Quantification of intravoxel incoherent motion with optimized b-values using deep neural network

Self-supervised learning for denoising of multidimensional MRI data

PURPOSE

To develop a fast denoising framework for high-dimensional MRI data based on a self-supervised learning scheme, which does not require ground truth clean image.

THEORY AND METHODS

Quantitative MRI faces limitations in SNR, because the variation of signal amplitude in a large set of images is the key mechanism for quantification. In addition, the complex non-linear signal models make the fitting process vulnerable to noise. To address these issues, we propose a fast deep-learning framework for denoising, which efficiently exploits the redundancy in multidimensional MRI data. A self-supervised model was designed to use only noisy images for training, bypassing the challenge of clean data paucity in clinical practice. For validation, we used two different datasets of simulated magnetization transfer contrast MR fingerprinting (MTC-MRF) dataset and in vivo DWI image dataset to show the generalizability.

RESULTS

The proposed method drastically improved denoising performance in the presence of mild-to-severe noise regardless of noise distributions compared to previous methods of the BM3D, tMPPCA, and Patch2self. The improvements were even pronounced in the following quantification results from the denoised images.

CONCLUSION

The proposed MD-S2S (Multidimensional-Self2Self) denoising technique could be further applied to various multi-dimensional MRI data and improve the quantification accuracy of tissue parameter maps.

Calculation of intravoxel incoherent motion parameter maps using a kernelized total difference-based method

Quantitative analysis of diffusion-weighted magnetic resonance imaging (DW-MRI) has been explored for many clinical applications since its development. In particular, the intravoxel incoherent motion (IVIM) model for DW-MRI has been commonly utilized in various organs. However, because of the presence of excessive noise, the IVIM parameter maps obtained from pixel-wise fitting are often unreliable. In this study, we propose a kernelized total difference-based curve-fitting method to estimate the IVIM parameters. Simulated DW-MRI data at five signal-to-noise ratios (i.e., 10, 20, 30, 50, and 100) and real abdominal DW-MRI data acquired on a 1.5-T MRI scanner with nine b-values (i.e., 0, 10, 25, 50, 100, 200, 300, 400, and 500 s/mm2) and six diffusion-encoding gradient directions were used to evaluate the performance of the proposed method. The results were compared with those obtained by three existing methods: trust-region reflective (TRR) algorithm, Bayesian probability (BP), and deep neural network (DNN). Our simulation results showed that the proposed method outperformed the other three comparing methods in terms of root-mean-square error. Moreover, the proposed method could preserve small details in the estimated IVIM parameter maps. The experimental results showed that, compared with the TRR method, the proposed method as well as the BP (and DNN) method could reduce the overestimation of the pseudodiffusion coefficient and improve the quality of IVIM parameter maps. For all studied abdominal organs except the pancreas, both the proposed method and the BP method could provide IVIM parameter estimates close to the reference values; the former had higher precision. The kernelized total difference-based curve-fitting method has the potential to improve the reliability of IVIM parametric imaging.

Comparison of methods for intravoxel incoherent motion parameter estimation in the brain from flow-compensated and non-flow-compensated diffusion-encoded data

PURPOSE

Joint analysis of flow-compensated (FC) and non-flow-compensated (NC) diffusion MRI (dMRI) data has been suggested for increased robustness of intravoxel incoherent motion (IVIM) parameter estimation. For this purpose, a set of methods commonly used or previously found useful for IVIM analysis of dMRI data obtained with conventional diffusion encoding were evaluated in healthy human brain.

METHODS

Five methods for joint IVIM analysis of FC and NC dMRI data were compared: (1) direct non-linear least squares fitting, (2) a segmented fitting algorithm with estimation of the diffusion coefficient from higher b-values of NC data, (3) a Bayesian algorithm with uniform prior distributions, (4) a Bayesian algorithm with spatial prior distributions, and (5) a deep learning-based algorithm. Methods were evaluated on brain dMRI data from healthy subjects and simulated data at multiple noise levels. Bipolar diffusion encoding gradients were used with b-values 0-200 s/mm2 and corresponding flow weighting factors 0-2.35 s/mm for NC data and by design 0 for FC data. Data were acquired twice for repeatability analysis.

RESULTS

Measurement repeatability as well as estimation bias and variability were at similar levels or better with the Bayesian algorithm with spatial prior distributions and the deep learning-based algorithm for IVIM parametersD $$ D $$ andf $$ f $$ , and for the Bayesian algorithm only forv d $$ {v}_d $$ , relative to the other methods.

CONCLUSION

A Bayesian algorithm with spatial prior distributions is preferable for joint IVIM analysis of FC and NC dMRI data in the healthy human brain, but deep learning-based algorithms appear promising.

