Dr. Alison Q. O’Neil

Principal Scientist – Canon Medical Research Europe

Bio: Alison Q. O’Neil is a Principal Scientist in the AI Research Team at Canon Medical Research Europe and is an Honorary Research Fellow at the University of Edinburgh. She received her Engineering Doctorate (EngD) degree from Heriot-Watt University in 2016. Since 2015, she has worked as an industrial research scientist for Canon, in the Image Analysis and AI Research teams. Alison has 5 patents granted and has authored over 40 technical publications. She leads a team working on machine learning for imaging and precision medicine applications, in the domains of medical image analysis, natural language processing (NLP), and multimodal AI. Her interests include robust representation learning, multimodal learning, learning with less supervision, and the integration of causal techniques and knowledge graphs with deep learning methods.

Keynote talk: Knowledge-driven semantic representations for predictable predictions

Clinicians are increasingly embracing an AI-powered healthcare future. However, few AI models have yet been deployed in clinical practice. An important barrier is lack of robustness of current AI models, leading to unreliable and potentially unsafe predictions for out-of-distribution data. This talk will explore how enforcing semantic representations which reflect clinically meaningful concepts such as anatomical structures and markers of disease can allow greater controllability and transparency in the context of imaging and multimodal healthcare applications. 

Prof. Dr. Tal Arbel

Professor of Electrical and Computer Engineering – McGill University

Bio: Tal Arbel is a Professor in the Department of Electrical and Computer Engineering, where she is the Director of the Probabilistic Vision Group and Medical Imaging Lab in the Centre for Intelligent Machines, McGill University. She is a Canada CIFAR AI Chair – Mila (Montreal Institute for Learning Algorithms).  Arbel’s research focuses on development of probabilistic, deep learning methods in computer vision and medical image analysis, for a wide range of real-world applications, with a focus on neurological diseases. She is a recipient of the 2019 McGill Engineering Christophe Pierre Research Award. She regularly serves on the organizing team of major international conferences in computer vision and in medical image analysis (e.g.  MICCAI, MIDL, ICCV, CVPR). She is currently the Editor-in-Chief and co-founder of the arXiv overlay journal: Machine Learning for Biomedical Imaging (MELBA).

Keynote talk: Towards Safe and Trustworthy Causal Models for Image-Based Personalized Medicine

Precision medicine involves choosing a treatment that best balances efficacy against side effects for the individual, where decisions rely on statistics across demographic groups and established clinical markers. Causal models for personalized medicine based on patient images have enormous potential to improve healthcare and drug development, by providing personalized predictions of future patient outcomes and treatment responses across heterogenous populations. Yetthese models currently remain underexplored, and open challenges presented by real clinical contexts hinder their safe clinical deployment. In this talk, we describe the first causal inference model for personalized medicine from medical images acquired from patients with neurological disease during a series of clinical trials. We then describe strategies to improve the safety and reliability of the models, such as building uncertainty-aware causal models for image-based personalized medicine.