Toward safer ophthalmic artificial intelligence via distributed validation on real-world data.

PURPOSE OF REVIEW: The current article provides an overview of the present approaches to algorithm validation, which are variable and largely self-determined, as well as solutions to address inadequacies. RECENT FINDINGS: In the last decade alone, numerous machine learning applications have been proposed for ophthalmic diagnosis or disease monitoring. Remarkably, of these, less than 15 have received regulatory approval for implementation into clinical practice. Although there exists a vast pool of structured and relatively clean datasets from which to develop and test algorithms in the computational 'laboratory', real-world validation remains key to allow for safe, equitable, and clinically reliable implementation. Bottlenecks in the validation process stem from a striking paucity of regulatory guidance surrounding safety and performance thresholds, lack of oversight on critical postdeployment monitoring and context-specific recalibration, and inherent complexities of heterogeneous disease states and clinical environments. Implementation of secure, third-party, unbiased, pre and postdeployment validation offers the potential to address existing shortfalls in the validation process. SUMMARY: Given the criticality of validation to the algorithm pipeline, there is an urgent need for developers, machine learning researchers, and end-user clinicians to devise a consensus approach, allowing for the rapid introduction of safe, equitable, and clinically valid machine learning implementations.

Democratizing Artificial Intelligence Imaging Analysis With Automated Machine Learning: Tutorial.

Deep learning-based clinical imaging analysis underlies diagnostic artificial intelligence (AI) models, which can match or even exceed the performance of clinical experts, having the potential to revolutionize clinical practice. A wide variety of automated machine learning (autoML) platforms lower the technical barrier to entry to deep learning, extending AI capabilities to clinicians with limited technical expertise, and even autonomous foundation models such as multimodal large language models. Here, we provide a technical overview of autoML with descriptions of how autoML may be applied in education, research, and clinical practice. Each stage of the process of conducting an autoML project is outlined, with an emphasis on ethical and technical best practices. Specifically, data acquisition, data partitioning, model training, model validation, analysis, and model deployment are considered. The strengths and limitations of available code-free, code-minimal, and code-intensive autoML platforms are considered. AutoML has great potential to democratize AI in medicine, improving AI literacy by enabling "hands-on" education. AutoML may serve as a useful adjunct in research by facilitating rapid testing and benchmarking before significant computational resources are committed. AutoML may also be applied in clinical contexts, provided regulatory requirements are met. The abstraction by autoML of arduous aspects of AI engineering promotes prioritization of data set curation, supporting the transition from conventional model-driven approaches to data-centric development. To fulfill its potential, clinicians must be educated on how to apply these technologies ethically, rigorously, and effectively; this tutorial represents a comprehensive summary of relevant considerations.

Evaluating an automated machine learning model that predicts visual acuity outcomes in patients with neovascular age-related macular degeneration.

PURPOSE: Neovascular age-related macular degeneration (nAMD) is a major global cause of blindness. Whilst anti-vascular endothelial growth factor (anti-VEGF) treatment is effective, response varies considerably between individuals. Thus, patients face substantial uncertainty regarding their future ability to perform daily tasks. In this study, we evaluate the performance of an automated machine learning (AutoML) model which predicts visual acuity (VA) outcomes in patients receiving treatment for nAMD, in comparison to a manually coded model built using the same dataset. Furthermore, we evaluate model performance across ethnic groups and analyse how the models reach their predictions. METHODS: Binary classification models were trained to predict whether patients' VA would be 'Above' or 'Below' a score of 70 one year after initiating treatment, measured using the Early Treatment Diabetic Retinopathy Study (ETDRS) chart. The AutoML model was built using the Google Cloud Platform, whilst the bespoke model was trained using an XGBoost framework. Models were compared and analysed using the What-if Tool (WIT), a novel model-agnostic interpretability tool. RESULTS: Our study included 1631 eyes from patients attending Moorfields Eye Hospital. The AutoML model (area under the curve [AUC], 0.849) achieved a highly similar performance to the XGBoost model (AUC, 0.847). Using the WIT, we found that the models over-predicted negative outcomes in Asian patients and performed worse in those with an ethnic category of Other. Baseline VA, age and ethnicity were the most important determinants of model predictions. Partial dependence plot analysis revealed a sigmoidal relationship between baseline VA and the probability of an outcome of 'Above'. CONCLUSION: We have described and validated an AutoML-WIT pipeline which enables clinicians with minimal coding skills to match the performance of a state-of-the-art algorithm and obtain explainable predictions.

