On the use of Mahalanobis distance for out-of-distribution detection with neural networks for medical imaging.
Harry AnthonyKonstantinos KamnitsasPublished in: CoRR (2023)
Keyphrases
- medical imaging
- mahalanobis distance
- neural network
- kl divergence
- medical images
- x ray
- image registration
- image analysis
- image processing and analysis
- anatomical structures
- medical image processing
- data distribution
- image processing
- ct images
- noise model
- imaging modalities
- covariance matrix
- image guided surgery
- pattern recognition
- image segmentation
- euclidean distance
- morphological image analysis
- metric learning
- distance metric
- biomedical engineering
- computer vision
- remote sensing
- volumetric data
- gaussian distribution
- object detection
- feature vectors