Quantitative DCE-MRI prediction of breast cancer recurrence following neoadjuvant chemotherapy: a preliminary study.

INTRODUCTION: Breast cancer patients treated with neoadjuvant chemotherapy (NACT) are at risk of recurrence depending on clinicopathological characteristics. This preliminary study aimed to investigate the predictive performances of quantitative dynamic contrast-enhanced (DCE) MRI parameters, alone and in combination with clinicopathological variables, for prediction of recurrence in patients treated with NACT. METHODS: Forty-seven patients underwent pre- and post-NACT MRI exams including high spatiotemporal resolution DCE-MRI. The Shutter-Speed model was employed to perform pharmacokinetic analysis of the DCE-MRI data and estimate the Ktrans, ve, kep, and τi parameters. Univariable logistic regression was used to assess predictive accuracy for recurrence for each MRI metric, while Firth logistic regression was used to evaluate predictive performances for models with multi-clinicopathological variables and in combination with a single MRI metric or the first principal components of all MRI metrics. RESULTS: Pre- and post-NACT DCE-MRI parameters performed better than tumor size measurement in prediction of recurrence, whether alone or in combination with clinicopathological variables. Combining post-NACT Ktrans with residual cancer burden and age showed the best improvement in predictive performance with ROC AUC = 0.965. CONCLUSION: Accurate prediction of recurrence pre- and/or post-NACT through integration of imaging markers and clinicopathological variables may help improve clinical decision making in adjusting NACT and/or adjuvant treatment regimens to reduce the risk of recurrence and improve survival outcome.

Pharmacokinetic modeling of dynamic contrast-enhanced MRI using a reference region and input function tail.

PURPOSE: Quantitative analysis of dynamic contrast-enhanced MRI (DCE-MRI) requires an arterial input function (AIF) which is difficult to measure. We propose the reference region and input function tail (RRIFT) approach which uses a reference tissue and the washout portion of the AIF. METHODS: RRIFT was evaluated in simulations with 100 parameter combinations at various temporal resolutions (5-30 s) and noise levels (σ = 0.01-0.05 mM). RRIFT was compared against the extended Tofts model (ETM) in 8 studies from patients with glioblastoma multiforme. Two versions of RRIFT were evaluated: one using measured patient-specific AIF tails, and another assuming a literature-based AIF tail. RESULTS: RRIFT estimated the transfer constant Ktrans and interstitial volume ve with median errors within 20% across all simulations. RRIFT was more accurate and precise than the ETM at temporal resolutions slower than 10 s. The percentage error of Ktrans had a median and interquartile range of -9 ± 45% with the ETM and -2 ± 17% with RRIFT at a temporal resolution of 30 s under noiseless conditions. RRIFT was in excellent agreement with the ETM in vivo, with concordance correlation coefficients (CCC) of 0.95 for Ktrans , 0.96 for ve , and 0.73 for the plasma volume vp using a measured AIF tail. With the literature-based AIF tail, the CCC was 0.89 for Ktrans , 0.93 for ve and 0.78 for vp . CONCLUSIONS: Quantitative DCE-MRI analysis using the input function tail and a reference tissue yields absolute kinetic parameters with the RRIFT method. This approach was viable in simulation and in vivo for temporal resolutions as low as 30 s.

Diffusion tensor imaging for characterizing tumor microstructure and improving diagnostic performance on breast MRI: a prospective observational study.

