eduzhai > Applied Sciences > Engineering >

Histopathological imaging features- versus molecular measurements-based cancer prognosis modeling

  • Save

... pages left unread,continue reading

Document pages: 17 pages

Abstract: For most if not all cancers, prognosis is of significant importance, andextensive modeling research has been conducted. With the genetic nature ofcancer, in the past two decades, multiple types of molecular data (such as geneexpressions and DNA mutations) have been explored. More recently,histopathological imaging data, which is routinely collected in biopsy, hasbeen shown as informative for modeling prognosis. In this study, using the TCGALUAD and LUSC data as a showcase, we examine and compare modeling lung canceroverall survival using gene expressions versus histopathological imagingfeatures. High-dimensional regularization methods are adopted for estimationand selection. Our analysis shows that gene expressions have slightly betterprognostic performance. In addition, most of the gene expressions are found tobe weakly correlated imaging features. It is expected that this study canprovide some insight into utilizing the two types of important data in cancerprognosis modeling and into lung cancer overall survival.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×