MSACL 2019 US Abstract

High-Throughput Metabolomics Combined with Artificial Intelligence for Prognostic and Diagnostic Assessments

Pier-Luc Plante, Élina Francovic-Fontaine, Francis Brière, Nancy Boucher and Jacques Corbeil
Laval University, Quebec, Canada

Warning: Undefined variable $headshot in /var/www/html/view_abstract/view_abstract_in_program.php on line 704

Jacques Corbeil (Presenter) Laval University

Presenter Bio: Prof. Jacques Corbeil focuses on using the latest techniques in bioinformatics and machine learning to assist diagnostic, prognostic and response to treatment. Dr. Corbeil is using state-of-the-art instrumentation and methodologies to facilitate the interpretation of complex data. Dr. Corbeil operates at the interface of computer sciences and omic sciences. He uses high throughput metabolomics coupled with artificial intelligence to assist in disease diagnostic and prognostic. Dr.Corbeil obtained his Ph.D. from New South Wales University, moved to the University of California, San Diego for his postdoctoral studies, rising to the rank of associate professor. He is now a professor of medicine in the department of molecular medicine at Laval University, Quebec, Canada and holds the Canada Research Chair in Medical Genomics.

Relevant Financial Disclosures (within past 24 months) Committee/Board/Advisory Board Compute Canada Salary evelo biosciences, Genocean Discussing Products? NO, presenter will NOT present educational content related to products/services of the ineligible company(ies) listed above. Conflict mitigation NOT REQUIRED.

Fast ionization technologies such as Laser Diode Thermal Desorption (LDTD) and Desorption ElectroSpray Ionisation (DESI) enables the analysis of samples at an incomparable speed. Combining these techniques with machine learning algorithms enables the creation of new prognostic and diagnostic tools that are in line with the needs of the clinic: fast, low cost and accurate. As a proof of concept, we accurately differentiated patients with influenza-like symptoms that were actually infected with influenza virus or not. The process can be completed in minutes using the LDTD ion source, a high-resolution mass spectrometer and state-of-the-art machine learning algorithms directly using an extract of a nasopharyngeal swab without the need for any culture. The identification of the peaks is assisted by combining m/z and ionic mobility values using a deep learning algorithm.