, area, and vascular index in a fovea-centered region of interest. We analyzed segmentation agreement (AUC, Dice) and choroid metrics agreement (Pearson, Spearman, mean absolute error [MAE]) in internal and external test sets. We compared Choroidalyzer to two manual graders on a small subset of external test images and examined cases of high error. Results: Choroidalyzer took 0.299 seconds per image on a standard laptop and achieved excellent region (Dice: internal 0.9789, external 0.9749), very good vessel segmentation performance (Dice: internal 0.8817, external 0.8703), and excellent fovea location prediction (MAE: internal 3.9 pixels, external 3.4 pixels). For thickness, area, and vascular index, Pearson correlations were 0.9754, 0.9815, and 0.8285 (internal)/0.9831, 0.9779, 0.7948 (external), respectively (all P < 0.0001). Choroidalyzer's agreement with graders was comparable to the intergrader agreement across all metrics. Conclusions: Choroidalyzer is an open-source, end-to-end pipeline that accurately segments the choroid and reliably extracts thickness, area, and vascular index. Especially choroidal vessel segmentation is a difficult and subjective task, and fully automatic methods like Choroidalyzer could provide objectivity and standardization.

PallorMetrics: Software for Automatically Quantifying Optic Disc Pallor in Fundus Photographs, and Associations With Peripapillary RNFL Thickness.

Purpose: We sough to develop an automatic method of quantifying optic disc pallor in fundus photographs and determine associations with peripapillary retinal nerve fiber layer (pRNFL) thickness. Methods: We used deep learning to segment the optic disc, fovea, and vessels in fundus photographs, and measured pallor. We assessed the relationship between pallor and pRNFL thickness derived from optical coherence tomography scans in 118 participants. Separately, we used images diagnosed by clinical inspection as pale (n = 45) and assessed how measurements compared with healthy controls (n = 46). We also developed automatic rejection thresholds and tested the software for robustness to camera type, image format, and resolution. Results: We developed software that automatically quantified disc pallor across several zones in fundus photographs. Pallor was associated with pRNFL thickness globally (ß = -9.81; standard error [SE] = 3.16; P < 0.05), in the temporal inferior zone (ß = -29.78; SE = 8.32; P < 0.01), with the nasal/temporal ratio (ß = 0.88; SE = 0.34; P < 0.05), and in the whole disc (ß = -8.22; SE = 2.92; P < 0.05). Furthermore, pallor was significantly higher in the patient group. Last, we demonstrate the analysis to be robust to camera type, image format, and resolution. Conclusions: We developed software that automatically locates and quantifies disc pallor in fundus photographs and found associations between pallor measurements and pRNFL thickness. Translational Relevance: We think our method will be useful for the identification, monitoring, and progression of diseases characterized by disc pallor and optic atrophy, including glaucoma, compression, and potentially in neurodegenerative disorders.

Ultra-Widefield Imaging of the Retinal Macrovasculature in Parkinson Disease Versus Controls With Normal Cognition Using Alpha-Shapes Analysis.

Purpose: To investigate retinal vascular characteristics using ultra-widefield (UWF) scanning laser ophthalmoscopy in Parkinson disease (PD). Methods: Individuals with an expert-confirmed clinical diagnosis of PD and controls with normal cognition without PD underwent Optos California UWF imaging. Patients with diabetes, uncontrolled hypertension, glaucoma, dementia, other movement disorders, or known retinal or optic nerve pathology were excluded. Images were analyzed using Vasculature Assessment and Measurement Platform for Images of the Retina (VAMPIRE-UWF) software, which describes retinal vessel width gradient and tortuosity, provides vascular network fractal dimensions, and conducts alpha-shape analysis to further characterize vascular morphology (complexity, Opαmin; spread, OpA). Results: In the PD cohort, 53 eyes of 38 subjects were assessed; in the control cohort, 51 eyes of 33 subjects were assessed. Eyes with PD had more tortuous retinal arteries in the superotemporal quadrant (P = 0.043). In eyes with PD, alpha-shape analysis revealed decreased OpA, indicating less retinal vasculature spread compared to controls (P = 0.032). Opαmin was decreased in PD (P = 0.044), suggesting increased vascular network complexity. No differences were observed in fractal dimension in any region of interest. Conclusions: This pilot study suggests that retinal vasculature assessment on UWF images using alpha-shape analysis reveals differences in retinal vascular network spread and complexity in PD and may be a more sensitive metric compared to fractal dimension. Translational Relevance: Retinal vasculature assessment using these novel methods may be useful in understanding ocular manifestations of PD and the development of retinal biomarkers.

Evaluation of an Automated Choroid Segmentation Algorithm in a Longitudinal Kidney Donor and Recipient Cohort.

