A Teleophthalmology Support System Based on the Visibility of Retinal Elements Using the CNNs.

This paper proposes a teleophthalmology support system in which we use algorithms of object detection and semantic segmentation, such as faster region-based CNN (FR-CNN) and SegNet, based on several CNN architectures such as: Vgg16, MobileNet, AlexNet, etc. These are used to segment and analyze the principal anatomical elements, such as optic disc (OD), region of interest (ROI) composed by the macular region, real retinal region, and vessels. Unlike the conventional retinal image quality assessment system, the proposed system provides some possible reasons about the low-quality image to support the operator of an ophthalmoscope and patient to acquire and transmit a better-quality image to central eye hospital for its diagnosis. The proposed system consists of four steps: OD detection, OD quality analysis, obstruction detection of the region of interest (ROI), and vessel segmentation. For the OD detection, artefacts and vessel segmentation, the FR-CNN and SegNet are used, while for the OD quality analysis, we use transfer learning. The proposed system provides accuracies of 0.93 for the OD detection, 0.86 for OD image quality, 1.0 for artefact detection, and 0.98 for vessel segmentation. As the global performance metric, the kappa-based agreement score between ophthalmologist and the proposed system is calculated, which is higher than the score between ophthalmologist and general practitioner.

Assessment of Postural Compliance After Pneumatic Retinopexy.

PURPOSE: We describe the functioning of a novel device, aimed to assess patient head position after a pneumatic retinopexy. METHODS: We enrolled patients with the clinical diagnosis of rhegmatogenous retinal detachment. All patients were asked to wear a specially designed headband with a monitoring device composed of an accelerometer, gyroscope, and magnetometer, powered by a 3.7V lithium battery. Every 200 ms, the device measured neck flexion and extension, left and right rotation, and left and right flexion. Patients were asked to come back the next morning for follow-up and headband retrieving. RESULTS: The device was worn an average of 19.17 ± 2.1 hours and performed a mean number of 57,670 ± 8663 measurements without power failures or program errors. An acceptable head position was kept for a mean of 3.33 ± 1.8 hours. The hardest axis to maintain was the right and left flexion of the neck (5.5 ± 2.54 hours of acceptable positioning). CONCLUSION: Real-time monitoring of patient head position after a vitreoretinal procedure is feasible. Maintaining a fixed head position for more than 5 consecutive hours is difficult to achieve and physicians should consider this difficulty when planning treatment. TRANSLATIONAL RELEVANCE: In addition to a significant improvement to the basic design of similar devices, our device allows for assessment of patient adherence to postoperative instructions objectively for the first time to our knowledge. This information could be used in the future to elaborate more detailed position nomograms to improve outcomes.