Prediction of malignant breast lesions from mri features: a comparison of artificial neural network and logistic regression techniques
McLaren CE et al. - In a study to compare artificial neural network (ANN) and logistic regression techniques to predict malignant breast lesions from MRI features, it was found that the diagnostic performance of models selected by ANN and logistic regression was similar. Analytic methods were found to be roughly equivalent in terms of predictive ability when a small number of variables were chosen. Robust ANN methodology uses a sophisticated nonlinear model, while logistic regression analysis provides insightful information to enhance interpretation of the model features. Methods- The study included 43 malignant and 28 benign histologically proven lesions.
- 8 morphologic parameters, 10 gray-level co-occurrence matrix texture features, and 14 Laws texture features were obtained using automated lesion segmentation and quantitative feature extraction.
- ANN and logistic regression analysis were compared for selection of the best predictors of malignant lesions among normalized features.
Results- Using ANN, the final 4 selected features were compactness, energy, homogeneity, and Law_LS, with an area under the receiver-operating characteristic curve (AUC) of 0.82 and accuracy of 0.76.
- Diagnostic performance of these 4 features computed on the basis of logistic regression yielded an AUC of 0.80, similar to that of ANN.
- Analysis also showed that the odds of a malignant lesion decreased by 48% for every increase of 1 standard deviation in the Law_LS feature, adjusted for differences in compactness, energy, and homogeneity.
- Using logistic regression with z-score transformation, a model composed of compactness, normalized radial length entropy, and gray-level sum average was selected, and it had the highest overall accuracy, 0.75, among all models, with an AUC of 0.77.
- When logistic modeling of transformations using the Box-Cox method was performed, the most parsimonious model with predictors compactness and Law_LS had an AUC of 0.79.
[more...]
|
|
|
