Signal Intensity Time Courses Research Articles

Tissue response curve-shape analysis of dynamic glucose-enhanced and dynamic contrast-enhanced magnetic resonance imaging in patients with brain tumor.

Dynamic glucose-enhanced (DGE) MRI is used to study the signal intensity time course (tissue response curve) after D-glucose injection. D-glucose has potential as a biodegradable alternative or complement to gadolinium-based contrast agents, with DGE being comparable with dynamic contrast-enhanced (DCE) MRI. However, the tissue uptake kinetics as well as the detection methods of DGE differ from DCE MRI, and it is relevant to compare these techniques in terms of spatiotemporal enhancement patterns. This study aims to develop a DGE analysis method based on tissue response curve shapes, and to investigate whether DGE MRI provides similar or complementary information to DCE MRI. Eleven patients with suspected gliomas were studied. Tissue response curves were measured for DGE and DCE MRI at 7 T and the area under the curve (AUC) was assessed. Seven types of response curve shapes were postulated and subsequently identified by deep learning to create color-coded "curve maps" showing the spatial distribution of different curve types. DGE AUC values were significantly higher in lesions than in normal tissue (p < 0.007). Furthermore, the distribution of curve types differed between lesions and normal tissue for both DGE and DCE. The DGE and DCE response curves in a 6-min postinjection time interval were classified as the same curve type in 20% of the lesion voxels, which increased to 29% when a 12-min DGE time interval was considered. While both DGE and DCE tissue response curve-shape analysis enabled differentiation of lesions from normal brain tissue in humans, their enhancements were neither temporally identical nor confined entirely to the same regions. Curve maps can provide accessible and intuitive information about the shape of DGE response curves, which is expected to be useful in the continued work towards the interpretation of DGE uptake curves in terms of D-glucose delivery, transport, and metabolism.

Healthy Kidney Segmentation in the Dce-Mr Images Using a Convolutional Neural Network and Temporal Signal Characteristics.

Quantification of renal perfusion based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) requires determination of signal intensity time courses in the region of renal parenchyma. Thus, selection of voxels representing the kidney must be accomplished with special care and constitutes one of the major technical limitations which hampers wider usage of this technique as a standard clinical routine. Manual segmentation of renal compartments—even if performed by experts—is a common source of decreased repeatability and reproducibility. In this paper, we present a processing framework for the automatic kidney segmentation in DCE-MR images. The framework consists of two stages. Firstly, kidney masks are generated using a convolutional neural network. Then, mask voxels are classified to one of three regions—cortex, medulla, and pelvis–based on DCE-MRI signal intensity time courses. The proposed approach was evaluated on a cohort of 10 healthy volunteers who underwent the DCE-MRI examination. MRI scanning was repeated on two time events within a 10-day interval. For semantic segmentation task we employed a classic U-Net architecture, whereas experiments on voxel classification were performed using three alternative algorithms—support vector machines, logistic regression and extreme gradient boosting trees, among which SVM produced the most accurate results. Both segmentation and classification steps were accomplished by a series of models, each trained separately for a given subject using the data from other participants only. The mean achieved accuracy of the whole kidney segmentation was 94% in terms of IoU coefficient. Cortex, medulla and pelvis were segmented with IoU ranging from 90 to 93% depending on the tissue and body side. The results were also validated by comparing image-derived perfusion parameters with ground truth measurements of glomerular filtration rate (GFR). The repeatability of GFR calculation, as assessed by the coefficient of variation was determined at the level of 14.5 and 17.5% for the left and right kidney, respectively and it improved relative to manual segmentation. Reproduciblity, in turn, was evaluated by measuring agreement between image-derived and iohexol-based GFR values. The estimated absolute mean differences were equal to 9.4 and 12.9 mL/min/1.73 m2 for scanning sessions 1 and 2 and the proposed automated segmentation method. The result for session 2 was comparable with manual segmentation, whereas for session 1 reproducibility in the automatic pipeline was weaker.

Dynamic contrast-enhanced MRI in oncology: how we do it.

