Predicting visual field global and local parameters from OCT measurements using explainable machine learning

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 16, 2025

Abstract Glaucoma is characterised by progressive vision loss due to retinal ganglion cell deterioration, leading gradual visual field (VF) impairment. The standard VF test may be impractical in some cases, where optical coherence tomography (OCT) can offer predictive insights into for multimodal diagnoses. However, predicting measures from OCT data remains challenging. To address this, five regression models were developed predict OCT, Shapley Additive exPlanations (SHAP) analysis was performed interpretability, and a clinical software tool called Predictor developed. evaluate the models, total of 268 glaucomatous eyes (86 early, 72 moderate, 110 advanced) 226 normal included. machine learning outperformed recent OCT-based prediction deep studies, with correlation coefficients 0.76, 0.80 0.76 mean deviation, index pattern respectively. Introducing pointwise normalisation step-size concept, absolute error 2.51 dB obtained sensitivity prediction, grayscale model yielded structural similarity 77%. SHAP-based provided critical most relevant features glaucoma diagnosis, showing promise assisting eye care practitioners through an explainable AI tool.

Language: Английский

---

OCT-based diagnosis of glaucoma and glaucoma stages using explainable machine learning

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 28, 2025

Abstract Glaucoma poses a growing health challenge projected to escalate in the coming decades. However, current automated diagnostic approaches on diagnosis solely rely black-box deep learning models, lacking explainability and trustworthiness. To address issue, this study uses optical coherence tomography (OCT) images develop an explainable artificial intelligence (XAI) tool for diagnosing staging glaucoma, with focus its clinical applicability. A total of 334 normal 268 glaucomatous eyes (86 early, 72 moderate, 110 advanced) were included, signal processing theory was employed, model interpretability rigorously evaluated. Leveraging SHapley Additive exPlanations (SHAP)-based global feature ranking partial dependency analysis (PDA) estimated decision boundary cut-offs machine (ML) novel algorithm developed implement XAI tool. Using selected features, ML models produce AUC 0.96 (95% CI: 0.95–0.98), 0.98 0.96–1.00) 1.00 1.00–1.00) respectively differentiating moderate advanced glaucoma patients. Overall, outperformed clinicians early stage overall 10.4 –11.2% higher accuracy. The user-friendly software shows potential as valuable eye care practitioners, offering transparent interpretable insights improve decision-making.

Language: Английский

---

RPE Curvature Can Screen for Early and Intermediate AMD

Investigative Ophthalmology & Visual Science, Journal Year: 2024, Volume and Issue: 65(2), P. 2 - 2

Published: Feb. 1, 2024

Purpose: Diagnosing AMD early optimizes clinical management. However, current diagnostic accuracy is limited by the subjectivity of qualitative measures used in practice. This study tests if RPE curvature could be an accurate, quantitative measure for diagnosis. Methods: Consecutive patients without or normal aging changes (n = 111), with 107), 102) and intermediate 114) were recruited. was calculated based on sinuosity method measuring river environmental science. Bruch's membrane manually segmented from optical coherence tomography B-scans then their lengths automatically extracted using customized MATLAB code. as a ratio to length. Diagnostic determined area under receiver operator characteristic curve (aROC). Results: foveal distinguish any eyes (early intermediate) those (non-AMD changes) acceptable (aROC 0.775). Similarly, identify all other groups 0.871) between 0.737). significantly associated known lesions: reticular pseudodrusen (P < 0.0001) drusen volume 0.0001), but not physiological variables such age, sex, ethnicity. Conclusions: simple, robust, biomarker that amenable automation enhance screening AMD.

Language: Английский

---

Viability of Performing Multiple 24-2 Visual Field Examinations at the Same Clinical Visit: The Frontloading Fields Study (FFS)

American Journal of Ophthalmology, Journal Year: 2021, Volume and Issue: 230, P. 48 - 59

Published: May 2, 2021

To assess the viability, in terms of time taken for testing and repeatability, frontloading (performing multiple perimetric examinations) a single clinic visit.Reliability enhancement analysis.A total 329 healthy glaucoma suspect subjects within undergoing using SITA-Faster twice each eye same session were included. Global indices, pointwise sensitivity probability scores, test duration, reliability metrics analysed.For both tests 9.1% right 6.7% left results unreliable, with 58.4% 67.5% eyes achieving reliable results, respectively; 83.8% all spent less than 20 minutes performing tests. Differences global scores showed no systematic or clinically significant difference between one two eye. There was also number locations identified as defective at P < .05 level Test that unreliable tended to show more instances failed "cluster" criterion not repeatable.Frontloading viable obtaining sets reliable, repeatable data conventional outputs, overcoming practical issues regarding low singleton confirmation visual field defects. Despite need remain cognisant SITA-Faster, this algorithm may be method meeting recommendations required make confident inferences about state progression.

Language: Английский

---

Optical coherence tomography angiography in the diagnosis of ocular disease

Clinical and Experimental Optometry, Journal Year: 2024, Volume and Issue: 107(5), P. 482 - 498

Published: March 7, 2024

Clinical imaging provided by optical coherence tomography (OCT) and its variant, OCT-angiography (OCT-A), has revolutionised eyecare practice. The techniques allow for the identification quantification of ocular structures, supporting diagnosis prognosis eye disease. In this review, an overview usefulness OCT-A in management a range conditions is when used isolation or combination with other modalities measures visual function (visual field results). capacity to identify quantify vasculature non-invasively, thereby assisting clinician determine efficacy intervention impacting retinal vasculature. Thus, additional clinically useful information can be obtained diseases involving such as those vessel occlusion, diabetic retinopathy, inherited dystrophy, age-related macular degeneration, choroidal neovascularisation optic nerve disorders. Through clinical case series, various are reviewed, impact discussed. Although great promise already management, there lack set standards characterise altered vascular features disease consequently prognostication, primarily due large-scale trials variability algorithms generating quantitative parameters.

