Magnetic resonance fingerprinting (MRF) is a magnetic resonance imaging (MRI)-based method that can provide quantitative maps of multiple tissue properties simultaneously from a single rapid ...acquisition. Tissue property maps are generated by matching the complex signal evolutions collected at the scanner to a dictionary of signals derived using the Bloch equation simulations. However, in some circumstances, the process of dictionary generation and signal matching can be time-consuming, reducing the utility of this technique. Recently, several groups have proposed using machine learning to accelerate the extraction of quantitative maps from the MRF data. This article will provide an overview of current research that combines MRF and machine learning, as well as present original research demonstrating how machine learning can speed up dictionary generation for cardiac MRF (cMRF).
This study aims to improve the accuracy and consistency of T1 and T2 measurements using cardiac MR Fingerprinting (cMRF) by investigating and accounting for the effects of confounding factors ...including slice profile, inversion and T2 preparation pulse efficiency, and B1+. The goal is to understand how measurements with different pulse sequences are affected by these factors. This can be used to determine which factors must be taken into account for accurate measurements, and which may be mitigated by the selection of an appropriate pulse sequence. Simulations were performed using a numerical cardiac phantom to assess the accuracy of over 600 cMRF sequences with different flip angles, TRs, and preparation pulses. A subset of sequences, including one with the lowest errors in T1 and T2 maps, was used in subsequent analyses. Errors due to non-ideal slice profile, preparation pulse efficiency, and B1+ were quantified in Bloch simulations. Corrections for these effects were included in the dictionary generation and demonstrated in phantom and in vivo cardiac imaging at 3 T. Neglecting to model slice profile and preparation pulse efficiency led to underestimated T1 and overestimated T2 for most cMRF sequences. Sequences with smaller maximum flip angles were less affected by slice profile and B1+. Simulating all corrections in the dictionary improved the accuracy of T1 and T2 phantom measurements, regardless of acquisition pattern. More consistent myocardial T1 and T2 values were measured using different sequences after corrections. Based on these results, a pulse sequence which is minimally affected by confounding factors can be selected, and the appropriate residual corrections included for robust T1 and T2 mapping.
Purpose
To develop a deep learning method for rapidly reconstructing T1 and T2 maps from undersampled electrocardiogram (ECG) triggered cardiac magnetic resonance fingerprinting (cMRF) images.
...Methods
A neural network was developed that outputs T1 and T2 values when given a measured cMRF signal time course and cardiac RR interval times recorded by an ECG. Over 8 million cMRF signals, corresponding to 4000 random cardiac rhythms, were simulated for training. The training signals were corrupted by simulated k‐space undersampling artifacts and random phase shifts to promote robust learning. The deep learning reconstruction was evaluated in Monte Carlo simulations for a variety of cardiac rhythms and compared with dictionary‐based pattern matching in 58 healthy subjects at 1.5T.
Results
In simulations, the normalized root‐mean‐square error (nRMSE) for T1 was below 1% in myocardium, blood, and liver for all tested heart rates. For T2, the nRMSE was below 4% for myocardium and liver and below 6% for blood for all heart rates. The difference in the mean myocardial T1 or T2 observed in vivo between dictionary matching and deep learning was 3.6 ms for T1 and −0.2 ms for T2. Whereas dictionary generation and pattern matching required more than 4 min per slice, the deep learning reconstruction only required 336 ms.
Conclusion
A neural network is introduced for reconstructing cMRF T1 and T2 maps directly from undersampled spiral images in under 400 ms and is robust to arbitrary cardiac rhythms, which paves the way for rapid online display of cMRF maps.
To develop a deep learning method for rapidly reconstructing T
and T
maps from undersampled electrocardiogram (ECG) triggered cardiac magnetic resonance fingerprinting (cMRF) images.
A neural network ...was developed that outputs T
and T
values when given a measured cMRF signal time course and cardiac RR interval times recorded by an ECG. Over 8 million cMRF signals, corresponding to 4000 random cardiac rhythms, were simulated for training. The training signals were corrupted by simulated k-space undersampling artifacts and random phase shifts to promote robust learning. The deep learning reconstruction was evaluated in Monte Carlo simulations for a variety of cardiac rhythms and compared with dictionary-based pattern matching in 58 healthy subjects at 1.5T.
In simulations, the normalized root-mean-square error (nRMSE) for T
was below 1% in myocardium, blood, and liver for all tested heart rates. For T
, the nRMSE was below 4% for myocardium and liver and below 6% for blood for all heart rates. The difference in the mean myocardial T
or T
observed in vivo between dictionary matching and deep learning was 3.6 ms for T
and -0.2 ms for T
. Whereas dictionary generation and pattern matching required more than 4 min per slice, the deep learning reconstruction only required 336 ms.
A neural network is introduced for reconstructing cMRF T
and T
maps directly from undersampled spiral images in under 400 ms and is robust to arbitrary cardiac rhythms, which paves the way for rapid online display of cMRF maps.
