Machine Learning Applications in Medicine - Is There Still Room for Signal Processing Innovations?
Tiago Falk
September 21, 2022, Wednesday, 3:00 PM - 4:00 PM EST
There is no doubt that machine learning, in particular deep learning, has revolutionized the medical field over the last decade. What used to be done with a Fourier or a wavelet transform has now been replaced by a convolutional neural network or a multi-head transformer. This raises the question: in the era of large (deep) models, is there still room for signal processing? In this talk, my task will be to convince you that YES, the research community still needs innovations in signal processing to make artificial intelligence, well, more intelligent. I hope to convince you of this by showcasing several real-world applications that we have tackled in the Lab, from monitoring nurse anxiety levels with wearables to detecting COVID-19 infections from coughs monitored via smartphones from home. In particular, I will show the impact that signal processing combined with machine learning can have on i) improving accuracy with very noisy data collected in highly ecological settings, ii) making machine learning tools more trustworthy and resilient against adversarial attacks and deep fakes, and iii) reducing the carbon footprint required to train large complex models, thus making existing tools "greener".
Dr. Falk received the BSc degree from the Federal University of Pernambuco, Brazil, in 2002, and the MSc and PhD degrees from Queen's University, Canada, in 2005 and 2008, respectively, all in electrical engineering. In 2007, he was a visiting Research Fellow at the Sound and Image Processing Lab, Royal Institute of Technology (KTH), Sweden, and in 2008 at the Quality and Usability Lab, Deutsche Telekom/TU Berlin, Germany. From 2009-2010 he was an NSERC Postdoctoral Fellow at Holland-Bloorview Kids Rehabilitation Hospital, affiliated with the University of Toronto. Since 2010, he has been an Assistant Professor at the Institut National de la Recherche Scientifique (INRS) in Montreal, Canada. Dr. Falk is also an Adjunct Scientist at Holland-Bloorview and a Research Advisor for InteraXon - Thought Controlled Technologies in Toronto.