Limits...
Pattern Recognition Approaches for Breast Cancer DCE-MRI Classification: A Systematic Review

View Article: PubMed Central - PubMed

ABSTRACT

We performed a systematic review of several pattern analysis approaches for classifying breast lesions using dynamic, morphological, and textural features in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Several machine learning approaches, namely artificial neural networks (ANN), support vector machines (SVM), linear discriminant analysis (LDA), tree-based classifiers (TC), and Bayesian classifiers (BC), and features used for classification are described. The findings of a systematic review of 26 studies are presented. The sensitivity and specificity are respectively 91 and 83 % for ANN, 85 and 82 % for SVM, 96 and 85 % for LDA, 92 and 87 % for TC, and 82 and 85 % for BC. The sensitivity and specificity are respectively 82 and 74 % for dynamic features, 93 and 60 % for morphological features, 88 and 81 % for textural features, 95 and 86 % for a combination of dynamic and morphological features, and 88 and 84 % for a combination of dynamic, morphological, and other features. LDA and TC have the best performance. A combination of dynamic and morphological features gives the best performance.

No MeSH data available.


Sensitivity and specificity plotted in receiver operating characteristic space for individual studies; sROC curves are plotted from data points for each feature
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5016558&req=5

Fig5: Sensitivity and specificity plotted in receiver operating characteristic space for individual studies; sROC curves are plotted from data points for each feature

Mentions: The included studies also considered different features. In this second systematic meta-analysis, we considered only the studies that reported a detailed description of used features (22 of the 26 included studies). The most used features were dynamic features, followed closely by morphological features, and then textural features (Table 2). The results of this second meta-analysis are shown in Figs. 4 and 5, which respectively show the forest plot and the sROC curves.Table 2


Pattern Recognition Approaches for Breast Cancer DCE-MRI Classification: A Systematic Review
Sensitivity and specificity plotted in receiver operating characteristic space for individual studies; sROC curves are plotted from data points for each feature
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5016558&req=5

Fig5: Sensitivity and specificity plotted in receiver operating characteristic space for individual studies; sROC curves are plotted from data points for each feature
Mentions: The included studies also considered different features. In this second systematic meta-analysis, we considered only the studies that reported a detailed description of used features (22 of the 26 included studies). The most used features were dynamic features, followed closely by morphological features, and then textural features (Table 2). The results of this second meta-analysis are shown in Figs. 4 and 5, which respectively show the forest plot and the sROC curves.Table 2

View Article: PubMed Central - PubMed

ABSTRACT

We performed a systematic review of several pattern analysis approaches for classifying breast lesions using dynamic, morphological, and textural features in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Several machine learning approaches, namely artificial neural networks (ANN), support vector machines (SVM), linear discriminant analysis (LDA), tree-based classifiers (TC), and Bayesian classifiers (BC), and features used for classification are described. The findings of a systematic review of 26 studies are presented. The sensitivity and specificity are respectively 91 and 83 % for ANN, 85 and 82 % for SVM, 96 and 85 % for LDA, 92 and 87 % for TC, and 82 and 85 % for BC. The sensitivity and specificity are respectively 82 and 74 % for dynamic features, 93 and 60 % for morphological features, 88 and 81 % for textural features, 95 and 86 % for a combination of dynamic and morphological features, and 88 and 84 % for a combination of dynamic, morphological, and other features. LDA and TC have the best performance. A combination of dynamic and morphological features gives the best performance.

No MeSH data available.