SERS-CNN approach for non-invasive and non-destructive monitoring of stem cell growth on a universal substrate through an analysis of the cultivation medium

The development of advanced methods of SERS-CNN data analysis seems to provide a perfect analytical system that is capable of solving the sophisticated task of determining the species and the behavior of microorganisms. Unlike the widely-used analytical approach, machine learning allows precise analysis even of very complex spectra of biological samples, and can provide precise decisions for a specific biochemical or microbiological task. In this article, we show for the first time the utilization of the SERS-CNN approach for remote observation of mesenchymal stem cell behavior. Our approach is based on SERS measurements of the biochemical changes taking place in the surrounding culture media due to stem cell proliferation and their biochemical activity. The cells were cultivated on various substrates supporting random or oriented cell growth, and also on surface-toxic substrates. SERS-CNN analysis reveals the ability to perform remote non-invasive estimation (i.e. using the surrounding medium analysis) of the degree of cell survival and the proliferation rate, using Raman measurements and advanced spectra data processing. It should be noted that the proposed approach makes it possible to analyze cell behavior without disrupting cell growth, and it can also be performed by untrained staff with the use of widely-available equipment.

Automated summary

The development of advanced methods of SERS-CNN data analysis provides an effective solution for determining the species and behavior of microorganisms. Machine learning allows precise analysis of complex spectra of biological samples and provides accurate decisions for specific tasks. This article demonstrates the application of the SERS-CNN approach in remote observation of mesenchymal stem cell behavior, enabling non-invasive estimation of cell survival and proliferation rate using Raman measurements and advanced spectra data processing.