Centre National De La Recherche Scientifique

Universitè De Montpellier

Horiba Abx Sas

Franck Nicoud

Damien Isebe

Simon Mendez

Pierre Taraconat

A medical analysis device with cellular impedance signal processing features a memory arranged to receive pulse data sets, with each pulse data set including impedance value data that are associated each time with a time marker. The pulse data sets together represent a curve of cellular impedance values that are measured as a cell passes through a polarised opening. The device further includes a classifier in communication with the memory. The classifier may feature a convolutional neural network that receives the pulse data sets as input. The classifier is provided with at least one convolutional layer, which convolutional layer has a depth greater than or equal to 3, and at least two fully connected layers, in addition to an output layer rendering a cell classification from which a pulse data set is derived.

Growing community of inventors

MONTPELLIER, France

Franck Nicoud