Convolutional neural network for accelerating the computation of the extended Tofts model in dynamic contrast-enhanced magnetic resonance imaging

Deep learning-based super-resolution and denoising algorithm improves reliability of dynamic contrast-enhanced MRI in diffuse glioma

Dynamic contrast-enhanced MRI (DCE-MRI) is increasingly used to non-invasively image blood-brain barrier leakage, yet its clinical utility has been hampered by issues such as noise and partial volume artifacts. In this retrospective study involving 306 adult patients with diffuse glioma, we applied deep learning-based super-resolution and denoising (DLSD) techniques to enhance the signal-to-noise ratio (SNR) and resolution of DCE-MRI. Quantitative analysis comparing standard DCE-MRI (std-DCE) and DL-enhanced DCE-MRI (DL-DCE) revealed that DL-DCE achieved significantly higher SNR and contrast-to-noise ratio (CNR) compared to std-DCE (SNR, 52.09 vs 27.21; CNR, 9.40 vs 4.71; P < 0.001 for all). Diagnostic performance assessed by the area under the receiver operating characteristic curve (AUROC) showed improved differentiation of WHO grades based on a pharmacokinetic parameter [Formula: see text] (AUC, 0.88 vs 0.83, P = 0.02), while remaining comparable to std-DCE in other parameters. Analysis of arterial input function (AIF) reliability demonstrated that [Formula: see text] exhibited superior agreement compared to [Formula: see text], as indicated by mostly higher intraclass correlation coefficients (Time to peak, 0.79 vs 0.43, P < 0.001). In conclusion, DLSD significantly enhances both the image quality and reliability of DCE-MRI in patients with diffuse glioma, while maintaining or improving diagnostic performance.

Dynamic Contrast Enhanced MRI Mapping of Vascular Permeability for Evaluation of Breast Cancer Neoadjuvant Chemotherapy Response Using Image-to-Image Conditional Generative Adversarial Networks

Dynamic contrast enhanced (DCE) MRI is a non-invasive imaging technique that has become a quantitative standard for assessing tumor microvascular permeability. Through the application of a pharmacokinetic (PK) model to a series of T1-weighed MR images acquired after an injection of a contrast agent, several vascular permeability parameters can be quantitatively estimated. These parameters, including Ktrans, a measure of capillary permeability, have been widely implemented for assessing tumor vascular function as well as tumor therapeutic response. However, conventional PK modeling for translation of DCE MRI to PK vascular permeability parameter maps is complex and time-consuming for dynamic scans with thousands of pixels per image. In recent years, image-to-image conditional generative adversarial network (cGAN) is emerging as a robust approach in computer vision for complex cross-domain translation tasks. Through a sophisticated adversarial training process between two neural networks, image-to-image cGANs learn to effectively translate images from one domain to another, producing images that are indistinguishable from those in the target domain. In the present study, we have developed a novel image-to-image cGAN approach for mapping DCE MRI data to PK vascular permeability parameter maps. The DCE-to-PK cGAN not only generates high-quality parameter maps that closely resemble the ground truth, but also significantly reduces computation time over 1000-fold. The utility of the cGAN approach to map vascular permeability is validated using open-source breast cancer patient DCE MRI data provided by The Cancer Imaging Archive (TCIA). This data collection includes images and pathological analyses of breast cancer patients acquired before and after the first cycle of neoadjuvant chemotherapy (NACT). Importantly, in good agreement with previous studies leveraging this dataset, the percentage change of vascular permeability Ktrans derived from the DCE-to-PK cGAN enables early prediction of responders to NACT.

Unpaired deep learning for pharmacokinetic parameter estimation from dynamic contrast-enhanced MRI without AIF measurements

DCE-MRI provides information about vascular permeability and tissue perfusion through the acquisition of pharmacokinetic parameters. However, traditional methods for estimating these pharmacokinetic parameters involve fitting tracer kinetic models, which often suffer from computational complexity and low accuracy due to noisy arterial input function (AIF) measurements. Although some deep learning approaches have been proposed to tackle these challenges, most existing methods rely on supervised learning that requires paired input DCE-MRI and labeled pharmacokinetic parameter maps. This dependency on labeled data introduces significant time and resource constraints and potential noise in the labels, making supervised learning methods often impractical. To address these limitations, we present a novel unpaired deep learning method for estimating pharmacokinetic parameters and the AIF using a physics-driven CycleGAN approach. Our proposed CycleGAN framework is designed based on the underlying physics model, resulting in a simpler architecture with a single generator and discriminator pair. Crucially, our experimental results indicate that our method does not necessitate separate AIF measurements and produces more reliable pharmacokinetic parameters than other techniques.

