The growth and development of open-source software and digitized and publicly available data sources have made the application of machine learning (ML) methods to laboratory medicine highly accessible. A recent IFCC working group has published guidelines for using machine learning in laboratory medicine. In the first of two webinars, two working group members will describe the basis and application of the guidelines. Then, real examples from the literature will illustrate how the guidelines can avoid misusing machine learning techniques or misinterpreting outcomes from those techniques. All laboratory professionals will benefit from a basic understanding of ML concepts, as these technologies increasingly become part of our workflows. As with other emerging technologies in our laboratories, some of us will seek a more advanced understanding to contribute to the development and technical evaluation of these methods, engaging in driving this new area forward as collaborators in method development and implementation. More of us will be consumers and decisions makers, called upon to critically review and generally evaluate ML-based products to ensure their safe, effective, and sustained use. Broad recognition of common failures and accepted best practices by our community will improve the quality of science and set higher standards for ML innovations being applied in clinical laboratories.

This webinar comprises of three following presentations of 20 min each followed by 20 min of panel discussion at the end.

Chair: Prof. Tony Badrick

Talk 1- "The IFCC Guidelines Document" -  Prof. Tony Badrick

Talk 2- "How to evaluate machine learning models in Laboratory Medicine?" - Dr. Andreas Bietenbeck

Please feel free to share this event in your social media with #IFCCLive Webinars.

Click here to convert to your time zone