Emergence of non-artificial intelligence digital health innovations in ophthalmology: A systematic review

Enhanced diplopia detection and binocular single vision assessment through virtual reality: A comprehensive study

This study aimed to develop a novel virtual reality (VR)-based binocular single vision (BSV) testing system for the quantitative assessment of diplopia and to evaluate its diagnostic accuracy and stability through clinical research. We first developed a VR-based BSV testing apparatus (VR-BSVT) using Oculus Quest 2 VR glasses and Unity software. The system provides three parameters for assessing subjects' binocular single vision function, and hence their diplopia: VR-BSVF (Virtual Reality-Based Binocular Single Vision Field area), VR-BSVD (Virtual Reality-Based Binocular Single Vision Distance), and VR-BAR (Virtual Reality-Based Binocular Single Vision Field area ratio). Subsequently, we conducted a clinical control study to systematically evaluate the accuracy and stability of VR-BSVT in the quantitative assessment of diplopia. In this comparative study, we recruited 31 visually healthy subjects and 35 patients diagnosed with diplopia. Each participant underwent two VR-BSVT assessments. The diagnostic accuracy of VR-BSVT in identifying diplopia was analyzed using receiver operating characteristic (ROC) curves, Spearman's rank correlation coefficient, and Bland-Altman analyses. Intraclass correlation coefficient (ICC) was employed to measure the diagnostic stability of VR-BSVT. Through human-computer interaction, VR-BSVT could rapidly detect diplopia and assess binocular single vision function, allowing for the detection of diplopia at different test distances. Among the 66 individuals who participated in the study, results from Intraclass correlation coefficient (ICC) for different test distances showed no significant differences in VR-BAR measurements at both near and far distances between healthy volunteers and patients with diplopia (P = 0.988), indicating good stability of VR-BSVT in diagnosing diplopia. Additionally, the VR-BSVF and VR-BSVD metrics were significantly reduced in the diplopia group compared to the healthy controls (P < 0.01). ROC analysis indicated that VR-BSVT could accurately discriminate patients with diplopia.The Bland-Altman plot revealed a 95% agreement range spanning from - 17.70 to 22.86. These results suggest that VR-BSVT has good precision in diagnosing diplopia. The VR-BSVT developed in this study achieves rapid, accurate, and stable detection and assessment of clinical diplopia, and utilizes virtual reality technology to detect diplopia over a larger visual space. With its compactness and portability, VR-BSVT holds promise for facilitating home healthcare and telemedicine in the future.

Digital health and wearable devices for retinal disease monitoring

Digital health is wielding a growing influence across all areas of healthcare, encompassing various facets such as telemedicine, artificial intelligence (AI), and electronic healthcare records. In Ophthalmology, digital health innovations can be broadly divided into four categories: (i) self-monitoring home devices and apps, (ii) virtual and augmented reality visual aids, (iii) AI software, and (iv) wearables. Wearable devices can work in the background, collecting large amounts of objective data while we do our day-to-day activities, which may be ecologically more valid and meaningful to patients than that acquired in traditional hospital settings. They can be a watch, wristband, piece of clothing, glasses, cane, smartphone in our pocket, earphones, or any other device with a sensor that we carry with us. Focusing on retinal diseases, a key challenge in developing novel therapeutics has been to prove a meaningful benefit in patients' lives and the creation of objective patient-centred endpoints in clinical trials. In this review, we will discuss wearable devices collecting different aspects of visual behaviour, visual field, central vision, and functional vision, as well as their potential implementation as outcome measures in research/clinical trial settings. The healthcare landscape is facing a paradigm shift. Clinicians have a key role of collaborating with the development and fine-tuning of digital health innovations, as well as identifying opportunities where they can be leveraged to enhance our understanding of retinal diseases and improve patient outcomes.

