Machine Learning-Enhanced Dynamic and Continuous Monitoring of Cellular Spatio-Temporal Organization

Understanding the dynamics of spatio-temporal organization of mixtures of cell types is essential for a wide array of physiological and pathological scenarios including organ development, cancer cell invasion of healthy tissues, and vascular disorders such as neointimal hyperplasia.
Today, dynamic monitoring of cellular spatiotemporal organization is conducted is through live-cell microscopic imaging; however, this modality has significant limitations due to cellular photo-damage.

We propose to use impedance-based sensors developed by the startup company Sensome, the industrial partner of the project, to non-invasively and continuously monitor cellular spatio-temporal organization. Sensome has developed state-of-the-art impedance sensors that allow discrimination among different cell types. We propose to develop a platform that allows noninvasive and continuous monitoring of cellular spatio-temporal organization using machine learning-enabled impedance spectroscopy. The platform will consist of a substrate compatible with cell culture that incorporates an array of impedance sensors. We will use this platform to monitor the dynamics of mixtures of smooth muscle cells and endothelial cells as well as mixtures of normal and cancerous cells over long periods of time. To maximize the accuracy of the interpretation of the results, we will use machine learning tools that improve our ability to identify the cell types of interest. We will complement the measurements with mathematical models that will shed insight into cellular dynamics.

The proposed platform will be the first of its kind. The results of the research will significantly advance our understanding of the dynamics of cellular spatio-temporal organization and will shed light onto how different cell types interact. From a technology transfer perspective, the platform that will be developed could provide the basis for a system that can be potentially commercialized in various academic and industrial settings.