![Drawing depicting AI and Healthcare.](/sites/g/files/bhrskd331/files/styles/max_width_l_1470px/public/2024-01/machinemdstoryimage.png?itok=NRT_AGXn)
Artificial Intelligence (AI) will rapidly transform our healthcare system. Already AI devices and applications are being used to support patients in prevention and health promotion, track patient data, triage care, scan medical images, diagnose disease, and make treatment decisions. Across the continuum of care, AI is predicted to assist and potentially substitute for human caregivers, medical service providers, diagnosticians and expert decision-makers. This stands to radically improve healthcare by improving efficiency and timeliness of care. At the same, there is great uncertainty as to whether Canada’s legal and governance structures are equipped to address the diverse concerns raised by AI health technologies. These include concerns related to product safety and efficacy, algorithmic bias, privacy, informed consent to care involving AI, and impact on the provider-patient relationship.
Part of Machine MD team’s work includes examining real AI technologies, the practical issues they raise, and their current treatment in Canadian and foreign law. This approach moves beyond abstract concerns into concrete realities, helping to inform law reform with a better understanding of real-world applications. The goal is to support beneficial AI technology innovation while minimizing associated risks through appropriate legal governance. In keeping with this aim, in 2022, the Machine MD team partnered with CIFAR to host a series of online case study events. Each event assembled an interdisciplinary group of experts in AI, law, ethics, policy, and medicine to discuss the regulatory issues raised by a specific AI technology: The OR Black Box; The Suicide Artificial Intelligence Prediction Heuristic; Digital Twins; and Cardiac Arrest Prediction.
Case Study 1: The OR Black Box
A. Goldenberg, C. Régis, C. M. Flood, T. Scassa, F. Rudzicz, N. Cortez, I. Stedman and F. Martin-Bariteau
Download the Case Study 1 report (PDF, 694 KB)
The report is also available in French: Case Study 1 report in French (PDF, 705 KB).
Case Study 2: The Suicide Artificial Intelligence Prediction Heuristic
C. Mercer, S. Nunnelley, A. Goldenberg, C. Régis, C. M. Flood, T. Scassa, Z. Kaminsky, J. Chandler, N. Martin and R. Cartagena
Download the Case Study 2 report (PDF, 644 KB)
The report is also available in French: Case Study 2 report in French (PDF, 662 KB).
Case Study 3: Digital Twins
C. Mercer, S. Nunnelley, A. Goldenberg, C. Régis, C. M. Flood, T. Scassa, A. El Saddik, L. Khoury, and K. Dewhirst
Download the Case Study 3 report (PDF, 682 KB)
The report is also available in French: Case Study 3 report in French (PDF, 707 KB).
Case Study 4: Cardiac Arrest Prediction
S. Nunnelley, A. Goldenberg, C. Régis, C. M. Flood, T. Scassa, A. Ferron Parayre, P. Déziel, and V. Gruben
Download the Case Study 4 report (PDF, 809 KB)
The report is also available in French: Case Study 4 report in French (PDF, 770 KB).
These workshops were co-hosted by CIFAR and the Canadian Institutes of Health Research (CIHR)-funded Machine MD: How Should We Regulate AI in Health Care? project, with support from the Alex Trebek Forum for Dialogue. It is part of CIFAR’s AI & Society Program. CIFAR’s leadership of the Pan-Canadian AI Strategy is funded by the Government of Canada.
More details about our AI + Health research program can be found at aisociety.ca/health.