Contrast reduction in coronary catheterization procedures through Artificial Intelligence (AI) analysis of coronary angiograms.

This invention is a software that integrates artificial intelligence (AI) to improve the way contrast media is injected into coronary arteries during medical procedures. It focuses on using the first contrast injection to determine the exact amount needed for further injections in either the left or right coronary artery. This means that the software can predict how much contrast media is needed for each patient, which can be given to an automatic injection system for use by passing specific parameters (volume, flow rate). By tailoring the contrast volume to the patient, the software aims to reduce the amount of contrast media used, lowering the risk of side effects.

Contrast media can cause adverse effects, including contrast-induced acute kidney injury (CI-AKI), which affects up to 8% of patients undergoing coronary catheterization. The risk is higher in patients with conditions like diabetes, hypertension, and chronic kidney disease. Developing CI-AKI can lead to serious complications such as increased risk of bleeding, heart attacks, and even death. It also results in longer hospital stays and higher costs, with an average increase of 3.36 days and $9,448 per patient. In severe cases, about 0.3% of patients might need dialysis, significantly increasing both the hospital stay and costs.

As far as the inventors know, this AI-guided system is the first of its kind to provide specific recommendations for contrast injections in the heart’s arteries, with no existing patents found that conflict with this invention.

Faculty Supervisor:

Muhammad Mamdani

Student:

Partner:

Sunnybrook Research Institute

Discipline:

Computer science

Sector:

Health and Related Sciences & Technology

University:

University of Toronto

Program: