You can request an alternate format of this email by contacting the sender.

Ottawa Medical AI Research Institute

Welcome to the November 2026 edition of the Ottawa Medical AI Research Institute's monthly newsletter. Here, we share updates, insights and articles, and it's a helpful way to stay connected to what's happening in medical AI at the University of Ottawa and its affiliated hospitals and institutes.

Join Us November 25 for a Workshop on AI and Medical Imaging

Join us in person for the final session of OMARI's fall seminar series, From Pixels to Predictions: Developing AI Models for Medical Image Interpretation, on Wednesday, November 25, featuring Dr. Kevin Dick and Youssef Megahed.

This hands-on workshop will provide a technical overview of building computer vision models for medical imaging, from data preparation to clinical interpretation. Participants will explore key architectures, including convolutional neural networks, vision transformers and hybrid designs, along with challenges involving scarce labels, class imbalance and domain shift.

The session will also cover explainability, including Class Activation Mapping (CAM) and its variants, with examples from ongoing work on MetaCAM++. Interactive demonstrations on participants' own devices will connect clinical intuition with model behaviour and support trustworthy, interpretable AI in medicine.

Date: Wednesday, November 25, 2026
Time: 5 to 7 pm
Location: Roger Guindon Hall, room 2021, University of Ottawa
Format: In person

Register for the November 25 seminar

Beyond the AUC: How to Evaluate a Clinical Prediction Model

A high AUC can make a clinical prediction model look impressive. Short for the area under the receiver operating characteristic curve, AUC measures discrimination: how well a model ranks people with a particular outcome above those without it. It does not establish whether individual estimates are accurate or whether using the model improves clinical decisions.

That distinction framed the Ottawa Medical AI Research Institute’s second seminar, held in person on October 28, 2026. In Beyond the AUC: Developing and Validating Clinical Prediction Models, Dr. Steven Hawken, senior scientist at the Ottawa Hospital Research Institute and associate professor at the University of Ottawa, examined what researchers and clinicians need to know before relying on a model’s predictions. An opening slide posed a question: would an AUC of 0.95 be enough to use a model in a clinic?

Through simulated examples and analyses of real newborn screening data, the presentation showed why answering that question requires more than a performance score.

Read the full article, Beyond the AUC: How to Evaluate a Clinical Prediction Model

New uOttawa Centre for Applied AI Brings Expertise to Health Care Research

The University of Ottawa's Faculty of Engineering has launched the Centre for Applied Artificial Intelligence (CAAI), bringing together approximately 50 experts to help organizations develop and apply AI solutions to real-world challenges. Directed by Professor Lionel Briand, the Centre works with academic, industry and government partners, with a focus on developing trustworthy AI that addresses safety, privacy, security and regulatory requirements.

Health care is one of the Centre's 13 application domains. Its research spans human movement analysis, biomedical evidence processing, agentic health-care systems, affective computing and remote physiological sensing. Specific capabilities include using AI to assess joint health and predict gait, automate the search and synthesis of scientific and clinical literature, and develop conversational and recommender tools to support triage, care navigation and clinical decision-making with human oversight. Researchers are also exploring how language, behaviour and wearable sensor data can support physiological sensing and mental health assessment.

For researchers and clinicians across uOttawa's affiliated hospitals and institutes, the Centre offers opportunities to connect with engineering expertise in areas relevant to medical AI development, evaluation and implementation. Its website identifies the researchers working in each area and provides information about potential collaborations.

Read the uOttawa article

Explore the Centre for Applied Artificial Intelligence website

The Ottawa Hospital Announces $10 Million Investment in Precision Medicine

The Ottawa Hospital and CAIVAN have announced a partnership to establish the CAIVAN Centre for Precision Medicine, supported by a founding investment of $10 million.

The new centre will bring together clinical data, technology, science and clinical expertise to advance precision medicine. Artificial intelligence will be central to its work, helping researchers use clinical data to better understand patients, improve predictions and develop more precise approaches to diagnosis, treatment and prevention. The centre will initially focus on building a trusted research data environment that brings together anonymized health data to support research and clinical innovation. 

The initiative aims to develop AI-enabled clinical decision intelligence that can identify patterns, synthesize information and generate increasingly personalized predictions to support clinical decisions. Initial areas of focus will include pharmacogenomics and oncology, with plans to expand into other clinical areas. It represents a new investment in the hospital's capacity to connect AI research with clinical practice.

Read the announcement

Learn more about the centre's plans

uOttawa Research Aims to Improve Patient Scheduling and Reduce Wait Times

Patient scheduling and capacity planning are closely connected, yet health-care organizations often make these decisions separately. New research at the University of Ottawa's Telfer School of Management aims to bring them together.

Professor Jonathan Patrick has received a Natural Sciences and Engineering Research Council (NSERC) Discovery Grant for a project titled Scheduling and Capacity Planning Using Approximate Dynamic Programming. The research is developing a unified approach that links capacity allocation, scheduling decisions and wait-time targets. It combines approximate dynamic programming with machine learning and neural networks to account for the complexity and changing conditions of health-care services. The goal is to help organizations allocate resources more effectively, respond to changing demands and reduce patient waiting times.

Read the uOttawa article

Privacy Regulators Call for Responsible AI in Health and Research 

Canada’s federal, provincial and territorial privacy regulators have issued a joint resolution on the responsible use of AI in health care and health research. The resolution calls for AI systems to be privacy-protective, valid and reliable, safe, transparent, accountable and respectful of human rights throughout their full life cycle. 