Contrast-agent-based perfusion MRI code repository and testing framework: ISMRM Open Science Initiative for Perfusion Imaging (OSIPI)

PURPOSE

Software has a substantial impact on quantitative perfusion MRI values. The lack of generally accepted implementations, code sharing and transparent testing reduces reproducibility, hindering the use of perfusion MRI in clinical trials. To address these issues, the ISMRM Open Science Initiative for Perfusion Imaging (OSIPI) aimed to establish a community-led, centralized repository for sharing open-source code for processing contrast-based perfusion imaging, incorporating an open-source testing framework.

METHODS

A repository was established on the OSIPI GitHub website. Python was chosen as the target software language. Calls for code contributions were made to OSIPI members, the ISMRM Perfusion Study Group, and publicly via OSIPI websites. An automated unit-testing framework was implemented to evaluate the output of code contributions, including visual representation of the results.

RESULTS

The repository hosts 86 implementations of perfusion processing steps contributed by 12 individuals or teams. These cover all core aspects of DCE- and DSC-MRI processing, including multiple implementations of the same functionality. Tests were developed for 52 implementations, covering five analysis steps. For T1 mapping, signal-to-concentration conversion and population AIF functions, different implementations resulted in near-identical output values. For the five pharmacokinetic models tested (Tofts, extended Tofts-Kety, Patlak, two-compartment exchange, and two-compartment uptake), differences in output parameters were observed between contributions.

CONCLUSIONS

The OSIPI DCE-DSC code repository represents a novel community-led model for code sharing and testing. The repository facilitates the re-use of existing code and the benchmarking of new code, promoting enhanced reproducibility in quantitative perfusion imaging.

Normalizing Flow-Based Distribution Estimation of Pharmacokinetic Parameters in Dynamic Contrast-Enhanced Magnetic Resonance Imaging

OBJECTIVE

The pharmacokinetic (PK) parameters estimated from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) provide valuable information for clinical research and diagnosis. However, these estimated PK parameters suffer from many sources of variability. Thus, the estimation of the posterior distributions of these PK parameters could provide a way to simultaneously quantify the values and uncertainties of the PK parameters. Our objective is to develop an efficient and flexible method to more closely approximate and estimate the underlying posterior distributions of the PK parameters.

METHODS

The normalizing flow model-based parameters distribution estimation neural network (FPDEN) is proposed to adaptively learn and estimate the posterior distributions of the PK parameters. The maximum likelihood estimation (MLE) loss is directly constructed based on the parameter distributions learned by the normalizing flow model, rather than pre-defined distributions.

RESULTS

Experimental analysis shows that the proposed method can improve parameter estimation accuracy. Moreover, the uncertainty derived from the parameter distribution constitutes an effective indicator to exclude unreliable parametric results. A successful demonstration is the improved classification performance of the glioma World Health Organization (WHO) grading task, specifically in terms of distinguishing between low and high grades, as well as between Grade III and Grade IV.

CONCLUSION

The FPDEN method offers improved accuracy for estimation of PK parameters and boosts the performance of the glioma grading task.

SIGNIFICANCE

By enhancing the precision and reliability of DCE-MRI, the proposed method promotes its further applications in clinical practice.

Deconvolution-Based Pharmacokinetic Analysis to Improve the Prediction of Pathological Information of Breast Cancer

Pharmacokinetic (PK) parameters, revealing changes in the tumor microenvironment, are related to the pathological information of breast cancer. Tracer kinetic models (e.g., Tofts-Kety model) with a nonlinear least square solver are commonly used to estimate PK parameters. However, the method is sensitive to noise in images. To relieve the effects of noise, a deconvolution (DEC) method, which was validated on synthetic concentration-time series, was proposed to accurately calculate PK parameters from breast dynamic contrast-enhanced magnetic resonance imaging. A time-to-peak-based tumor partitioning method was used to divide the whole tumor into three tumor subregions with different kinetic patterns. Radiomic features were calculated from the tumor subregion and whole tumor-based PK parameter maps. The optimal features determined by the fivefold cross-validation method were used to build random forest classifiers to predict molecular subtypes, Ki-67, and tumor grade. The diagnostic performance evaluated by the area under the receiver operating characteristic curve (AUC) was compared between the subregion and whole tumor-based PK parameters. The results showed that the DEC method obtained more accurate PK parameters than the Tofts method. Moreover, the results showed that the subregion-based Ktrans (best AUCs = 0.8319, 0.7032, 0.7132, 0.7490, 0.8074, and 0.6950) achieved a better diagnostic performance than the whole tumor-based Ktrans (AUCs = 0.8222, 0.6970, 0.6511, 0.7109, 0.7620, and 0.5894) for molecular subtypes, Ki-67, and tumor grade. These findings indicate that DEC-based Ktrans in the subregion has the potential to accurately predict molecular subtypes, Ki-67, and tumor grade.