Quantitative Analysis of OCT for Neovascular Age-Related Macular Degeneration Using Deep Learning.

PURPOSE: To apply a deep learning algorithm for automated, objective, and comprehensive quantification of OCT scans to a large real-world dataset of eyes with neovascular age-related macular degeneration (AMD) and make the raw segmentation output data openly available for further research. DESIGN: Retrospective analysis of OCT images from the Moorfields Eye Hospital AMD Database. PARTICIPANTS: A total of 2473 first-treated eyes and 493 second-treated eyes that commenced therapy for neovascular AMD between June 2012 and June 2017. METHODS: A deep learning algorithm was used to segment all baseline OCT scans. Volumes were calculated for segmented features such as neurosensory retina (NSR), drusen, intraretinal fluid (IRF), subretinal fluid (SRF), subretinal hyperreflective material (SHRM), retinal pigment epithelium (RPE), hyperreflective foci (HRF), fibrovascular pigment epithelium detachment (fvPED), and serous PED (sPED). Analyses included comparisons between first- and second-treated eyes by visual acuity (VA) and race/ethnicity and correlations between volumes. MAIN OUTCOME MEASURES: Volumes of segmented features (mm3) and central subfield thickness (CST) (µm). RESULTS: In first-treated eyes, the majority had both IRF and SRF (54.7%). First-treated eyes had greater volumes for all segmented tissues, with the exception of drusen, which was greater in second-treated eyes. In first-treated eyes, older age was associated with lower volumes for RPE, SRF, NSR, and sPED; in second-treated eyes, older age was associated with lower volumes of NSR, RPE, sPED, fvPED, and SRF. Eyes from Black individuals had higher SRF, RPE, and serous PED volumes compared with other ethnic groups. Greater volumes of the majority of features were associated with worse VA. CONCLUSIONS: We report the results of large-scale automated quantification of a novel range of baseline features in neovascular AMD. Major differences between first- and second-treated eyes, with increasing age, and between ethnicities are highlighted. In the coming years, enhanced, automated OCT segmentation may assist personalization of real-world care and the detection of novel structure-function correlations. These data will be made publicly available for replication and future investigation by the AMD research community.

Automated deep learning in ophthalmology: AI that can build AI.

PURPOSE OF REVIEW: The purpose of this review is to describe the current status of automated deep learning in healthcare and to explore and detail the development of these models using commercially available platforms. We highlight key studies demonstrating the effectiveness of this technique and discuss current challenges and future directions of automated deep learning. RECENT FINDINGS: There are several commercially available automated deep learning platforms. Although specific features differ between platforms, they utilise the common approach of supervised learning. Ophthalmology is an exemplar speciality in the area, with a number of recent proof-of-concept studies exploring classification of retinal fundus photographs, optical coherence tomography images and indocyanine green angiography images. Automated deep learning has also demonstrated impressive results in other specialities such as dermatology, radiology and histopathology. SUMMARY: Automated deep learning allows users without coding expertise to develop deep learning algorithms. It is rapidly establishing itself as a valuable tool for those with limited technical experience. Despite residual challenges, it offers considerable potential in the future of patient management, clinical research and medical education. VIDEO ABSTRACT: http://links.lww.com/COOP/A44.

Clinician-driven artificial intelligence in ophthalmology: resources enabling democratization.