For researchers and developers, that means applying privacy by design, limiting the personal information used, knowing where training data came from, using de-identified or synthetic data where possible, testing for bias and inaccurate outputs, and conducting ongoing privacy, security and AI-specific risk assessments. Researchers should also be prepared to explain how risks and benefits have been assessed, how privacy and bias risks will be mitigated, how human oversight will be maintained and how research findings will be shared transparently. 

The resolution also calls for clearer national guidance for AI-assisted health research. It recommends updating the Tri-Council Policy Statement to address issues such as re-identification risk, synthetic data and the underrepresentation of diverse populations, while strengthening the capacity of research ethics boards to assess AI-related research.

Read the full FPT resolution

Applications are Open for the Next Startup Garage Exploration Circuit 

Applications are open for the next cohort of the Startup Garage Exploration Circuit! This is your chance to turn your medical AI venture into something real with serious support behind you.

Here's what you get: Up to $12,000 in non-dilutive funding; 1:1 mentorship from seasoned professionals; expert support from lawyers, accountants, HR, brand and leadership specialists; strategic guidance toward product-market fit and investment readiness; and workspace at the eHub.

This opportunity is for uOttawa students, researchers, and alumni from any faculty; and current students or recent grads from any Ontario post-secondary institution who have a serious interest in advancing their invention or innovation for commercial or social impact.

Deadline to apply: November 30

Apply now!

ThinkRare Receives $1 Million to Support National Expansion

ThinkRare, an AI-driven tool developed at the CHEO Research Institute, has received $1 million in funding over one year from Conscience to support expansion to additional Canadian children's hospitals.

Already used in clinics at CHEO, ThinkRare uses a rule-based algorithm to analyze electronic medical records and help identify children who may have an undiagnosed rare genetic disease. Genetics specialists can then review identified patients and determine whether further evaluation is appropriate.

Implementation plans are being developed at McMaster Children's Hospital, Alberta Children's Hospital and Stollery Children's Hospital.

The funding will also support further work on disease-specific algorithms, potential connections with rare disease registries and evaluation of the economic impact of earlier diagnosis.

Read about ThinkRare's expansion

New Federal Guidance Clarifies the Use of Generative AI in Research Funding Applications

The Canadian Institutes of Health Research (CIHR), Natural Sciences and Engineering Research Council (NSERC), Social Sciences and Humanities Research Council (SSHRC) and Canada Foundation for Innovation (CFI) have updated their guidance on the use of generative AI in research funding applications and peer review. Applicants may use generative AI to help draft, summarize or translate application materials and are not required to disclose its use. However, they remain responsible for verifying the accuracy of all information, checking sources and citations, and protecting confidential or proprietary information.

The guidance makes an important distinction between applicants and reviewers. Peer reviewers are prohibited from using generative AI to review funding applications, including tools procured and securely deployed by their own institutions. This restriction covers application materials, draft reviews, notes and feedback, including their use for translation. The updated guidance and accompanying FAQs provide practical information to help researchers understand what is permitted and their responsibilities when using generative AI.

Read the updated guidance and consult the FAQs.

CARE-AI Framework Guides Responsible Health AI

Researchers involved in the CARE-AI Framework, including Dr. Lyn Sonnenberg and Dr. Jerry Maniate, have developed practical guidance to support responsible AI use across health professions education, research and clinical care.

The framework addresses values, competence, accountability and structural equity. Its accompanying implementation guide offers context-specific guidance for educational, research, clinical and administrative settings, including issues involving bias, privacy, transparency and accountability.

A recently published article from Bruyère Health explains how the framework and guide can help organizations approach AI implementation responsibly. The article highlights how patient partners helped shape CARE-AI from the outset. Dr. Sonnenberg explains that their involvement was intended to keep patients central at every stage of the framework. Dr. Maniate emphasizes the guide’s concrete examples, which people can apply directly to their work at different stages of AI adoption. The article also stresses that responsibility for safe and equitable AI use rests with institutions as well as individuals.

Read the article, CARE-AI Framework guides responsible health AI

Ottawa Applied Medical AI Summit Returns in Spring 2027

The Ottawa Medical AI Research Institute (OMARI) is planning the return of the Ottawa Applied Medical AI Summit in April 2027, following its inaugural event in May 2026. This full-day, in-person event will bring together researchers, clinicians, students, faculty and experts from across the health care, technology and research communities.

The 2027 Summit will focus on implementation — evaluation, deployment, governance, and monitoring — of AI models in research, education, and clinical settings.

More details about the program, speakers and registration will be announced in the coming months. In the meantime, revisit the 2026 Summit, including the event report, presentation materials and session recordings, which explore how LLMs are moving from promising prototypes to safe, effective and well-governed tools for real-world use.

Explore the 2026 Summit report, materials and videos

JMIR is the official journal of OMARI

Share, Subscribe and Contribute


Please share this newsletter with anyone who may be interested.

Not yet subscribed? Join our mailing list.

Visit the OMARI website regularly for updates and resources.

Make sure to follow us on LinkedIn.

Know of initiatives, events or resources we should feature on the OMARI website or in a future edition? Email iriamo-omari@uottawa.ca.

Please reply to this message with “Unsubscribe” if you do not wish to receive these emails. If you would prefer the French version of this newsletter, use this subscription form.