Dynamic contrast-enhanced case-control analysis in 3T MRI of prostate cancer can help to characterize tumor aggressiveness

doi:10.1016/j.ejrad.2016.09.022

European Journal of Radiology

Volume 85, Issue 11, November 2016, Pages 2119-2126

https://doi.org/10.1016/j.ejrad.2016.09.022 Get rights and content

Highlights

•
Curve types showed no statistical association with healthy/tumor peripheral areas.
•
K^trans, v_e, upslope and AUC showed significant differences in controls vs. tumors.
•
The global diagnostic performance of standard MRI perfusion parameters is poor.
•
Normalized K^trans, upslope and AUC had good diagnostic accuracy for tumor grading.

Abstract

Purpose

The aim of this work is to establish normality and tumor tissue ranges for perfusion parameters from dynamic contrast-enhanced (DCE) MR of the peripheral prostate at 3T and to compare the diagnostic performance of quantitative and semi-quantitative parameters.

Materials and methods

Thirty-six patients with prostate carcinomas (18 Gleason-6, 15 Gleason-7, and 3 Gleason-8) and 33 healthy subjects were included. Image analysis workflow comprised four steps: manual segmentation of whole prostate and lesions, series registration, voxelwise T1 mapping and calculation of pharmacokinetic and semi-quantitative parameters.

Results

K^trans, v_e, upslope and AUC60 showed statistically significant differences between healthy peripheral areas and tumors. Curve type showed no association with healthy/tumor peripheral areas (chi-square = 0.702). Areas under the ROC curves were 0.64 (95% CI: 0.54–0.75), 0.70 (0.60–0.80), 0.62 (0.51–0.72) and 0.63 (0.52–0.74) for K^trans, v_e, upslope and AUC60, respectively. The optimal cutoff values were: K^trans = 0.21 min⁻¹ (sensitivity = 0.61, specificity = 0.64), v_e = 0.36 (0.63, 0.71), upslope = 0.59 (0.59, 0.59) and AUC60 = 2.4 (0.63, 0.64). Significant differences were found between Gleason scores 6 and 7 for normalized K^trans, upslope and AUC60, with good diagnostic accuracy (area under ROC curve 0.80, 95% CI: 0.60–1.00).

Conclusion

Quantitative (K^trans and v_e) and semi-quantitative (upslope and AUC60) perfusion parameters showed significant differences between tumors and control areas in the peripheral prostate. Normalized K^trans, upslope and AUC60 values might characterize tumor aggressiveness.

Introduction

Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is an interesting technique to non-invasively evaluate prostate cancer and establish perfusion differences between tumor grade and normal tissue [1]. It has been included in the qualitative assessment of the malignancy probability of a certain region through the PI-RADS scoring system [2], [3]. Currently, in version 2, the role of DCE-MRI is secondary to diffusion imaging and it focuses on subjectively assessing the enhancement characteristics of the peripheral zone.

Quantitative parameters obtained from pharmacokinetic models [4], [5] applied to DCE-MRI series have demonstrated superior performance to differentiate cancerous from normal tissue prostate in comparison to simpler enhancement analysis such as semi-quantitative or qualitative parameters [6], [7], as they provide physiological information related to capillary permeability, blood fraction and interstitial space volume.

Several research groups have published pharmacokinetic parameters results both from the normal and pathological prostate gland (Table 1). There are striking differences in the results across the studies, with overlaps between normal (0.36 ± 0.60, mean ± standard deviation) and carcinoma (0.55 ± 0.59) K^trans values. Strict comparisons among results are limited due to variability in methodologies. Notwithstanding, at least four main sources of variability can be identified: MRI acquisition protocol, type of gadolinium-based contrast agent, signal analysis methods and statistical approaches.

The variability in image analysis methods also relates to the selection of the arterial input function (AIF) (individual manual, individual automatic, population-averaged or reference); the selection of the pharmacokinetic model; the selection of the region of interest (whole prostate, only the hot spots or only central/peripheral gland) and the signal intensity to contrast agent concentration conversion strategy. The measured properties can also be analyzed on a voxel-by-voxel basis or after averaging the enhancement curve over manually defined ROIs. The statistical approach has also methodological differences, such as the use of different statistical descriptors either from the whole ROI or from some percentile of the data distribution. Standards need to be reached with regard to image and series acquisition, and data processing [23], [24], [25], a fact that may limit a widespread clinical use and multicenter comparisons of these parameters.