BACKGROUND: Diffusion-weighted imaging (DWI) can increase breast MRI diagnostic specificity due to the tendency of malignancies to restrict diffusion. Diffusion tensor imaging (DTI) provides further information over conventional DWI regarding diffusion directionality and anisotropy. Our study evaluates DTI features of suspicious breast lesions detected on MRI to determine the added diagnostic value of DTI for breast imaging. METHODS: With IRB approval, we prospectively enrolled patients over a 3-year period who had suspicious (BI-RADS category 4 or 5) MRI-detected breast lesions with histopathological results. Patients underwent multiparametric 3 T MRI with dynamic contrast-enhanced (DCE) and DTI sequences. Clinical factors (age, menopausal status, breast density, clinical indication, background parenchymal enhancement) and DCE-MRI lesion parameters (size, type, presence of washout, BI-RADS category) were recorded prospectively by interpreting radiologists. DTI parameters (apparent diffusion coefficient [ADC], fractional anisotropy [FA], axial diffusivity [λ1], radial diffusivity [(λ2 + λ3)/2], and empirical difference [λ1 - λ3]) were measured retrospectively. Generalized estimating equations (GEE) and least absolute shrinkage and selection operator (LASSO) methods were used for univariate and multivariate logistic regression, respectively. Diagnostic performance was internally validated using the area under the curve (AUC) with bootstrap adjustment. RESULTS: The study included 238 suspicious breast lesions (95 malignant, 143 benign) in 194 women. In univariate analysis, lower ADC, axial diffusivity, and radial diffusivity were associated with malignancy (OR = 0.37-0.42 per 1-SD increase, p < 0.001 for each), as was higher FA (OR = 1.45, p = 0.007). In multivariate analysis, LASSO selected only ADC (OR = 0.41) as a predictor for a DTI-only model, while both ADC (OR = 0.41) and FA (OR = 0.88) were selected for a model combining clinical and imaging parameters. Post-hoc analysis revealed varying association of FA with malignancy depending on the lesion type. The combined model (AUC = 0.81) had a significantly better performance than Clinical/DCE-MRI-only (AUC = 0.76, p < 0.001) and DTI-only (AUC = 0.75, p = 0.002) models. CONCLUSIONS: DTI significantly improves diagnostic performance in multivariate modeling. ADC is the most important diffusion parameter for distinguishing benign and malignant breast lesions, while anisotropy measures may help further characterize tumor microstructure and microenvironment.

Quantitative 3D dynamic contrast-enhanced (DCE) MR imaging of carotid vessel wall by fast T1 mapping using Multitasking.

PURPOSE: To develop a dynamic contrast-enhanced (DCE) MRI method capable of high spatiotemporal resolution, 3D carotid coverage, and T1-based quantification of contrast agent concentration for the assessment of carotid atherosclerosis using a newly developed Multitasking technique. METHODS: 5D imaging with 3 spatial dimensions, 1 T1 recovery dimension, and 1 DCE time dimension was performed using MR Multitasking based on low-rank tensor modeling, which allows direct T1 quantification with high spatiotemporal resolution (0.7 mm isotropic and 595 ms, respectively). Saturation recovery preparations followed by 3D segmented fast low angle shot readouts were implemented with Gaussian-density random 3D Cartesian sampling. A bulk motion removal scheme was developed to improve image quality. The proposed protocol was tested in phantom and human studies. In vivo scans were performed on 14 healthy subjects and 7 patients with carotid atherosclerosis. Kinetic parameters including area under the concentration versus time curve (AUC), vp , Ktrans , and ve were evaluated for each case. RESULTS: Phantom experiments showed that T1 measurements using the proposed protocol were in good agreement with reference value ( R2=0.96 ). In vivo studies demonstrated that AUC, vp , and Ktrans in the patient group were significantly higher than in the control group (0.63 ± 0.13 versus 0.42 ± 0.12, P < 0.001; 0.14 ± 0.05 versus 0.11 ± 0.03, P = 0.034; and 0.13 ± 0.04 versus 0.08 ± 0.02, P < 0.001, respectively). Results from repeated subjects showed good interscan reproducibility (intraclass correlation coefficient: vp , 0.83; Ktrans , 0.87; ve , 0.92; AUC, 0.94). CONCLUSION: Multitasking DCE is a promising approach for quantitatively assessing the vascularity properties of the carotid vessel wall.

A one-step biomarker quantification methodology for DCE-MRI of adnexal masses: Capturing kinetic pattern from early to late enhancement.

