RESUMO
Purpose/Background: We evaluate how a deep learning model can be applied to extract refractive error metrics from pupillary red reflex images taken by a low-cost handheld fundus camera. This could potentially provide a rapid and economical vision-screening method, allowing for early intervention to prevent myopic progression and reduce the socioeconomic burden associated with vision impairment in the later stages of life. Methods: Infrared and color images of pupillary crescents were extracted from eccentric photorefraction images of participants from Choithram Hospital in India and Dargaville Medical Center in New Zealand. The pre-processed images were then used to train different convolutional neural networks to predict refractive error in terms of spherical power and cylindrical power metrics. Results: The best-performing trained model achieved an overall accuracy of 75% for predicting spherical power using infrared images and a multiclass classifier. Conclusions: Even though the model's performance is not superior, the proposed method showed good usability of using red reflex images in estimating refractive error. Such an approach has never been experimented with before and can help guide researchers, especially when the future of eye care is moving towards highly portable and smartphone-based devices.
RESUMO
SIGNIFICANCE: We demonstrate that the visual field defects in patients with tilted disc syndrome can be reduced or eliminated by neutralizing the peripheral scotoma in the area of posterior retinal bowing, which may allow differentiation between a congenital anomaly and acquired pathology. PURPOSE: Tilted disc syndrome is a congenital and unchanging condition that may present with visual field defects mimicking loss seen in neurological diseases, such as transsynaptic retrograde degeneration. Our purpose was to systematically investigate the ability of a neutralized peripheral refraction to eliminate refractive visual field defects seen in tilted disc syndrome. This was compared with the same technique performed on patients with neurological deficits. METHODS: The Humphrey Field Analyzer was used to measure sensitivities across the 30-2 test grid in 14 patients with tilted disc syndrome using four refractive corrections: habitual near correction and with an additional -1.00, -2.00 or -3.00 D negative lens added as correction lenses. Peripheral refractive errors along the horizontal meridian were determined using peripheral retinoscopy and thus allowed calculation of residual peripheral refraction with different levels of refractive correction. Visual field defects were assessed qualitatively and quantitatively using sensitivities and probability scores in both patient groups. RESULTS: A smaller residual refractive error after the application of negative addition lenses correlated with improvement in visual field defects in terms of sensitivity and probability scores in patients with tilted disc syndrome. Patients with established neurological deficits (retrograde degeneration) showed improvement in sensitivities but not in probability scores. CONCLUSIONS: Neutralizing the refractive error at the region of posterior retinal bowing due to tilted disc syndrome reduces the apparent visual field defect. This may be a useful and rapid test to help differentiate between tilted disc syndrome and other pathological causes of visual field defects such as neurological deficits.
