Classification of Sodium MRI Data of Cartilage with Machine Learning and Logistic Regression

Keywords

accuracy acquisition agar annual applied articular best bottom cartilage characteristic chosen class classification classifications classify combination combinations combined combining component composed concentration concentrations consecutive content control controls cover datasets defined deviation diagnosing differentiating discriminant distance distribution efficient eliminating empirical euclidean eugenics except expected expense fisher fluid function funding generate gold healthy identify improvement improvements include institute inversion investigation joint kernel knee known learning lewis linear list logistic loss machine majority measured medical metric much naive narrowing nearest needed neighbor neighbors optimization original osteoarthritis partial patients phantoms placed positive predictors presence principal prior probability processing programming progress quadratic quantification radial radiology randomly recovery regression remaining rule selection sensitivity slices slightly sodium space specificity statistical stepwise subject subjects support suppressed suppression table theory tissue training trans true type variable vector volume volunteers