Evaluation and comparison of ocular biometric parameters obtained with Tomey OA-2000 in silicone oil-filled aphakic eyes.

PURPOSE: To evaluate a new non-contact instrument (OA-2000) measuring the ocular biometry parameters of silicone oil (SO)-filled aphakic eyes, as compared with IOLMaster 700. METHODS: Forty SO-filled aphakic eyes of 40 patients were enrolled in this cross-sectional clinical trial. The axial length (AL), central corneal thickness (CCT), keratometry ((flattest keratometry) Kf and (steep keratometry, 90° apart from Kf) Ks), and axis of the Kf (Ax1) were measured with OA-2000 and IOLMaster 700. The coefficient of variation (CoV) was calculated to assess the repeatability. The correlation was evaluated by the Pearson coefficient. Bland-Altman analysis and paired t test were used to analyze the agreements and differences of parameters measured by the two devices, respectively. RESULTS: The mean AL obtained with the OA-2000 was 23.57 ± 0.93 mm (range: 21.50 to 25.68 mm), and that obtained with the IOLMaster 700 was 23.69 ± 0.94 mm (range: 21.85 to 25.86 mm), resulting in a mean offset of 0.124 ± 0.125 mm (p < 0.001). The mean offset of CCT measured by OA-2000 and IOLMaster 700 was 14.6 ± 7.5 µm (p < 0.001). However, the Kf, Ks and Ax1 values from the two devices were comparable (p > 0.05). All the measured parameters of the two devices showed strong linear correlations (all r ≥ 0.966). The Bland-Altman analysis showed a narrow 95% limits of agreement (LoA) of Kf, Ks and AL, but 95%LoA of CCT and Ax1 was wide, which were - 29.3 ~ 0.1 µm and-25.9 ~ 30.7°respectively. The CoVs of the biometric parameters obtained with OA-2000 were lower than 1%. CONCLUSION: In SO-filled aphakic eyes, the ocular parameters (including AL, Kf, Ks, Ax1, and CCT) measured by the OA-2000 and IOLMaster 700 had a good correlation. Two devices had an excellent agreement on ocular biometric measurements of Kf, Ks and AL. The OA-2000 provided excellent repeatability of ocular parameters in SO-filled aphakic eyes.

Automated multidimensional deep learning platform for referable diabetic retinopathy detection: a multicentre, retrospective study.

OBJECTIVE: To develop and validate a real-world screening, guideline-based deep learning (DL) system for referable diabetic retinopathy (DR) detection. DESIGN: This is a multicentre platform development study based on retrospective, cross-sectional data sets. Images were labelled by two-level certificated graders as the ground truth. According to the UK DR screening guideline, a DL model based on colour retinal images with five-dimensional classifiers, namely image quality, retinopathy, maculopathy gradability, maculopathy and photocoagulation, was developed. Referable decisions were generated by integrating the output of all classifiers and reported at the image, eye and patient level. The performance of the DL was compared with DR experts. SETTING: DR screening programmes from three hospitals and the Lifeline Express Diabetic Retinopathy Screening Program in China. PARTICIPANTS: 83 465 images of 39 836 eyes from 21 716 patients were annotated, of which 53 211 images were used as the development set and 30 254 images were used as the external validation set, split based on centre and period. MAIN OUTCOMES: Accuracy, F1 score, sensitivity, specificity, area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), Cohen's unweighted κ and Gwet's AC1 were calculated to evaluate the performance of the DL algorithm. RESULTS: In the external validation set, the five classifiers achieved an accuracy of 0.915-0.980, F1 score of 0.682-0.966, sensitivity of 0.917-0.978, specificity of 0.907-0.981, AUROC of 0.9639-0.9944 and AUPRC of 0.7504-0.9949. Referable DR at three levels was detected with an accuracy of 0.918-0.967, F1 score of 0.822-0.918, sensitivity of 0.970-0.971, specificity of 0.905-0.967, AUROC of 0.9848-0.9931 and AUPRC of 0.9527-0.9760. With reference to the ground truth, the DL system showed comparable performance (Cohen's κ: 0.86-0.93; Gwet's AC1: 0.89-0.94) with three DR experts (Cohen's κ: 0.89-0.96; Gwet's AC1: 0.91-0.97) in detecting referable lesions. CONCLUSIONS: The automatic DL system for detection of referable DR based on the UK guideline could achieve high accuracy in multidimensional classifications. It is suitable for large-scale, real-world DR screening.

Automatic detection of 39 fundus diseases and conditions in retinal photographs using deep neural networks.

Retinal fundus diseases can lead to irreversible visual impairment without timely diagnoses and appropriate treatments. Single disease-based deep learning algorithms had been developed for the detection of diabetic retinopathy, age-related macular degeneration, and glaucoma. Here, we developed a deep learning platform (DLP) capable of detecting multiple common referable fundus diseases and conditions (39 classes) by using 249,620 fundus images marked with 275,543 labels from heterogenous sources. Our DLP achieved a frequency-weighted average F1 score of 0.923, sensitivity of 0.978, specificity of 0.996 and area under the receiver operating characteristic curve (AUC) of 0.9984 for multi-label classification in the primary test dataset and reached the average level of retina specialists. External multihospital test, public data test and tele-reading application also showed high efficiency for multiple retinal diseases and conditions detection. These results indicate that our DLP can be applied for retinal fundus disease triage, especially in remote areas around the world.