High-Resolution Direct Ophthalmoscopy With an Unmodified iPhone X

Accuracy of Artificial Intelligence in Estimating Best-Corrected Visual Acuity From Fundus Photographs in Eyes With Diabetic Macular Edema

Importance

Best-corrected visual acuity (BCVA) is a measure used to manage diabetic macular edema (DME), sometimes suggesting development of DME or consideration of initiating, repeating, withholding, or resuming treatment with anti-vascular endothelial growth factor. Using artificial intelligence (AI) to estimate BCVA from fundus images could help clinicians manage DME by reducing the personnel needed for refraction, the time presently required for assessing BCVA, or even the number of office visits if imaged remotely.

Objective

To evaluate the potential application of AI techniques for estimating BCVA from fundus photographs with and without ancillary information.

Design, Setting, and Participants

Deidentified color fundus images taken after dilation were used post hoc to train AI systems to perform regression from image to BCVA and to evaluate resultant estimation errors. Participants were patients enrolled in the VISTA randomized clinical trial through 148 weeks wherein the study eye was treated with aflibercept or laser. The data from study participants included macular images, clinical information, and BCVA scores by trained examiners following protocol refraction and VA measurement on Early Treatment Diabetic Retinopathy Study (ETDRS) charts.

Main Outcomes

Primary outcome was regression evaluated by mean absolute error (MAE); the secondary outcome included percentage of predictions within 10 letters, computed over the entire cohort as well as over subsets categorized by baseline BCVA, determined from baseline through the 148-week visit.

Results

Analysis included 7185 macular color fundus images of the study and fellow eyes from 459 participants. Overall, the mean (SD) age was 62.2 (9.8) years, and 250 (54.5%) were male. The baseline BCVA score for the study eyes ranged from 73 to 24 letters (approximate Snellen equivalent 20/40 to 20/320). Using ResNet50 architecture, the MAE for the testing set (n = 641 images) was 9.66 (95% CI, 9.05-10.28); 33% of the values (95% CI, 30%-37%) were within 0 to 5 letters and 28% (95% CI, 25%-32%) within 6 to 10 letters. For BCVA of 100 letters or less but more than 80 letters (20/10 to 20/25, n = 161) and 80 letters or less but more than 55 letters (20/32 to 20/80, n = 309), the MAE was 8.84 letters (95% CI, 7.88-9.81) and 7.91 letters (95% CI, 7.28-8.53), respectively.

Conclusions and Relevance

This investigation suggests AI can estimate BCVA directly from fundus photographs in patients with DME, without refraction or subjective visual acuity measurements, often within 1 to 2 lines on an ETDRS chart, supporting this AI concept if additional improvements in estimates can be achieved.

Prospective evaluation of medical student accuracy conducting direct ophthalmoscopy with an unmodified iPhone X

Purpose

To evaluate fundus examination accuracy of medical students when using an unmodified iPhone X or a direct ophthalmoscope in comparison to a staff ophthalmologist’s retinal examination.

Methods

In this prospective comparative analysis, patients underwent dilated fundus examination by novice medical trainees using either an unmodified iPhone X or standard direct ophthalmoscope. The primary outcome was the mean difference and degree of agreement in cup-to-disc ratio between student examination and the staff ophthalmologist’s cup-to-disc observation.

Results

A total of 18 medical students conducted 230 retinal examinations, 117 with the iPhone X and 113 with the direct ophthalmoscope. A greater proportion of students were unable to report cup-to-disc ratio using the iPhone X (81.2%) vs direct ophthalmoscope (30.1%). Student examination of cup-to-disc ratio led to a systematic bias (95% limits of agreement) of + 0.16 (−0.22 to + 0.54) and + 0.10 (−0.36 to + 0.56) with the iPhone X and direct ophthalmoscope, respectively. iPhone X and direct ophthalmoscope student observation concordance for optic disc colour (88.7 and 82.4%, respectively) and contour (68.3 and 74.2%, respectively) demonstrated low agreement with staff ophthalmologist findings. Student iPhone X observations demonstrated lower agreement with staff findings compared to direct ophthalmoscope observations for spontaneous venous pulsations (Cohen’s Kappa = −0.044 vs 0.099).