Magnetic resonance imaging (MRI) is particularly attractive for clinical application in perfusion imaging thanks to the absence of ionizing radiation and limited volumes of contrast agent (CA) necessary. Dynamic contrast-enhanced MRI (DCE-MRI) involves sequentially acquiring T1-weighted images through an organ of interest during the passage of a bolus administration of CA. It is a particularly flexible approach to perfusion imaging as the signal intensity time course allows not only rapid qualitative assessment, but also quantitative measures of intrinsic perfusion and permeability parameters. We examine aspects of the T1-weighted image series acquisition, CA administration, post-processing that constitute a DCE-MRI study in clinical practice, before considering some heuristics that may aid in interpreting the resulting contrast enhancement time series. While qualitative DCE-MRI has a well-established role in the diagnostic assessment of a range of tumours, and a central role in MR mammography, clinical use of quantitative DCE-MRI remains limited outside of clinical trials. The recent publication of proposals for standardized acquisition and analysis protocols for DCE-MRI by the Quantitative Imaging Biomarker Alliance may be an opportunity to consolidate and advance clinical practice.

Quantitative Mapping of Specific Ventilation in the Human Lung using Proton Magnetic Resonance Imaging and Oxygen as a Contrast Agent.

Specific ventilation imaging (SVI) is a functional magnetic resonance imaging technique capable of quantifying specific ventilation - the ratio of the fresh gas entering a lung region divided by the region's end-expiratory volume - in the human lung, using only inhaled oxygen as a contrast agent. Regional quantification of specific ventilation has the potential to help identify areas of pathologic lung function. Oxygen in solution in tissue shortens the tissue's longitudinal relaxation time (T1), and thus a change in tissue oxygenation can be detected as a change in T1-weighted signal with an inversion recovery acquired image. Following an abrupt change between two concentrations of inspired oxygen, the rate at which lung tissue within a voxel equilibrates to a new steady-state reflects the rate at which resident gas is being replaced by inhaled gas. This rate is determined by specific ventilation. To elicit this sudden change in oxygenation, subjects alternately breathe 20-breath blocks of air (21% oxygen) and 100% oxygen while in the MRI scanner. A stepwise change in inspired oxygen fraction is achieved through use of a custom three-dimensional (3D)-printed flow bypass system with a manual switch during a short end-expiratory breath hold. To detect the corresponding change in T1, a global inversion pulse followed by a single shot fast spin echo sequence was used to acquire two-dimensional T1-weighted images in a 1.5T MRI scanner, using an eight-element torso coil. Both single slice and multi-slice imaging are possible, with slightly different imaging parameters. Quantification of specific ventilation is achieved by correlating the time-course of signal intensity for each lung voxel with a library of simulated responses to the air/oxygen stimulus. SVI estimations of specific ventilation heterogeneity have been validated against multiple breath washout and proved to accurately determine the heterogeneity of the specific ventilation distribution.

Differentiation of High-grade Gliomas from Brain Metastases Using Tissue Similarity Maps (TSMs) Based Relative Cerebral Blood Volume Values

Background: Perfusion Weighted Imaging (PWI) has been used to differentiate the solitary metastases from high-grade glioma. However, the CBV value obtained from classical algorithm can be affected by a number of factors. Tissue Similarity Map (TSM) is a new algorithm depending solely on the signal intensity time course but does not require the use of Concentration Time Curve (CTC), which is different from the classical rCBV algorithm. Objective: The purpose of this study is to investigate the diagnostic utility of TSM based rCBV values in differentiating between high-grade gliomas and solitary metastases. Methods: The preoperative MR PWI studies of 30 patients with a solitary cerebral neoplasm were retrospectively analyzed. All PWI data were calculated for TSM algorithm with MATLAB 7.6 (Math Works, Natick, MA). The TSM based rCBV (rCBVTSM) were measured in the parenchyma of tumor and peritumoral regions. All specimens were histopathologically categorized as high-grade gliomas or metastases and were correlated with corresponding rCBVTSM values. Results: All PWI data could be post-processed by using the TSM post-processing software. The differences in the rCBVTSM-T between high-grade gliomas and metastases were not statistically significant (p=0.299). While the rCBVTSM-P of high-grade gliomas was higher than that of metastases (p=0.040). ROC curve analysis showed significant AUC of rCBVTSM-P for differentiating highgrade gliomas from metastases (p=0.020). The rCBVTSM-P threshold of 1.725 was found to be a significant cutoff value for high-grade glioma/metastases prediction with 0.619 sensitivity and 0.889 specificity. Conclusion: The rCBVTSM-P derived from MR PWI may be helpful in differentiating high-grade gliomas from solitary metastases. Keywords: TSM, perfusion imaging, cerebral blood volume, glioma, neoplasms metastasis, differentiation diagnosis.

