This paper explores the use of machine learning to differentiate healthy from apoptotic cells using only forward (FSC) and side (SSC) scatter flow cytometry data. Colorectal cancer HCT116 cells treated with miR-34a were classified using Annexin V/propidium iodide staining. Six features (FSC and SSC area, height, and width) were used to train various machine learning models. A multilayer perceptron (MLP) model achieved the best performance, with high precision, recall, F-value, and area under the ROC curve for live cell classification. This model offers an automated, reliable, and stain-free alternative to conventional flow cytometry gating.

Yi Li, Chance M. Nowak, Uyen Pham, Khai Nguyen, Leonidas Bleris