Ontario’s autism assessment system is under strain. While early identification is critical for helping children and families access support during key developmental windows, timely diagnosis remains difficult and long waits are common. That is the urgent problem behind a project exploring whether AI could help systematically identify children who may need autism assessment and facilitate timely diagnosis. On its own, the initiative is a strong medical AI project tied to a real need in pediatric care. But it also creates an opportunity to reflect on a wider set of questions. Moving from research to real-world use means thinking through issues such as model explainability, data governance, workflow, scale, data quality, stigma, equity, and transparency.

March 18, 2026 – The Ottawa Hospital is the first hospital in the world to evaluate the Extubation Advisor, an AI and machine learning-enabled prediction tool that helps ICU teams decide when patients are ready to breathe on their own. The device continuously monitors and analyzes a patient’s pattern of variation of their respiratory rate during trials of minimal support from the ventilator. It then uses AI to translate that altered breathing pattern into a prediction to help determine if the patient could be safely removed from the ventilator.

February 18, 2026 – A clinic visit asks a lot from everyone in the room. A patient is describing symptoms, answering questions and often being examined, all in a short window. The clinician is trying to understand the problem, decide what to do next and keep track of key details. Then comes the documentation. Notes still need to be completed, reviewed and filed. That work can pile up, especially when the number of visits per shift keeps climbing. For Dr. Venkatesh Thiruganasambandamoorthy, the pressure shows up as a tradeoff, especially during a busy shift. “Information could get lost in transit because I can either take copious notes or I can focus fully on the patient and be truly engaged.” His interest in exploring the use of an AI scribe wasn’t driven by curiosity about AI. It came from a familiar problem: even with dictation, he still had to come back after seeing a patient and write the story of the interaction. 

February 17, 2026 – For people with advanced chronic kidney disease, starting dialysis does not have to happen in an emergency room. Ideally, the transition is planned. Patients begin dialysis as outpatients, in a controlled setting, and those people tend to do well. When clinicians can see what is coming, they can prepare with patients, arrange access and make sure the transition is planned rather than rushed. Yet many patients still “crash” onto dialysis. They arrive in hospital very sick and need urgent treatment. Those unplanned starts are linked with worse outcomes. Roughly 40% of patients begin dialysis this way. That pattern is what led Dr. Gregory Hundemer, a nephrologist and clinician-researcher at The Ottawa Hospital and the University of Ottawa, to ask a straightforward question: can we tell who is likely to need dialysis soon, and act before it becomes urgent?

January 19, 2026 – In November 2025, we launched our series, Pathways from AI Research to Software as a Medical Device (SaMD). The first article explained why regulation matters early in research, introduced AI as SaMD and outlined how Health Canada’s risk-based classification framework works. In the second article in the series, we zoom in on Class I SaMD, the category where many research-based AI tools actually fall. The article explains in simple terms what Class I means, how intended use (not technical complexity) drives classification and what researchers need to understand about Class I requirements, from documentation and labelling to ongoing risk management. Through concrete examples, this can help research teams move forward with greater clarity as AI tools begin to transition toward real-world use.

December 16, 2025 – At the Equity in Health Systems (EqHS), housed within the Bruyère Health Research Institute, a project called Health CARE‑AI is giving health teams a clearer way to talk about ethics and equity when they use AI. Health CARE‑AI (Contextual, Accountable and Responsible Ethics for Artificial Intelligence in Healthcare) is a professionalism and ethics framework that sets out practical principles for AI use in health education, research and care. The initiative received support from the Faculty of Medicine’s AI Seed Funding Program in 2024, which helped the team build an evidence‑based foundation for the framework.

November 20, 2025 – Artificial Intelligence (AI) is rapidly evolving and no longer confined to controlled research settings but increasingly entering and shaping clinical practice. It holds enormous transformative potential for medicine, but translating academic research into real-world clinical application requires more than technical and clinical expertise. A new OMARI blog series explores the pathways from AI research to Software as a Medical Device (SaMD), focusing on the regulatory, technical, and institutional practices that help bridge the gap. In the first article in the series, titled Regulatory Considerations when Bridging the Research-Deployment Gap, we outline why regulation matters early in research, explain the concept of AI as SaMD, offer examples of various SaMD classifications in practice, highlight evolving guidance from health regulators unique to AI and look at what's next.

November 19, 2025 – At Bruyère Health’s Centretown Site, a robotic device called the ALTA Platform® is changing a task that happens hundreds of times a week: moving patients in and out of beds. Transfers that used to need two or three people can now be done by one trained team member at the push of a button. As a first adopter and evaluation site for a Canadian AI-driven innovation, Bruyère Health is helping to turn promising tools into real practice - another way hospitals can support development while engaging in practical research.

November 18, 2025 – Systematic reviews can be slow, error-prone and hard to scale. They often become the bottleneck before evidence-based projects even start. DistillerSR is being rolled out to help fix that, so that teams can move through the first step faster, keep reviews current over time and streamline research and guideline development work.

October 17, 2025 - The CHEO Research Institute has developed a rules-based AI algorithm, ThinkRare, that uses routinely collected clinical information from around 300,000 patient charts to flag children and youth who may have an undiagnosed rare genetic disorder. When the criteria are met, clinicians are prompted to consider referral and genetic testing, helping families get to informed care sooner.

October 17, 2025 - Starting in early November 2025, Cosmos, a platform with robust tools and a rich pool of de-identified health data from 1,809 hospitals, 41,500+ clinics and 300 million patients across participating Epic sites in Canada and worldwide, will begin rolling out at The Ottawa Hospital (TOH) and Atlas Alliance hospitals. Access to Cosmos will help drive meaningful discovery and advance research projects while supporting more informed clinical decision-making and enabling more personalized patient care.