PURPOSE: To develop a one-step quantification approach that accounts for joint preprocessing and quantification of whole-range kinetics (early and late-phase washout) of dynamic contrast-enhanced (DCE) MRI of indeterminate adnexal masses. METHODS: Preoperative DCE-MRI of 43 (24 benign, 19 malignant) sonographically indeterminate adnexal masses were analyzed prospectively. A five-parameter sigmoid function was implemented to model the enhancement curves calculated within regions of interest. Diagnostic performance of five-parameter sigmoid model parameters (P1 through P5 ) was compared with pharmacokinetic (PK) modeling, semiquantitative analysis, and three-parameter sigmoid. Statistical analysis was performed using two-tailed student's t-test. RESULTS: The results revealed that P2 , representing the enhancement amplitude, is significantly higher, and P5 , indicating the terminal phase, is generally negative in malignant lesions (P < 0.001). P2 (sensitivity = 79%, specificity = 87.5%, accuracy = 84%, area under the receiver operating characteristic curve = 91%) outperforms classification performances of PK and semiquantitative parameters. A combination of P2 and P5 shows comparable performance (sensitivity = 79%, specificity = 87.5%, accuracy = 84%, area under the receiver operating characteristic curve = 92%) to that of the combination of PK parameters, whereas the five-parameter sigmoid function maintains fewer assumptions than PK. CONCLUSIONS: The presented one-step quantification approach is helpful for accurate discrimination of benign from malignant indeterminate adnexal masses. Accordingly, P2 has considerably high diagnostic performance and terminal slope (P5 ), as a previously overlooked feature, contributes more than widely accepted early-enhancement kinetic features. Magn Reson Med 79:1165-1171, 2018. © 2017 International Society for Magnetic Resonance in Medicine.

Characterizing and eliminating errors in enhancement and subtraction artifacts in dynamic contrast-enhanced breast MRI: Chemical shift artifact of the third kind.

PURPOSE: To characterize errors in enhancement in breast dynamic contrast-enhanced (DCE) MRI studies as a function of echo time and determine the source of dark band artifacts in clinical subtraction images. METHODS: Computer simulations, oil and water substitute (methylene chloride), as well as an American College of Radiology quality control phantom were tested. Routine clinical DCE breast MRI study was bracketed with (accelerated) in-phase DCE acquisitions in five patients. RESULTS: Simulation results demonstrated up to -160% suppression of the expected enhancement caused by differential enhancement of fat and water. Two-dimensional gradient-recalled echo and fat-suppressed 3D GRE phantom imaging confirmed the simulation results and showed that fat suppression does not eliminate the artifact. In vivo in-phase DCE images showed increased enhancement consistent with predictions and also confirmed increased spatial blurring on in-phase 3D gradient-recalled echo images. Combined multi-dimensional partial Fourier and parallel imaging provided a time-equivalent in-phase DCE MRI acquisition. CONCLUSION: Errors in expected enhancement occur in DCE breast MRI subtraction images because of differential enhancement of fat and water and incomplete fat signal suppression. These errors can lead to artificial suppression of enhancement as well as dark band artifacts on subtraction images. These artifacts can be eliminated with a time-equivalent in-phase fat-suppressed 3D gradient-recalled echo sequence. Understanding chemical shift artifact of the third kind, a unique artifact of artificial enhancement suppression in the presence of intravoxel fat and water signal, will aid DCE breast MRI image interpretation. In-phase acquisitions (combined with simultaneous minimum echo time or opposed-phase echoes) may facilitate qualitative, quantitative and longitudinal analysis of contrast enhancement. Magn Reson Med 79:2277-2289, 2018. © 2017 International Society for Magnetic Resonance in Medicine.

Correlation between voxel-wise enhancement parameters on DCE-MRI and pathological prognostic factors in invasive breast cancers.