The aim of this study was to establish prospectively ranges of normality and tumor-specific values for the standardized DCE-MRI pharmacokinetic parameters of the prostate at 3T, and to compare the performance of the quantitative and semi-quantitative parameters. A secondary aim was to give further evidence of the potential of the pharmacokinetic parameters as imaging biomarkers. Recommendations proposed by the QIBA group [26] have been followed wherever possible.

Section snippets

Subjects

The Ethics Committee approved this study. All patients signed an informed consent for the inclusion of their anonymized data in the study.

The initial group consisted of 133 eligible patients. The inclusion criterion for the tumor group was histological confirmation of prostate cancer. The inclusion criteria for the healthy group were asymptomatic patients; stable PSA (<2.5 ng/ml) in at least two controls; negative digital rectal examination; no significant increase in PSA and negative digital

Comparison between peripheral healthy and tumor ROIs

K^trans, v_e, upslope and AUC60 showed statistically significant differences for all the statistical descriptors. Table 3 shows a summary of the results.

The contingency table for the curve type showed no significant association between healthy/tumor peripheral areas and the three types of curve (chi-square = 0.702, AUROC curve 0.53). Healthy peripheral areas showed 12%, 69% and 19% of curve types I, II and III, respectively. Tumor areas showed 11%, 63% and 26% of curves I, II and III, respectively.

Quantitative analysis

The values for K^trans were in the range of other studies [8], [10], [11], [17], [20], [22], with mean values of 0.2 min⁻¹ for the healthy peripheral gland and 0.4 min⁻¹ for the tumor. Franiel et al. [13] and Vos et al. [18] obtained slightly higher values for both regions. Other works reported comparatively lower values for both regions [14], [16], [19], [21]. On the other side, Lüdemann et al. [9] reported much higher values.

For k_ep, no significant differences between healthy and tumor

Conclusions

The results presented in this study contribute to the standardization of tumor and normal prostate values for quantitative parameters obtained from prostate DCE-MR images. It has been demonstrated that quantitative (K^trans and v_e) and semi-quantitative (upslope and AUC60) perfusion parameters show statistically significant differences between tumors and control areas in the peripheral prostate. However, the overall performance of individual DCE-MRI parameters remains relatively poor to

Conflict of interest

The authors declare no conflict of interest.

Funding

This work was supported by a grant from GUERBET (investigator-initiated study). GUERBET did not have any role in the study design, implementation and discussions on the study results.

References (30)

P. Puech et al.
Prostate MRI: can we do without DCE sequences in 2013?
Diagn. Interv. Imaging
(2013)
J.C. Weinreb et al.
PI-RADS Prostate Imaging - Reporting and Data System: 2015 version 2
Eur. Urol.
(2016)
P. Kozlowski et al.
Combined prostate diffusion tensor imaging and dynamic contrast-enhanced MRI at 3T–quantitative correlation with biopsy
Magn. Reson. Imaging
(2010)
E.K. Vos et al.
Assessment of prostate cancer aggressiveness using dynamic contrast-enhanced MRI at 3T
Eur. Urol.
(2013)
F.M. Fennessy et al.
Practical considerations in T1 mapping of prostate for dynamic contrast-enhancement pharmacokinetic analyses
Magn. Reson. Imaging
(2012)
A. Fedorov et al.
The role of pathology correlation approach in prostate cancer index lesion detection and quantitative analysis with multiparametric MRI
Acad. Radiol.
(2015)
A. Fedorov et al.
A comparison of two methods for estimating DCE-MRI parameters via individual and cohort based AIFs in prostate cancer: a step towards practical implementation
Magn. Reson. Imaging
(2014)
J.O. Barentsz et al.
European Society of Urogenital Radiology, ESUR prostate MR guidelines 2012
Eur. Radiol.
(2012)
P.S. Tofts et al.
Estimating kinetic parameters from dynamic contrast-enhanced T(1)-weighted MRI of a diffusable tracer: standardized quantities and symbols
J. Magn. Reson. Imaging
(1999)
T. Franiel et al.
Dynamic contrast-enhanced MRI and pharmacokinetic models in prostate cancer
Eur. Radiol.
(2011)

J.J. Futterer et al.

