Mixture of expert artificial neural networks with ensemble training for reduction of various sources of false positives in CAD.
Kenji SuzukiLifeng HeShweta KhankariLiang GeJoel VercelesAbraham H. DachmanPublished in: Medical Imaging: Computer-Aided Diagnosis (2007)
Keyphrases
- object oriented
- false positives
- artificial neural networks
- computer aided design
- false negative
- neural network
- training set
- detection rate
- false positive rate
- number of false positives
- feed forward neural networks
- true positive
- evolutionary artificial neural networks
- multi layer perceptron
- low false positive rate
- benchmark classification problems
- blind separation
- false alarms
- training data
- radial basis function
- ensemble learning
- source separation
- training process
- ensemble methods
- mixture model
- computational intelligence
- genetic algorithm
- genetic algorithm ga
- domain experts
- back propagation
- feedforward artificial neural networks
- information sources
- feedforward neural networks
- multiple sources
- decision trees
- false detections
- learning algorithm
- computer vision
- ann models
- image sequences
- data sources
- training samples
- using artificial neural networks
- multilayer perceptron