High-sensitivity TMS/fMRI of the Human Motor Cortex Using a Dedicated Multichannel MR Coil

PurposeTo validate a novel setup for concurrent TMS/fMRI in the human motor cortex based on a dedicated, ultra-thin, multichannel receive MR coil positioned between scalp and TMS system providing greatly enhanced sensitivity compared to the standard birdcage coil setting. MethodsA combined TMS/fMRI design was applied over the primary motor cortex based on 1Hz stimulation with stimulation levels of 80%, 90%, 100%, and 110% of the individual active motor threshold, respectively. Due to the use of a multichannel receive coil we were able to use multiband-accelerated (MB=2) EPI sequences for the acquisition of functional images. Data were analysed with SPM12 and BOLD-weighted signal intensity time courses were extracted in each subject from two local maxima (individual functional finger tapping localiser, fixed MNI coordinate of the hand knob) next to the hand area of the primary motor cortex (M1) and from the global maximum. ResultsWe report excellent image quality without noticeable signal dropouts or image distortions. Parameter estimates in the three peak voxels showed monotonically ascending activation levels over increasing stimulation intensities. Across all subjects, mean BOLD signal changes for 80%, 90%, 100%, 110% of the individual active motor threshold were 0.43%, 0.63%, 1.01%, 2.01% next to the individual functional finger tapping maximum, 0.73%, 0.91%, 1.34%, 2.21% next to the MNI-defined hand knob and 0.88%, 1.09%, 1.65%, 2.77% for the global maximum, respectively. ConclusionOur results show that the new setup for concurrent TMS/fMRI experiments using a dedicated MR coil array allows for high-sensitivity fMRI particularly at the site of stimulation. Contrary to the standard birdcage approach, the results also demonstrate that the new coil can be successfully used for multiband-accelerated EPI acquisition. The gain in flexibility due to the new coil can be easily combined with neuronavigation within the MR scanner to allow for accurate targeting in TMS/fMRI experiments.

Diagnosis of Spinal Lesions Using Heuristic and Pharmacokinetic Parameters Measured by Dynamic Contrast-Enhanced MRI

This study aimed to evaluate the diagnostic performance of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in differentiation of four spinal lesions by using heuristic and pharmacokinetic parameters analyzed from DCE signal intensity time course. DCE-MRI of 62 subjects with confirmed myeloma (n = 9), metastatic cancer (n = 22), lymphoma (n = 7), and inflammatory tuberculosis (TB) (n = 24) in the spine were analyzed retrospectively. The region of interest was placed on strongly enhanced tissues. The DCE time course was categorized as the "wash-out," "plateau," or "persistent enhancement" pattern. The maximum enhancement, steepest wash-in enhancement, and wash-out slope using the signal intensity at 67 seconds after contrast injection as reference were measured. The Tofts 2-compartmental pharmacokinetic model was applied to obtain Ktrans and kep. Pearson correlation between heuristic and pharmacokinetic parameters was evaluated, and receiver operating characteristic curve analysis was performed for pairwise group differentiation. The mean wash-out slope was -22% ± 10% for myeloma, 1% ± 0.4% for metastatic cancer, 3% ± 3% for lymphoma, and 7% ± 10% for TB, and it could significantly distinguish myeloma from metastasis (area under the curve [AUC] = 0.884), lymphoma (AUC = 1.0), and TB (AUC = 1.0) with P = .001, and distinguish metastasis from TB (AUC = 0.741) with P = .005. The kep and wash-out slope were highly correlated (r = 0.92), and they showed a similar diagnostic performance. The Ktrans was significantly correlated with the maximum enhancement (r = 0.71) and the steepest wash-in enhancement (r = 0.85), but they had inferior diagnostic performance compared to the wash-out slope. DCE-MRI may provide additional diagnostic information, and a simple wash-out slope had the best diagnostic performance. The heuristic and pharmacokinetic parameters were highly correlated.

Feasibility of in vivo three-dimensional T 2* mapping using dicarboxy-PROXYL and CW-EPR-based single-point imaging.