Prostate cancer localization with dynamic contrast-enhanced MR imaging and proton MR spectroscopic imaging

Radiology

(2006)

D. Bonekamp et al.

Advancements in MR imaging of the prostate: from diagnosis to interventions

Radiographics

(2011)

I. Ocak et al.

Dynamic contrast-enhanced MRI of prostate cancer at 3T: a study of pharmacokinetic parameters

AJR Am. J. Roentgenol.

(2007)

L. Lüdemann et al.

Simultaneous quantification of perfusion and permeability in the prostate using dynamic contrast-enhanced MRI with and inversion-prepared dual-contrast sequence

Ann. Biomed. Eng.

(2009)

A.S. Jackson et al.

Dynamic contrast-enhanced MRI for prostate cancer localization

Br. J. Radiol.

(2009)

Cited by (28)

Sequential multiblock partial least squares discriminant analysis for assessing prostate cancer aggressiveness with multiparametric magnetic resonance imaging
2022, Chemometrics and Intelligent Laboratory Systems
In current radiology practice, multi-parametric magnetic resonance imaging (mpMRI) has recently become a key tool in diagnostic and therapeutic decisions. Although it is based on the subjective assessment of T2-weighted images, as well as perfusion-weighted and diffusion-weighted sequences, further quantitative parameters can also be derived from them for improving lesion phenotyping. Despite these parameters are usually exploited in a univariate way, ignoring the benefits of a real multivariate approach, still it is the gold standard imaging technique to assess prostate cancer location and probability of malignancy. In this paper, pharmacokinetic (perfusion) and exponential (diffusion) clinical models, as well as latent variable-based multivariate statistical models like multivariate curve resolution-alternating least squares (MCR-ALS), have been calculated and analyzed with sequential multi block-partial least squares discriminant analysis (SMB-PLS-DA) including technique-block differentiation, in order to better assess for cancer aggressiveness based on Gleason scales. The best prediction result was achieved by the ordered combination of diffusion blocks (MCR-ALS and exponential models) and normalized T2 values. The perfusion blocks did not improve the results obtained by diffusion and T2-weighted based parameters alone, so they can be removed from the SMB-PLS-DA model.
Quantitative diffusion-weighted imaging and dynamic contrast-enhanced MR imaging for assessment of tumor aggressiveness in prostate cancer at 3T
2021, Magnetic Resonance Imaging
To compare diffusion-weighted imaging (DWI) and dynamic contrast-enhanced MR imaging (DCE-MRI) for characterization of prostate cancer (PC).
104 PC patients who underwent prostate multiparametric MRI at 3T including DWI and DCE-MRI before MRI-guided biopsy or radical prostatectomy. Apparent diffusion coefficient (ADC) with histogram analysis (mean, 0–25th percentile, skewness, and kurtosis), intravoxel incoherent motion model including D and f; stretched exponential model including distributed diffusion coefficient (DDC) and a; and permeability parameters including K^trans, K_ep, and V_e were obtained from a region of interest placed on the dominant tumor of each patient.
ADC_mean, ADC₀_–₂₅, D, DDC, and V_e were significantly lower and K_ep was significantly higher in GS ≥ 3 + 4 tumors (n = 89) than in GS = 3 + 3 tumors (n = 15), and also in GS ≥ 4 + 3 tumors (n = 57) than in GS ≤ 3 + 4 tumors (n = 47) (P < 0.001 to P = 0.040). f was significantly lower in GS ≥ 4 + 3 tumors than in GS ≤ 3 + 4 tumors (P = 0.022), but there was no significant difference between GS = 3 + 3 tumors and GS ≥ 3 + 4 tumors, or between the remaining metrics in both comparisons. In metrics with area under the curve (AUC) >0.80, there was a significant difference in AUC between ADC₀_–₂₅ and D, and DDC for separating GS ≤ 3 + 4 tumors from GS ≥ 4 + 3 tumors (P = 0.040 and P = 0.022, respectively). There were no significant differences between metrics with AUC > 0.80 for separating GS = 3 + 3 tumors from GS ≥ 3 + 4 tumors. ADC₀_–₂₅ had the highest correlation with Gleason grade (ρ = −0.625, P < 0.001).
DWI and DCE-MRI showed no apparent clinical superiority of non-Gaussian models or permeability MRI over the mono-exponential model for assessment of tumor aggressiveness in PC.
The Utility of Diffusion-Weighted Imaging and Perfusion Magnetic Resonance Imaging Parameters for Detecting Clinically Significant Prostate Cancer
2019, Canadian Association of Radiologists Journal
Citation Excerpt :
The Ktrans for low-grade tumours reported in their protocol was also similar to ours (0.56 ± 0.12), but for high-grade tumours, the Ktrans value (0.60 ± 0.17) was lower than that of the present study. Sanz-Requina et al [11] studied a similar small group and reported no clinically significant differences between GS of 6 and GS of 7 for Ktrans, Kep, or Ve. Similar to the studies described above, Oto et al [10] reported no significant correlation between GS and quantitative perfusion parameters and did not clarify the mean or median values for the subgroups.
To establish the diagnostic performance of the parameters obtained from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and diffusion-weighted imaging at 3T in discriminating between non-clinically significant prostate cancers (ncsPCa, Gleason score [GS] < 7) and clinically significant prostate cancers (csPCa, GS ≥ 7) in the peripheral zone.