PURPOSE: To investigate the correlation between enhancement parameters on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and pathologic prognostic factors in invasive breast cancers (BCs). MATERIALS AND METHODS: A total of 25 invasive BCs were included: 22 invasive ductal, 2 invasive lobular and 1 invasive mucinous. The tumor volume was segmented using a semi-automatic software (Olea Sphere). The following voxel-wise enhancement parameters were extracted: (1) time to peak enhancement; (2) signal intensity at peak (SIP); (3) peak enhancement percentage (PEP); (4) post-initial enhancement percentage (PIEP). The following pathological prognostic factors were considered for potential correlation: tumor (pT) and nodal (pN) stage, grading, perivascular/perineural invasion, estrogen/progesterone receptor status, Ki-67 proliferation, and HER2 expression. Spearman and Pearson correlation coefficients were calculated according with type of variable and data distribution. RESULTS: Tumor volume was 2.8 ± 2.0 cm3 (mean ± standard deviation [SD]). Mean SIP correlated with pT (ρ = 0.424, p = 0.035); mean PEP correlated with HER2 overexpression (ϕ = 0.471, p = 0.017) and pT (ρ = 0.449, p = 0.024). The percentage of voxels with fast PEP directly correlated with pT (ρ = 0.482, p = 0.015) and pN (ρ = 0.446, p = 0.026), while the percentage of voxels with slow PEP inversely correlated with pT (ρ = -0.421, p = 0.039) and pN (ρ = -0.481, p = 0.015). Segmentation time was 14.6 ± 1.3 min (mean ± SD). CONCLUSION: In invasive BCs, DCE-MRI voxel-wise enhancement parameters correlated with HER2, pT, and pN.

Increased robustness in reference region model analysis of DCE MRI using two-step constrained approaches.

PURPOSE: Reference region models (RRMs) can quantify tumor perfusion in dynamic contrast-enhanced MRI without an arterial input function. Inspection of the RRM reveals that one of the free parameters in the fit is uniquely linked to the reference region and is common to all voxels. A two-step approach is proposed that takes this constraint into account. METHODS: Three constrained RRM (CRRM) approaches were devised and evaluated. Simulations were performed to compare their accuracy and precision over a range of noise and temporal resolutions. The CRRM was also applied on a virtual phantom that simulates different perfusion values. In vivo evaluation was performed on data from breast cancer and soft tissue sarcoma. RESULTS: In simulations, the CRRM consistently improved precision and had better accuracy at low signal-to-noise ratio (SNR). In virtual phantom, the CRRMs were able to fit voxels that had similar kinetics to the reference tissue, whereas the unconstrained models failed to accurately fit these voxels. In the in vivo data, the constrained approaches produced parameter maps that had less variability and were in better agreement with the Tofts model. CONCLUSION: These findings indicate that the two-step fitting approach of the CRRM can reduce the variability of perfusion estimates for quantifying perfusion with dynamic contrast-enhanced (DCE) MRI. Magn Reson Med 78:1547-1557, 2017. © 2016 International Society for Magnetic Resonance in Medicine.

Direct estimation of tracer-kinetic parameter maps from highly undersampled brain dynamic contrast enhanced MRI.

PURPOSE: The purpose of this work was to develop and evaluate a T1 -weighted dynamic contrast enhanced (DCE) MRI methodology where tracer-kinetic (TK) parameter maps are directly estimated from undersampled (k,t)-space data. THEORY AND METHODS: The proposed reconstruction involves solving a nonlinear least squares optimization problem that includes explicit use of a full forward model to convert parameter maps to (k,t)-space, utilizing the Patlak TK model. The proposed scheme is compared against an indirect method that creates intermediate images by parallel imaging and compressed sensing before to TK modeling. Thirteen fully sampled brain tumor DCE-MRI scans with 5-second temporal resolution are retrospectively undersampled at rates R = 20, 40, 60, 80, and 100 for each dynamic frame. TK maps are quantitatively compared based on root mean-squared-error (rMSE) and Bland-Altman analysis. The approach is also applied to four prospectively R = 30 undersampled whole-brain DCE-MRI data sets. RESULTS: In the retrospective study, the proposed method performed statistically better than indirect method at R ≥ 80 for all 13 cases. This approach provided restoration of TK parameter values with less errors in tumor regions of interest, an improvement compared to a state-of-the-art indirect method. Applied prospectively, the proposed method provided whole-brain, high-resolution TK maps with good image quality. CONCLUSION: Model-based direct estimation of TK maps from k,t-space DCE-MRI data is feasible and is compatible up to 100-fold undersampling. Magn Reson Med 78:1566-1578, 2017. © 2016 International Society for Magnetic Resonance in Medicine.

Dynamic contrast-enhanced MRI for automatic detection of foci of residual or recurrent disease after prostatectomy.

