Agenda - Day 2

Gain insight into today’s sessions so you can get the most out of your conference experience.

Mental health data represents one of healthcare’s most valuable yet underutilized resources, characterized by unprecedented complexity, fragmentation, and unique governance challenges that traditional healthcare frameworks cannot address. Move beyond data collection toward data activation by creating trusted infrastructures that accelerate discovery, enable precision mental health interventions, and build the foundation for responsible AI innovation. Master the success factors to:

• Navigate the distinctive challenges of sensitive data, from managing privacy concerns and reducing stigma to handling the inherent heterogeneity that makes this data both powerful and problematic.
• Execute successful multi-modal data integration to develop more holistic pictures of your patients.
• Build stakeholder trust while maximizing data utility.

Achieve mental health innovation by transforming fragmented data into meaningful, equitable, and actionable insight.

Health organizations across Canada are under growing pressure to scale beyond pilot projects, modernize aging infrastructure, and support increasingly complex reporting and payment models. Too many initiatives stall because systems are fragmented, data is inconsistent, and teams are burdened by manual processes that cannot keep up with clinical or regulatory demands. Cut through operational complexity and equip your organization to:

• Shift from capital-intensive legacy systems to SaaS models that deliver predictable costs, continuous innovation, and faster deployment cycles.
• Create a unified data foundation that integrates with multiple EHRs, normalizes terminology, and provides consistent, high-quality information for population health and value-based care.
• Enable real-time monitoring of key indicators—attachment, quality measures, risk scoring—while reducing the time and resources spent on CIHI reporting and other regulatory submissions.
• Deploy managed population health capabilities that support alternative payment models, improve patient navigation, and free clinical teams from repetitive administrative work.

Scale value-based care by turning perpetual pilots into reliable, measurable system performance.

Take a deep dive down the innovation rabbit hole in one of our roundtable discussions. Share common challenges and best practices with your healthcare data peers on a topic of your choosing:

1. The Role of Patients as Data Stewards: Explore new models for engaging patients in the governance, access, and stewardship of their own data — from consent design to co-ownership and dynamic permissions.
2. Scaling AI Ethics Oversight without Slowing Innovation: Discuss practical ways to build lean, scalable AI governance frameworks that manage risk while keeping innovation timelines intact.
3. Safe Entry Points for Generative AI in Healthcare
   What GenAI use cases are emerging in Canadian healthcare? Explore low-risk, high-value pilots (e.g. summarization, document classification, patient comms) and the governance required to scale safely.
4. Future Metrics —Measuring Value, Not Just Volume: Rethink performance and success metrics for healthcare data initiatives, prioritizing equity, outcomes, experience, and long-term system value over traditional activity-based KPIs.
5. Making Insight Flow —Reducing Friction from Insight to Action in Clinical & Admin Settings. Bridge the final mile between analytics and operational change by identifying barriers to adoption, workflow misalignment, and credibility gaps.

Hospital Montfort began its data transformation with a clear-eyed assessment of its current state: fragmented governance, limited scalability, and inconsistent data literacy. To address these gaps, Montfort designed a strategic roadmap anchored by a robust governance framework, phased policy development, and technology modernization. Adopt best practices to:

• Establish a centralized data governance model by defining clear roles and responsibilities and implementing a Microsoft Fabric Lakehouse architecture for scalable, real-time analytics.
• Optimize an AI-ready, compliant, and agile data ecosystem — enabling advanced analytics, responsible AI deployment, and continuous innovation across the organization.

Master your healthcare data strategy with a governance-first approach that enables innovation while ensuring trust, compliance, and scalability.

As Canada accelerates AI adoption in healthcare, a defining question emerges: should innovation come before privacy? With Bill C-27, PHIPA, and HIA reshaping the privacy landscape, healthcare leaders must balance progress with protection.

Advocates for stronger privacy argue that trust is the foundation of transformation — without it, adoption and data quality collapse. Others contend that excessive regulation stifles innovation, delaying AI-driven insights that could save lives. This panel explores how Canada can navigate these tensions through:

• Embedding privacy-by-design in AI and analytics.
• Advancing de-identification, consent, and portability
• Introducing AI transparency and auditability to sustain trust.

