April 11, 2019

Development of an automated system guided by artificial intelligence that may accelerate discovery of drugs to repair damaged nerve cells


MS Society funded researcher, Dr. Timothy Kennedy, McGill University, developed an automated self-guided system to better understand specialized cells in the central nervous system, called oligodendrocytes. Oligodendrocytes are critical because they produce a protective layer, myelin, around nerve fibers. In multiple sclerosis (MS), myelin becomes damaged and is inadequately repaired. Without myelin the communication between nerve cells is disrupted, and the body does not adequately transmit the instructions necessary to perform basic functions.

Overview of Research:

In this study, researchers developed a new automated and self-guided system to assess the ability of oligodendrocytes to produce myelin on artificial nerve-like fibers. The system is designed to detect myelin formation using automated microscopy and measured with a new machine learning algorithm (a type of artificial intelligence that guides software to self-learn in order to complete a task), eliminating the need for human analysis, which is time consuming and slow. The researchers tested this new method against existing approaches and demonstrated that the system has the same accuracy of a human expert, while dramatically enhancing the speed of analysis, and removing human bias and variability of measurements. This new system and approach can be expanded and used to identify new drugs that stimulate oligodendrocyte activity and enhance myelin production – dramatically accelerating the discovery of drugs that have the potential to repair damaged nerve cells.

Read more about this research published in Communications Biology.

*Funding for this project was provided by the MS Society of Canada, the International Progressive Multiple Sclerosis Alliance, and the McGill University Healthy Brains for Healthy Lives initiative.