Free-breathing and instantaneous abdominal T2 mapping via single-shot multiple overlapping-echo acquisition and deep learning reconstruction

ObjectivesTo develop a real-time abdominal T-2 mapping method without requiring breath-holding or respiratory-gating.MethodsThe single-shot multiple overlapping-echo detachment (MOLED) pulse sequence was employed to achieve free-breathing T-2 mapping of the abdomen. Deep learning was used to untangle the non-linear relationship between the MOLED signal and T-2 mapping. A synthetic data generation flow based on Bloch simulation, modality synthesis, and randomization was proposed to overcome the inadequacy of real-world training set.ResultsThe results from simulation and in vivo experiments demonstrated that our method could deliver high-quality T-2 mapping. The average NMSE and R-2 values of linear regression in the digital phantom experiments were 0.0178 and 0.9751. Pearson's correlation coefficient between our predicted T-2 and reference T-2 in the phantom experiments was 0.9996. In the measurements for the patients, real-time capture of the T-2 value changes of various abdominal organs before and after contrast agent injection was realized. A total of 33 focal liver lesions were detected in the group, and the mean and standard deviation of T-2 values were 141.1 +/- 50.0 ms for benign and 63.3 +/- 16.0 ms for malignant lesions. The coefficients of variance in a test-retest experiment were 2.9%, 1.2%, 0.9%, 3.1%, and 1.8% for the liver, kidney, gallbladder, spleen, and skeletal muscle, respectively.ConclusionsFree-breathing abdominal T-2 mapping is achieved in about 100 ms on a clinical MRI scanner. The work paved the way for the development of real-time dynamic T-2 mapping in the abdomen.

Automated summary

This study developed a real-time abdominal T-2 mapping method that does not require breath-holding or respiratory-gating. The single-shot multiple overlapping-echo detachment (MOLED) pulse sequence was used for free-breathing T-2 mapping of the abdomen. Deep learning was utilized to untangle the non-linear relationship between the MOLED signal and T-2 mapping. A synthetic data generation flow based on Bloch simulation, modality synthesis, and randomization was proposed to overcome the inadequacy of real-world training set.