Investigation of K-means and fuzzy K-means clustering for the analysis of mass spectrometry imaging data
MetadataShow full item record
Mass spectrometry imaging (MSI) is an experimental technique used to measure molecular composition across the surface of a sample, such as a tissue slice. MSI can simultaneously measure hundreds to thousands of molecules, and link those molecular profiles with their spatial location in the sample. However, MSI datasets can be very large, and identifying potentially important biological patterns is a challenging problem. Many types of explorative data analysis have been applied to MSI datasets, and in particular, k-means clustering has recently gained attention for this application. In this study, we examine the effects of different parameters on the performance of basic k-means and fuzzy k-means clustering in identifying biologically relevant patterns in MSI datasets.