A combined convolutional and recurrent neural network for enhanced glaucoma detection.

Glaucoma, a leading cause of blindness, is a multifaceted disease with several patho-physiological features manifesting in single fundus images (e.g., optic nerve cupping) as well as fundus videos (e.g., vascular pulsatility index). Current convolutional neural networks (CNNs) developed to detect glaucoma are all based on spatial features embedded in an image. We developed a combined CNN and recurrent neural network (RNN) that not only extracts the spatial features in a fundus image but also the temporal features embedded in a fundus video (i.e., sequential images). A total of 1810 fundus images and 295 fundus videos were used to train a CNN and a combined CNN and Long Short-Term Memory RNN. The combined CNN/RNN model reached an average F-measure of 96.2% in separating glaucoma from healthy eyes. In contrast, the base CNN model reached an average F-measure of only 79.2%. This proof-of-concept study demonstrates that extracting spatial and temporal features from fundus videos using a combined CNN and RNN, can markedly enhance the accuracy of glaucoma detection.

Clinical model assisting with the collaborative care of glaucoma patients and suspects.

BACKGROUND: Optimizing patient management will reduce unnecessary vision loss in glaucoma through early detection. One method is the introduction of collaborative care schemes between optometrists and ophthalmologists. DESIGN: We conducted a retrospective study to evaluate the impact of the Centre for Eye Health (CFEH) on glaucoma patient outcomes and management in primary optometric care. PARTICIPANTS: Patients referred to CFEH by optometrists for a glaucoma assessment were eligible for this study if written consent was provided (500 participants were randomly chosen). METHODS: Clinical data were classified according to disease risk and implemented patient care and analysed against the original diagnosis and patient parameters, followed by statistical analysis. MAIN OUTCOME MEASURES: Two main parameters were evaluated; suitable referral of patients for glaucoma condition assessment and appropriate implementation of follow-up care. RESULTS: The majority of patients referred for glaucoma assessment (86.2%) were classified as glaucoma suspects or likely to have glaucoma, indicating suitable referral of patients for a CFEH evaluation. Further, the involvement of CFEH resulted in a false positive rate of 7.8% for those patients who proceeded to ophthalmological care. However, long-term optometric patient care was not maintained for up to a third of primarily lower risk patients. CONCLUSIONS: The investigated collaborative eye health-care model led to a substantial improvement in appropriate referrals of glaucoma patients to ophthalmologists and could be suitable for optimizing patient care and utilization of resources. Improvement in follow-up of patients by optometrists is required to minimize inappropriately discontinued patient care.