Deep learning with ECG data in the ICU: From modelling to actionable AI (NISS-CANSSI Collaborative Data Science Series)

Overview:

Deep learning with ECG data in the ICU: From modelling to actionable AI is part of the NISS-CANSSI Collaborative Data Science Series, which highlights cutting-edge applications of data science across disciplines. This session will examine how deep learning methods can be leveraged to analyze electrocardiogram (ECG) data collected in intensive care units (ICUs), where rapid, reliable interpretation of patient information is crucial. The discussion will span the full pipeline—from methodological advances in modeling ECG signals to the translation of AI-driven insights into tools that can support real-time decision-making at the bedside. Together, the speakers will bridge perspectives from computer science and clinical practice, offering insights into both the technical challenges of modeling high-dimensional physiological time-series data and the practical considerations required to make AI trustworthy, interpretable, and actionable in critical care environments.

Speakers

Parvin Mousavi, Director, School of Computing at Queen’s University

David Maslove, Associate Professor & Clinician Scientist, Departments of Medicine and Critical Care Medicine, Queen’s University; & Internist and Intensivist, Kingston Health Sciences Centre

Moderator

Joel Dubin, Statistics & Actuarial Sciences ​Health Data Science Lab (HDSL) Lead, University of Waterloo

About the Speakers

David Maslove is a Clinician Scientist in the Departments of Medicine and Critical Care Medicine at Queen’s University, and an Internist and Intensivist at Kingston Health Sciences Centre. His research focuses on the use of physiologic and genomic data to advance precision medicine in the ICU. Dr. Maslove completed medical school and residency in Internal Medicine at the University of Toronto. He trained in Critical Care Medicine at Stanford University where he also completed graduate studies in Biomedical Informatics. He is a member of the Canadian Critical Care Trials Group, and the Society of Critical Care Medicine, and since 2018 has been the Associate Editor for Data Science for Critical Care Medicine. See Profile

Parvin Mousavi is the Director of the School of Computing at Queen’s University. Her research interests are in Computer-aided diagnosis and interventions. These include: Machine learning techniques for in silico inference and prediction Analysis of ultrasound images and signals for enhancement of cancer detection Image-aided, computer-assisted diagnosis of disease Ultrasound-guided interventions Knowledge discovery from high throughput biological data Quantitative modeling and reverse engineering of gene regulatory networks Analysis, segmentation and classification of fluorescence microscopy images Chromosome and cell imaging. See Profile

About the Moderator

Dr. Joel Dubin is a leading methodological statistician whose work focuses on longitudinal data analysis, especially multivariate and time-varying outcomes. He develops tools for modeling multiple physiological measurements over time—such as heart rate, respiratory rate, or blood pressure—using advanced techniques like curve-based methods, derivatives, and lagged effects. He also works on change-point and latent response models, prediction models that leverage subject similarity, and methods to handle missingness and complexity in real-world health data. His research spans a range of applications including intensive care, electronic health records, mobile health, child and aging populations, nephrology, cancer, nutrition, smoking cessation, and environmental health. Dr. Dubin received his M.S. in Applied Statistics from Villanova University, then worked in health services research at the U.S. Veterans Affairs and the MD Anderson Cancer Center. He earned his Ph.D. in Statistics from UC-Davis, followed by a faculty appointment at Yale. In 2005 he joined the University of Waterloo with a joint appointment in Statistics & Actuarial Science and Health Studies & Gerontology (now the School of Public Health Sciences). See Profile

About the NISS-CANSSI
Collaborative Data Science Web Series:

The NISS-CANSSI Collaborative Data Science initiative that the National Institute of Statistical Sciences (NISS) in collaboration with the Canadian Statistical Sciences Institute (CANSSI) brings together experts from various fields to tackle complex data challenges through interdisciplinary teamwork and innovative methodologies.

Goals of the Initiative

The goal is to foster progress in:

• Developing new ideas for experimental and observational data-driven learning and discovery that address key questions at the cutting edge of science and scientific deduction;
• Quantifying and summarizing uncertainty in data-driven theories, as well as complex Data Science models, algorithms, and workflows; and
• Establishing new practices for scientific reproducibility and replicability through Data Science.

Featured Webinars

Data Science Techniques for Control of Assistive Devices After Neurological Injury

Date: Thursday, June 12, 2025 at 1-2pm ET
Speakers: Lauren Wengerd, Ohio State University, Depart of Rehabilitation Science and Dave Friedenberg, Battelle; Moderator: Nancy McMillan, Battelle

From MPEG-4 to Deep Learning: Transforming Audio-Visual Analytics for Healthcare and Beyond

Date: Thursday, May 8, 2025 at 1-2pm ET
Speakers: An-Chao Tsai, Department of Computer Science and Artificial Intelligence, National Pingtung University and Anand Paul, LSU Health-New Orleans; Moderator: Qingzhao Yu, Associate Dean for Research at the School of Public Health, Louisiana State University Health, New Orleans

Astronomy & Cosmic Emulation