Purpose: To accelerate magnetic resonance fingerprinting (MRF) by developing a flexible deep learning reconstruction method. Materials and Methods: Synthetic data were used to train a deep learning model. The trained model was then applied to MRF for different organs and diseases. Iterative reconstruction was performed outside the deep learning model, allowing a changeable encoding matrix, i.e., with flexibility of choice for image resolution, radiofrequency coil, k-space trajectory, and undersampling mask. In vivo experiments were performed on normal brain and prostate cancer volunteers to demonstrate the model performance and generalizability. Results: In 400-dynamics brain MRF, direct nonuniform Fourier transform caused a slight increase of random fluctuations on the T2 map. These fluctuations were reduced with the proposed method. In prostate MRF, the proposed method suppressed fluctuations on both T1 and T2 maps. Conclusion The deep learning and iterative MRF reconstruction method described in this study was flexible with different acquisition settings such as radiofrequency coils. It is generalizable for different In vivo applications.

Objective To explore the predictive value of the model for end-stage liver disease (MELD) score, energy metabolism and serum thyroid hormone levels on the severity and prognosis of patients with liver failure and their correlation. Methods This study collected clinicopathological data from 60 liver failure patients, e.g., end-stage liver disease (MELD) score, energy metabolism, and serum thyroid hormone levels. The χ 2 test was performed to analyze the categorical variables, while the Mann-Whitney U test and independent sample t test were performed to assess the continuous variables between the two groups. Spearman correlation coefficient test was used to evaluate correlation of each index. The receiver operating characteristic (ROC) curve was used to analyze the optimal cut-off points of serum total triiodothyronine (TT3) and free triiodothyronine (FT3) levels in predicting prognosis of the patients. Results The rates of low TT3 and FT3 levels in liver failure patients were 78.2% and 69.1%, respectively, whereas the low TT3 rates were 95.2% and 67.6% and the low FT3 rates were 90.5% and 55.9% in survival and non-survival groups of patients, respectively (both P < 0.05). Moreover, the MELD score was significantly higher in the non-survival patients than in survival patients [26.0(21.0-29.0) vs 21.0 (19.0-24.0), Z =-3.396, P =0.001], while TT3 and FT3 levels were significantly lower in the non-survival patients than in the survival patients [0.69(0.62-0.73) vs 0.83(0.69-0.94) and 2.17(1.99-2.31) vs 2.54(2.12-2.86), respectively; Z =-2.884、-2.876, all P < 0.01]. The MELD score was negatively associated with serum TT3, FT3, and thyroid stimulating hormone (TSH) levels and the respiratory quotient (RQ) ( r =-0.487、-0.329、-0.422、-0.350, all P < 0.01), whereas the RQ was associated with serum TT3 and FT3 levels ( r =0.271、0.265, all P < 0.05). The optimal cutoff values in predicting the severity and survival of patients was 0.75 nmol/L and 2.37pmol/L with the sensitivity values of 67.6% and 64.7% and the specificity of 90.5% and 81.0%, respectively. Conclusion Abnormal thyroid hormone levels and low respiratory quotient could be used to predict the severity and prognosis of patients with liver failure.