PURPOSE: This study aimed to develop an automated procedure for identifying suspicious foci of residual/recurrent disease in the prostate bed using dynamic contrast-enhanced-MRI (DCE-MRI) in prostate cancer patients after prostatectomy. MATERIALS AND METHODS: Data of 22 patients presenting for salvage radiotherapy (RT) with an identified gross tumor volume (GTV) in the prostate bed were analyzed retrospectively. An unsupervised pattern recognition method was used to analyze DCE-MRI curves from the prostate bed. Data were represented as a product of a number of signal-vs.-time patterns and their weights. The temporal pattern, characterized by fast wash-in and gradual wash-out, was considered the "tumor" pattern. The corresponding weights were thresholded based on the number (1, 1.5, 2, 2.5) of standard deviations away from the mean, denoted as DCE1.0, …, DCE2.5, and displayed on the T2-weighted MRI. The resultant four volumes were compared with the GTV and maximum pre-RT prostate-specific antigen (PSA) level. Pharmacokinetic modeling was also carried out. RESULTS: Principal component analysis determined 2-4 significant patterns in patients' DCE-MRI. Analysis and display of the identified suspicious foci was performed in commercial software (MIM Corporation, Cleveland, OH, USA). In general, DCE1.0/DCE1.5 highlighted larger areas than GTV. DCE2.0 and GTV were significantly correlated (r = 0.60, p < 0.05). DCE2.0/DCA2.5 were also significantly correlated with PSA (r = 0.52, 0.67, p < 0.05). Ktrans for DCE2.5 was statistically higher than the GTV's Ktrans (p < 0.05), indicating that the automatic volume better captures areas of malignancy. CONCLUSION: A software tool was developed for identification and visualization of the suspicious foci in DCE-MRI from post-prostatectomy patients and was integrated into the treatment planning system.

Apparent diffusion coefficient values may help predict which MRI-detected high-risk breast lesions will upgrade at surgical excision.

PURPOSE: To investigate whether diffusion-weighted imaging (DWI) features could assist in determining which high-risk lesions identified on dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) and diagnosed on core needle biopsy (CNB) will upgrade to malignancy on surgical excision. MATERIALS AND METHODS: This Institutional Review Board (IRB)-approved prospective study included participants with MRI-detected Breast Imaging Reporting and Data System (BI-RADS) 4 or 5 lesions with high-risk pathology on CNB who underwent surgical excision. Twenty-three high-risk lesions detected on 3T breast MRI in 20 women (average age = 54 ± 9 years) were evaluated, of which six lesions (26%) upgraded to malignancy at surgery. DCE, DWI characteristics, and clinical factors were compared between high-risk lesions that upgraded to malignancy on surgical excision and those that did not. Logistic regression modeling was performed to identify features that optimally predicted upgrade to malignancy, with performance described using area under the receiver operating characteristic curve (AUC). RESULTS: High-risk lesions that upgraded on excision demonstrated lower apparent diffusion coefficient (ADC) than those that did not (median, 1.08 × 10-3 mm2 /s vs.1.39 × 10-3 mm2 /s, P = 0.046), and a trend of greater maximum lesion size (median, 24 mm vs. 8 mm, P = 0.053). There were no significant differences in lesion type (mass vs. nonmass enhancement, P = 1.0) or kinetic features (P = 0.78 for peak initial enhancement; P = 1.0 for worst curve type) among the high-risk cohorts. A model incorporating maximum lesion size and ADC provided optimal performance to predict upgrade to malignancy (AUC = 0.89). CONCLUSION: ADC and maximum lesion size on MRI show promise for predicting which MRI-detected high-risk lesions will upgrade to malignancy at surgical excision. LEVEL OF EVIDENCE: 2 Technical Efficacy: Stage 3 J. Magn. Reson. Imaging 2017;46:1028-1036.

Lymphoma and inflammation in the orbit: Diagnostic performance with diffusion-weighted imaging and dynamic contrast-enhanced MRI.

