Limits...
Maxdenominator Reweighted Sparse Representation for Tumor Classification

View Article: PubMed Central - PubMed

ABSTRACT

The classification of tumors is crucial for the proper treatment of cancer. Sparse representation-based classifier (SRC) exhibits good classification performance and has been successfully used to classify tumors using gene expression profile data. In this study, we propose a three-step maxdenominator reweighted sparse representation classification (MRSRC) method to classify tumors. First, we extract a set of metagenes from the training samples. These metagenes can capture the structures inherent to the data and are more effective for classification than the original gene expression data. Second, we use a reweighted regularization method to obtain the sparse representation coefficients. Reweighted regularization can enhance sparsity and obtain better sparse representation coefficients. Third, we classify the data by utilizing a maxdenominator residual error function. Maxdenominator strategy can reduce the residual error and improve the accuracy of the final classification. Extensive experiments using publicly available gene expression profile data sets show that the performance of MRSRC is comparable with or better than many existing representative methods.

No MeSH data available.


Comparison of prediction accuracy on four binary classification datasets by varying the number of samples from per subclass.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5385541&req=5

f2: Comparison of prediction accuracy on four binary classification datasets by varying the number of samples from per subclass.

Mentions: The average prediction accuracies of the classification of the balanced training set and test set are shown in Fig. 2. The MRSRC exhibits encouraging performance. The MRSRC achieves the best classification accuracy in all cases for the colon cancer data (Fig. 2b). Although gliomas are difficult to classify, the proposed approach can still achieve the highest classification accuracy via 21 (81%) samples per subclass used for training (Fig. 2c). MRSRC also achieves relatively high prediction accuracies in most cases in the acute leukemia and DLBCL data sets (Fig. 2a and d). Notably, the classification accuracies of LDA + SVM and ICA + SVM drop quickly as the number of samples considered for training increases. These results are consistent with the observations in the literature22.


Maxdenominator Reweighted Sparse Representation for Tumor Classification
Comparison of prediction accuracy on four binary classification datasets by varying the number of samples from per subclass.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5385541&req=5

f2: Comparison of prediction accuracy on four binary classification datasets by varying the number of samples from per subclass.
Mentions: The average prediction accuracies of the classification of the balanced training set and test set are shown in Fig. 2. The MRSRC exhibits encouraging performance. The MRSRC achieves the best classification accuracy in all cases for the colon cancer data (Fig. 2b). Although gliomas are difficult to classify, the proposed approach can still achieve the highest classification accuracy via 21 (81%) samples per subclass used for training (Fig. 2c). MRSRC also achieves relatively high prediction accuracies in most cases in the acute leukemia and DLBCL data sets (Fig. 2a and d). Notably, the classification accuracies of LDA + SVM and ICA + SVM drop quickly as the number of samples considered for training increases. These results are consistent with the observations in the literature22.

View Article: PubMed Central - PubMed

ABSTRACT

The classification of tumors is crucial for the proper treatment of cancer. Sparse representation-based classifier (SRC) exhibits good classification performance and has been successfully used to classify tumors using gene expression profile data. In this study, we propose a three-step maxdenominator reweighted sparse representation classification (MRSRC) method to classify tumors. First, we extract a set of metagenes from the training samples. These metagenes can capture the structures inherent to the data and are more effective for classification than the original gene expression data. Second, we use a reweighted regularization method to obtain the sparse representation coefficients. Reweighted regularization can enhance sparsity and obtain better sparse representation coefficients. Third, we classify the data by utilizing a maxdenominator residual error function. Maxdenominator strategy can reduce the residual error and improve the accuracy of the final classification. Extensive experiments using publicly available gene expression profile data sets show that the performance of MRSRC is comparable with or better than many existing representative methods.

No MeSH data available.