Twenty-six male patients with peripheral zone prostate cancer (PCa) who had undergone 3T multiparametric magnetic resonance imaging (MRI) scan prior to biopsy were included in the study and evaluated retrospectively. The GS was obtained by both standard 12-core transrectal ultrasound guided biopsy and targeted MRI-US fusion biopsy and then confirmed by prostatectomy, if available. For each confirmed tumour focus, DCE-derived quantitative perfusion metrics (Ktrans, Kep, Ve, initial area under the curve [AUC]), the apparent diffusion coefficient (ADC) value, and normalized versions of quantitative metrics were measured and correlated with the GS.
Ktrans had the highest diagnostic accuracy value of 82% among the DCE-MRI parameters (AUC 0.90), and ADC had the strongest diagnostic accuracy value of 87% among the overall parameters (AUC 0.92). The combination of ADC and Ktrans have higher diagnostic performance with the area under the receiver operating characteristic curve being 0.98 (sensitivity 0.94; specificity 0.89; accuracy 0.92) compared to the individual evaluation of each parameter alone.The GS showed strong negative correlations with ADC (r = −0.72) and normalized ADC (r = −0.69) as well as a significant positive correlation with Ktrans (r = 0.69).
The combination of Ktrans and ADC and their normalized versions may help differentiate between ncsPCa from csPCa in the peripheral zone.
Établir les performances diagnostiques des paramètres obtenus par imagerie par résonance magnétique dynamique de contraste (IRM-DSC) et par imagerie pondérée en diffusion à 3T dans la distinction des cancers de la prostate sans importance clinique (CPsic; score de Gleason [SG] < 7) des cancers de la prostate d'importance clinique (CPic; SG ≥ 7) dans la zone périphérique.
Trente-six patients de sexe masculin atteints d'un cancer de la prostate dans la zone périphérique ayant bénéficié d'un examen d'imagerie par résonance magnétique (IRM) multiparamétrique à 3T préalable à une biopsie ont été intégrés dans l’étude et évalués rétrospectivement. Le score de Gleason (SG) a été mesuré par les deux techniques conventionnelles de biopsie à 12 prélèvements: une biopsie contrôlée sous échographie et une biopsie ciblée par fusion IRM/Echo. Il a ensuite été confirmé par prostatectomie, si pratiquée. Pour chaque tumeur confirmée évaluée, les paramètres associés à la perfusion quantitative dérivée de la DSC (constante de transfert [Ktrans], constante d’échange [Kpe], volume interstitiel [Ve] et aire sous la courbe [ASC]), le coefficient de diffusion apparent (CDA) et les versions normalisées des paramètres quantitatifs ont été mesurés et corrélés avec le SG.
La Ktrans était la valeur associée à la précision diagnostique la plus élevée (82 %) parmi les paramètres d’IRM-DSC (ASC = 0,90), et le CDA avait le pouvoir de précision diagnostique le plus élevé (87 %) parmi l'ensemble des paramètres (ASC = 0,92). La combinaison du CDA et de la Ktrans avait la performance diagnostique la plus élevée, avec une courbe des caractéristiques de performance (ROC) de 0,98 (sensibilité de 0,94; spécificité de 0,89; précision de 0,92) par rapport à l’évaluation individuelle de chacun des paramètres. Le SG présentait une forte corrélation négative avec le CDA (r = –0,72) et le CDA normalisé (r = –0,69) et une corrélation positive importante avec la Ktrans (r = 0,69).
La combinaison de la constante de transfert (Ktrans) et du coefficient de corrélation apparent (CDA), ainsi que leurs versions normalisées, pourrait contribuer à la différenciation des CPsic des CPic, dans la zone périphérique.
Can DCE-MRI reduce the number of PI-RADS v.2 false positive findings? Role of quantitative pharmacokinetic parameters in prostate lesions characterization
2019, European Journal of Radiology
Citation Excerpt :
Nevertheless, these results could be influenced by the great numeric difference among our GS classes, with GS 7 representing 74% of all neoplastic findings. Cho E et al. [25] and Sanz-requena R et al. [27] in their studies proposed a Ktrans cut-off value for the prediction of significant PCa (184 × 10−3/min and 210 × 10−3/min, respectively); these results are highly comparable with our findings (Ktrans cut-off: 191 × 10−3/min). In our study we were unable to find a satisfying cut-off for Kep values.
To test the potential impact of pharmacokinetic parameters, derived from DCE-MRI analysis, on the diagnostic performance of PI-RADSv.2 classification in prostate lesions characterization.
Among patients who underwent multiparametric prostate MRI (mpMRI) (January 2016-March 2018) followed by histological evaluation (targeted biopsies/prostatectomy), 103 men were retrospectively selected. For each patient the index lesion was identified and pharmacokinetic parameters (Ktrans, Kep, Ve, Vp) were assessed. MRI diagnostic performance in the detection of significant tumors [Gleason Score (GS)≥7] was assessed, considering PI-RADS≥3 as positive.
GS ≥ 7 (n = 59) showed higher Ktrans (p < 0.01) and Kep (p = 0.01) compared to GS < 7. At ROC curve analysis, a Ktrans cut-off of 191 × 10⁻³/min was identified to predict the presence of GS ≥ 7 (AUC:0.75; sensitivity:95%; specificity:61%). Sensitivity and PPV of mpMRI using PI-RADSv.2 were 98% and 61%. Reclassifying PI-RADS≥3 lesions according to Ktrans cut-off, 22 false positives were shifted to true negatives with 3 false negative findings; PPV raised to 79%. Appling Ktrans cut-off to PI-RADS 3 lesions of peripheral zone (n = 18), 12 true negatives, 4 true positives, 2 false positives were identified.