The aim of this study was to demonstrate the feasibility of in vivo three-dimensional (3D) relaxation time T 2* mapping of a dicarboxy-PROXYL radical using continuous-wave electron paramagnetic resonance (CW-EPR) imaging. Isotopically substituted dicarboxy-PROXYL radicals, 3,4-dicarboxy-2,2,5,5-tetra(2H3)methylpyrrolidin-(3,4-2H2)-(1-15N)-1-oxyl (2H,15N-DCP) and 3,4-dicarboxy-2,2,5,5-tetra(2H3)methylpyrrolidin-(3,4-2H2)-1-oxyl (2H-DCP), were used in the study. A clonogenic cell survival assay was performed with the 2H-DCP radical using squamous cell carcinoma (SCC VII) cells. The time course of EPR signal intensities of intravenously injected 2H,15N-DCP and 2H-DCP radicals were determined in tumor-bearing hind legs of mice (C3H/HeJ, male, n=5). CW-EPR-based single-point imaging (SPI) was performed for 3D T 2* mapping. 2H-DCP radical did not exhibit cytotoxicity at concentrations below 10mM. The in vivo half-life of 2H,15N-DCP in tumor tissues was 24.7±2.9min (mean±standard deviation [SD], n=5). The in vivo time course of the EPR signal intensity of the 2H,15N-DCP radical showed a plateau of 10.2±1.2min (mean±SD) where the EPR signal intensity remained at more than 90% of the maximum intensity. During the plateau, in vivo 3D T 2* maps with 2H,15N-DCP were obtained from tumor-bearing hind legs, with a total acquisition time of 7.5min. EPR signals of 2H,15N-DCP persisted long enough after bolus intravenous injection to conduct in vivo 3D T 2* mapping with CW-EPR-based SPI.

Characterisation of Lesions after Stereotactic Radiosurgery for Brain Metastases: Impact of Delayed Contrast Magnetic Resonance Imaging

AimsTo investigate if brain metastases and radiation injuries after stereotactic radiosurgery (SRS) have different signal intensity (SI) time courses up to 55 min after contrast agent application and if delayed contrast magnetic resonance imaging (MRI) contributes to improve diagnostic accuracy. Materials and methodsThirty-four consecutive patients treated with SRS for cerebral metastases were prospectively enrolled in the study. T1-weighted images were acquired on a 3-Tesla MR unit at three time points, at 2 (TP1), 15 (TP2) and 55 (TP3) min after administering contrast agent. A simultaneous, matched-pairs approach was used for region of interest analysis of the entire contrast-enhancing lesion (SI-e), the centre (SI-c), the border of the lesion (SI-b) and the adjacent non-contrast-enhancing tissue (SI-p). SIs of brain metastases and radiation injuries after SRS were compared using a two-level, linear, mixed-effects regression model. ResultsIn total, 41 lesions were analysed: 16 metastases and 25 radiation injuries. The SI time course of SI-e, SI-c and SI-b proved to be significantly different for both entities (P < 0.001) from TP2 to TP3. The SI of 39/41 lesions increased from TP1 to TP2 for the three parameters. Radiation injuries showed a further signal increase at least for SI-c from TP2 to TP3, whereas for all the three parameters SI decreased in all metastases. ConclusionBrain metastases and radiation injuries after SRS have a characteristic and statistically significantly different SI time course on sequential gadolinium enhancement MRI when late MR studies are included.

Technical Pitfalls of Signal Truncation in Perfusion MRI of Glioblastoma.

Dynamic susceptibility contrast (DSC) perfusion-weighted imaging (PWI) is widely used in clinical settings for the radiological diagnosis of brain tumor. The signal change in brain tissue in gradient echo-based DSC PWI is much higher than in spin echo-based DSC PWI. Due to its exquisite sensitivity, gradient echo-based sequence is the preferred method for imaging of all tumors except those near the base of the skull. However, high sensitivity also comes with a dynamic range problem. It is not unusual for blood volume to increase in gene-mediated cytotoxic immunotherapy-treated glioblastoma patients. The increase of fractional blood volume sometimes saturates the MRI signal during first-pass contrast bolus arrival and presents signal truncation artifacts of various degrees in the tumor when a significant amount of blood exists in the image pixels. It presents a hidden challenge in PWI, as this signal floor can be either close to noise level or just above and can go no lower. This signal truncation in the signal intensity time course is a significant issue that deserves attention in DSC PWI. In this paper, we demonstrate that relative cerebral blood volume and relative cerebral blood flow (rCBF) are underestimated due to signal truncation in DSC perfusion, in glioblastoma patients. We propose the use of second-pass tissue residue function in rCBF calculation using least-absolute-deviation deconvolution to avoid the underestimation problem.

