WASHINGTON (dpa-AFX) - Researchers at UC San Francisco have developed a machine learning algorithm that enhances 3T MRI scans by generating synthetic images resembling those produced by 7 Tesla (7T) MRIs.
The algorithm improves the visualization of pathological tissue, offering greater clinical insights, and marks a step toward applying synthetic 7T MRI models in medical practice.
Studies suggest that ultra-high-field 7T MRIs offer better resolution and clinical benefits compared to 3T MRIs, especially for detailed brain imaging, crucial for identifying and monitoring pathological tissue.
UCSF researchers collected MRI data from patients with mild traumatic brain injury (TBI) and designed three neural network models to enhance images and perform 3D segmentation using synthetic 7T MRIs created from standard 3T images.
Senior study author Dr. Reza Abbasi-Asl, UCSF Assistant Professor of Neurology, stated, 'Our research introduces a machine-learning model that synthesizes high-quality MRIs from lower-quality ones. We show how this AI system enhances the visualization of brain abnormalities in traumatic brain injury, underscoring the potential of AI to elevate the quality of medical images captured by less advanced systems.'
Copyright(c) 2024 RTTNews.com. All Rights Reserved
Copyright RTT News/dpa-AFX
© 2024 AFX News
