Retrieving clinically relevant diabetic retinopathy images using a multi-class multiple-instance framework.
Parag S. ChandakkarRagav VenkatesanBaoxin LiPublished in: Medical Imaging: Computer-Aided Diagnosis (2013)
Keyphrases
- multi class
- multiple instance learning
- multiple instance
- clinically relevant
- image annotation
- support vector machine
- image data
- object recognition
- input image
- image classification
- multi task
- real valued
- pairwise
- image features
- base classifiers
- image understanding
- edge detection
- binary classification
- data sets
- feature selection
- learning problems
- segmentation method
- image retrieval
- similarity measure
- decision trees
- diabetic retinopathy
- machine learning