PURPOSE OF REVIEW: This article aims to discuss the current state of resources enabling the democratization of artificial intelligence (AI) in ophthalmology. RECENT FINDINGS: Open datasets, efficient labeling techniques, code-free automated machine learning (AutoML) and cloud-based platforms for deployment are resources that enable clinicians with scarce resources to drive their own AI projects. SUMMARY: Clinicians are the use-case experts who are best suited to drive AI projects tackling patient-relevant outcome measures. Taken together, open datasets, efficient labeling techniques, code-free AutoML and cloud platforms break the barriers for clinician-driven AI. As AI becomes increasingly democratized through such tools, clinicians and patients stand to benefit greatly.

The retina revolution: signaling pathway therapies, genetic therapies, mitochondrial therapies, artificial intelligence.

PURPOSE OF REVIEW: The aim of this article is to review and discuss the history, current state, and future implications of promising biomedical offerings in the field of retina. RECENT FINDINGS: The technologies discussed are some of the more recent promising biomedical developments within the field of retina. There is a US Food and Drug Administration-approved gene therapy product and artificial intelligence device for retina, with many other offerings in the pipeline. SUMMARY: Signaling pathway therapies, genetic therapies, mitochondrial therapies, and artificial intelligence have shaped retina care as we know it and are poised to further impact the future of retina care. Retina specialists have the privilege and responsibility of shaping this future for the visual health of current and future generations.

STEM CELL THERAPIES, GENE-BASED THERAPIES, OPTOGENETICS, AND RETINAL PROSTHETICS: Current State and Implications for the Future.

PURPOSE: To review and discuss current innovations and future implications of promising biotechnology and biomedical offerings in the field of retina. We focus on therapies that have already emerged as clinical offerings or are poised to do so. METHODS: Literature review and commentary focusing on stem cell therapies, gene-based therapies, optogenetic therapies, and retinal prosthetic devices. RESULTS: The technologies discussed herein are some of the more recent promising biotechnology and biomedical developments within the field of retina. Retinal prosthetic devices and gene-based therapies both have an FDA-approved product for ophthalmology, and many other offerings (including optogenetics) are in the pipeline. Stem cell therapies offer personalized medicine through novel regenerative mechanisms but entail complex ethical and reimbursement challenges. CONCLUSION: Stem cell therapies, gene-based therapies, optogenetics, and retinal prosthetic devices represent a new era of biotechnological and biomedical progress. These bring new ethical, regulatory, care delivery, and reimbursement challenges. By addressing these issues proactively, we may accelerate delivery of care to patients in a safe, efficient, and value-based manner.

Image quality assessment of retinal fundus photographs for diabetic retinopathy in the machine learning era: a review.

This study aimed to evaluate the image quality assessment (IQA) and quality criteria employed in publicly available datasets for diabetic retinopathy (DR). A literature search strategy was used to identify relevant datasets, and 20 datasets were included in the analysis. Out of these, 12 datasets mentioned performing IQA, but only eight specified the quality criteria used. The reported quality criteria varied widely across datasets, and accessing the information was often challenging. The findings highlight the importance of IQA for AI model development while emphasizing the need for clear and accessible reporting of IQA information. The study suggests that automated quality assessments can be a valid alternative to manual labeling and emphasizes the importance of establishing quality standards based on population characteristics, clinical use, and research purposes. In conclusion, image quality assessment is important for AI model development; however, strict data quality standards must not limit data sharing. Given the importance of IQA for developing, validating, and implementing deep learning (DL) algorithms, it's recommended that this information be reported in a clear, specific, and accessible way whenever possible. Automated quality assessments are a valid alternative to the traditional manual labeling process, and quality standards should be determined according to population characteristics, clinical use, and research purpose.

Clinician-Driven AI: Code-Free Self-Training on Public Data for Diabetic Retinopathy Referral.

