A simple flow-cytometry based image processing algorithm for analysing cell-death mechanisms in radiotherapy

In-vitro testing of novel photodynamic therapy/radiotherapy procedures relies heavily on the use of different assays to fully probe various parameters such as cytotoxicity or cell-death pathways. These assays utilise sometimes expensive dyes or antibodies, along with cumbersome sample preparation for flow-cytometry. In this work, we propose a novel image-processing algorithm that uses the flow cytometry plots obtained through a Propidium Iodide based live-dead assay on cancerous and non-cancerous cells to deduce the possible cell-death mechanisms in the process of radiotherapy. Propidium Iodide (PI) is a membrane-impermeable dye taken up by those cells with loss of cell membrane integrity, and does not give any information about the integrity of intracellular components or cellular death pathways. In our novel image-processing algorithm, we determine the centroid of the Forward Scatter (FSC) and the Side Scatter (SSC) cytometer plots of such a PI assay, after suitable clustering. This algorithm is initially applied to an unirradiated control cell population where the FSC centroid gives an estimate of the mean cell size, while the SSC centroid gives the baseline granularity of the cell population. Thereafter, the centroids of the FSC and the SSC plots are calculated for the irradiated cell population, and the deviation in these centroids calculated. These differences are correlated to change in average cell size and denaturation/granularity, and serve as a useful substitute for the cell death mechanism. This can potentially pave the way for in-situ qualitative cell-death analysis in large-volume in-vitro studies in a cost-effective manner.