Artikel
Deep learning applications with OCT in AMD
Meeting Abstract
Suche in Medline nach
Autoren
Veröffentlicht: | 5. Februar 2020 |
---|
Gliederung
Text
Background: Deep learning applications show great promise in their ability to perform automated diagnosis and segmentation. Large databases of images are required to train and build these algorithms.
Methods: Multiple deep learning algorithms were used for automated diagnosis, feature segmentation, and prediction.
Results: Algorithms were validated and were successfully trained using large datasets.
Conclusion: Deep learning holds great promise in their ability to learn features in a data driven fashion.