Importance: Democratizing artificial intelligence (AI) enables model development by clinicians with a lack of coding expertise, powerful computing resources, and large, well-labeled data sets. Objective: To determine whether resource-constrained clinicians can use self-training via automated machine learning (ML) and public data sets to design high-performing diabetic retinopathy classification models. Design, Setting, and Participants: This diagnostic quality improvement study was conducted from January 1, 2021, to December 31, 2021. A self-training method without coding was used on 2 public data sets with retinal images from patients in France (Messidor-2 [n = 1748]) and the UK and US (EyePACS [n = 58 689]) and externally validated on 1 data set with retinal images from patients of a private Egyptian medical retina clinic (Egypt [n = 210]). An AI model was trained to classify referable diabetic retinopathy as an exemplar use case. Messidor-2 images were assigned adjudicated labels available on Kaggle; 4 images were deemed ungradable and excluded, leaving 1744 images. A total of 300 images randomly selected from the EyePACS data set were independently relabeled by 3 blinded retina specialists using the International Classification of Diabetic Retinopathy protocol for diabetic retinopathy grade and diabetic macular edema presence; 19 images were deemed ungradable, leaving 281 images. Data analysis was performed from February 1 to February 28, 2021. Exposures: Using public data sets, a teacher model was trained with labeled images using supervised learning. Next, the resulting predictions, termed pseudolabels, were used on an unlabeled public data set. Finally, a student model was trained with the existing labeled images and the additional pseudolabeled images. Main Outcomes and Measures: The analyzed metrics for the models included the area under the receiver operating characteristic curve (AUROC), accuracy, sensitivity, specificity, and F1 score. The Fisher exact test was performed, and 2-tailed P values were calculated for failure case analysis. Results: For the internal validation data sets, AUROC values for performance ranged from 0.886 to 0.939 for the teacher model and from 0.916 to 0.951 for the student model. For external validation of automated ML model performance, AUROC values and accuracy were 0.964 and 93.3% for the teacher model, 0.950 and 96.7% for the student model, and 0.890 and 94.3% for the manually coded bespoke model, respectively. Conclusions and Relevance: These findings suggest that self-training using automated ML is an effective method to increase both model performance and generalizability while decreasing the need for costly expert labeling. This approach advances the democratization of AI by enabling clinicians without coding expertise or access to large, well-labeled private data sets to develop their own AI models.

Retinal Optical Coherence Tomography Features Associated With Incident and Prevalent Parkinson Disease.

BACKGROUND AND OBJECTIVES: Cadaveric studies have shown disease-related neurodegeneration and other morphological abnormalities in the retina of individuals with Parkinson disease (PD); however, it remains unclear whether this can be reliably detected with in vivo imaging. We investigated inner retinal anatomy, measured using optical coherence tomography (OCT), in prevalent PD and subsequently assessed the association of these markers with the development of PD using a prospective research cohort. METHODS: This cross-sectional analysis used data from 2 studies. For the detection of retinal markers in prevalent PD, we used data from AlzEye, a retrospective cohort of 154,830 patients aged 40 years and older attending secondary care ophthalmic hospitals in London, United Kingdom, between 2008 and 2018. For the evaluation of retinal markers in incident PD, we used data from UK Biobank, a prospective population-based cohort where 67,311 volunteers aged 40-69 years were recruited between 2006 and 2010 and underwent retinal imaging. Macular retinal nerve fiber layer (mRNFL), ganglion cell-inner plexiform layer (GCIPL), and inner nuclear layer (INL) thicknesses were extracted from fovea-centered OCT. Linear mixed-effects models were fitted to examine the association between prevalent PD and retinal thicknesses. Hazard ratios for the association between time to PD diagnosis and retinal thicknesses were estimated using frailty models. RESULTS: Within the AlzEye cohort, there were 700 individuals with prevalent PD and 105,770 controls (mean age 65.5 ± 13.5 years, 51.7% female). Individuals with prevalent PD had thinner GCIPL (-2.12 µm, 95% CI -3.17 to -1.07, p = 8.2 × 10-5) and INL (-0.99 µm, 95% CI -1.52 to -0.47, p = 2.1 × 10-4). The UK Biobank included 50,405 participants (mean age 56.1 ± 8.2 years, 54.7% female), of whom 53 developed PD at a mean of 2,653 ± 851 days. Thinner GCIPL (hazard ratio [HR] 0.62 per SD increase, 95% CI 0.46-0.84, p = 0.002) and thinner INL (HR 0.70, 95% CI 0.51-0.96, p = 0.026) were also associated with incident PD. DISCUSSION: Individuals with PD have reduced thickness of the INL and GCIPL of the retina. Involvement of these layers several years before clinical presentation highlight a potential role for retinal imaging for at-risk stratification of PD.

