Fang Liu1, Georges El Fakhri1, Martin Torriani1, Richard Kijowski2, and Miho Tanaka3
1Radiology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, United States, 2New York University School of Medicine, New York, NY, United States, 3Orthopaedic Surgery, Massachusetts General Hospital and Harvard Medical School, Boston, MA, United States
1Radiology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, United States, 2New York University School of Medicine, New York, NY, United States, 3Orthopaedic Surgery, Massachusetts General Hospital and Harvard Medical School, Boston, MA, United States
A model-guided self-supervised deep learning approach was
demonstrated in accelerated T1/T2 mapping of the whole knee joint and proven to
outperform other state-of-the-art reconstruction methods.
Figure 1: The schematic demonstration of the CNN framework implementing RELAX. A
cyclic workflow was constructed to enforce self-supervised learning. The
physics models and additional constraints can be incorporated into the
framework to guide the learning of CNN mapping function to extract the latent
image parameter maps from undersampled images.
Figure 3: Comparison of T1
maps generated from different reconstruction methods for another testing knee
dataset at R=5. The deep learning-based methods, including both MANTIS and
RELAX, removed most of the artifacts and showed a similar reconstruction
performance, which outperformed conventional constrained reconstruction k-t
SLR. The absolute error maps were amplified by five times for display purposes
to show the method difference.