T1-weighted Dynamic Contrast-enhanced MR Imaging of the Lung in Asthma: Semiquantitative Analysis for the Assessment of Contrast Agent Kinetic Characteristics

To evaluate the contrast agent kinetics of dynamic contrast material-enhanced (DCE) magnetic resonance (MR) imaging in healthy lungs and asthmatic lungs by using non-model-based semiquantitative parameters and to explore the relationships with pulmonary function testing and eosinophil level. The study was approved by the National Research Ethical Committee (reference no. 11/NW/0387), and written informed consent was obtained from all individuals. Ten healthy subjects and 30 patients with asthma underwent pulmonary function tests, blood and sputum eosinophil counts, and 1.5-T DCE MR imaging within 7 days. Semiquantitative parameters of contrast agent kinetics were calculated from the relative signal intensity-time course curves on a pixel-by-pixel basis and were summarized by using whole-lung median values. The distribution heterogeneity was assessed by using the regional coefficient of variation. DCE MR imaging readouts were compared between groups by using one-way analysis of variance, and the relationships with pulmonary function testing and eosinophil counts were assessed by using Pearson correlation analysis. Asthmatic patients showed significantly lower peak enhancement (P < .001) and initial areas under the relative signal intensity curve in the first 60 seconds (P = .002) and significantly reduced late-phase washout slope (P = .002) when compared with healthy control subjects. The distribution heterogeneity of bolus arrival time (P = .029), time to peak (P = .008), upslope of the first-pass peak (P = .011), and late-phase washout slope (P = .032), estimated by using the median coefficient of variation, were significantly higher in asthmatic patients than in healthy control subjects. These imaging readouts also showed significant linear correlations with measurements of pulmonary function testing but not with eosinophil level in patients with asthma. The contrast agent kinetic characteristics of T1-weighted DCE MR images of asthmatic lungs are different from those of healthy lungs and are related to measurements of pulmonary function testing but not to eosinophil level.

Using Edge Voxel Information to Improve Motion Regression for rs-fMRI Connectivity Studies

Recent fMRI studies have outlined the critical impact of in-scanner head motion, particularly on estimates of functional connectivity. Common strategies to reduce the influence of motion include realignment as well as the inclusion of nuisance regressors, such as the 6 realignment parameters, their first derivatives, time-shifted versions of the realignment parameters, and the squared parameters. However, these regressors have limited success at noise reduction. We hypothesized that using nuisance regressors consisting of the principal components (PCs) of edge voxel time series would be better able to capture slice-specific and nonlinear signal changes, thus explaining more variance, improving data quality (i.e., lower DVARS and temporal SNR), and reducing the effect of motion on default-mode network connectivity. Functional MRI data from 22 healthy adult subjects were preprocessed using typical motion regression approaches as well as nuisance regression derived from edge voxel time courses. Results were evaluated in the presence and absence of both global signal regression and motion censoring. Nuisance regressors derived from signal intensity time courses at the edge of the brain significantly improved motion correction compared to using only the realignment parameters and their derivatives. Of the models tested, only the edge voxel regression models were able to eliminate significant differences in default-mode network connectivity between high- and low-motion subjects regardless of the use of global signal regression or censoring.

Spatiotemporal phase unwrapping for real-time phase-contrast flow MRI.

To develop and evaluate a practical phase unwrapping method for real-time phase-contrast flow MRI using temporal and spatial continuity. Real-time phase-contrast MRI of through-plane flow was performed using highly undersampled radial FLASH with phase-sensitive reconstructions by regularized nonlinear inversion. Experiments involved flow in a phantom and the human aorta (10 healthy subjects) with and without phase wrapping for velocity encodings of 100 cm·s(-1) and 200 cm·s(-1) . Phase unwrapping was performed for each individual cardiac cycle and restricted to a region of interest automatically propagated to all time frames. The algorithm exploited temporal continuity in forward and backward direction for all pixels with a "continuous" representation of blood throughout the entire cardiac cycle (inner vessel lumen). Phase inconsistencies were corrected by a comparison with values from direct spatial neighbors. The latter approach was also applied to pixels exhibiting a discontinuous signal intensity time course due to movement-induced spatial displacements (peripheral vessel zone). Phantom and human flow MRI data were successfully unwrapped. When halving the velocity encoding, the velocity-to-noise ratio (VNR) increased by a factor of two. The proposed phase unwrapping method for real-time flow MRI allows for measurements with reduced velocity encoding and increased VNR.

