Unsupervised Cell Identification on Multidimensional X-ray Fluorescence Datasets
|Title||Unsupervised Cell Identification on Multidimensional X-ray Fluorescence Datasets|
|Publication Type||Journal Article|
|Year of Publication||2013|
|Authors||Wang, S, Ward, J, Leyffer, S, Wild, SM, Jacobsen, C, Vogt, S|
|Journal||Journal of Synchrotron Radiation|
We introduce a novel approach to locate, identify, and refine positions and whole areas of cell structures based on elemental contents measured by X-ray fluorescence microscopy. We show that by initializing with only a handful of prototypical cell regions, this approach can obtain consistent cell populations, even when cells are partially overlapping, without training by explicit annotation. It is robust both to different measurements on the same sample and to different initializations. This effort provides a versatile framework to identify targeted cellular structures from datasets too complex for manual analysis, like most X-ray fluorescence microscopy data. We also discuss possible future extensions.