Improved predictive diagnosis of diabetic macular edema based on hybrid models: An observational study.

Diabetic Macular Edema (DME) is the most common sight-threatening complication of type 2 diabetes. Optical Coherence Tomography (OCT) is the most useful imaging technique to diagnose, follow up, and evaluate treatments for DME. However, OCT exam and devices are expensive and unavailable in all clinics in low- and middle-income countries. Our primary goal was therefore to develop an alternative method to OCT for DME diagnosis by introducing spectral information derived from spontaneous electroretinogram (ERG) signals as a single input or combined with fundus that is much more widespread. Baseline ERGs were recorded in 233 patients and transformed into scalograms and spectrograms via Wavelet and Fourier transforms, respectively. Using transfer learning, distinct Convolutional Neural Networks (CNN) were trained as classifiers for DME using OCT, scalogram, spectrogram, and eye fundus images. Input data were randomly split into training and test sets with a proportion of 80 %-20 %, respectively. The top performers for each input type were selected, OpticNet-71 for OCT, DenseNet-201 for eye fundus, and non-evoked ERG-derived scalograms, to generate a combined model by assigning different weights for each of the selected models. Model validation was performed using a dataset alien to the training phase of the models. None of the models powered by mock ERG-derived input performed well. In contrast, hybrid models showed better results, in particular, the model powered by eye fundus combined with mock ERG-derived information with a 91 % AUC and 86 % F1-score, and the model powered by OCT and mock ERG-derived scalogram images with a 93 % AUC and 89 % F1-score. These data show that the spontaneous ERG-derived input adds predictive value to the fundus- and OCT-based models to diagnose DME, except for the sensitivity of the OCT model which remains the same. The inclusion of mock ERG signals, which have recently been shown to take only 5 min to record in daylight conditions, therefore represents a potential improvement over existing OCT-based models, as well as a reliable and cost-effective alternative when combined with the fundus, especially in underserved areas, to predict DME.

Evaluation of an AI system for the detection of diabetic retinopathy from images captured with a handheld portable fundus camera: the MAILOR AI study.

OBJECTIVES: To evaluate the performance of an artificial intelligence (AI) system (Pegasus, Visulytix Ltd., UK*) at the detection of diabetic retinopathy (DR) from images captured by a handheld portable fundus camera. METHODS: A cohort of 6404 patients (~80% with diabetes mellitus) was screened for retinal diseases using a handheld portable fundus camera (Pictor Plus, Volk Optical Inc., USA) at the Mexican Advanced Imaging Laboratory for Ocular Research. The images were graded for DR by specialists according to the Scottish DR grading scheme. The performance of the AI system was evaluated, retrospectively, in assessing referable DR (RDR) and proliferative DR (PDR) and compared with the performance on a publicly available desktop camera benchmark dataset. RESULTS: For RDR detection, Pegasus performed with an 89.4% (95% CI: 88.0-90.7) area under the receiver operating characteristic (AUROC) curve for the MAILOR cohort, compared with an AUROC of 98.5% (95% CI: 97.8-99.2) on the benchmark dataset. This difference was statistically significant. Moreover, no statistically significant difference was found in performance for PDR detection with Pegasus achieving an AUROC of 94.3% (95% CI: 91.0-96.9) on the MAILOR cohort and 92.2% (95% CI: 89.4-94.8) on the benchmark dataset. CONCLUSIONS: Pegasus showed good transferability for the detection of PDR from a curated desktop fundus camera dataset to real-world clinical practice with a handheld portable fundus camera. However, there was a substantial, and statistically significant, decrease in the diagnostic performance for RDR when using the handheld device.

A comparative analysis of avoidable causes of childhood blindness in Malaysia with low income, middle income and high income countries.

To determine the avoidable causes of childhood blindness in Malaysia and to compare this to other middle income countries, low income countries and high income countries. Data were obtained from a school of the blind study by Patel et al. and analysed for avoidable causes of childhood blindness. Six other studies with previously published data on childhood blindness in Bangladesh, Ethiopia, Nigeria, Indonesia, China and the United Kingdom were reviewed for avoidable causes. Comparisons of data and limitations of the studies are described. Prevalence of avoidable causes of childhood blindness in Malaysia is 50.5 % of all the cases of childhood blindness, whilst in the poor income countries such as Bangladesh, Ethiopia, Nigeria and Indonesia, the prevalence was in excess of 60 %. China had a low prevalence, but this is largely due to the fact that most schools were urban, and thus did not represent the situation of the country. High income countries had the lowest prevalence of avoidable childhood blindness. In middle income countries, such as Malaysia, cataract and retinopathy of prematurity are the main causes of avoidable childhood blindness. Low income countries continue to struggle with infections such as measles and nutritional deficiencies, such as vitamin A, both of which are the main contributors to childhood blindness. In high income countries, such as the United Kingdom, these problems are almost non-existent.

A myopic shift in Australian Aboriginals: 1977-2000.

BACKGROUND/AIMS: The prevalence of myopia has been reported to have increased in a number of population groups. We compared the refraction of Australian Aboriginal adults in 2000 with data collected in 1977 to assess whether there had been a change in the prevalence of myopia. METHOD: Australian Aboriginal adults aged 20 to 30 years old were selected from Central Australian communities in 2000. Refraction was determined by noncycloplegic autorefraction. This was compared to mydriatic retinoscopy data collected in 1977. "Observer trials" were undertaken to assess the comparability of noncycloplegic autorefraction measurements and cycloplegic retinoscopy. Spherical equivalence cylinder and spheric were determined for all right and left eyes and compared using an analysis of variance. RESULTS: A total of 128 adults (58 males, 70 females) were examined in 2000 and compared with 161 adults (107 males, 54 females) examined in 1977. The mean spherical equivalent in 2000 was -0.55 D +/- 0.88 D and in 1977 was +0.54 D +/- 0.81 D. The difference of -1.09 D was highly significant (F = 126, P < .001). Intraclass correlation coefficients showed good agreement between noncycloplegic autorefraction and cycloplegic retinoscopy. Neither gender, schooling, nor diabetes was associated with an increased risk of myopia. CONCLUSIONS: There appears to have been a significant shift toward myopia in Australian Aboriginals between 1977 and 2000. The cause of this myopic shift is unknown but mirrors that observed in other populations in recent years.

Assessment of trachoma prevalence in a mobile population in Central Australia.

Trachoma is reported to be hyperendemic in Australia. This study was conducted in a desert area of Central Australia to implement and evaluate the WHO SAFE strategy to control trachoma. The aim of the study was to obtain baseline trachoma prevalence data and to determine whether a single annual visit is adequate for a treatment program targeting households with active cases in a highly mobile population. All registered residents of two Aboriginal communities were eligible for examination. Four visits over the course of 13 months were made to the communities for ocular examinations of residents present at the time of the visit. Examination, diagnosis, and grading of trachoma followed WHO guidelines. The overall examination rate was 75%, refusal rate was <1%, but approximately 50% of community residents were absent during the examination period. Prevalence varied on each visit, but the overall prevalence of active trachoma was 49% over the 13-month period. Children less than 10 years of age had the highest prevalence of active trachoma (79%), over the course of the 13 months, yet the prevalence at any one visit was approximately 60%. Trachomatous scarring was present in 23% of the population. These results suggest that many cases of active trachoma may be missed if a prevalence survey is conducted at only one point in time. Multiple examinations should be conducted to adequately establish prevalence in the population. Antibiotic treatment and health promotion campaigns need to be developed in consideration of local community dynamics.