AI For Health - February 2020

When: February 24th, Time: 4-6pm
Where: Université de Montréal - Pavillon Roger-Gaudry. Salle M-415

Agenda:

3:30 PM - Doors open
4:00 PM - Talks starts
6:00 PM - Pizza
7:00 PM - Doors close

Speakers:

Reza Forghani, Department of Radiology at McGill

Title: Artificial Intelligence Applications in Healthcare and Life Sciences: Pushing the Boundaries

Summary: There is great interest in applications of AI in life sciences and especially in healthcare. Although this technology has great potential for future transformation of healthcare processes, there are major basic and structural barriers in our healthcare system and research processes that need to be recognized and overcome for successful research and development and especially development of AI applications that will be adopted and can succeed in clinical practice. In this presentation, I will provide an overview of the Augmented Intelligence & Precision Health Laboratory’s vision for a multi-disciplinary collaborative facility to help accelerate development and implementation of different medical AI tools. This will include a high-level discussion of structural challenges, opportunities, as well as some of the major projects and initiatives being undertaken by the laboratory.

And

Mathieu Blanchette from the School of Computer Science (McGill Center for Bioinformatics) at McGill University.

Title: Learning from watching genomes evolve

Summary: The genomes of more than one hundred vertebrate genomes are now largely sequenced. How can one make use of this massive amount of evolutionary information to better understand the origin and function of portions of the human genome? In this talk, I will first discuss how the genomes of ancestral mammalian species can be reconstructed with surprisingly high accuracy from the genomes of extant species. I will then present how inferred ancestral sequences can be used to improve the detection of ancient evolutionary events such as transposable element and pseudogene insertions that have shaped mammalian genomes. Finally, I will then introduce algorithmic and machine learning approaches that make use of inferred ancestral DNA sequences to improve the accuracy of transcription factor binding site and micro-RNA target site prediction.