A Fast Deep Image Prior Reconstruction for Cardiac MR Fingerprinting Using Meta-Learning Acceleration

Background: High-quality cardiac MR Fingerprinting (MRF) maps can be obtained using a Deep Image Prior (DIP-MRF) reconstruction, which combines a low-rank model with zero-shot deep learning 1 .However, this technique requires long reconstruction times (>40min on a GPU) since training is performed in a scanspecific manner.This work introduces an approach to shorten the reconstruction time to 5min per slice using meta-learning with scanspecific fine-tuning.Methods: DIP-MRF employs: (1)a U-Net (Image Reconstruction Network or IRN) that outputs subspace images from a fixed noise input tensor; (2)a convolutional network (Coil Estimation Network or CEN) that refines sensitivity maps from ESPIRiT 2 , and (3)a fullyconnected network (Parameter Estimation Network or PEN) that estimates T1, T2, and M0 maps 3 .These networks are trained de novo to enforce consistency with k-space data from a single scan.Here, the following approach is proposed to accelerate DIP-MRF (Figure 1).Step0: The IRN and CEN are pre-trained using scans from multiple subjects (using the same input tensor for the IRN) by minimizing the MSE relative to target images and sensitivity maps from the original (slow) DIP-MRF.Rather than simple pre-training, this step uses REPTILE 4 , a meta-learning algorithm that allows pre-trained networks adapt to new datasets in fewer stepsStep1: The pre-trained IRN and CEN are fine-tuned for a specific scan.To accelerate this process, the loss does not include a forward physics model but instead minimizes the MSE between networkgenerated and zero-filled (undersampled) images in the subspaceStep2: Fine-tuning is then performed using a loss incorporating the forward physics model to enforce consistency with acquired kspace dataStep3: The PEN is trained in a self-supervised manner to estimate quantitative maps Note that the original DIP-MRF method is equivalent to only performing Steps 2-3.Data were collected at 1.5T (MAGNETOM Sola, Siemens) using a 2D breathheld cardiac MRF sequence in 28 healthy subjects at a mid-short-axis slice 5 .Meta-learning was performed using data from 20 subjects; the remaining 8 were used for testing, with 1000 epochs for Step1 and 10 epochs for Step2.For comparison, the original DIP-MRF was performed with 300 epochs.Results: MRF maps after each stage of the accelerated reconstruction are shown in Figure 2.After Step0, outlines of anatomical structures can be observed.After Step1, higher-quality maps are obtained, although substantial aliasing artifacts remain.After Step2, these artifacts are largely suppressed, yielding maps comparable to the original DIP-MRF technique in only 5min (compared to 40min).Subtle spatial blurring is observed relative to the original method, which can be mitigated by increasing the epochs in Step2 to 100 (Figure 3) at the expense of a longer reconstruction time (18min).Conclusion: This work introduces a meta-learning strategy to accelerate DIP-MRF, which could enable rapid, inline deep learning reconstruction of cardiac MRF maps.