Conclusion

Amongst medical trainees, optic disc visualization using an unmodified iPhone X was inferior to the direct ophthalmoscope. When able to visualize the optic nerve head, there was no significant difference in reported cup-to-disc ratio between modalities. However, both modalities demonstrated poor reliability in comparison to staff ophthalmologist findings.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10792-022-02377-4.

Smartphone Ophthalmoscopy: is there a place for it?

Smartphone technology is advancing at a rapid pace. Their role in day-to-day life is becoming more and more intricate and irreplaceable. Of late, they have gained immense importance in different medical specialities where they possess an active ability to guide the clinician. This is particularly evident in ophthalmology, where the constantly evolving camera-illumination systems and the artificial intelligence integrated technology have unravelled many novel observations for non-contact posterior segment imaging. The scope of this review is to highlight the role of smartphones as ophthalmoscopes (direct as well as indirect). Nevertheless, their limitations and future directions are also stated here with the intention of making progress in the field of smartphone fundus imaging.

Teleophthalmology and Artificial Intelligence As Game Changers in Ophthalmic Care After the COVID-19 Pandemic

The current COVID-19 pandemic has boosted a sudden demand for telemedicine due to quarantine and travel restrictions. The exponential increase in the use of telemedicine is expected to affect ophthalmology drastically. The aim of this review is to discuss the utility, effectiveness and challenges of teleophthalmological new tools for eyecare delivery as well as its implementation and possible facilitation with artificial intelligence. We used the terms: “teleophthalmology,” “telemedicine and COVID-19,” “retinal diseases and telemedicine,” “virtual ophthalmology,” “cost effectiveness of teleophthalmology,” “pediatric teleophthalmology,” “Artificial intelligence and ophthalmology,” “Glaucoma and teleophthalmology” and “teleophthalmology limitations” in the database of PubMed and selected the articles being published in the course of 2015-2020. After the initial search, 321 articles returned as relevant. A meticulous screening followed and eventually 103 published manuscripts were included and used as our references. Emerging in the market, teleophthalmology is showing great potential for the future of ophthalmological care, benefiting both patients and ophthalmologists in times of pandemics. The spectrum of eye diseases that could benefit from teleophthalmology is wide, including mostly retinal diseases such as diabetic retinopathy, retinopathy of prematurity, age-related macular degeneration but also glaucoma and anterior segment conditions. Simultaneously, artificial intelligence provides ways of implementing teleophthalmology easier and with better outcomes, contributing as significant changing factors for ophthalmology practice after the COVID-19 pandemic.

Clinical Role of Smartphone Fundus Imaging in Diabetic Retinopathy and Other Neuro-retinal Diseases

Purpose: In today's life, many electronic gadgets have the potential to become invaluable health care devices in future. The gadgets in this category include smartphones, smartwatches, and others. Till now, smartphone role has been highlighted on many occasions in different areas, and they continue to possess immense role in clinical documentation, clinical consultation, and digitalization of ocular care. In last one decade, many treatable conditions including diabetic retinopathy, glaucoma, and other pediatric retinal diseases are being imaged using smartphones.Methods: To comprehend this cumulative knowledge, a detailed medical literature search was conducted on PubMed/Medline, Scopus, and Web of Science till February 2021.Results: The included literature revealed a definitive progress in posterior segment imaging. From simple torch light with smartphone examination to present day compact handy devices with artificial intelligence integrated software's have changed the very perspectives of ocular imaging in ophthalmology. The consistently reproducible results, constantly improving imaging techniques, and most importantly their affordable costs have renegotiated their role as effective screening devices in ophthalmology. Moreover, the obtained field of view, ocular safety, and their key utility in non-ophthalmic specialties are also growing.Conclusions: To conclude, smartphone imaging can now be considered as a quick, cost-effective, and digitalized tool for posterior segment screenings, however, their definite role in routine ophthalmic clinics is yet to be established.

Digital ophthalmoscopy: through a non-specialist lens

The ophthalmoscope has been a mainstay for non-specialist physicians to review a range of disorders affecting patient fundi. Its design has remained unchanged since its inception and while relatively easy to operate, gaining mastery of it can prove challenging. Training is limited by its single-user interface with the lack of attachments to augment its design and therefore facilitate skill acquisition. Documentation is variable depending on the competence of the individual to satisfactorily see enough of the fundi, and the accuracy with which they can describe their findings. Several groups have looked to find adaptions to smartphones in order to develop user-friendly and affordable alternative to the traditional ophthalmoscope. A digital solution has the potential to provide greater versatility by enhancing documentation, improved training pathways and facilitating more efficient care, with options for geographically disparate healthcare settings to access teleophthalmology input.

