Smart MEMS Instrumentation for Biophysical flow Cytometry with Statistical Learning – CYTOMEMS

The objective of CYTOMEMS is to demonstrate the first smart MEMS equipment performing high content biophysical characterization of cells in flow for their classification by statistical learning.

Cell characterization is carried out by a bioMEMS device incorporating a microchannel for the passage of cells and equipped with fixed and mobile electrodes enabling both electrical and mechanical measurements of these cells in flux. The position of the mechanical sensor is tuned in real time to characterize the cell under controlled deformations knowing the cell size from upstream electrical measurement.
After a phase of training on on different cell lines, cell identification is performed by statistical classification analysing a comprehensive set of biophysical (electrical and mechanical) parameters.

The electronic system computes in real time key cell characteristics to adjust and optimize the sensor configuration and sort the cell in the classes determined by the statistical learning.

The proof of concept will be established by sorting targeted cell populations (for example monocyte) from heterogeneous cell solution (leukocytes). A specific application aims identifying and isolating rare event (identification of cancer cells) from a mixed cell selection.

The work flow includes 3 scientific tasks:
(1) the MEMS part with the development of improved versions of the device and intensive cell characterization campaigns to feed statistical learning algorithms;
(2) The mathematical and numerical work including the statistical learning from cell multiparametric characterization and the development and coding of the classification algorithm;
(3) The electronic data processing with low noise techniques for the acquisition of the MEMS measurements, the integration of the classification code and the generation of analog and digital output signals for the real time reconfiguration of the sensor and the physical sorting of cells in the device outlets.

A coordination task organizes the communication between the teams and the validation steps during the project and in its final phase. An international symposium on biophysical cytometry and cellular identification by statistical and artificial intelligence approaches is planned at the end of the project.

The complementary consortium gathers 2 academic laboratories with recognized MEMS background and technology facilities, a company providing start of the art electronic equipment and an academic research team specialist in models for data analysis and learning.