Improved Quantification of Dynamic Contrast Enhanced MR Imaging (DCE-MRI) in Adnexal Masses by Elastic Registration Method

Background: Ovarian cancer is the fifth commonest cancer affecting women in the developed world and is the primary indication for gynecological surgery. Quantitative dynamic contrast enhanced MRI (DCE-MRI) has been shown beneficial to differentiate malignant and benign tumors and is increasingly used as the investigational biomarker of response in clinical studies based on the measurement of enhancement characteristics. One of the major assumptions in quantification of DCE-MRI in abdominal organs is spatially-fixed region of interest over the time course of contrast agent passage. However, there are two types of motion occurring in the image series, which could invalidate this assumption and thus the quantification outcome: one of them from complex motion resulting from breathing, pulsation and the natural movement of the organ of interest, and the other one from the motion of the contrast agent. Thus, the accurate quantification of DCE-MR image series highly depends on minimization of motion artifacts. Objectives: To this end, proper registration of images that have been acquired in different time points is essential for deriving diagnostic information to produce a dataset without motion artifact. Here, we have proposed a registration approach for accurate quantification of DCE-MRI in ovary, employing elastic registration to account for spatially-varying intensity changes within the registration framework. Patients and Methods: Data Acquisition: DCE-MR images of eighteen patients (10 benign and 8 malignant tumors based on histopathology) diagnosed with solid or solid/cystic ovarian masses were acquired on a 3T MR scanner (Siemens MAGNETOM Tim TRIO) using a surface phased-array coil, TE/TR = 1.74/5 ms, flip angle = 60, image matrix = 156192, FOV=2323 cm2, slice thickness = 5 mm, number of measurements = 52 at 6 s/volume, number of slices = 16. The acquisition was performed before and immediately after injection of 0.2 mL/kg of Gadolinium (DOTAREM; Guerbet, Aulnay, France), followed by injection of 20 cc normal saline solution with 3 mL/min injection rate. : Image registration: The pre-contrast image is taken as the reference and the consequent images are aligned with the reference image. We have employed elastic registration algorithm developed by Periaswamy and Farid. The geometric transformation is a local affine model with a global smoothness constraint. Intensity variations are modeled with local changes in brightness and contrast. The mean squared error metric was applied to the intensity values to correct the nonlinear distortion. A least-squares technique was used to minimize the error function which is linear in model parameters. An iterative nonlinear minimization scheme is used to minimize the nonlinear error function. : Quantification: As proposed, time-to-peak (TTP) and wash-in-rate (WIR), defined as (SImax-SI0)/TTP, can be used to distinguish between benign and malignant ovarian masses. Results: The impact of the registration methods was shown on the selected regions-of-interest (ROIs) located in solid part of the tumor in malignant forms. As a result, the elastic registration algorithm significantly improves the signal intensity curves, especially in psoas that inhibited much distortion. In addition, the mean and the median of the standard deviation within the ROIs selected on tumor and psoas over the time courses of contrast agent passage were computed, which suggests that the elastic registration (E-Reg) method significantly improves the signal intensity-time courses in contrast to the unregistered (UnReg) images. Similar results are obtained for the other data sets. Also, quantitative parameters were calculated for unregistered and elastic registered images. It can be inferred that the value of mean to standard deviation ratio of the parameters increased after registration, which would improve the characterization of benign from malignant ovarian masses. Conclusions: From the results attained in this work, it can be concluded that the outcome of ovarian cancer characterization could benefit from elastic registration approach and hence this method can be reliably used for quantification of DCE-MR images of adnexal tumors.

Pharmacokinetic Approach for Dynamic Breast MRI to Indicate Signal Intensity Time Curves of Benign and Malignant Lesions by Using the Tumor Flow Residence Time