Clinically useful smartphone ophthalmic imaging techniques

Imaging devices in ophthalmology are numerous, and most of them are sophisticated and specialized for specific regions of the eye. In addition, these are fixed and involve close interaction of the patient and the examiner; therefore, simple, portable and tele facility-imbibed imaging tools can be considered optimal alternatives to routine exercises. In the last 10 years, utility of smartphones in ophthalmology is being continuously explored to unearth their potential benefits. In this direction, a smartphone device with/without simple attachments has been noted to aid in detailed, high-quality imaging of the ocular adnexa, cornea, angle, iris, lens, optic disc, and the retina including its periphery. In addition, such utility has also been extended in strabismology workup and intraocular pressure measurements. Hence, using these clinician friendly tools and techniques or by devising newer and more comprehensive tool kits, ophthalmic care can be well-managed with apt use of technology. Also, the smartphone companies are encouraged to collaborate with the medical experts to endeavor more, and help and serve the people better.

Telemedicine for neuro-ophthalmology: challenges and opportunities

PURPOSE OF REVIEW

Telemedicine for neuro-ophthalmology (tele-neuro-ophthalmology) has the potential to increase access to neuro-ophthalmic care by improving efficiency and decreasing the need for long-distance travel for patients. Requirements for decreased person-to-person contacts during the COVID-19 pandemic accelerated adoption of tele-neuro-ophthalmology. This review highlights the challenges and opportunities with tele-neuro-ophthalmology.

RECENT FINDINGS

Tele-neuro-ophthalmology programs can be used for triage, diagnostic consultation, and long-term treatment monitoring. Formats include telephone appointments, interprofessional collaborations, remote data interpretation, online asynchronous patient communication, and video visits. Barriers to long-term implementation of tele-neuro-ophthalmology arise from data quality, patient engagement, workflow integration, state and federal regulations, and reimbursement. General neurologists may collaborate with local eye care providers for ophthalmic examination, imaging, and testing to facilitate efficient and effective tele-neuro-ophthalmology consultation.

SUMMARY

Tele-neuro-ophthalmology has tremendous potential to improve patient access to high-quality cost-effective neuro-ophthalmic care. However, many factors may impact its long-term sustainability.

Smartphone Disc Photography Versus Standard Stereoscopic Disc Photography as a Teaching Tool

Optic disc photographic assessment was performed using an unmodified iPhone Xs Max. A video recording of the disc and peripapillary area was obtained in 5 patients. Good quality screenshots were acquired from the video, and, from them, a single best quality image was selected. This was compared with their respective standard optic disc stereo photograph. A glaucomatologist blinded to the type of image acquisition studied these images and concluded that both images could help well in the identification of glaucomatous optic disc features; however, the stereo photograph had better clarity than the iPhone static image. Nevertheless, the iPhone offered the advantage of real-time video assessment of the disc and peripapillary area and could serve as a practical and handy tool for telemedicine and teaching purposes.

The impact of artificial intelligence in the diagnosis and management of glaucoma

Deep learning (DL) is a subset of artificial intelligence (AI), which uses multilayer neural networks modelled after the mammalian visual cortex capable of synthesizing images in ways that will transform the field of glaucoma. Autonomous DL algorithms are capable of maximizing information embedded in digital fundus photographs and ocular coherence tomographs to outperform ophthalmologists in disease detection. Other unsupervised algorithms such as principal component analysis (axis learning) and archetypal analysis (corner learning) facilitate visual field interpretation and show great promise to detect functional glaucoma progression and differentiate it from non-glaucomatous changes when compared with conventional software packages. Forecasting tools such as the Kalman filter may revolutionize glaucoma management by accounting for a host of factors to set target intraocular pressure goals that preserve vision. Activation maps generated from DL algorithms that process glaucoma data have the potential to efficiently direct our attention to critical data elements embedded in high throughput data and enhance our understanding of the glaucomatous process. It is hoped that AI will realize more accurate assessment of the copious data encountered in glaucoma management, improving our understanding of the disease, preserving vision, and serving to enhance the deep bonds that patients develop with their treating physicians.