• Politics
  • Diversity, equity and inclusion
  • Financial Decision Making
  • Telehealth
  • Patient Experience
  • Leadership
  • Point of Care Tools
  • Product Solutions
  • Management
  • Technology
  • Healthcare Transformation
  • Data + Technology
  • Safer Hospitals
  • Business
  • Providers in Practice
  • Mergers and Acquisitions
  • AI & Data Analytics
  • Cybersecurity
  • Interoperability & EHRs
  • Medical Devices
  • Pop Health Tech
  • Precision Medicine
  • Virtual Care
  • Health equity

Machine Learning Analyzes MRI Scans 186-Times Faster than Humans


Using machine learning, an MRI scan can be analyzed in approximately four seconds.

Headshot courtesy of the American Heart Association.

Artificial intelligence (AI) can read cardiac MRI scans 186-times faster than humans, with comparable precision, according to the findings of a study published in the American Heart Association journal

Charlotte Ministy, Ph.D.

Charlotte Ministy, Ph.D.

Circulation: Cardiovascular Imaging.

Analyzing heart function on cardiac MRI scans takes approximately 13 minutes for humans. Using machine learning, an MRI scan can be analyzed in approximately four seconds, the researchers found.

“Cardiovascular MRI offers unparalleled image quality for assessing heart structure and function; however, current manual analysis remains basic and outdated,” said Charlotte Manisty, Ph.D., senior lecturer at University College London. “Automated machine-learning techniques offer the potential to change this and radically improve efficiency, and we look forward to further research that could validate its superiority to human analysis.”

A research team trained a neural network to read the cardiac MRI scans and results of nearly 600 patients. To be included, patients needed to be over 18 years old undergoing cardiovascular magnetic resonance with balanced steady-state free precession cine imagining on two occasions within a time frame where biological change was not anticipated.

Researchers compared the AI’s precision to an expert and trainee on 110 separate patients from multiple centers. Experts confidently detected a 9% change in ejection fraction, which was similar to using the AI — there was no significant decrease in accuracy.

Patients included had a wide range of heart diseases, which allowed the researchers to demonstrate that the greatest sources of measurement error come from human factors, Manisty said.

“This indicates that automated techniques are at least as good as humans, with the potential soon to be ‘superhuman’ — transforming clinical and research measurement precision,” she concluded.

Get the best insights in digital health directly to your inbox.


Analysis: AI Is as Effective as Humans at Diagnosing Disease

Machine Learning Helps Diagnose Patients with Head and Neck Cancers

AI Identifies Risk of Fatal Heart Attack 5 Years Before the Event

Recent Videos
Image: Ron Southwick, Chief Healthcare Executive
George Van Antwerp, MBA
Edmondo Robinson, MD
Craig Newman
Related Content
© 2024 MJH Life Sciences

All rights reserved.