The objective of this study was to evaluate a novel pharmacokinetic approach integrating a tumor model in a whole-body pharmacokinetic model to simulate contrast media-induced signal intensity time curves of breast tumors on dynamic contrast-enhanced magnetic resonance mammography. A recently developed, whole-body pharmacokinetic model, which describes the distribution and excretion of renally discharged contrast media, has been expanded by integrating a tumor model. The parameters of the general approach including exchange between plasma and interstitium were set as fixed values; only 2 tumor-specific parameters, blood volume fraction Vblood and blood flow kt, were varied. These parameters were adjusted with regard to signal intensity time course data of histologically verified benign and malignant mass-like breast lesions on clinical magnetic resonance imaging examinations (1.5 T) using 2 different contrast media (gadopentetate dimeglumine and gadoterate meglumine) and 2 application doses (0.1 and 0.2 mmol kg body weight). Thus, measured signal intensity time curves were compared with simulated gadolinium (Gd) concentration time curves calculated by the pharmacokinetic model. Benign lesions showed continuous signal increase; malignant tumors presented fast initial signal increase followed by washout effect. According to the pharmacokinetic approach, the variation of the Vblood/kt ratio, which defined the tumor flow residence time τr, led to Gd concentration time curves congruent with the shapes of the measured signal intensity time curves. Low values of τr were characteristic for malignant tumors, and high values were typical for benign lesions; τr of 200 seconds best separated malignant from benign tumors. Thus, the dynamic magnetic resonance imaging data can be well approximated by the pharmacokinetic model considering 2 contrast media and application doses. The calculated Gd concentration time curves of 0.1 mmol kg body weight gadopentetate dimeglumine and gadoterate meglumine overlapped for benign lesions; the curve of gadoterate meglumine was by a factor of 0.8 below the curve of gadopentetate dimeglumine for malignant tumors. Doubling the application dose of gadopentetate dimeglumine from 0.1 to 0.2 mmol kg led to an increase in the Gd concentration time curves for benign lesions but not for malignant tumors. High Gd concentrations with values greater than 1 mmol L were calculated in the vessels of the malignant tumors, outside the determined range of the linear relationship between Gd concentration and signal intensity due to saturation effects. On the basis of this pharmacokinetic model, the contrast media-induced time curves on dynamic contrast-enhanced magnetic resonance mammography can be classified by a single kinetic parameter, the tumor flow residence time τr, into benign (τr >200 seconds) and malignant (τr <200 seconds) curve shapes. Possible clinical application of this model is to create pharmacokinetic maps, displaying tumor flow residence times, for computer-assisted diagnosis, which may be integrated into clinical routine for the diagnosis of breast lesions.

Comparison of dual-echo DSC-MRI- and DCE-MRI-derived contrast agent kinetic parameters

The application of dynamic susceptibility contrast (DSC) MRI methods to assess brain tumors is often confounded by the extravasation of contrast agent (CA). Disruption of the blood–brain barrier allows CA to leak out of the vasculature leading to additional T1, T2 and T2⁎ relaxation effects in the extravascular space, thereby affecting the signal intensity time course in a complex manner. The goal of this study was to validate a dual-echo DSC-MRI approach that separates and quantifies the T1 and T2⁎ contributions to the acquired signal and enables the estimation of the volume transfer constant, Ktrans, and the volume fraction of the extravascular extracellular space, ve. To test the validity of this approach, DSC-MRI- and dynamic contrast enhanced (DCE) MRI-derived Ktrans and ve estimates were spatially compared in both 9L and C6 rat brain tumor models. A high degree of correlation (concordance correlation coefficients >0.83, Pearson's r>0.84) and agreement was found between the DSC-MRI- and DCE-MRI-derived measurements. These results indicate that dual-echo DSC-MRI can be used to simultaneously extract reliable DCE-MRI kinetic parameters in brain tumors in addition to conventional blood volume and blood flow metrics.

Improving suspicious breast lesion characterization using semi‐automatic lesion fractional volume washout kinetic analysis

Although breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) demonstrates high sensitivity for malignant tumor detection, a major limitation is the relative low specificity, resulting in many false-positive diagnoses of suspicious lesions (BI-RADS assessment of 4 or 5) in clinical practice and consequently producing a relatively low positive predictive value (PPV) for biopsies. The most enhanced areas in the malignant tumors show a typical washout (WO) kinetic feature for the postcontrast signal intensity time courses and also correlate with microvessel density. Benign proliferative breast diseases can also produce the WO curve, yielding an equivocal kinetic behavior for the benign lesions and rendering their diagnoses as suspicious lesions in clinical practice. Considering that tumor angiogenesis is essential to an aggressive cancer tumor growth, the authors hypothesize that the WO volume fraction, i.e., the total volume of the WO voxels that demonstrate the WO curve within the tumor, is relatively large for malignant tumors in comparison to that for benign lesions. In this study, the authors present a lesion fractional volume WO kinetic analysis for improving the characterization of suspicious breast lesions. A method to automatically detect the boundary of a manually selected contrast-enhanced lesion was introduced and tested, utilizing the signal intensity difference between the contrast-enhanced lesion and its surrounding tissues. The kinetic features of the postcontrast signal intensity time courses were quantitatively analyzed voxel-by-voxel with emphasis on the examination of the WO behavior. The WO volume fraction relative to the whole lesion volume was introduced and tested as a biomarker for improving the characterization of suspicious breast lesions. The sample for this test consisted of 28 suspicious lesions with correlative histopathology reports available. The lesions included 10 malignant tumors and 18 benign lesions, yielding a 35.7% PPV of the biopsies. The semi-automatic method produced an objective volume of interest for each lesion with voxelwise-quantified kinetic features. With an optimal choice of kinetic analysis, the mean and standard deviation of the WO volume fraction were 59.1 ± 13.1 (%) with the range from 41.0% to 80.7% for the malignant tumors and 31.4 ± 20.5 (%) with the range between 3.3% and 71.6% for the benign lesions, respectively. The WO volume fraction was significantly larger (p < 0.0004) for the malignant tumors than for the benign lesions. While maintaining the same sensitivity for malignant tumors, using the WO volume fraction as an additional biomarker would characterize 14 out of the 18 benign lesions as benign, potentially resulting in an 100% improvement rate in the PPV of the biopsies (from 35.7% to 71.4%) and consequently a 77.8% reduction rate in potentially unnecessary biopsies (from 18 to 4). The significantly larger WO volume fraction for the malignant tumors was probably related to the increased vascularity associated with tumor angiogenesis. The results suggest that the WO volume fraction biomarker has potential to improve the computer-based assessment of breast MRI by greatly increasing the PPV of breast biopsies and potentially significantly reducing the number of unnecessary biopsies without compromising sensitivity.

