OCT-based deep learning algorithm for the evaluation of treatment indication with anti-vascular endothelial growth factor medications.

PURPOSE: Intravitreal injections with anti-vascular endothelial growth factor (anti-VEGF) medications have become the standard of care for their respective indications. Optical coherence tomography (OCT) scans of the central retina provide detailed anatomical data and are widely used by clinicians in the decision-making process of anti-VEGF indication. In recent years, significant progress has been made in artificial intelligence and computer vision research. We trained a deep convolutional artificial neural network to predict treatment indication based on central retinal OCT scans without human intervention. METHOD: A total of 183,402 retinal OCT B-scans acquired between 2008 and 2016 were exported from the institutional image archive of a university hospital. OCT images were cross-referenced with the electronic institutional intravitreal injection records. OCT images with a following intravitreal injection during the first 21 days after image acquisition were assigned into the 'injection' group, while the same amount of random OCT images without intravitreal injections was labeled as 'no injection'. After image preprocessing, OCT images were split in a 9:1 ratio to training and test datasets. We trained a GoogLeNet inception deep convolutional neural network and assessed its performance on the validation dataset. We calculated prediction accuracy, sensitivity, specificity, and receiver operating characteristics. RESULTS: The deep convolutional neural network was successfully trained on the extracted clinical data. The trained neural network classifier reached a prediction accuracy of 95.5% on the images in the validation dataset. For single retinal B-scans in the validation dataset, a sensitivity of 90.1% and a specificity of 96.2% were achieved. The area under the receiver operating characteristic curve was 0.968 on a per B-scan image basis, and 0.988 by averaging over six B-scans per examination on the validation dataset. CONCLUSION: Deep artificial neural networks show impressive performance on classification of retinal OCT scans. After training on historical clinical data, machine learning methods can offer the clinician support in the decision-making process. Care should be taken not to mistake neural network output as treatment recommendation and to ensure a final thorough evaluation by the treating physician.

Ranibizumab in neovascular age-related macular degeneration: a 5-year follow-up.

PURPOSE: Our aim was to evaluate an optical coherence tomography (OCT) and visual acuity (VA)-guided, variable-dosing regimen with intravitreal ranibizumab injection for treating patients with neovascular age-related macular degeneration (AMD) from 2007 to 2012. DESIGN: This was a retrospective clinical study of 5 years follow-up in a tertiary eye center. PATIENTS AND METHODS: In this study, 66 patients with neovascular AMD (mean age of 74 years, SD 8.7 years) were included. We investigated the development of best-corrected visual acuity (BCVA), the number of intravitreal injections, and the central retinal thickness measured with OCT (OCT Spectralis) over 5 years of intravitreal treatment. RESULTS: The mean number of intravitreal ranibizumab injections over 5 years was 8.8. The mean BCVA before therapy was 0.4 logarithm of the minimum angle of resolution (logMAR). After 5 years of therapy, the mean BCVA was 0.6 logMAR. In all, 16% of treated patients had stable VA over 5 years and 10% of study eyes approved their VA. The mean OCT-measured central retinal thickness at the beginning of this study was 295 µm; after 5 years of treatment, the mean central retinal thickness was 315 µm. There was an increase in central retinal thickness in 47.5% of examined eyes. CONCLUSION: Other studies showed VA improvement in OCT-guided variable-dosing regimens. Our study revealed a moderate decrease in VA after a total mean injection number as low as 8.8 injections over 5 years. In OCT, an increase in central retinal thickness over 5 years could be observed. Probably, this is due to deficient treatment when comparing the total injection number to other treatment regimens. Anti-VEGF therapy helps to keep the VA stable for a period of time, but cannot totally stop the progression of the disease completely. Patients with late stages of neovascular AMD can maintain VA even if they are relatively undertreated.