AI Model Translates Cine MRI Images into LGE-like Images
A deep-learning AI model presented at the International Society of Magnetic Resonance in Medicine could reduce the need for contrast agents in cardiac MRIs. The algorithm, developed by Pengfang Qian, PhD, of ShanghaiTech University, can predict late gadolinium enhancement (LGE) in acute myocardial infarction patients without the use of contrast agents.
Non-Invasive and Contrast-Free Method for Diagnosing Myocardial Diseases
Cardiac MRI traditionally uses late gadolinium enhancement to detect injured areas in the heart muscle. However, this technique is not suitable for patients with contraindications to gadolinium contrast agents. Qian’s team developed a new technique called “Cine Generated Enhancement” (CGE), which generates LGE images using contrast-free cine images captured rapidly over time.
Effective Approach for Generating Enhanced MRI Images
The researchers aligned contrast-free cine and LGE images and trained a neural network to translate cine images into LGE-like images. The study compared the quality of CGE images to standard LGE imaging in acute myocardial infarction patients, showing good agreement between the two methods in measuring scar size and transmurality.
Promising Results and Future Directions
Although CGE images slightly underestimated scar size and transmurality markers, there was a significant correlation between CGE and LGE measurements. Future work will focus on post-processing corrections to improve accuracy and expand the patient population for further validation.
Conclusion
As an expert in artificial intelligence journalism, I am intrigued by the potential of this AI model to revolutionize cardiac MRI imaging by reducing the dependence on contrast agents. What are your thoughts on the use of AI in medical imaging applications? Feel free to share your insights in the comments section below.
IntelliPrompt curated this article: Read the full story at the original source by clicking here