Accurate parameter estimation using scan-specific unsupervised deep learning for relaxometry and MR fingerprinting
This work addresses the challenge of accurate and robust parameter estimation in MRI relaxometry and fingerprinting, which is crucial for medical imaging applications, though it appears incremental as it builds on existing deep learning and simulation techniques.
The authors tackled the problem of relaxation parameter estimation in MRI by proposing an unsupervised convolutional neural network that incorporates signal relaxation and Bloch simulations, achieving significantly improved quantification accuracy and robustness to noise compared to standard methods in simulations and in vivo data for T2 and T2* mapping, with concrete gains in T1 and T2 mapping from undersampled data.
We propose an unsupervised convolutional neural network (CNN) for relaxation parameter estimation. This network incorporates signal relaxation and Bloch simulations while taking advantage of residual learning and spatial relations across neighboring voxels. Quantification accuracy and robustness to noise is shown to be significantly improved compared to standard parameter estimation methods in numerical simulations and in vivo data for multi-echo T2 and T2* mapping. The combination of the proposed network with subspace modeling and MR fingerprinting (MRF) from highly undersampled data permits high quality T1 and T2 mapping.