Characterization of breast lesions using multi-parametric diffusion MRI and machine learning.

. To investigate quantitative imaging markers based on parameters from two diffusion-weighted imaging (DWI) models, continuous-time random-walk (CTRW) and intravoxel incoherent motion (IVIM) models, for characterizing malignant and benign breast lesions by using a machine learning algorithm.. With IRB approval, 40 women with histologically confirmed breast lesions (16 benign, 24 malignant) underwent DWI with 11-values (50 to 3000 s/mm) at 3T. Three CTRW parameters,,, andand three IVIM parameters,, andwere estimated from the lesions. A histogram was generated and histogram features of skewness, variance, mean, median, interquartile range; and the value of the 10%, 25% and 75% quantiles were extracted for each parameter from the regions-of-interest. Iterative feature selection was performed using the Boruta algorithm that uses the Benjamin Hochberg False Discover Rate to first determine significant features and then to apply the Bonferroni correction to further control for false positives across multiple comparisons during the iterative procedure. Predictive performance of the significant features was evaluated using Support Vector Machine, Random Forest, Naïve Bayes, Gradient Boosted Classifier (GB), Decision Trees, AdaBoost and Gaussian Process machine learning classifiers.. The 75% quantile, and median of; 75% quantile ofmean, median, and skewness ofkurtosis of; and 75% quantile ofwere the most significant features. The GB differentiated malignant and benign lesions with an accuracy of 0.833, an area-under-the-curve of 0.942, and an F1 score of 0.87 providing the best statistical performance (-value < 0.05) compared to the other classifiers.. Our study has demonstrated that GB with a set of histogram features from the CTRW and IVIM model parameters can effectively differentiate malignant and benign breast lesions.