User-Centred Design and Development of a Smartphone Application (OverSight) for Digital Phenotyping in Ophthalmology

BACKGROUND

Visual impairment can significantly impact an individual's daily activities. Patients require regular monitoring, typically occurring within hospital eye services. Capacity constraints have necessitated innovative solutions to improve patient care. Existing digital solutions rely on task-based digital home monitoring such as visual acuity testing. These require active involvement from patients and do not typically offer an indication of quality of life. Digital phenotyping refers to the use of personal digital devices to quantify passive behaviour for detecting clinically significant changes in vision and act as biomarkers for disease. Its uniqueness lies in the ability to detect changes passively. The objective was to co-design an accessible smartphone app (OverSight) for the purposes of digital phenotyping in people with sight impairment.

METHODS

Development of OverSight included stakeholder consultations following principles of user-centred design. Apple iOS software frameworks (HealthKit, ResearchKit, and SensorKit) and a SwiftUI developer toolkit were used to enable the collection of active and passive data streams. Accessibility and usability were assessed using the System Usability Scale (SUS) and feedback following a 3-month pilot study. Consultations with patients informed the design of OverSight, including preferred survey scheduling and the relevancy of patient support resources.

RESULTS

Twenty visually impaired participants (mean age 42 ± 19 years) were recruited to the pilot study. The average score on the SUS was 76.8 (±8.9), indicating good usability. There was a statistically significant moderate negative correlation between SUS scores and visual acuity in both the better (r = -0.494; p ≤ 0.001) and worse eye (r = -0.421; p ≤ 0.001).

CONCLUSIONS

OverSight offers promising potential for collecting patient-generated health data for the purposes of digital phenotyping in patients with eye disease. Through further testing and validation, this novel approach to patient care may ultimately provide opportunities for remote monitoring in ophthalmology.

Measuring the visual angle of polarization-related entoptic phenomena using structured light

The ability to perceive polarization-related entoptic phenomena arises from the dichroism of macular pigments held in Henle's fiber layer of the retina and can be inhibited by retinal diseases, such as age-related macular degeneration, which alters the structure of the macula. Structured light tools enable the direct probing of macular pigment density and retinal structure through the perception of polarization-dependent entoptic patterns. Here, we directly measure the visual angle of an entoptic pattern created through the illumination of the retina with a structured state of light and a perception task that is insensitive to corneal birefringence. The central region of the structured light stimuli was obstructed, with the size of the obstruction varying according to a psychophysical staircase. Two stimuli, one producing 11 azimuthal fringes and the other three azimuthal fringes, were presented to 24 healthy participants. The pattern with 11 azimuthal fringes produced an average visual angle threshold of 10° ± 1° and a 95% confidence interval (C.I.) of [6°, 14°]. For the pattern with three azimuthal fringes, a threshold extent of 3.6° ± 0.3° C.I. = [1.3°, 5.8°] was measured, a value similar to the published extent of Haidinger's brush (4°). The increase in apparent size and clarity of entoptic phenomena produced by the presented structured light stimuli offers the potential to detect the early signs of macular disease over perception tasks using uniform polarization stimuli.

Artificial intelligence and digital solutions for myopia

Myopia as an uncorrected visual impairment is recognized as a global public health issue with an increasing burden on health-care systems. Moreover, high myopia increases one's risk of developing pathologic myopia, which can lead to irreversible visual impairment. Thus, increased resources are needed for the early identification of complications, timely intervention to prevent myopia progression, and treatment of complications. Emerging artificial intelligence (AI) and digital technologies may have the potential to tackle these unmet needs through automated detection for screening and risk stratification, individualized prediction, and prognostication of myopia progression. AI applications in myopia for children and adults have been developed for the detection, diagnosis, and prediction of progression. Novel AI technologies, including multimodal AI, explainable AI, federated learning, automated machine learning, and blockchain, may further improve prediction performance, safety, accessibility, and also circumvent concerns of explainability. Digital technology advancements include digital therapeutics, self-monitoring devices, virtual reality or augmented reality technology, and wearable devices - which provide possible avenues for monitoring myopia progression and control. However, there are challenges in the implementation of these technologies, which include requirements for specific infrastructure and resources, demonstrating clinically acceptable performance and safety of data management. Nonetheless, this remains an evolving field with the potential to address the growing global burden of myopia.