Research Summer Student 2026

Union: Non-Union 
Number of Vacancies: 1
New or Replacement: New
Site: Toronto General Hospital
Department: Cardiology
Reports to: Dr. Chris McIntosh
Hours: 35
Salary: $17.60 - $20.91 per hour
Shifts: Monday to Friday
Status: Temporary Full Time
Closing Date: February 27th, 2026

Position Summary:
Traditional AI tools for assisting clinical care are restricted to a single modality (e.g., X-ray). Thus, they fail to take into account the more complete information available through additional data modalities, including electrocardiograms (ECG), bloodwork, previous patient reports, medical histories, and so on. 

Recent advances in multimodal large language models (MM-LLM) enable researchers to build models considering data from multiple data modalities and how they interrelate. Our group is building such models to advance cardiovascular care. 
 
Duties:

• Build and maintain reproducible data processing pipelines for multimodal clinical datasets (e.g., ECG-derived features, labs, clinical notes/reports) 
• Perform data quality assessment and cleaning (missingness, outliers, unit harmonization, cohort definition checks) 
• Support model development and evaluation in PyTorch (training runs, inference, metrics, error analysis) 
• Contribute to documentation of datasets and evaluation procedures (data dictionaries, dataset cards, experiment logs) 
• Summarize approaches to prospective validation and deployment of clinical AI systems, with attention to temporal validation, leakage prevention, and monitoring 
• Understand the development and application of multimodal AI models in healthcare 
• Learn about methods of prospective AI deployment 
• Contribute to the development and validation of a large multimodal AI model 
• Contribute to data cleaning and preparation 
• Survey of different prospective validations of cross-modal AI models 

• Enrolled in a Computer Science, Biomedical Engineering, Data Science, or related degree program 
• Strong Python programming skills and familiarity with software best practices (version control, readable code, debugging) 
• Experience in the development and application of machine learning in python using PyTorch 
• Coursework or equivalent experience in at least one of: machine learning, statistics, signal processing, or data engineering 
• Interest in applying ML to healthcare and comfort working with messy real-world data 

Why join UHN?

In addition to working alongside some of the most talented and inspiring healthcare professionals in the world, UHN offers a wide range of benefits, programs and perks. It is the comprehensiveness of these offerings that makes it a differentiating factor, allowing you to find value where it matters most to you, now and throughout your career at UHN.

• Competitive offer packages
• Government organization and a member of the Healthcare of Ontario Pension Plan (HOOPP https://hoopp.com/)
• Close access to Transit and UHN shuttle service
• A flexible work environment
• Opportunities for development and promotions within a large organization
• Additional perks (multiple corporate discounts including: travel, restaurants, parking, phone plans, auto insurance discounts, on-site gyms, etc.)

Current UHN employees must have successfully completed their probationary period, have a good employee record along with satisfactory attendance in accordance with UHN's attendance management program, to be eligible for consideration.

All applications must be submitted before the posting close date.

UHN uses email to communicate with selected candidates.  Please ensure you check your email regularly.

Please be advised that a Criminal Record Check may be required of the successful candidate. Should it be determined that any information provided by a candidate be misleading, inaccurate or incorrect, UHN reserves the right to discontinue with the consideration of their application.

UHN is an equal opportunity employer committed to an inclusive recruitment process and workplace. Requests for accommodation can be made at any stage of the recruitment process. Applicants need to make their requirements known.

We thank all applicants for their interest, however, only those selected for further consideration will be contacted.