New meaning for NLP: the trials and tribulations of natural language processing with GPT-3 in ophthalmology.

Natural language processing (NLP) is a subfield of machine intelligence focused on the interaction of human language with computer systems. NLP has recently been discussed in the mainstream media and the literature with the advent of Generative Pre-trained Transformer 3 (GPT-3), a language model capable of producing human-like text. The release of GPT-3 has also sparked renewed interest on the applicability of NLP to contemporary healthcare problems. This article provides an overview of NLP models, with a focus on GPT-3, as well as discussion of applications specific to ophthalmology. We also outline the limitations of GPT-3 and the challenges with its integration into routine ophthalmic care.

Wide-Field Swept-Source Optical Coherence Tomography of Optic Disc Pit Maculopathy Demonstrates Connection Between Vitreous and Subretinal Fluid.

The source of subretinal or intraretinal fluid in patients with optic disc pit maculopathy (ODP-M) remains unclear and is often thought to be either vitreous or cerebrospinal fluid.1 Here, we present the case of a 40-year-old man who developed ODP-M. Further imaging with wide-field swept-source optical coherence tomography demonstrated that the macular fluid was tracking from a nasal optic disc pit with superonasal communication to the vitreous. This suggests that swept-source optical coherence tomography can be a useful tool for determining the origin of macular fluid in patients with ODP-M. [Ophthalmic Surg Lasers Imaging Retina 2022;53:579-581.].

Multimodal imaging reveals retinoschisis masquerading as retinal detachment in patients with choroideremia.

Purpose: To report three cases of retinoschisis in patients with intermediate to advanced choroideremia. Observations: Three patients were referred for evaluation of retinal detachment in the context of an inherited retinal degenerative disease. In all three cases, patients carried variants in the CHM gene suspected to be pathogenic and exhibited the characteristic findings of choroideremia, including pigment clumping and chorioretinal atrophy with scleral exposure and prominent choroidal vessels. Interestingly, these patients were also found to have areas of typical retinoschisis and cystoid degeneration located in the outer plexiform layer of the mid periphery or macula. Retinoschisis was confirmed by spectral domain optical coherence tomography (SD-OCT). Conclusions/Importance: This paper draws attention to the occurrence of retinoschisis in patients with choroideremia. OCT can be used to confirm the presence of retinoschisis rather than retinal detachment, as the clinical exam findings that distinguish the two conditions are not helpful in the setting of advanced chorioretinal atrophy. Although it remains unclear whether patients with choroideremia as a group are at increased risk of retinoschisis, it is possible that abnormal vesicular traffic in the RPE and photoreceptors could contribute to abnormalities in cell adhesion and the extracellular matrix. As gene therapy by subretinal injection of adeno-associated virus becomes the standard of care to slow down or arrest retinal degeneration in choroideremia, it will be critical to carefully screen these patients for retinoschisis prior to surgical intervention and to incorporate any such findings into surgical planning.

Modeling absolute zone size in retinopathy of prematurity in relation to axial length.