PURPOSE: To investigate the diagnostic performance of diffusion-weighted imaging (DWI), dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI), and the combination of both in the differential diagnosis of lymphoma and inflammation in the orbit. MATERIALS AND METHODS: This retrospective study was approved by the Institutional Review Board and the informed consent requirement was waived. A total of 53 patients underwent preoperative 3T MRI. Parameters of DWI and DCE MRI were evaluated in these 30 patients with orbital lymphoma and 23 patients with orbital inflammation. The diagnostic performance was evaluated using receiver operating characteristic curve analysis. RESULTS: Apparent diffusion coefficient (ADC) values and parameters derived from DCE MRI of orbital lymphoma and orbital inflammation differed significantly (ADC, Tmax , contrast index [CI], enhancement ratio [ER], and washout ratio [WR]: P < 0.001, P = 0.008, P < 0.001, P < 0.001, and P = 0.005 for reviewer 1, respectively; P < 0.001, P = 0.004, P < 0.001, P < 0.001, and P < 0.001 for reviewer 2, respectively). Sensitivity, specificity, and accuracy values of DWI were 76.67%, 100%, and 86.79% for reviewer 1; 70%, 95.65%, and 81.13% for reviewer 2, respectively. The combination of both were 90%, 86.96%, and 88.68% for reviewer 1; 93.33%, 78.26%, and 86.79% for reviewer 2, respectively. The combination of both was significantly superior to DWI for differentiation of orbital lymphoma from orbital inflammation (P = 0.016 for reviewer 1; P = 0.001 for reviewer 2). CONCLUSION: The combination of DWI and DCE MRI can improve diagnostic performance in differentiating lymphoma from inflammation in the orbit compared with DWI alone. LEVEL OF EVIDENCE: 3 J. MAGN. RESON. IMAGING 2017;45:1438-1445.

Gadolinium-free extracellular MR contrast agent for tumor imaging.

PURPOSE: To evaluate a new formulation of manganese porphyrin as a potential gadolinium (Gd)-free extracellular magnetic resonance imaging (MRI) contrast agent for dynamic contrast-enhanced (DCE) MRI of tumors. MATERIALS AND METHODS: A previously reported new contrast agent, MnTCP, was evaluated in six female tumor-bearing nude rats. MRI was performed on a 3 T clinical scanner 3 to 4 weeks after inoculation of breast tumor cells in the mammary fat pads. Gd-DTPA was injected intravenously, followed by injection of MnTCP at least 2 hours later (both at 0.05 mmol/kg). T1 relaxation time measurements and DCE-MRI were performed. RESULTS: Enhancement and clearance patterns were visually similar between MnTCP and Gd-DTPA. However, relative R1 increases in all 11 tumors were larger for MnTCP over 60 minutes postcontrast, the difference being significant as late as 20 minutes (R1post /R1pre = 1.42 ± 0.15 for MnTCP vs. 1.20 ± 0.08 for Gd-DTPA, P < 0.05). R1 -related effects for MnTCP were largely reduced after 60 minutes (R1post /R1pre = 1.13 ± 0.07) and completely gone within 24 hours (R1post /R1pre = 0.97 ± 0.06). DCE-MRI revealed a consistently larger (1.5 to over 2-fold) peak enhancement and higher values of the steepest slope, time-to-peak, and AUC60 in all tumors with MnTCP (P < 0.01). CONCLUSION: MnTCP is an alternative to extracellular Gd agents for tumor imaging, offering sensitive detection and rapid renal clearance.

Gadolinium-free T1 contrast agents for MRI: tunable pharmacokinetics of a new class of manganese porphyrins.

PURPOSE: To evaluate a new class of manganese porphyrins with tunable pharmacokinetics as potential gadolinium (Gd)-free T1 agents for contrast-enhanced magnetic resonance imaging (MRI). MATERIALS AND METHODS: Two new contrast agents, MnTCP and MnP2, were evaluated in four female rats. MRI was performed daily up to 3 days postinjection (0.05 mmol/kg) on a 3 T clinical scanner. T1 relaxation times and dynamic contrast-enhanced MRI were performed to assess contrast enhancement and clearance in blood, heart, liver, kidney, and muscle. RESULTS: Relative T1 decreases were similar for MnTCP and Gd-DTPA in all tissues but were significantly larger (P < 0.05) for MnP2 in blood, heart, kidney, and liver (2-6-fold larger). Clearance of MnTCP was similar to Gd-DTPA, with T1 returning to baseline by 40 minutes and complete elimination in 1 day. MnP2 was cleared from blood after 2 days and sustained a lowered T1 in other tissues for at least 1 hour (P < 0.05). The maximum enhancement, slope, and time-to-peak were similar between contrast agents. Only the parameter AUC60 differed, with MnP2 yielding the largest AUC60 values primarily through longer retention in tissue. CONCLUSION: MnTCP and MnP2 offer distinct applications as Gd-free T1 contrast agents. MnTCP behaves like a Gd-DTPA analog, while MnP2 provides significantly greater and longer positive signal enhancement.

