Séminaire du CIMMUL – Zahra Yazdani

Extracting structural properties from living neuronal networks using machine learning and graph theory

Zahra Yazdani

Centre de recherche CERVO
Université Laval

Résumé

More than ten percent of people in G7 nations are affected by brain diseases, yet current treatments only manage symptoms without addressing the root causes. The Neuro-CERVO Alliance for Drug Discovery is working to change this by using artificial intelligence to find biomarkers for personalized treatments. I’m part of a project within this initiative that examines neuronal cell cultures derived from human induced pluripotent stem cells to understand their key structural and functional properties.

In my presentation, I’ll detail how we use digital holography microscopy alongside machine learning techniques and graph theory to examine cultured neuronal networks. This innovative approach meticulously processes images of living neurons in culture, segmenting out neuronal processes and cell bodies, to construct a detailed graph model. This model, referred to as the graph fingerprint of the culture, represents cell bodies as vertices and their potential interconnections as edges. Through this model, we can identify various graph features to understand the network organization at both local and global levels. Our analysis of rat-derived neuronal networks has uncovered key graph features, such as Density and Modularity, which are critical for classifying neural networks into their developmental stages.

These findings not only shed light on the development of neuronal networks at a microscopic level but also pave the way for future research on the functional and structural organization of neuronal networks, especially in cultures derived from patients with early-stage brain disorders. »

_

Le séminaire aura lieu au local 2880 du pavillon Alexandre-Vachon et en ligne.

Pour rejoindre la réunion Zoom :
https://ulaval.zoom.us/j/62680136430?pwd=eldBYjdNTG5QR2VxTTFqbVM4UGVRZz09

Meeting ID: 626 8013 6430
Passcode: 693150