Deep learning achieves perfect anomaly detection on 108, 308 retinal images including unlearned diseases.
Ayaka SuzukiYoshiro SuzukiPublished in: CoRR (2020)
Keyphrases
- anomaly detection
- deep learning
- retinal images
- unsupervised learning
- diabetic retinopathy
- retinal vessel
- blood vessels
- intrusion detection
- optic disk
- network intrusion detection
- detecting anomalies
- optic disc
- anomalous behavior
- intrusion detection system
- mental models
- network traffic
- machine learning
- weakly supervised
- supervised learning
- detect anomalies
- pattern recognition
- one class support vector machines
- feature selection
- data mining
- region of interest
- higher order
- object recognition
- negative selection algorithm