Congratulations to the four DMCBH members who have received funding from the UBC AI and Health Network’s Fellows Program!

In total, 11 research projects have been funded through the inaugural UBC AI and Health Network Fellows Program, which supports researchers harnessing artificial intelligence (AI) to transform health research and improve patient care across BC.

Launched earlier this year with support from a transformative $22.5 million gift from the Gordon B. Shrum Charitable Fund to the UBC Faculty of Medicine, the UBC AI and Health Network unites UBC’s strengths in AI, health research and biomedical innovation to support health system transformation through AI-driven innovation.

The Network’s Fellows Program enables principal investigators (PIs) to recruit, mentor and support postdoctoral and clinical fellows conducting interdisciplinary research at the nexus of AI and health. Each award provides $100,000 towards the hiring of a postdoctoral or clinical fellow.

The selected projects for 2025 align with the Network’s core focus areas, spanning clinical practice, to translational sciences, data sharing, end-user acceptance and training. Each project advances the Network’s mission to develop and deploy innovative AI tools that enhance timely access to equitable healthcare.

Funded through the UBC AI and Health Network, with generous support from Canada’s Immuno-Engineering and Biomanufacturing Hub (CIEBH) and the BC Cancer Foundation, the Fellows Program is part of a broader effort to expand B.C.’s leadership in AI-enabled health innovation.

Learn more about the projects that are being led by DMCBH members:

AI Integrated Diagnostic Platform for Mild Traumatic Brain Injury

Principal investigator: Naznin Virji-Babul, Faculty of Medicine

Description: Concussions, or mild traumatic brain injuries (mTBI), are common yet difficult to diagnose accurately. Current assessments depend on patients’ self-reported symptoms and clinician judgment, which can be inconsistent and delay care.

This project will develop an AI-powered diagnostic tool that integrates electroencephalography (EEG), a non-invasive measure of brain activity, with clinical data such as symptoms, injury details, and medical history. Using advanced agentic AI and Transformer-based models, the system will learn to detect and monitor concussions more accurately and objectively than current approaches. Interpretable AI models and clinician-friendly tools will be designed to integrate seamlessly into healthcare settings while maintaining strong privacy protections.

Ultimately, this research aims to transform concussion diagnosis and management by providing faster, more consistent, and evidence-based assessments, improving outcomes for patients, and supporting more effective, data-driven care.

 

An Explainable AI Framework for Collaborative Assessment of Motor Disorders using De-Identified 3D Biomarkers

Principal investigator: Timothy Murphy, Faculty of Medicine

Description: Parkinson’s disease (PD) affects movement and is currently assessed using clinical rating scales that rely on human observation, which can be subjective and inconsistent. Although AI has revolutionized many areas of medicine, its use in assessing movement disorders like PD is still limited, mainly due to patient privacy concerns and the lack of transparent, explainable AI models. This project will develop an AI tool that can analyze patient movement from video in a secure and understandable way. The system uses advanced computer vision and 3D modeling to represent a person’s motion through an anonymized digital avatar, protecting privacy while revealing how the AI reaches its conclusions.

Clinicians can review and interact with these visualizations, helping the AI improve through feedback. The result will be a privacy-preserving, explainable AI platform that supports more objective and collaborative diagnosis of Parkinson’s disease.

 

Decoding Alzheimer’s Disease via Spatial Redoxomics and Interpretable Multimodal AI

Principal investigator: Xin Tang, Faculty of Science
Co-investigators: Freda Miller, Faculties of Medicine and Science, Margo Seltzer, Faculty of Science

Description: Alzheimer’s disease (AD) arises from a complex mix of biological changes in the brain – including oxidative stress, which disrupts the cell’s chemical balance, and the accumulation of amyloid-β (Aβ) plaques that damage neurons. However, scientists still do not fully understand how these changes interact with the brain’s many genes and cell types.

This project will use AI to uncover these relationships by analyzing data from lab-grown human “mini-brains” known as organoids. Using cutting-edge imaging and spatial genomics tools, researchers can measure oxidative stress, amyloid plaque formation, and gene activity in the same cells. By combining these rich data sources, the team will develop an interpretable AI model that links oxidative damage and Aβ accumulation to specific genes and cells.

This work will generate powerful new tools for studying AD and identify potential biomarkers and therapeutic targets, helping pave the way toward earlier diagnosis and more effective treatments.

 

DiagTrace: Making Cancer Diagnosis Traceable with Knowledge Graphs and Chain-of-Reasoning

Principal investigator: Xiaoxiao Li, Faculty of Applied Science
Co-investigators: Zu-Hua Gao, Faculty of Medicine, Gang Wang, Faculty of Medicine

Description: Cancer care depends on information from medical reports and patient updates, but much of it is trapped in free text that requires time-consuming manual review. DiagTrace uses explainable AI to make this information clear, traceable, and actionable. It builds a structured “knowledge graph” linking key details such as tumor site, stage, and biomarkers from BC Cancer reports, with each fact cited to its source.

Using this foundation, DiagTrace creates concise, auditable summaries that show how conclusions are reached (e.g., lesion → biomarker → therapy eligibility) and flag uncertainties or missing data. The system combines advanced language models with strong privacy and fairness safeguards, integrating patient-reported symptoms and preferences. Co-designed with clinicians, DiagTrace acts as an assistant, helping doctors by triaging reports and providing clear, evidence-based summaries to improve the speed and clarity of cancer care. The project will deliver a validated prototype and a toolkit for scaling AI-enabled diagnostics across cancer centers.