Intravoxel incoherent motion magnetic resonance imaging reconstruction from highly under-sampled diffusion-weighted PROPELLER acquisition data via physics-informed residual feedback unrolled network

Objective. The acquisition of diffusion-weighted images for intravoxel incoherent motion (IVIM) imaging is time consuming. This work aims to accelerate the scan through a highly under-sampling diffusion-weighted turbo spin echo PROPELLER (DW-TSE-PROPELLER) scheme and to develop a reconstruction method for accurate IVIM parameter mapping from the under-sampled data.Approach.The proposed under-sampling DW-TSE-PROPELLER scheme for IVIM imaging is that a few blades perb-value are acquired and rotated along theb-value dimension to cover high-frequency information. A physics-informed residual feedback unrolled network (PIRFU-Net) is proposed to directly estimate distortion-free and artifact-free IVIM parametric maps (i.e., the perfusion-free diffusion coefficientDand the perfusion fractionf) from highly under-sampled DW-TSE-PROPELLER data. PIRFU-Net used an unrolled convolution network to explore data redundancy in the k-q space to remove under-sampling artifacts. An empirical IVIM physical constraint was incorporated into the network to ensure that the signal evolution curves along theb-value follow a bi-exponential decay. The residual between the realistic and estimated measurements was fed into the network to refine the parametric maps. Meanwhile, the use of synthetic training data eliminated the need for genuine DW-TSE-PROPELLER data.Main results.The experimental results show that the DW-TSE-PROPELLER acquisition was six times faster than full k-space coverage PROPELLER acquisition and within a clinically acceptable time. Compared with the state-of-the-art methods, the distortion-freeDandfmaps estimated by PIRFU-Net were more accurate and had better-preserved tissue boundaries on a simulated human brain and realistic phantom/rat brain/human brain data.Significance.Our proposed method greatly accelerates IVIM imaging. It is capable of directly and simultaneously reconstructing distortion-free, artifact-free, and accurateDandfmaps from six-fold under-sampled DW-TSE-PROPELLER data.

Image downsampling expedited adaptive least-squares (IDEAL) fitting improves intravoxel incoherent motion (IVIM) analysis in the human kidney

PURPOSE

To improve the reliability of intravoxel incoherent motion (IVIM) model parameter estimation for the DWI in the kidney using a novel image downsampling expedited adaptive least-squares (IDEAL) approach.

METHODS

The robustness of IDEAL was investigated using simulated DW-MRI data corrupted with different levels of Rician noise. Subsequently, the performance of the proposed method was tested by fitting bi- and triexponential IVIM model to in vivo renal DWI data acquired on a clinical 3 Tesla MRI scanner and compared to conventional approaches (fixed D* and segmented fitting).

RESULTS

The numerical simulations demonstrated that the IDEAL algorithm provides robust estimates of the IVIM parameters in the presence of noise (SNR of 20) as indicated by relatively low absolute percentage bias (maximal sMdPB <20%) and normalized RMSE (maximal RMSE <28%). The analysis of the in vivo data showed that the IDEAL-based IVIM parameter maps were less noisy and more visually appealing than those obtained using the fixed D* and segmented methods. Further, coefficients of variation for nearly all IVIM parameters were significantly reduced in cortex and medulla for IDEAL-based biexponential (coefficients of variation: 4%-50%) and triexponential (coefficients of variation: 7.5%-75%) IVIM modelling compared to the segmented (coefficients of variation: 4%-120%) and fixed D* (coefficients of variation: 17%-174%) methods, reflecting greater accuracy of this method.

CONCLUSION

The proposed fitting algorithm yields more robust IVIM parameter estimates and is less susceptible to poor SNR than the conventional fitting approaches. Thus, the IDEAL approach has the potential to improve the reliability of renal DW-MRI analysis for clinical applications.

An unsupervised convolutional neural network method for estimation of intravoxel incoherent motion parameters

Objective. Intravoxel incoherent motion (IVIM) imaging obtained by fitting a biexponential model to multiple b-value diffusion-weighted magnetic resonance imaging (DW-MRI) has been shown to be a promising tool for different clinical applications. Recently, several deep neural network (DNN) methods were proposed to generate IVIM imaging. Approach. In this study, we proposed an unsupervised convolutional neural network (CNN) method for estimation of IVIM parameters. We used both simulated and real abdominal DW-MRI data to evaluate the performance of the proposed CNN-based method, and compared the results with those obtained from a non-linear least-squares fit (TRR, trust-region reflective algorithm) and a feed-forward backward-propagation DNN-based method. Main results. The simulation results showed that both the DNN- and CNN-based methods had lower coefficients of variation than the TRR method, but the CNN-based method provided more accurate parameter estimates. The results obtained from real DW-MRI data showed that the TRR method produced many biased IVIM parameter estimates that hit the upper and lower parameter bounds. In contrast, both the DNN- and CNN-based methods yielded less biased IVIM parameter estimates. Overall, the perfusion fraction and diffusion coefficient obtained from the DNN- and CNN-based methods were close to literature values. However, compared with the CNN-based method, both the TRR and DNN-based methods tended to yield increased pseudodiffusion coefficients (55%–180%). Significance. Our preliminary results suggest that it is feasible to estimate IVIM parameters using CNN.

