DeepCEST 3T: Robust MRI parameter determination and uncertainty quantification with neural networks—application to CEST imaging of the human brain at 3T

• Published in: Magnetic resonance in medicine Vol. 84; no. 1; pp. 450 - 466
• Main Authors: Glang, Felix, Deshmane, Anagha, Prokudin, Sergey, Martin, Florian, Herz, Kai, Lindig, Tobias, Bender, Benjamin, Scheffler, Klaus, Zaiss, Moritz
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.07.2020
• Subjects: Algorithms; APT; Brain; Brain - diagnostic imaging; Brain cancer; Brain Neoplasms - diagnostic imaging; Brain tumors; chemical exchange saturation transfer (CEST); Computational neuroscience; deepCEST; Humans; Image contrast; Image Interpretation, Computer-Assisted; Image reconstruction; Inhomogeneity; Magnetic Resonance Imaging; Mathematical models; Neural networks; Neural Networks, Computer; Neuroimaging; NOE; Parameter estimation; Parameter robustness; Parameter sensitivity; Parameter uncertainty; probabilistic neural network; Robustness (mathematics); Semisolids; Tumors; Uncertainty; uncertainty quantification; deepCEST; uncertainty quantification; probabilistic neural network; APT; NOE; chemical exchange saturation transfer (CEST)
• ISSN: 0740-3194; 1522-2594; 1522-2594
• DOI: 10.1002/mrm.28117

Purpose Calculation of sophisticated MR contrasts often requires complex mathematical modeling. Data evaluation is computationally expensive, vulnerable to artifacts, and often sensitive to fit algorithm parameters. In this work, we investigate whether neural networks can provide not only fast model fitting results, but also a quality metric for the predicted values, so called uncertainty quantification, investigated here in the context of multi‐pool Lorentzian fitting of CEST MRI spectra at 3T. Methods A deep feed‐forward neural network including a probabilistic output layer allowing for uncertainty quantification was set up to take uncorrected CEST‐spectra as input and predict 3T Lorentzian parameters of a 4‐pool model (water, semisolid MT, amide CEST, NOE CEST), including the B0 inhomogeneity. Networks were trained on data from 3 subjects with and without data augmentation, and applied to untrained data from 1 additional subject and 1 brain tumor patient. Comparison to conventional Lorentzian fitting was performed on different perturbations of input data. Results The deepCEST 3T networks provided fast and accurate predictions of all Lorentzian parameters and were robust to input perturbations because of noise or B0 artifacts. The uncertainty quantification detected fluctuations in input data by increase of the uncertainty intervals. The method generalized to unseen brain tumor patient CEST data. Conclusions The deepCEST 3T neural network provides fast and robust estimation of CEST parameters, enabling online reconstruction of sophisticated CEST contrast images without the typical computational cost. Moreover, the uncertainty quantification indicates if the predictions are trustworthy, enabling confident interpretation of contrast changes.