Treatment outcomes in retinopathy of prematurity (ROP) are closely correlated with the location (i.e. zone) of disease, with more posterior zones having poorer outcomes. The most posterior zone, Zone I, is defined as a circle centered on the optic nerve with radius twice the distance from nerve to fovea, or subtending an angle of 30 degrees. Because the eye enlarges and undergoes refractive changes during the period of ROP screening, the absolute area of Zone I according to these definitions may likewise change. It is possible that these differences may confound accurate assessment of risk in patients with ROP. In this study, we estimated the area of Zone I in relation to different ocular parameters to determine how variability in the size and refractive power of the eye may affect zoning. Using Gaussian optics, a model was constructed to calculate the absolute area of Zone I as a function of corneal power, anterior chamber depth, lens power, lens thickness, and axial length (AL), with Zone I defined as a circle with radius set by a 30-degree visual angle. Our model predicted Zone I area to be most sensitive to changes in AL; for example, an increase of AL from 14.20 to 16.58 mm at postmenstrual age 32 weeks was calculated to expand the area of Zone I by up to 72%. These findings motivate several hypotheses which upon future testing may help optimize treatment decisions for ROP.

Re-evaluating diabetic papillopathy using optical coherence tomography and inner retinal sublayer analysis.

BACKGROUND/OBJECTIVES: To re-evaluate diabetic papillopathy using optical coherence tomography (OCT) for quantitative analysis of the peripapillary retinal nerve fibre layer (pRNFL), macular ganglion cell layer (mGCL) and inner nuclear layer (mINL) thickness. SUBJECTS/METHODS: In this retrospective observational case series between June 2008 and July 2019 at Moorfields Eye hospital, 24 eyes of 22 patients with diabetes and optic disc swelling with confirmed diagnosis of NAION or diabetic papillopathy by neuro-ophthalmological assessment were included for evaluation of the pRNFL, mGCL and mINL thicknesses after resolution of optic disc swelling. RESULTS: The mean age of included patients was 56.5 (standard deviation (SD) ± 14.85) years with a mean follow-up duration of 216 days. Thinning of pRNFL (mean: 66.26, SD ± 31.80 µm) and mGCL (mean volume: 0.27 mm3, SD ± 0.09) were observed in either group during follow-up, the mINL volume showed no thinning with 0.39 ± 0.05 mm3. The mean decrease in visual acuity was 4.13 (SD ± 14.27) ETDRS letters with a strong correlation between mGCL thickness and visual acuity (rho 0.74, p < 0.001). CONCLUSION: After resolution of acute optic disc swelling, atrophy of pRNFL and mGCL became apparent in all cases of diabetic papillopathy and diabetic NAION, with preservation of mINL volumes. Analysis of OCT did not provide a clear diagnostic distinction between both entities. We suggest a diagnostic overlay with the degree of pRNFL and mGCL atrophy of prognostic relevance for poor visual acuity independent of the semantics of terminology.

Reinforcement learning in ophthalmology: potential applications and challenges to implementation.

Reinforcement learning is a subtype of machine learning in which a virtual agent, functioning within a set of predefined rules, aims to maximise a specified outcome or reward. This agent can consider multiple variables and many parallel actions at once to optimise its reward, thereby solving complex, sequential problems. Clinical decision making requires physicians to optimise patient outcomes within a set practice framework and, thus, presents considerable opportunity for the implementation of reinforcement learning-driven solutions. We provide an overview of reinforcement learning, and focus on potential applications within ophthalmology. We also explore the challenges associated with development and implementation of reinforcement learning solutions and discuss possible approaches to address them.

Acute iridocyclitis and cystoid macular edema related to kinked Hydrus® Microstent in advanced glaucoma.

A 58-year-old male who underwent cataract extraction with combined intraocular lens and Hydrus® Microstent (Ivantis Inc, Irvine, CA, US) implantation 2 years ago in the right eye (OD) due to advanced glaucoma presented with blurry vision in right eye (OD) for 3 months. The visual acuity was 20/60 and slit-lamp examination indicated mild anterior chamber inflammation with unexposed, functioning tube shunt superotemporally in OD. Optical coherence tomography demonstrated cystoid macular edema (CME) with subretinal fluid. Fluorescein angiography demonstrated petaloid pattern leakage of CME. Gonioscopy revealed a kinked appearance of a Hydrus® Microstent protruding into the anterior chamber and causing iris chafing. Topical ketorolac tromethamine and prednisolone acetate were started. At the 2nd month of follow-up, the anterior chamber was quiet, and the CME resolved completely. Protruded kinked Hydrus® Microstent may lead to acute iridocyclitis and CME through iris chafing, which may be responsive to topical anti-inflammatory drops.