A supervised deep neural network approach with standardized targets for enhanced accuracy of IVIM parameter estimation from multi-SNR images

Extraction of intravoxel incoherent motion (IVIM) parameters from noisy diffusion-weighted (DW) images using a biexponential fitting model is computationally challenging, and the reliability of the estimated perfusion-related quantities represents a limitation of this technique. Artificial intelligence can overcome the current limitations and be a suitable solution to advance use of this technique in both preclinical and clinical settings. The purpose of this work was to develop a deep neural network (DNN) approach, trained on numerical simulated phantoms with different signal to noise ratios (SNRs), to improve IVIM parameter estimation. The proposed approach is based on a supervised fully connected DNN having 3 hidden layers, 18 inputs and 3 targets with standardized values. 14 × 10³ simulated DW images, based on a Shepp-Logan phantom, were randomly generated with varying SNRs (ranging from 10 to 100). 7 × 10³ images (1000 for each SNR) were used for training. Performance accuracy was assessed in simulated images and the proposed approach was compared with the state-of-the-art Bayesian approach and other DNN algorithms. The DNN approach was also evaluated in vivo on a high-field MRI preclinical scanner. Our DNN approach showed an overall improvement in accuracy when compared with the Bayesian approach and other DNN methods in most of the simulated conditions. The in vivo results demonstrated the feasibility of the proposed approach in real settings and generated quantitative results comparable to those obtained using the Bayesian and unsupervised approaches, especially for D and f, and with lower variability in homogeneous regions. The DNN architecture proposed in this work outlines two innovative features as compared with other studies: (1) the use of standardized targets to improve the estimation of parameters, and (2) the implementation of a single DNN to enhance the IVIM fitting at different SNRs, providing a valuable alternative tool to compute IVIM parameters in conditions of high background noise.

Registration and quantification network (RQnet) for IVIM-DKI analysis in MRI

PURPOSE

A deep learning method is proposed for aligning diffusion weighted images (DWIs) and estimating intravoxel incoherent motion-diffusion kurtosis imaging parameters simultaneously.

METHODS

We propose an unsupervised deep learning method that performs 2 tasks: registration and quantification for intravoxel incoherent motion-diffusion kurtosis imaging analysis. A common registration method in diffusion MRI is based on minimizing dissimilarity between various DWIs, which may result in registration errors due to different contrasts in different DWIs. We designed a novel unsupervised deep learning method for both accurate registration and quantification of various diffusion parameters. In order to generate motion-simulated training data and test data, 17 volunteers were scanned without moving their heads, and 4 volunteers moved their heads during the scan in a 3 Tesla MRI. In order to investigate the applicability of the proposed method to other organs, kidney images were also obtained. We compared the registration accuracy of the proposed method, statistical parametric mapping, and a deep learning method with a normalized cross-correlation loss. In the quantification part of the proposed method, a deep learning method that considered the diffusion gradient direction was used.

RESULTS

Simulations and experimental results showed that the proposed method accurately performed registration and quantification for intravoxel incoherent motion-diffusion kurtosis imaging analysis. The registration accuracy of the proposed method was high for all b values. Furthermore, quantification performance was analyzed through simulations and in vivo experiments, where the proposed method showed the best performance among the compared methods.

CONCLUSION

The proposed method aligns the DWIs and accurately quantifies the intravoxel incoherent motion-diffusion kurtosis imaging parameters.

