Eye Disease Prediction Using Deep Learning and Attention on Oct Scans
SN Computer Science, ISSN: 2661-8907, Vol: 5, Issue: 8
2024
Metric Options: CountsSelecting the 1-year or 3-year option will change the metrics count to percentiles, illustrating how an article or review compares to other articles or reviews within the selected time period in the same journal. Selecting the 1-year option compares the metrics against other articles/reviews that were also published in the same calendar year. Selecting the 3-year option compares the metrics against other articles/reviews that were also published in the same calendar year plus the two years prior.
Example: if you select the 1-year option for an article published in 2019 and a metric category shows 90%, that means that the article or review is performing better than 90% of the other articles/reviews published in that journal in 2019. If you select the 3-year option for the same article published in 2019 and the metric category shows 90%, that means that the article or review is performing better than 90% of the other articles/reviews published in that journal in 2019, 2018 and 2017.
Citation Benchmarking is provided by Scopus and SciVal and is different from the metrics context provided by PlumX Metrics.
Example: if you select the 1-year option for an article published in 2019 and a metric category shows 90%, that means that the article or review is performing better than 90% of the other articles/reviews published in that journal in 2019. If you select the 3-year option for the same article published in 2019 and the metric category shows 90%, that means that the article or review is performing better than 90% of the other articles/reviews published in that journal in 2019, 2018 and 2017.
Citation Benchmarking is provided by Scopus and SciVal and is different from the metrics context provided by PlumX Metrics.
Article Description
For many years, eye disorders have presented formidable obstacles; nevertheless, recent technological developments have created new opportunities for both diagnosis and therapy. This field relies heavily on deep learning and machine learning techniques, especially when paired with optical coherent tomography (OCT) imaging. We suggest a brand-new technique for accurately identifying eye conditions from OCT images. With the use of our technique, patients can be categorized as either disease-free (normal eyes) or as having a particular ailment, such as Normal, Diabetic Macular Edema (DME), Choroidal Neovascularization (CNV) and Drusen. In this study, we present an end-to-end online application for effective eye illness prediction using machine learning and deep learning approaches. Patients can upload their raw OCT scanned images using the application, and a trained custom UNet model is used to segment the images. Following segmentation, the images are fed into an InceptionV3 network that has been improved with a self-attention layer. This self-attention method uses each model’s feature maps to increase classification accuracy. The output of the deep learning model is combined to forecast and categorize different eye conditions. The efficiency and optimal performance of the application have been ensured by extensive experimentation and optimization. Our findings show that the suggested strategy successfully predicts eye diseases. The created web application has a great deal of promise for prompt intervention and early detection, which should improve the results of eye healthcare. The outcome of inception v3 model as 0.92 of precision, 0.96 of recall, 0.96 of accuracy and 0.94 of f1 score.
Bibliographic Details
Springer Science and Business Media LLC
Provide Feedback
Have ideas for a new metric? Would you like to see something else here?Let us know