Diagnosis of multiple cancer types by shrunken centroids of gene expression

TLDR

The technique is general and can be used in many other classification problems. We have devised an approach to cancer class prediction from gene expression profiling, based on an enhancement of the simple nearest prototype (centroid) classifier. Our method of nearest shrunken centroids identifies subsets of genes that best characterize each class. By shrinking the prototypes, the classifier achieves higher accuracy than competing methods and efficiently identifies genes for classifying small round blue cell tumors and leukemias.

Abstract

We have devised an approach to cancer class prediction from gene expression profiling, based on an enhancement of the simple nearest prototype (centroid) classifier. We shrink the prototypes and hence obtain a classifier that is often more accurate than competing methods. Our method of "nearest shrunken centroids" identifies subsets of genes that best characterize each class. The technique is general and can be used in many other classification problems. To demonstrate its effectiveness, we show that the method was highly efficient in finding genes for classifying small round blue cell tumors and leukemias.