# UBC researchers develop AI model to detect heart disease risk in breast cancer patients

> A new AI model from UBC Okanagan and BC Cancer flags cardiovascular risk in breast cancer patients using routine CT scans already taken for treatment planning. It outperformed traditional risk tools in accuracy, with no extra tests needed.

- Type: AI news
- Page: https://completeaitraining.com/news/ubc-researchers-develop-ai-model-to-detect-heart-disease/
- Published: 2026-05-24
- Updated: 2026-05-25
- Author: Jeroen Erne
- Jobs: Science and research
- Reading time: 1 min

### AI Model Flags Heart Disease Risk in Breast Cancer Patients

Researchers at UBC Okanagan and BC Cancer - Kelowna have developed an AI model that identifies breast cancer patients at high risk for cardiovascular disease by analyzing routine chest CT scans alongside clinical health records.

Breast cancer patients face elevated cardiovascular risk compared to the general population. Cardiovascular disease kills an estimated 17.9 million people annually worldwide, making it a critical concern for oncology clinicians.

"Breast cancer patients already face immense challenges," said Dr. Mohammad Shehata, a computer science professor at UBC Okanagan. "Cardiovascular disease is a critical and often overlooked health threat they endure."

#### How the Model Works

All breast cancer patients undergoing radiation therapy receive chest CT scans for treatment planning. The new model repurposes these existing scans to assess cardiovascular risk without requiring additional imaging.

The AI system combines CT imaging data with electronic health records-including age, hypertension status, diabetes history, and family background-to generate personalized risk assessments. Traditional risk models rely solely on clinical factors like age and symptom presence.

"What makes this approach powerful is the use of multimodal artificial intelligence," said Shehata. "By integrating imaging and clinical data, the model can capture complex patterns that traditional tools simply miss."

#### Performance and Clinical Value

The model demonstrated higher predictive accuracy than existing methods, according to research published in Radiotherapy and Oncology. It identifies subtle structural changes in the heart visible on CT scans that correlate with cardiovascular risk.

"By combining routinely collected CT imaging with clinical health records, we can detect risk earlier and more accurately than ever before-without adding extra burden to patients or the health-care system," said Dr. Rasika Rajapakshe, a senior medical physicist at BC Cancer - Kelowna.

Early identification of high-risk patients allows clinicians to intervene sooner and tailor treatment accordingly. The approach operates as a non-invasive tool for predicting cardiovascular-related mortality in a population already managing cancer treatment.

The [AI for Healthcare](https://completeaitraining.com/tag/healthcare/) application demonstrates how [AI Research](https://completeaitraining.com/tag/research/) addresses specific clinical gaps by combining multiple data types into a single predictive framework.

---

This is the Markdown form of https://completeaitraining.com/news/ubc-researchers-develop-ai-model-to-detect-heart-disease/. The whole site is mapped at https://completeaitraining.com/llms.txt.