TIST-Net: style transfer in dynamic contrast enhanced MRI using spatial and temporal information.

Objective.Training deep learning models for image registration or segmentation of dynamic contrast enhanced (DCE) MRI data is challenging. This is mainly due to the wide variations in contrast enhancement within and between patients. To train a model effectively, a large dataset is needed, but acquiring it is expensive and time consuming. Instead, style transfer can be used to generate new images from existing images. In this study, our objective is to develop a style transfer method that incorporates spatio-temporal information to either add or remove contrast enhancement from an existing image.Approach.We propose a temporal image-to-image style transfer network (TIST-Net), consisting of an auto-encoder combined with convolutional long short-term memory networks. This enables disentanglement of the content and style latent spaces of the time series data, using spatio-temporal information to learn and predict key structures. To generate new images, we use deformable and adaptive convolutions which allow fine grained control over the combination of the content and style latent spaces. We evaluate our method, using popular metrics and a previously proposed contrast weighted structural similarity index measure. We also perform a clinical evaluation, where experts are asked to rank images generated by multiple methods.Main Results.Our model achieves state-of-the-art performance on three datasets (kidney, prostate and uterus) achieving an SSIM of 0.91 ± 0.03, 0.73 ± 0.04, 0.88 ± 0.04 respectively when performing style transfer between a non-enhanced image and a contrast-enhanced image. Similarly, SSIM results for style transfer from a contrast-enhanced image to a non-enhanced image were 0.89 ± 0.03, 0.82 ± 0.03, 0.87 ± 0.03. In the clinical evaluation, our method was ranked consistently higher than other approaches.Significance.TIST-Net can be used to generate new DCE-MRI data from existing images. In future, this may improve models for tasks such as image registration or segmentation by allowing small training datasets to be expanded.

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.

Analysis of parametric histogram from dynamic contrast-enhanced MRI: application in evaluating brain tumor response to radiotherapy.

Dynamic contrast-enhanced MRI (DCE MRI) has been used to study tumor response to treatment for many years. In this study, the modified full width at half-maximum (mFWHM), calculated from the wash-in slope histogram, is proposed as a parameter for the evaluation of changes in tumor heterogeneity which respond to radiotherapy. Twenty-five patients with brain tumors were evaluated and divided into the nonresponder group (n = 11) and the responder group (n = 14) according to the Response Evaluation Criteria in Solid Tumors (RECIST). All selected tumors were evaluated by mFWHM ratios of post- to pre-therapy (the ratio was defined as the therapeutic mFWHM ratio, TMR). The changes in kurtosis of the histograms and the averaged K(trans) within a tumor were also calculated for comparison. The receiver operating characteristic analysis and Kaplan-Meier curves were used to examine the diagnosis ability. The TMR values were significantly higher in nonresponders than in responders (p < 0.001). When compared with the other two parameters, the proposed method also demonstrated better sensitivity and specificity. When adopting the TMR for the estimation of prognosis after therapy, there was a significant difference between the population survival curves. In conclusion, the derived mFWHM reflects tumor heterogeneity, and the ability to depict patient survival probability from TMR corresponds well with that from RECIST. The results reveal that, in brain tumors, progression may be exhibited not only by tumor size, but also by tumor heterogeneity.

Deep Learning-based Deformable Registration of Dynamic Contrast-Enhanced MR Images of the Kidney.