The aim of this study is to shorten the breathhold and diastolic acquisition window in cardiac magnetic resonance fingerprinting (MRF) for simultaneous T
, T
, and proton spin density (M
) mapping to ...improve scan efficiency and reduce motion artifacts. To this end, a novel reconstruction was developed that combines low-rank subspace modeling with a deep image prior, termed DIP-MRF. A system of neural networks is used to generate spatial basis images and quantitative tissue property maps, with training performed using only the undersampled k-space measurements from the current scan. This approach avoids difficulties with obtaining
MRF training data, as training is performed
for each acquisition. Calculation of the forward model during training is accelerated by using GRAPPA operator gridding to shift spiral k-space data to Cartesian grid points, and by using a neural network to rapidly generate fingerprints in place of a Bloch equation simulation. DIP-MRF was evaluated in simulations and at 1.5 T in a standardized phantom, 18 healthy subjects, and 10 patients with suspected cardiomyopathy. In addition to conventional mapping, two cardiac MRF sequences were acquired, one with a 15-heartbeat(HB) breathhold and 254 ms acquisition window, and one with a 5HB breathhold and 150 ms acquisition window. In simulations, DIP-MRF yielded decreased nRMSE compared to dictionary matching and a sparse and locally low rank (SLLR-MRF) reconstruction. Strong correlation (R
> 0.999) with T
and T
reference values was observed in the phantom using the 5HB/150 ms scan with DIP-MRF. DIP-MRF provided better suppression of noise and aliasing artifacts
, especially for the 5HB/150 ms scan, and lower intersubject and intrasubject variability compared to dictionary matching and SLLR-MRF. Furthermore, it yielded a better agreement between myocardial T
and T
from 15HB/254 ms and 5HB/150 ms MRF scans, with a bias of -9 ms for T
and 2 ms for T
. In summary, this study introduces an extension of the deep image prior framework for cardiac MRF tissue property mapping, which does not require pre-training with
scans, and has the potential to reduce motion artifacts by enabling a shortened breathhold and acquisition window.
Objective
This study combines a deep image prior with low-rank subspace modeling to enable real-time (free-breathing and ungated) functional cardiac imaging on a commercial 0.55 T scanner.
Materials ...and methods
The proposed low-rank deep image prior (LR-DIP) uses two u-nets to generate spatial and temporal basis functions that are combined to yield dynamic images, with no need for additional training data. Simulations and scans in 13 healthy subjects were performed at 0.55 T and 1.5 T using a golden angle spiral bSSFP sequence with images reconstructed using
l
1
-ESPIRiT, low-rank plus sparse (L + S) matrix completion, and LR-DIP. Cartesian breath-held ECG-gated cine images were acquired for reference at 1.5 T. Two cardiothoracic radiologists rated images on a 1–5 scale for various categories, and LV function measurements were compared.
Results
LR-DIP yielded the lowest errors in simulations, especially at high acceleration factors (R
≥
8). LR-DIP ejection fraction measurements agreed with 1.5 T reference values (mean bias − 0.3% at 0.55 T and − 0.2% at 1.5 T). Compared to reference images, LR-DIP images received similar ratings at 1.5 T (all categories above 3.9) and slightly lower at 0.55 T (above 3.4).
Conclusion
Feasibility of real-time functional cardiac imaging using a low-rank deep image prior reconstruction was demonstrated in healthy subjects on a commercial 0.55 T scanner.
Multiparametric quantitative imaging is gaining increasing interest due to its widespread advantages in clinical applications. Magnetic resonance fingerprinting is a recently introduced approach of ...fast multiparametric quantitative imaging. In this article, magnetic resonance fingerprinting acquisition, dictionary generation, reconstruction, and validation are reviewed.
Cardiovascular magnetic resonance (CMR) is an established imaging modality with proven utility in assessing cardiovascular diseases. The ability of CMR to characterize myocardial tissue using T
- and ...T
-weighted imaging, parametric mapping, and late gadolinium enhancement has allowed for the non-invasive identification of specific pathologies not previously possible with modalities like echocardiography. However, CMR examinations are lengthy and technically complex, requiring multiple pulse sequences and different anatomical planes to comprehensively assess myocardial structure, function, and tissue composition. To increase the overall impact of this modality, there is a need to simplify and shorten CMR exams to improve access and efficiency, while also providing reproducible quantitative measurements. Multiparametric MRI techniques that measure multiple tissue properties offer one potential solution to this problem. This review provides an in-depth look at one such multiparametric approach, cardiac magnetic resonance fingerprinting (MRF). The article is structured as follows. First, a brief review of single-parametric and (non-Fingerprinting) multiparametric CMR mapping techniques is presented. Second, a general overview of cardiac MRF is provided covering pulse sequence implementation, dictionary generation, fast k-space sampling methods, and pattern recognition. Third, recent technical advances in cardiac MRF are covered spanning a variety of topics, including simultaneous multislice and 3D sampling, motion correction algorithms, cine MRF, synthetic multicontrast imaging, extensions to measure additional clinically important tissue properties (proton density fat fraction, T
*, and T
), and deep learning methods for image reconstruction and parameter estimation. The last section will discuss potential clinical applications, concluding with a perspective on how multiparametric techniques like MRF may enable streamlined CMR protocols. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY: Stage 1.