Purpose: To evaluate the performance of an automated choroid segmentation algorithm in optical coherence tomography (OCT) data using a longitudinal kidney donor and recipient cohort. Methods: We assessed 22 donors and 23 patients requiring renal transplantation over up to 1 year posttransplant. We measured choroidal thickness (CT) and area and compared our automated CT measurements to manual ones at the same locations. We estimated associations between choroidal measurements and markers of renal function (estimated glomerular filtration rate [eGFR], serum creatinine, and urea) using correlation and linear mixed-effects (LME) modeling. Results: There was good agreement between manual and automated CT. Automated measures were more precise because of smaller measurement error over time. External adjudication of major discrepancies was in favor of automated measures. Significant differences were observed in the choroid pre- and posttransplant in both cohorts, and LME modeling revealed significant linear associations observed between choroidal measures and renal function in recipients. Significant associations were mostly stronger with automated CT (eGFR, P < 0.001; creatinine, P = 0.004; urea, P = 0.04) compared to manual CT (eGFR, P = 0.002; creatinine, P = 0.01; urea, P = 0.03). Conclusions: Our automated approach has greater precision than human-performed manual measurements, which may explain stronger associations with renal function compared to manual measurements. To improve detection of meaningful associations with clinical endpoints in longitudinal studies of OCT, reducing measurement error should be a priority, and automated measurements help achieve this. Translational Relevance: We introduce a novel choroid segmentation algorithm that can replace manual grading for studying the choroid in renal disease and other clinical conditions.

An Open-Source Deep Learning Algorithm for Efficient and Fully Automatic Analysis of the Choroid in Optical Coherence Tomography.

Purpose: To develop an open-source, fully automatic deep learning algorithm, DeepGPET, for choroid region segmentation in optical coherence tomography (OCT) data. Methods: We used a dataset of 715 OCT B-scans (82 subjects, 115 eyes) from three clinical studies related to systemic disease. Ground-truth segmentations were generated using a clinically validated, semiautomatic choroid segmentation method, Gaussian Process Edge Tracing (GPET). We finetuned a U-Net with the MobileNetV3 backbone pretrained on ImageNet. Standard segmentation agreement metrics, as well as derived measures of choroidal thickness and area, were used to evaluate DeepGPET, alongside qualitative evaluation from a clinical ophthalmologist. Results: DeepGPET achieved excellent agreement with GPET on data from three clinical studies (AUC = 0.9994, Dice = 0.9664; Pearson correlation = 0.8908 for choroidal thickness and 0.9082 for choroidal area), while reducing the mean processing time per image on a standard laptop CPU from 34.49 ± 15.09 seconds using GPET to 1.25 ± 0.10 seconds using DeepGPET. Both methods performed similarly according to a clinical ophthalmologist who qualitatively judged a subset of segmentations by GPET and DeepGPET, based on smoothness and accuracy of segmentations. Conclusions: DeepGPET, a fully automatic, open-source algorithm for choroidal segmentation, will enable researchers to efficiently extract choroidal measurements, even for large datasets. As no manual interventions are required, DeepGPET is less subjective than semiautomatic methods and could be deployed in clinical practice without requiring a trained operator. Translational Relevance: DeepGPET addresses the lack of open-source, fully automatic, and clinically relevant choroid segmentation algorithms, and its subsequent public release will facilitate future choroidal research in both ophthalmology and wider systemic health.

Retinal Vascular Changes in Alzheimer's Dementia and Mild Cognitive Impairment: A Pilot Study Using Ultra-Widefield Imaging.

Purpose: Retinal microvascular abnormalities measured on retinal images are a potential source of prognostic biomarkers of vascular changes in the neurodegenerating brain. We assessed the presence of these abnormalities in Alzheimer's dementia and mild cognitive impairment (MCI) using ultra-widefield (UWF) retinal imaging. Methods: UWF images from 103 participants (28 with Alzheimer's dementia, 30 with MCI, and 45 with normal cognition) underwent analysis to quantify measures of retinal vascular branching complexity, width, and tortuosity. Results: Participants with Alzheimer's dementia displayed increased vessel branching in the midperipheral retina and increased arteriolar thinning. Participants with MCI displayed increased rates of arteriolar and venular thinning and a trend for decreased vessel branching. Conclusions: Statistically significant differences in the retinal vasculature in peripheral regions of the retina were observed among the distinct cognitive stages. However, larger studies are required to establish the clinical importance of our findings. UWF imaging may be a promising modality to assess a larger view of the retinal vasculature to uncover retinal changes in Alzheimer's disease. Translational Relevance: This pilot work reports an investigation into which retinal vasculature measurements may be useful surrogate measures of cognitive decline, as well as technical developments (e.g., measurement standardization), that are first required to establish their recommended use and translational potential.