Respiratory motion is a major contributor to bias in quantitative analysis of magnetic resonance imaging (MRI) acquisitions. Deformable registration of three-dimensional (3D) dynamic contrast-enhanced (DCE) MRI data improves estimation of kidney kinetic parameters. In this study, we proposed a deep learning approach with two steps: a convolutional neural network (CNN) based affine registration network, followed by a U-Net trained for deformable registration between two MR images. The proposed registration method was applied successively across consecutive dynamic phases of the 3D DCE-MRI dataset to reduce motion effects in the different kidney compartments (i.e., cortex, medulla). Successful reduction in the motion effects caused by patient respiration during image acquisition allows for improved kinetic analysis of the kidney. Original and registered images were analyzed and compared using dynamic intensity curves of the kidney compartments, target registration error of anatomical markers, image subtraction, and simple visual assessment. The proposed deep learning-based approach to correct motion effects in abdominal 3D DCE-MRI data can be applied to various kidney MR imaging applications.

Quantitative transport mapping (QTM) for differentiating benign and malignant breast lesion: Comparison with traditional kinetics modeling and semi-quantitative enhancement curve characteristics.

PURPOSE: To test the feasibility of using quantitative transport mapping (QTM) method, which is based on the inversion of transport equation using spatial deconvolution without any arterial input function, for automatically postprocessing dynamic contrast enhanced MRI (DCE-MRI) to differentiate malignant and benign breast tumors. MATERIALS AND METHODS: Breast DCE-MRI data with biopsy confirmed malignant (n = 13) and benign tumors (n = 13) was used to assess QTM velocity (|u|) and diffusion coefficient (D), volume transfer constant (Ktrans), volume fraction of extravascular extracellular space (Ve) from kinetics method, and traditional enhancement curve characteristics (ECC: amplitude A, wash-in rate α, wash-out rate ß). A Mann-Whitney U test and receiver operating characteristic curve (ROC) analysis were performed to assess the diagnostic performance of these parameters for distinguishing between benign and malignant tumors. RESULTS: Between malignant and benign tumors, there was a significant difference in |u| and Ktrans, (p = 0.0066, 0.0274, respectively), but not in D, Ve, A, α and ß (p = 0.1119, 0.2382, 0.4418,0.2592 and 0.9591, respectively). ROC area-under-the-curve was 0.82, 0.75 (95% confidence level 0.60-0.95, 0.51-0.90) for |u| and Ktrans, respectively. CONCLUSION: QTM postprocesses DCE-MRI automatically through deconvolution in space and time to solve the inverse problem of the transport equation. Comparing with traditional kinetics method and ECC, QTM method showed better diagnostic accuracy in differentiating benign from malignant breast tumors in this study.

The kinetic analysis of breast cancer: An investigation of the optimal temporal resolution for dynamic contrast-enhanced MR imaging.

INTRODUCTION: There is wide agreement that morphologic features and enhancement kinetics should be evaluated for MRI of the breast, although there has been no clear consensus concerning optimal temporal resolutions. The objective of this study was to investigate the optimal temporal resolution for the kinetic analysis of breast cancers. METHODS: Thirty-four patients with 34 enhancing lesions of breast cancer who underwent dynamic contrast-enhanced MRI (DCE-MRI) on a 3.0-T scanner were included in this retrospective study. DCE-MRI was performed with an original temporal resolution of 10-s, and the values of pharmacokinetic parameters (Ktrans, Ve, Kep, and area under the curve (AUC)) were compared with selected data of 30-s and 60-s time intervals. RESULTS: Among the 34 lesions, 10 showed a wash out pattern, 16 showed a plateau pattern, and 8 showed a persistent enhancement pattern. The Ktrans value in the wash-out pattern was significantly higher than that of other time-intensity curve patterns (p < 0.01). The Kep and AUC also showed significant differences between the wash-out pattern and other types (p < 0.01). On comparing the perfusion parameters among different temporal resolutions, simulations showed that only the AUC differed significantly between the data acquired at a 10-s temporal resolution and that acquired at a 60-s time interval (p < 0.01). Although the comparison of the AUC between the 30-s and 60-s data also showed significant differences (p = 0.01), there was no significant difference between the 10-s and 30-s data (p = 0.17). CONCLUSIONS: DCE-MRI with a temporal resolution of 30-s preserves the kinetic information. Further prospective studies will be needed to investigate the trade-off between temporal and spatial resolution in DCE-MRI.