Despite its high sensitivity prostate mpMRI generates many false positive cases: Ktrans in addition to PIRADS v.2 seems to improve MRI-PPV and may help in avoiding redundant biopsies.
Hierarchical segmentation using equivalence test (HiSET): Application to DCE image sequences
2019, Medical Image Analysis
Citation Excerpt :
DCE (dynamic contrast enhanced) imaging using computed tomography (CT), magnetic resonance imaging (MRI) or ultrasound imaging (US) appears promising as it can monitor the local changes in microcirculation secondary to the development of new vessels (neo-angiogenesis). DCE-MRI and DCE-CT (called also CT-perfusion) have been extensively tested alone or in combination with other techniques (Winfield et al., 2016) in pathological conditions such as cancer, ischemia and inflammation, in various tissues including brain (Bergamino et al., 2014), breast (Chen et al., 2010), prostate (Sanz-Requena et al., 2016), heart (Bakir et al., 2016; Nee et al., 2009), kidney (Woodard et al., 2015), liver (Raj and Juluru, 2009; Chen and Shih, 2014), genital organs, gastrointestinal tract, bone (Michoux et al., 2012) and placenta (Frias et al., 2015). They show a great potential to: 1/ detect and characterize lesions (Sanz-Requena et al., 2016; Ferre et al., 2016; Bhooshan et al., 2010); 2/ personalize treatment including new targeted drugs, radiotherapy and mini invasive surgery; 3/ monitor and optimize treatments during the follow-up (Padhani and Khan, 2010; Wang et al., 2014) or after heart, liver or kidney transplants (Khalifa et al., 2013).
Dynamical contrast enhanced (DCE) imaging allows non invasive access to tissue micro-vascularization. It appears as a promising tool to build imaging biomarkers for diagnostic, prognosis or anti-angiogenesis treatment monitoring of cancer. However, quantitative analysis of DCE image sequences suffers from low signal to noise ratio (SNR). SNR may be improved by averaging functional information in a large region of interest when it is functionally homogeneous.
We propose a novel method for automatic segmentation of DCE image sequences into functionally homogeneous regions, called DCE-HiSET. Using an observation model which depends on one parameter a and is justified a posteriori, DCE-HiSET is a hierarchical clustering algorithm. It uses the p-value of a multiple equivalence test as dissimilarity measure and consists of two steps. The first exploits the spatial neighborhood structure to reduce complexity and takes advantage of the regularity of anatomical features, while the second recovers (spatially) disconnected homogeneous structures at a larger (global) scale.
Given a minimal expected homogeneity discrepancy for the multiple equivalence test, both steps stop automatically by controlling the Type I error. This provides an adaptive choice for the number of clusters. Assuming that the DCE image sequence is functionally piecewise constant with signals on each piece sufficiently separated, we prove that DCE-HiSET will retrieve the exact partition with high probability as soon as the number of images in the sequence is large enough. The minimal expected homogeneity discrepancy appears as the tuning parameter controlling the size of the segmentation. DCE-HiSET has been implemented in C++ for 2D and 3D image sequences with competitive speed.
Robust and efficient pharmacokinetic parameter non-linear least squares estimation for dynamic contrast enhanced MRI of the prostate
2018, Magnetic Resonance Imaging
To describe an efficient numerical optimization technique using non-linear least squares to estimate perfusion parameters for the Tofts and extended Tofts models from dynamic contrast enhanced (DCE) MRI data and apply the technique to prostate cancer.
Parameters were estimated by fitting the two Tofts-based perfusion models to the acquired data via non-linear least squares. We apply Variable Projection (VP) to convert the fitting problem from a multi-dimensional to a one-dimensional line search to improve computational efficiency and robustness. Using simulation and DCE-MRI studies in twenty patients with suspected prostate cancer, the VP-based solver was compared against the traditional Levenberg-Marquardt (LM) strategy for accuracy, noise amplification, robustness to converge, and computation time.
The simulation demonstrated that VP and LM were both accurate in that the medians closely matched assumed values across typical signal to noise ratio (SNR) levels for both Tofts models. VP and LM showed similar noise sensitivity. Studies using the patient data showed that the VP method reliably converged and matched results from LM with approximate 3 × and 2 × reductions in computation time for the standard (two-parameter) and extended (three-parameter) Tofts models. While LM failed to converge in 14% of the patient data, VP converged in the ideal 100%.
The VP-based method for non-linear least squares estimation of perfusion parameters for prostate MRI is equivalent in accuracy and robustness to noise, while being more reliably (100%) convergent and computationally about 3 × (TM) and 2 × (ETM) faster than the LM-based method.