Language: Английский

---

The Frontloading Fields Study: The Impact of False Positives and Seeding Point Errors on Visual Field Reliability When Using SITA-Faster

Translational Vision Science & Technology, Journal Year: 2022, Volume and Issue: 11(2), P. 20 - 20

Published: Feb. 10, 2022

The purpose of this study was to evaluate the impact two conventional reliability criteria (false positives [FPs] and seeding point errors [SPEs]) concurrent effect low sensitivity points (≤19 dB) on intrasession SITA-Faster visual field (VF) result correlations.There were 2320 VF results from 1160 eyes healthy, glaucoma suspects, subjects with that separated into "both reliable" or "reliable-unreliable" pairs. (mean deviation pointwise sensitivity) analyzed against spectrum FP rates SPE, without censorship ≤19 dB. Segmental linear regression used identify critical where significantly different between tests due levels.There a significant, but small (0.09 dB per 1% exceeding 12%) increase in mean deviation, an number showing >3 (0.25-0.28 locations 12%). SPEs almost exclusively related decrease at primary did not significant differences other indices. Censoring improved correlation reliable unreliable results.Current for judging (FP rate >15% SPE) can lead data being erroneously excluded, as many do show compared those deemed "reliable." improves correlations results.We provide guidelines assessing FP, interpretation.

Language: Английский

---

High-Density Optical Coherence Tomography Analysis Provides Insights Into Early/Intermediate Age-Related Macular Degeneration Retinal Layer Changes

Investigative Ophthalmology & Visual Science, Journal Year: 2022, Volume and Issue: 63(5), P. 36 - 36

Published: May 27, 2022

Purpose: To topographically map all of the thickness differences in individual retinal layers between early/intermediate age-related macular degeneration (AMDearly/AMDint) and normal eyes to determine interlayer relationships. Methods: Ninety-six AMDtotal (48 AMDearly 48 AMDint) 96 from 192 participants were propensity-score matched by age, sex, refraction. Retrospective optical coherence tomography (OCT) cube scans acquired, high-density (60 × 60 0.01-mm2) grid thicknesses custom extracted for comparison corrected confounding. Resultant "normal differences" underwent cluster, correlation, dose–response analyses nerve fiber layer (RNFL), ganglion cell (GCL), inner plexiform (IPL), nuclear (INL), outer (OPL), + Henle's (ONL+HFL), segment (IS/OS) thickness, pigment epithelium (RPE) Bruch's membrane (BM) thickness. Results: clusters demonstrated extensively thinned RNFL, GCL, IPL, paracentral INL thickened elsewhere, with difference means ranging −8.13 µm (95% confidence interval [CI], −11.12 −5.13) 1.58 CI, 1.07–2.09) (P < 0.0001 P 0.05). Outer displayed OPL/ONL+HFL, central IS/OS, peripheral RPE–BM RPE–BM, −1.31 −2.06 −0.55) 2.99 0.97–5.01] <0.05). Effect sizes (−2.56 9.93 SD), cluster sizes, eccentricity effects varied. All correlations negligible moderate regardless AMD severity. Only was partly thicker greater severity (up 5.44 µm; 95% 4.88–6.00; 0.01). Conclusions: From early stage, demonstrate compared unique topographies across layers. Poor highlight that retina inadequately reflects complete health. The clinical importance OCT assessment requires further investigation.

Language: Английский

---

ChatGPT for addressing patient-centred frequently asked questions in glaucoma clinical practice

Ophthalmology Glaucoma, Journal Year: 2024, Volume and Issue: unknown

Published: Oct. 1, 2024

Language: Английский

---

The frontloading approach to meet guideline-recommended visual field testing for glaucoma: time and cost

Ophthalmology Glaucoma, Journal Year: 2025, Volume and Issue: unknown

Published: April 1, 2025

Language: Английский

---

Radial Peripapillary Capillary Plexus Sparing and Underlying Retinal Vascular Impairment in Intermediate Age-Related Macular Degeneration

Investigative Ophthalmology & Visual Science, Journal Year: 2021, Volume and Issue: 62(4), P. 2 - 2

Published: April 1, 2021

To examine location-specific retinal vascular changes in intermediate age-related macular degeneration (iAMD) using age-matched, high-density en face optical coherence tomography angiography (OCTA) cluster analysis.En OCTA images of the 6 × mm area were retrospectively acquired from 60 iAMD eyes and age-matched normal then subdivided into 126 (47.62 47.62 µm) grids within superficial deep complex. Grid-wise vessel perfusion (VP) compared between corresponding 10-yearly age cohort, forming difference plots. Difference plots further separated by normative topographical map spatial clusters (C1-6), derived normaldatabase (n = 236, 20-81 years old).Overall showed decreased VP (-12.19%) complex (-6.44%) to (P < 0.0001 both comparisons). Cluster-based highlighted nonuniform complex, with sparing at nasal macula (corresponding radial peripapillary capillary plexus) versus toward temporal foveal avascular zone (FAZ) (C1-6 all comparisons P 0.0001, except C1 vs. C2P > 0.99 C4 C5P 0.11). The displayed diffusely VP, greater FAZ 0.0001).High-density analysis suggests relative plexus impairment underlying vasculature, supporting potential anterograde transsynaptic iAMD. These data may better guide future diagnostic management protocol

Language: Английский

---