Multimodal, multitask, multiattention (M3) deep learning detection of reticular pseudodrusen: Toward automated and accessible classification of age-related macular degeneration.
Qingyu ChenTiarnan D. L. KeenanAlexis AllotYifan PengElvira AgrónAmitha DomalpallyCaroline C. W. KlaverDaniel T. LuttikhuizenMarcus H. ColyerCatherine A. CukrasHenry E. WileyM. Teresa MagoneChantal Cousineau-KriegerWai T. WongYingying ZhuEmily Y. ChewZhiyong LuPublished in: J. Am. Medical Informatics Assoc. (2021)
Keyphrases
- deep learning
- age related macular degeneration
- multi task
- unsupervised learning
- machine learning
- support vector
- multitask learning
- feature extraction
- retinal images
- classification accuracy
- feature selection
- support vector machine
- decision trees
- object detection
- multi task learning
- image classification
- training set
- d objects
- knowledge discovery
- text classification
- e learning
- learning tasks
- probabilistic model
- pairwise
- weakly supervised
- computer vision
- object recognition
- data sets