Exploring the potential of Physics-Informed Neural Networks to extract vascularization data from DCE-MRI in the presence of diffusion

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is widely used to assess tissue vascularization, particularly in oncological applications. However, the most widely used pharmacokinetic (PK) models do not account for contrast agent (CA) diffusion between neighboring voxels, which can limit the accuracy of the results, especially in cases of heterogeneous tumors. To address this issue, previous works have proposed algorithms that incorporate diffusion phenomena into the formulation. However, these algorithms often face convergence problems due to the ill-posed nature of the problem. In this work, we present a new approach to fitting DCE-MRI data that incorporates CA diffusion by using Physics-Informed Neural Networks (PINNs). PINNs can be trained to fit measured data obtained from DCE-MRI while ensuring the mass conservation equation from the PK model. We compare the performance of PINNs to previous algorithms on different 1D cases inspired by previous works from literature. Results show that PINNs retrieve vascularization parameters more accurately from diffusion-corrected tracer-kinetic models. Furthermore, we demonstrate the robustness of PINNs compared to other traditional algorithms when faced with noisy or incomplete data. Overall, our results suggest that PINNs can be a valuable tool for improving the accuracy of DCE-MRI data analysis, particularly in cases where CA diffusion plays a significant role.

Efficient estimation of pharmacokinetic parameters from breast dynamic contrast-enhanced MRI based on a convolutional neural network for predicting molecular subtypes

Objective. Tracer kinetic models allow for estimating pharmacokinetic (PK) parameters, which are related to pathological characteristics, from breast dynamic contrast-enhanced magnetic resonance imaging. However, existing tracer kinetic models subject to inaccuracy are time-consuming for PK parameters estimation. This study aimed to accurately and efficiently estimate PK parameters for predicting molecular subtypes based on convolutional neural network (CNN).Approach. A CNN integrating global and local features (GL-CNN) was trained using synthetic data where known PK parameters map was used as the ground truth, and subsequently used to directly estimate PK parameters (volume transfer constantKtransand flux rate constantKep) map. The accuracy assessed by the peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and concordance correlation coefficient (CCC) was compared between the GL-CNN and Tofts-based PK parameters in synthetic data. Radiomic features were calculated from the PK parameters map in 208 breast tumors. A random forest classifier was constructed to predict molecular subtypes using a discovery cohort (n= 144). The diagnostic performance evaluated on a validation cohort (n= 64) using the area under the receiver operating characteristic curve (AUC) was compared between the GL-CNN and Tofts-based PK parameters.Main results. The average PSNR (48.8884), SSIM (0.9995), and CCC (0.9995) between the GL-CNN-basedKtransmap and ground truth were significantly higher than those between the Tofts-basedKtransmap and ground truth. The GL-CNN-basedKtransobtained significantly better diagnostic performance (AUCs = 0.7658 and 0.8528) than the Tofts-basedKtransfor luminal B and HER2 tumors. The GL-CNN method accelerated the computation by speed approximately 79 times compared to the Tofts method for the whole breast of all patients.Significance. Our results indicate that the GL-CNN method can be used to accurately and efficiently estimate PK parameters for predicting molecular subtypes.