View all citing articles on Scopus

View full text

Dynamic contrast-enhanced case-control analysis in 3T MRI of prostate cancer can help to characterize tumor aggressiveness

Highlights

Abstract

Purpose

Materials and methods

Results

Conclusion

Introduction

Section snippets

Subjects

Comparison between peripheral healthy and tumor ROIs

Quantitative analysis

Conclusions

Conflict of interest

Funding

Diagn. Interv. Imaging

Eur. Urol.

Magn. Reson. Imaging

Eur. Urol.

Magn. Reson. Imaging

Acad. Radiol.

Magn. Reson. Imaging

European Society of Urogenital Radiology, ESUR prostate MR guidelines 2012

Eur. Radiol.

Estimating kinetic parameters from dynamic contrast-enhanced T(1)-weighted MRI of a diffusable tracer: standardized quantities and symbols

J. Magn. Reson. Imaging

Dynamic contrast-enhanced MRI and pharmacokinetic models in prostate cancer

Eur. Radiol.

Prostate cancer localization with dynamic contrast-enhanced MR imaging and proton MR spectroscopic imaging

Radiology

Advancements in MR imaging of the prostate: from diagnosis to interventions

Radiographics

Dynamic contrast-enhanced MRI of prostate cancer at 3T: a study of pharmacokinetic parameters

AJR Am. J. Roentgenol.

Simultaneous quantification of perfusion and permeability in the prostate using dynamic contrast-enhanced MRI with and inversion-prepared dual-contrast sequence

Ann. Biomed. Eng.

Dynamic contrast-enhanced MRI for prostate cancer localization

Br. J. Radiol.