Can Canada innovate responsibly — or must privacy always come first?

As the demand for evidence-based care and research accelerates, healthcare organizations are increasingly looking to empower clinicians and researchers with direct access to the data they need — without relying on overburdened technical teams. Build modern self-service analytics platforms that enable frontline staff and researchers to query, visualize, and model data — safely and at speed. Receive a blueprint to:

• Build intuitive data access tools that align with clinical workflows and research protocols
• Implement governance frameworks that support safe self-service use of sensitive health data
• Curate reusable data assets, definitions, and query templates to reduce duplication
• Enable responsible innovation by embedding ethics, privacy, and reproducibility into platform design

Transform self-service into it’s a strategic lever for unlocking the full value of health data while protecting what matters most.

Sunnybrook Health Sciences Centre has embarked on an ambitious journey to elevate its analytics maturity by building out a modern reporting infrastructure that supports both real-time decision-making and longer-term forecasting. Data staging and advanced ETL querying capabilities enable integration of data from multiple clinical and operational systems, in turn supporting advanced reporting in Power BI and the use of predictive modelling techniques. This session will review how our teams:

• Integrate millions of rows of data across multiple systems to enable real-time and retrospective analysis
• Streamline reporting using Power BI for both clinical teams and executive leadership
• Built predictive modelling capabilities to support operational and clinical planning
• Overcame the complexities of healthcare-specific data formats and reporting hierarchies

The Canadian Institute for Health Information (CIHI) is partnering with provinces and territories to unlock near real-time hospital data — enabling faster insights on emergency department trends, capacity challenges, and emerging health issues. This initiative marks a major step towards improving data timeliness and quality, easing coder workload, and enhancing clinical documentation through AI assisted automation and coding. Align yoursefl with:

• CIHI’s vision for fit-for-purpose, real-time hospital data flows.
• Collaborative models for innovation across jurisdictions.
• Provincial partner insights on how AI-assisted coding is reshaping data accuracy, efficiency, and readiness for analytics.

Building the real-time data backbone for a smarter, more responsive Canadian health system

With over 100,000 cases registered in its system, Acclaim Ability Management is turning disability and absence data into a proactive tool for healthcare workforce resilience. In a sector already strained by physical and mental health challenges, this session explores how data-driven insights can help predict, prevent, and manage employee disabilities—from repetitive strain to mental health-related absences. Take back to your organization strategies to:

• Use trend analysis to identify and address high-risk roles and conditions early
• Build predictive models that reduce avoidable absences and lost productivity
• Apply mental health data to shape more responsive employer supports
• Translate frontline case data into actionable workforce strategies

Empower HR, health and safety, and clinical leaders to act on data—not just react to it—by building systems of care that support those who deliver care.

Psoriatic Arthritis affects nearly 300,000 Canadians, yet treatment selection remains a trial-and-error process due to fragmented datasets, strict privacy constraints, and limited computational infrastructure. To address this, our team built a federated, multi-omic learning platform that brings models to the data—spanning genomics, proteomics, metabolomics, and clinical records across multiple hospitals. The approach preserves patient privacy, accelerates precision-medicine research, and introduces a revenue-sharing model that lets institutions participate directly in the value they help create.

Leave with a practical strategy to:

• Deploy federated learning across hospitals while keeping all patient data on-premise
• Combine multi-omic data streams to improve treatment-response prediction for complex diseases
• Leverage validated genomic markers (HLA-B27, HLA-C06, TNFAIP3, IL23R, TYK2) within scalable ML pipelines
• Achieve strong predictive performance using ensemble approaches, with retrospective AUCs up to 0.96
• Align hospitals, researchers, and industry partners using an equitable, revenue-sharing commercialization model

Fuel more efficient clinical research, stronger predictive models, and scalable precision-medicine programs through federated multi-omic AI.

Review the key solutions and takeaways from the conference. Source a summary of action points to implement in your work.