Quantitative effects of using compressed sensing in dynamic contrast enhanced MRI

Dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) involves the acquisition of images before, during and after the injection of a contrast agent. In order to perform quantitative modeling on the resulting signal intensity time course, data must be acquired rapidly, which compromises spatial resolution, signal to noise and/or field of view. One approach that may allow for gains in temporal or spatial resolution or signal to noise of an individual image is to use compressed sensing (CS) MRI. In this study, we demonstrate the accuracy of extracted pharmacokinetic parameters from DCE-MRI data obtained as part of pre-clinical and clinical studies in which fully sampled acquisitions have been retrospectively undersampled by factors of 2, 3 and 4 in Fourier space and then reconstructed with CS. The mean voxel-level concordance correlation coefficient for Ktrans (i.e. the volume transfer constant) obtained from the 2× accelerated and the fully sampled data is 0.92 and 0.90 for mouse and human data, respectively; for 3×, the results are 0.79 and 0.79, respectively; for 4×, the results are 0.64 and 0.70, respectively. The mean error in the tumor mean Ktrans for the mouse and human data at 2× acceleration is 1.8% and −4.2%, respectively; at 3×, 3.6% and −10%, respectively; at 4×, 7.8% and −12%, respectively. These results suggest that CS combined with appropriate reduced acquisitions may be an effective approach to improving image quality in DCE-MRI.

FMRI-Based Hierarchical SVM Model for the Classification and Grading of Liver Fibrosis

We present a novel method for the automatic classification and grading of liver fibrosis based on hepatic hemodynamic changes measured noninvasively from functional MRI (fMRI) scans combined with hypercapnia and hyperoxia. The supervised learning method automatically creates a classification and grading model for liver fibrosis grade from training datasets. It constructs a statistical model of liver fibrosis by evaluating the signal intensity time course and local variance in T2(*)-W fMRI scans acquired during the breathing of air, air-carbon dioxide, and carbogen with a hierarchical multiclass binary-based support vector machine (SVM) classifier. Two experimental studies on 162 slices from 34 mice with the hierarchical multiclass binary-based SVM classifier yield 96.9% separation accuracy between healthy and histological-based fibrosis graded subjects, and an overall accuracy of 75.3% for healthy, fibrotic, and cirrhotic subjects. These results outperform existing image-based methods that can discriminate between healthy and mild-grade fibrosis subjects.

Impaired adenosine-induced myocardial perfusion in children with mild/moderate aortic stenosis using quantitative magnetic resonance imaging

Method Children with mild or moderate AS and no coronary stenoses (mean Doppler pressure gradient 20.3 ± 10.4 torr, n=11) were compared to a control group with hemodynamically normal hearts (“C”, pressure gradient 3.2 ± 1.5 torr, n=20). Mean age was 8.4 ± 4.4 years. For resting rest and hyperemia acquisitions the LV myocardium was divided into two slices (apical and basal) and each slice into six regions. The time course of average regional signal intensity during contrast transit was used to quantify absolute flow (ml/g/min) by deconvolution of the tissue curves with the arterial input of contrast. A linear mixed effects statistical model was employed to account for inter-subject variance (AS Vs C) and intrasubject variance (resting/adenosine, slice, and region), and also included interaction terms between AS and C, and treatment.