Enablers and Barriers to Deployment of Smartphone-Based Home Vision Monitoring in Clinical Practice Settings.

IMPORTANCE: Telemedicine is accelerating the remote detection and monitoring of medical conditions, such as vision-threatening diseases. Meaningful deployment of smartphone apps for home vision monitoring should consider the barriers to patient uptake and engagement and address issues around digital exclusion in vulnerable patient populations. OBJECTIVE: To quantify the associations between patient characteristics and clinical measures with vision monitoring app uptake and engagement. DESIGN, SETTING, AND PARTICIPANTS: In this cohort and survey study, consecutive adult patients attending Moorfields Eye Hospital receiving intravitreal injections for retinal disease between May 2020 and February 2021 were included. EXPOSURES: Patients were offered the Home Vision Monitor (HVM) smartphone app to self-test their vision. A patient survey was conducted to capture their experience. App data, demographic characteristics, survey results, and clinical data from the electronic health record were analyzed via regression and machine learning. MAIN OUTCOMES AND MEASURES: Associations of patient uptake, compliance, and use rate measured in odds ratios (ORs). RESULTS: Of 417 included patients, 236 (56.6%) were female, and the mean (SD) age was 72.8 (12.8) years. A total of 258 patients (61.9%) were active users. Uptake was negatively associated with age (OR, 0.98; 95% CI, 0.97-0.998; P = .02) and positively associated with both visual acuity in the better-seeing eye (OR, 1.02; 95% CI, 1.00-1.03; P = .01) and baseline number of intravitreal injections (OR, 1.01; 95% CI, 1.00-1.02; P = .02). Of 258 active patients, 166 (64.3%) fulfilled the definition of compliance. Compliance was associated with patients diagnosed with neovascular age-related macular degeneration (OR, 1.94; 95% CI, 1.07-3.53; P = .002), White British ethnicity (OR, 1.69; 95% CI, 0.96-3.01; P = .02), and visual acuity in the better-seeing eye at baseline (OR, 1.02; 95% CI, 1.01-1.04; P = .04). Use rate was higher with increasing levels of comfort with use of modern technologies (ß = 0.031; 95% CI, 0.007-0.055; P = .02). A total of 119 patients (98.4%) found the app either easy or very easy to use, while 96 (82.1%) experienced increased reassurance from using the app. CONCLUSIONS AND RELEVANCE: This evaluation of home vision monitoring for patients with common vision-threatening disease within a clinical practice setting revealed demographic, clinical, and patient-related factors associated with patient uptake and engagement. These insights inform targeted interventions to address risks of digital exclusion with smartphone-based medical devices.

Predicting sex from retinal fundus photographs using automated deep learning.

Deep learning may transform health care, but model development has largely been dependent on availability of advanced technical expertise. Herein we present the development of a deep learning model by clinicians without coding, which predicts reported sex from retinal fundus photographs. A model was trained on 84,743 retinal fundus photos from the UK Biobank dataset. External validation was performed on 252 fundus photos from a tertiary ophthalmic referral center. For internal validation, the area under the receiver operating characteristic curve (AUROC) of the code free deep learning (CFDL) model was 0.93. Sensitivity, specificity, positive predictive value (PPV) and accuracy (ACC) were 88.8%, 83.6%, 87.3% and 86.5%, and for external validation were 83.9%, 72.2%, 78.2% and 78.6% respectively. Clinicians are currently unaware of distinct retinal feature variations between males and females, highlighting the importance of model explainability for this task. The model performed significantly worse when foveal pathology was present in the external validation dataset, ACC: 69.4%, compared to 85.4% in healthy eyes, suggesting the fovea is a salient region for model performance OR (95% CI): 0.36 (0.19, 0.70) p = 0.0022. Automated machine learning (AutoML) may enable clinician-driven automated discovery of novel insights and disease biomarkers.