Self-supervised neural network improves tri-exponential intravoxel incoherent motion model fitting compared to least-squares fitting in non-alcoholic fatty liver disease

Recent literature suggests that tri-exponential models may provide additional information and fit liver intravoxel incoherent motion (IVIM) data more accurately than conventional bi-exponential models. However, voxel-wise fitting of IVIM results in noisy and unreliable parameter maps. For bi-exponential IVIM, neural networks (NN) were able to produce superior parameter maps than conventional least-squares (LSQ) generated images. Hence, to improve parameter map quality of tri-exponential IVIM, we developed an unsupervised physics-informed deep neural network (IVIM3-NET). We assessed its performance in simulations and in patients with non-alcoholic fatty liver disease (NAFLD) and compared outcomes with bi-exponential LSQ and NN fits and tri-exponential LSQ fits. Scanning was performed using a 3.0T free-breathing multi-slice diffusion-weighted single-shot echo-planar imaging sequence with 18 b-values. Images were analysed for visual quality, comparing the bi- and tri-exponential IVIM models for LSQ fits and NN fits using parameter-map signal-to-noise ratios (SNR) and adjusted R2. IVIM parameters were compared to histological fibrosis, disease activity and steatosis grades. Parameter map quality improved with bi- and tri-exponential NN approaches, with a significant increase in average parameter-map SNR from 3.38 to 5.59 and 2.45 to 4.01 for bi- and tri-exponential LSQ and NN models respectively. In 33 out of 36 patients, the tri-exponential model exhibited higher adjusted R2 values than the bi-exponential model. Correlating IVIM data to liver histology showed that the bi- and tri-exponential NN outperformed both LSQ models for the majority of IVIM parameters (10 out of 15 significant correlations). Overall, our results support the use of a tri-exponential IVIM model in NAFLD. We show that the IVIM3-NET can be used to improve image quality compared to a tri-exponential LSQ fit and provides promising correlations with histopathology similar to the bi-exponential neural network fit, while generating potentially complementary additional parameters.

b value and first-order motion moment optimized data acquisition for repeatable quantitative intravoxel incoherent motion DWI

PURPOSE

To design a b value and first-order motion moment (M1 ) optimized data acquisition for repeatable intravoxel incoherent motion (IVIM) quantification in the liver.

METHODS

Cramer-Rao lower bound optimization was performed to determine optimal monopolar and optimal 2D samplings of the b-M1 space based on noise performance. Monte Carlo simulations were used to evaluate the bias and variability in estimates obtained using the proposed optimal samplings and conventional monopolar sampling. Diffusion MRI of the liver was performed in 10 volunteers using 3 IVIM acquisitions: conventional monopolar, optimized monopolar, and b-M1 -optimized gradient waveforms (designed based on the optimal 2D sampling). IVIM parameter maps of diffusion coefficient, perfusion fraction, and blood velocity SD were obtained using nonlinear least squares fitting. Noise performance (SDs), stability (outlier percentage), and test-retest or scan-rescan repeatability (intraclass correlation coefficients) were evaluated and compared across acquisitions.

RESULTS

Cramer-Rao lower bound and Monte Carlo simulations demonstrated improved noise performance of the optimal 2D sampling in comparison to monopolar samplings. Evaluating the designed b-M1 -optimized waveforms in healthy volunteers, significant decreases (p < 0.05) in the SDs and outlier percentages were observed for measurements of diffusion coefficient, perfusion fraction, and blood velocity SD in comparison to measurements obtained using monopolar samplings. Good-to-excellent repeatability (intraclass correlation coefficients ≥ 0.77) was observed for all 3 parameters in both the right and left liver lobes using the b-M1 -optimized waveforms.

CONCLUSIONS

2D b-M1 -optimized data acquisition enables repeatable IVIM quantification with improved noise performance. 2D acquisitions may advance the establishment of IVIM quantitative biomarkers for liver diseases.

Self-supervised IVIM DWI parameter estimation with a physics based forward model

PURPOSE

To assess the robustness and repeatability of intravoxel incoherent motion model (IVIM) parameter estimation for the diffusion-weighted MRI in the abdominal organs under the constraints of noisy diffusion signal using a novel neural network method.

METHODS

Clinically acquired abdominal scans of Crohn's disease patients were retrospectively analyzed with regions segmented in the kidney cortex, spleen, liver, and bowel. A novel IVIM parameter fitting method based on the principle of a physics guided self-supervised convolutional neural network that does not require reference parameter estimates for training was compared to a conventional non-linear least squares (NNLS) algorithm, and a voxelwise trained artificial neural network (ANN).

RESULTS

Results showed substantial increase in parameter robustness to the noise corrupted signal. In an intra-session repeatability experiment, the proposed method showed reduced coefficient of variation (CoV) over multiple acquisitions in comparison to conventional NLLS method and comparable performance to ANN. The use of D and f estimates from the proposed method led to the smallest misclassification error in linear discriminant analysis for characterization between normal and abnormal Crohn's disease bowel tissue. The fitting of D∗ parameter remains to be challenging.

CONCLUSION

The proposed method yields robust estimates of D and f IVIM parameters under the constraints of noisy diffusion signal. This indicates a potential for the use of the proposed method in conjunction with accelerated DW-MRI acquisition strategies, which would typically result in lower signal to noise ratio.