Deep learning reconstruction enables 5-minute MRI scans
Deep learning image reconstruction could enable techs to complete MRI scans of the knee in just a matter of minutes.
New research published in the American Journal of Roentgenology details a deep learning super-resolution image reconstruction technique that opens the door for accelerated imaging of the knee. Using the technique, which combines threefold parallel imaging (PI) and twofold simultaneous multislice (SMS) acceleration and DL super-resolution image reconstruction, researchers were able to complete imaging of the knee in just five minutes—a fraction of the time it typically takes to complete an MRI of the area.
The technique was retrospectively tested in a group of 124 adults with painful knee conditions who underwent arthroscopic surgery between October 2022 and July 2023. Seven musculoskeletal radiologists evaluated the images for the presence of artifacts, structural visibility, cruciate ligament tears, collateral ligament tears, meniscal tears, cartilage defects and fractures to determine if the technique impacted the diagnostic quality of the exams.
On a 5-point scale, readers categorized the resultant image quality as good or very good, scoring the exams 4-5 on average. Interreader agreement also was considered very good. The visibility of anatomical structures was very good, while the diagnostic performance for diagnosing arthroscopy-validated structural abnormalities was rated good to excellent. Sensitivity and specificity were high for all anatomical abnormalities as well.
What’s more, the reconstruction technique eliminated all motion artifacts and image noise.
Corresponding author Jan Fritz, MD, with the department of radiology at the Grossman School of Medicine, New York University, and colleagues suggest their findings offer validation of the technique, indicating clinical utility.
“Sixfold PI-SMS-accelerated DL super-resolution 3-T knee MRI provides high efficiency through short scan times and high diagnostic performance,” the group concluded.
Learn more about the team’s findings here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.