Retinal capillary microvessel morphology changes are associated with vascular damage and dysfunction in cerebral small vessel disease.

Cerebral small vessel disease (SVD) is a cause of stroke and dementia. Retinal capillary microvessels revealed by optical coherence tomography angiography (OCTA) are developmentally related to brain microvessels. We quantified retinal vessel density (VD) and branching complexity, investigating relationships with SVD lesions, white matter integrity on diffusion tensor imaging (DTI) and cerebrovascular reactivity (CVR) to CO2 in patients with minor stroke. We enrolled 123 patients (mean age 68.1 ± SD 9.9 years), 115 contributed retinal data. Right (R) and left (L) eyes are reported. After adjusting for age, eye disease, diabetes, blood pressure and image quality, lower VD remained associated with higher mean diffusivity (MD) (standardized ß; R -0.16 [95%CI -0.32 to -0.01]) and lower CVR (L 0.17 [0.03 to 0.31] and R 0.19 [0.02 to 0.36]) in normal appearing white matter (NAWM). Sparser branching remained associated with sub-visible white matter damage shown by higher MD (R -0.24 [-0.08 to -0.40]), lower fractional anisotropy (FA) (L 0.17 [0.01 to 0.33]), and lower CVR (R 0.20 [0.02 to 0.38]) in NAWM. OCTA-derived metrics provide evidence of microvessel abnormalities that may underpin SVD lesions in the brain.

Measuring axial length of the eye from magnetic resonance brain imaging.

BACKGROUND: Metrics derived from the human eye are increasingly used as biomarkers and endpoints in studies of cardiovascular, cerebrovascular and neurological disease. In this context, it is important to account for potential confounding that can arise from differences in ocular dimensions between individuals, for example, differences in globe size. METHODS: We measured axial length, a geometric parameter describing eye size from T2-weighted brain MRI scans using three different image analysis software packages (Mango, ITK and Carestream) and compared results to biometry measurements from a specialized ophthalmic instrument (IOLMaster 500) as the reference standard. RESULTS: Ninety-three healthy research participants of mean age 51.0 ± SD 5.4 years were analyzed. The level of agreement between the MRI-derived measurements and the reference standard was described by mean differences as follows, Mango - 0.8 mm; ITK - 0.5 mm; and Carestream - 0.1 mm (upper/lower 95% limits of agreement across the three tools ranged from 0.9 mm to - 2.6 mm). Inter-rater reproducibility was between - 0.03 mm and 0.45 mm (ICC 0.65 to 0.93). Intra-rater repeatability was between 0.0 mm and - 0.2 mm (ICC 0.90 to 0.95). CONCLUSIONS: We demonstrate that axial measurements of the eye derived from brain MRI are within 3.5% of the reference standard globe length of 24.1 mm. However, the limits of agreement could be considered clinically significant. Axial length of the eye obtained from MRI is not a replacement for the precision of biometry, but in the absence of biometry it could provide sufficient accuracy to act as a proxy. We recommend measuring eye axial length from MRI in studies that do not have biometry but use retinal imaging to study neurodegenerative changes so as to control for differing eye size across individuals.

Rationale and design of a longitudinal study of cerebral small vessel diseases, clinical and imaging outcomes in patients presenting with mild ischaemic stroke: Mild Stroke Study 3.

BACKGROUND: Cerebral small vessel disease is a major cause of dementia and stroke, visible on brain magnetic resonance imaging. Recent data suggest that small vessel disease lesions may be dynamic, damage extends into normal-appearing brain and microvascular dysfunctions include abnormal blood-brain barrier leakage, vasoreactivity and pulsatility, but much remains unknown regarding underlying pathophysiology, symptoms, clinical features and risk factors of small vessel disease.Patients and Methods: The Mild Stroke Study 3 is a prospective observational cohort study to identify risk factors for and clinical implications of small vessel disease progression and regression among up to 300 adults with non-disabling stroke. We perform detailed serial clinical, cognitive, lifestyle, physiological, retinal and brain magnetic resonance imaging assessments over one year; we assess cerebrovascular reactivity, blood flow, pulsatility and blood-brain barrier leakage on magnetic resonance imaging at baseline; we follow up to four years by post and phone. The study is registered ISRCTN 12113543. SUMMARY: Factors which influence direction and rate of change of small vessel disease lesions are poorly understood. We investigate the role of small vessel dysfunction using advanced serial neuroimaging in a deeply phenotyped cohort to increase understanding of the natural history of small vessel disease, identify those at highest risk of early disease progression or regression and uncover novel targets for small vessel disease prevention and therapy.