Impact of imperfect annotations on CNN training and performance for instance segmentation and classification in digital pathology.
Laura Gálvez JiménezChristine DecaesteckerPublished in: Comput. Biol. Medicine (2024)
Keyphrases
- training phase
- training set
- decision trees
- level set
- pattern recognition
- supervised learning
- support vector machine
- training samples
- support vector
- statistical classification
- classification method
- pixel classification
- metadata
- machine learning
- feature vectors
- semantic annotation
- medical images
- training process
- feature space
- discriminative classifiers
- lung nodules
- convolutional neural network
- feature extraction and classification
- classification performances
- classification algorithm
- segmentation method
- image classification
- classification accuracy
- image segmentation
- image processing
- image regions
- class labels
- segmentation algorithm
- feature set
- training dataset
- semi supervised
- image analysis
- multiscale
- feature extraction
- neural network