Professor John Kelleher, Chair of Computer Science at Trinity College Dublin and Director of the ADAPT Centre, delivered his inaugural lecture “Aging, Artificial Intelligence, and Outcomes: Modeling Age in Stroke Risk and Patient Prognosis” to a full house on 7 November in the Trinity Business School.
The event was attended by Provost, Dr Linda Doyle, the Dean of the Faculty of Science, Technology, Engineering, and Mathematics (STEM), Professor Sylvia Draper, along with members of Professor Kelleher’s family, friends, colleagues, and students.
In his lecture, Professor Kelleher reflected on more than two decades of research at the intersection of machine learning, natural language processing, and AI for health. He traced developments in the technology and unpacked complex computer science concepts, from weighted-sum functions and neural networks to the challenges of building interpretable, age-aware risk models for stroke.
Drawing on findings from collaborative research, he highlighted how age is often the dominant factor in cardiovascular risk prediction, yet also one of the most misunderstood. His work demonstrates that chronological age alone can obscure the real picture, and that more nuanced, data-driven approaches, such as hierarchical models, can lead to better, fairer predictions for patients of all ages.
Professor Kelleher also emphasised the value of interdisciplinary collaboration, acknowledging the statisticians, clinicians, and computer scientists who contributed to the research featured in the lecture. He closed by expressing deep gratitude for the honour of his role as Chair of Computer Science and a tribute to his family who have supported him on the journey. The lecture showcased the depth of Professor Kelleher’s expertise and the broader strengths of ADAPT and Trinity in shaping AI research that serves society.