Unified Bayesian network for uncertainty quantification of physiological parameters in dynamic contrast enhanced (DCE) MRI of the liver

Objective. Physiological parameter estimation is affected by intrinsic ambiguity in the data such as noise and model inaccuracies. The aim of this work is to provide a deep learning framework for accurate parameter and uncertainty estimates for DCE-MRI in the liver.Approach. Concentration time curves are simulated to train a Bayesian neural network (BNN). Training of the BNN involves minimization of a loss function that jointly minimizes the aleatoric and epistemic uncertainties. Uncertainty estimation is evaluated for different noise levels and for different out of distribution (OD) cases, i.e. where the data during inference differs strongly to the data during training. The accuracy of parameter estimates are compared to a nonlinear least squares (NLLS) fitting in numerical simulations andin vivodata of a patient suffering from hepatic tumor lesions.Main results. BNN achieved lower root-mean-squared-errors (RMSE) than the NLLS for the simulated data. RMSE of BNN was on overage of all noise levels lower by 33% ± 1.9% forktrans, 22% ± 6% forveand 89% ± 5% forvpthan the NLLS. The aleatoric uncertainties of the parameters increased with increasing noise level, whereas the epistemic uncertainty increased when a BNN was evaluated with OD data. For thein vivodata, more robust parameter estimations were obtained by the BNN than the NLLS fit. In addition, the differences between estimated parameters for healthy and tumor regions-of-interest were significant (p< 0.0001).Significance. The proposed framework allowed for accurate parameter estimates for quantitative DCE-MRI. In addition, the BNN provided uncertainty estimates which highlighted cases of high noise and in which the training data did not match the data during inference. This is important for clinical application because it would indicate cases in which the trained model is inadequate and additional training with an adapted training data set is required.

Radiomics characterization of tissues in an animal brain tumor model imaged using dynamic contrast enhanced (DCE) MRI

Here, we investigate radiomics-based characterization of tumor vascular and microenvironmental properties in an orthotopic rat brain tumor model measured using dynamic-contrast-enhanced (DCE) MRI. Thirty-two immune compromised-RNU rats implanted with human U-251N cancer cells were imaged using DCE-MRI (7Tesla, Dual-Gradient-Echo). The aim was to perform pharmacokinetic analysis using a nested model (NM) selection technique to classify brain regions according to vasculature properties considered as the source of truth. A two-dimensional convolutional-based radiomics analysis was performed on the raw-DCE-MRI of the rat brains to generate dynamic radiomics maps. The raw-DCE-MRI and respective radiomics maps were used to build 28 unsupervised Kohonen self-organizing-maps (K-SOMs). A Silhouette-Coefficient (SC), k-fold Nested-Cross-Validation (k-fold-NCV), and feature engineering analyses were performed on the K-SOMs' feature spaces to quantify the distinction power of radiomics features compared to raw-DCE-MRI for classification of different Nested Models. Results showed that eight radiomics features outperformed respective raw-DCE-MRI in prediction of the three nested models. The average percent difference in SCs between radiomics features and raw-DCE-MRI was: 29.875% ± 12.922%, p < 0.001. This work establishes an important first step toward spatiotemporal characterization of brain regions using radiomics signatures, which is fundamental toward staging of tumors and evaluation of tumor response to different treatments.

Dynamic contrast enhanced (DCE) MRI estimation of vascular parameters using knowledge-based adaptive models

We introduce and validate four adaptive models (AMs) to perform a physiologically based Nested-Model-Selection (NMS) estimation of such microvascular parameters as forward volumetric transfer constant, Ktrans, plasma volume fraction, vp, and extravascular, extracellular space, ve, directly from Dynamic Contrast-Enhanced (DCE) MRI raw information without the need for an Arterial-Input Function (AIF). In sixty-six immune-compromised-RNU rats implanted with human U-251 cancer cells, DCE-MRI studies estimated pharmacokinetic (PK) parameters using a group-averaged radiological AIF and an extended Patlak-based NMS paradigm. One-hundred-ninety features extracted from raw DCE-MRI information were used to construct and validate (nested-cross-validation, NCV) four AMs for estimation of model-based regions and their three PK parameters. An NMS-based a priori knowledge was used to fine-tune the AMs to improve their performance. Compared to the conventional analysis, AMs produced stable maps of vascular parameters and nested-model regions less impacted by AIF-dispersion. The performance (Correlation coefficient and Adjusted R-squared for NCV test cohorts) of the AMs were: 0.914/0.834, 0.825/0.720, 0.938/0.880, and 0.890/0.792 for predictions of nested model regions, vp, Ktrans, and ve, respectively. This study demonstrates an application of AMs that quickens and improves DCE-MRI based quantification of microvasculature properties of tumors and normal tissues relative to conventional approaches.

