Skip to Main content Skip to Navigation
Conference papers

Classification de types de neurones à partir de signaux calciques

Abstract : Videos of calcium activities of mice striatum slices are recorded under stimulations by two-photon fluorescence microscopy. Neurons are selected by regions of interests (ROI) on the images and labelled into two classes: medium spiny neuron (MSN); interneurons (IN). Each ROI enables to obtain a neural signal. Features are extracted on these ROI and signals. A subset feature selection is performed with a quadratic discriminant analysis, to solve the supervised learning of the two classes of neuron of the striatum. It is shown, that a realistic evaluation of the database leads to a classification with an accuracy of 75 % for IN and 90% for MSN.
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02528364
Contributor : Guillaume Becq <>
Submitted on : Wednesday, April 1, 2020 - 5:57:35 PM
Last modification on : Wednesday, May 20, 2020 - 10:41:50 AM

File

becq400.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02528364, version 1

Collections

Citation

Guillaume Becq, Nagham Badreddine, Nicolas Tremblay, Florence Appaix, Gisela Zalcman, et al.. Classification de types de neurones à partir de signaux calciques. Gretsi 2019, Aug 2019, Lille, France. ⟨hal-02528364⟩

Share

Metrics

Record views

35

Files downloads

13