Multi-scale textural feature extraction and particle swarm optimization based model selection for false positive reduction in mammography.
Imad ZyoutJoanna CzajkowskaMarcin GrzegorzekPublished in: Comput. Medical Imaging Graph. (2015)
Keyphrases
- model selection
- false positives
- particle swarm optimization
- feature extraction
- multiscale
- false negative
- feature selection
- texture analysis
- cross validation
- wavelet transform
- detection rate
- false positive rate
- local binary pattern
- image processing
- bayesian learning
- parameter estimation
- true positive
- hyperparameters
- regression model
- variable selection
- texture features
- feature set
- mixture model
- machine learning
- sample size
- gaussian process
- error estimation
- low false positive rate
- dimension reduction
- model selection criteria
- selection criterion
- feature vectors
- statistical inference
- principal component analysis
- unsupervised learning
- gray level
- genetic algorithm
- dimensionality reduction
- support vector machine
- data sets
- textural features
- image sequences
- marginal likelihood
- image segmentation
- discriminant analysis