Transfer Learning Approach to Vascular Permeability Changes in Brain Metastasis Post-Whole-Brain Radiotherapy

The purpose of this study is to further validate the utility of our previously developed CNN in an alternative small animal model of BM through transfer learning. Unlike the glioma model, the BM mouse model develops multifocal intracranial metastases, including both contrast enhancing and non-enhancing lesions on DCE MRI, thus serving as an excellent brain tumor model to study tumor vascular permeability. Here, we conducted transfer learning by transferring the previously trained GBM CNN to DCE MRI datasets of BM mice. The CNN was re-trained to learn about the relationship between BM DCE images and target permeability maps extracted from the Extended Tofts Model (ETM). The transferred network was found to accurately predict BM permeability and presented with excellent spatial correlation with the target ETM PK maps. The CNN model was further tested in another cohort of BM mice treated with WBRT to assess vascular permeability changes induced via radiotherapy. The CNN detected significantly increased permeability parameter Ktrans in WBRT-treated tumors (p < 0.01), which was in good agreement with the target ETM PK maps. In conclusion, the proposed CNN can serve as an efficient and accurate tool for characterizing vascular permeability and treatment responses in small animal brain tumor models.

Radiomics can differentiate high-grade glioma from brain metastasis: a systematic review and meta-analysis

OBJECTIVE

(1) To evaluate the diagnostic performance of radiomics in differentiating high-grade glioma from brain metastasis and how to improve the model. (2) To assess the methodological quality of radiomics studies and explore ways of embracing the clinical application of radiomics.

METHODS

Studies using radiomics to differentiate high-grade glioma from brain metastasis published by 26 July 2021 were systematically reviewed. Methodological quality and risk of bias were assessed using the Radiomics Quality Score (RQS) system and Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool, respectively. Pooled sensitivity and specificity of the radiomics model were also calculated.

RESULTS

Seventeen studies combining 1,717 patients were included in the systematic review, of which 10 studies without data leakage suspicion were employed for the quantitative statistical analysis. The average RQS was 5.13 (14.25% of total), with substantial or almost perfect inter-rater agreements. The inclusion of clinical features in the radiomics model was only reported in one study, as was the case for publicly available algorithm code. The pooled sensitivity and specificity were 84% (95% CI, 80-88%) and 84% (95% CI, 81-87%), respectively. The performances of feature extraction from the volume of interest (VOI) or (semi) automatic segmentation in the radiomics models were superior to those of protocols employing region of interest (ROI) or manual segmentation.

CONCLUSION

Radiomics can accurately differentiate high-grade glioma from brain metastasis. The adoption of standardized workflow to avoid potential data leakage as well as the integration of clinical features and radiomics are advised to consider in future studies.

KEY POINTS

• The pooled sensitivity and specificity of radiomics for differentiating high-grade gliomas from brain metastasis were 84% and 84%, respectively. • Avoiding potential data leakage by adopting an intensive and standardized workflow is essential to improve the quality and generalizability of the radiomics model. • The application of radiomics in combination with clinical features in differentiating high-grade gliomas from brain metastasis needs further validation.

T2 mapping for the characterization of prostate lesions

Purpose

Purpose of this study is to evaluate the diagnostic accuracy of quantitative T2/ADC values in differentiating between PCa and lesions showing non-specific inflammatory infiltrates and atrophy, features of chronic prostatitis, as the most common histologically proven differential diagnosis.

Methods

In this retrospective, single-center cohort study, we analyzed 55 patients suspected of PCa, who underwent mpMRI (3T) including quantitative T2 maps before robot-assisted mpMRI-TRUS fusion prostate biopsy. All prostate lesions were scored according to PI-RADS v2.1. Regions of interest (ROIs) were annotated in focal lesions and normal prostate tissue. Quantitative mpMRI values from T2 mapping and ADC were compared using two-tailed t tests. Receiver operating characteristic curves (ROCs) and cutoff were calculated to differentiate between PCa and chronic prostatitis.

Results

Focal lesions showed significantly lower ADC and T2 mapping values than normal prostate tissue (p < 0.001). PCa showed significantly lower ADC and T2 values than chronic prostatitis (p < 0.001). ROC analysis revealed areas under the receiver operating characteristic curves (AUCs) of 0.85 (95% CI 0.74–0.97) for quantitative ADC values and 0.84 (95% CI 0.73–0.96) for T2 mapping. A significant correlation between ADC and T2 values was observed (r = 0.70; p < 0.001).

Conclusion

T2 mapping showed high diagnostic accuracy for differentiating between PCa and chronic prostatitis, comparable to the performance of ADC values.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00345-022-03991-8.