Flow Magnetic Resonance Research Articles

Magnetic resonance measurement of coronary blood flow

Magnetic resonance (MR) flow measurement in the coronary artery can be achieved with either a breath-hold acquisition or a respiration-triggered acquisition. MR measurements of cardiac output are significantly depressed during breath-holding at deep inspiration, but the advantage is that the breath-hold method requires less scan time. Blood flow in the coronary sinus reflects the global myocardial blood flow because it represents approximately 96% of the total myocardial blood flow of the left ventricle (LV). If blood flow in the coronary sinus is measured with phase-contrast cine magnetic resonance imaging (MRI) and LV myocardial mass is measured with cine MRI, both the total myocardial blood flow and the average coronary blood flow per gram of myocardial mass can be quantified. Coronary flow reserve with volumetric MR flow measurement is measured to be within 4.2-5.0-fold. The noninvasive MR measurement of coronary flow reserve has been shown to be useful in identifying the functional significance of stenoses in the left anterior descending artery. The sensitivity and specificity of MR coronary flow velocity reserve for identifying stenosis of 70% or greater in the left main or left anterior descending artery were 100% and 83%, respectively. The MR quantification of total coronary blood flow and coronary blood flow per gram of myocardial mass seems to be an ideal method for evaluating coronary hemodynamics and may be useful in evaluating endothelial dysfunction of the coronary circulation.

Renal disease: value of functional magnetic resonance imaging with flow and perfusion measurements.

To differentiate healthy kidneys from diseased kidneys, we propose a combined magnetic resonance (MR) examination that includes measurements of renal arterial blood flow and parenchymal perfusion. A total of 130 kidneys (patients/healthy volunteers: 83/47) were examined using renal artery MR flow measurements and renal parenchymal perfusion measurements, as well as contrast-enhanced MR angiography. Cine phase-contrast-flow measurements were performed using an ECG-gated fast low angle shot pulse sequence; perfusion was measured with an arterial spin labeling flow-sensitive alternating inversion recovery technique. Contrast-enhanced MR angiography was performed with a fast 3D gradient echo sequence in a single breath hold. For evaluation, kidneys were divided into groups based on nephrologic diagnosis of the patient. Recursive partitioning and Wilcoxon rank-sum tests were used to separate the different groups. Significant differences in mean renal artery flow and parenchymal perfusion were found in kidneys with renal artery stenosis as well as parenchymal disease as compared with healthy kidneys. Using a classification tree derived from the recursive partitioning, a specificity of 99% and sensitivity of 69% with a positive/negative predictive value of 97%/84% was achieved for the separation of healthy kidneys from kidneys with vascular, parenchymal or combined disease. The overall accuracy was 88%. The combination of cine PC flow measurements and MR perfusion measurements offers a comprehensive assessment of both renovascular and renoparenchymal disease and provide a noninvasive approach to differentiate between these kidneys and normal kidneys.

Arterial MR imaging phase-contrast flow measurement: improvements with varying velocity sensitivity during cardiac cycle.

To reduce noise in velocity images of magnetic resonance (MR) phase-contrast measurements, the authors implemented and evaluated a pulse sequence that enables automatic optimization of the velocity-encoding parameter V(enc) for individual heart phases in pulsatile flow on the basis of a rapid prescan. This sequence was prospectively evaluated by comparing velocity-to-noise ratios with those from a standard MR flow scan obtained in the carotid artery in eight volunteers. This sequence was shown to improve velocity-to-noise ratios by a factor of 2.0-6.0 in all but the systolic heart phase and was determined to be an effective technique for reducing noise in phase-contrast velocity measurements.

Vein graft function improvement after percutaneous intervention: evaluation with MR flow mapping.

To provide functional reference values in single and sequential vein grafts by using magnetic resonance (MR) flow mapping and to examine the effect of percutaneous intervention (PCI) on coronary artery bypass graft function. Fast MR flow mapping at baseline and during adenosine-induced stress was performed in 39 nonstenotic single vein grafts and 20 nonstenotic sequential vein grafts, as well as in 15 stenotic vein grafts before and 7.3 weeks +/- 1.5 after successful PCI. We evaluated the following parameters (in terms of mean values +/- SDs): average peak velocity (APV) at baseline, stress APV, and velocity reserve. Parameters in nonstenotic single and sequential vein grafts were compared by means of unpaired two-tailed Student t testing. To evaluate changes in velocities before and after PCI, a paired two-tailed Student t test was used. P <.05 was considered to indicate a statistically significant difference. Reference values in single vein grafts for baseline APV, stress APV, and velocity reserve were 8.6 cm/sec +/- 3.4, 20.2 cm/sec +/- 9.5, and 2.4 +/- 0.8, respectively. In sequential vein grafts, significantly higher values for baseline APV (12.2 cm/sec +/- 5.0) and stress APV (27.2 cm/sec +/- 10.6) but a similar velocity reserve (2.3 +/- 0.7) were found. Significant improvements were observed after PCI in baseline APV (before PCI: 9.2 cm/sec +/- 6.6; after PCI: 12.9 cm/sec +/- 7.9; P =.008) and stress APV (before PCI: 12.9 cm/sec +/- 6.3; after PCI: 27.1 cm/sec +/- 13.9; P <.001). No improvement in velocity reserve was observed. Significantly higher absolute velocity and flow values were observed in sequential versus single vein grafts, underscoring the need for separate functional reference values for different graft types. Graft function showed significant improvement after PCI to the point that it was restored or nearly restored to reference values.

A vascular stent as an active component for locally enhanced magnetic resonance imaging: initial in vivo imaging results after catheter-guided placement in rabbits.

A vascular stent constructed as a high frequency resonator improves the local signal-to-noise ratio at magnetic resonance (MR) imaging. After catheter placement and intravascular expansion, the stent can be used as an inductively coupled coil for MRI. The imaging properties of this balloon-expandable active MRI stent (AMRIS) were evaluated after x-ray fluoroscopy guided placement in the abdominal aorta of five rabbits using MR angiography (MRA) and flow measurements. The AMRIS was implanted in the abdominal aorta of five rabbits using a balloon catheter inserted through the common carotid artery. The rabbits were examined by MRA (3D fast low-angle shot) at 1.5 tesla before and after intravenous injection of an iron-oxide-based blood pool contrast medium (dose 50 micro mol Fe/kg) and flow measurements (ECG-triggered phase contrast cine gradient-echo sequence). Signal-to-noise ratios (SNR) were calculated and flow volume curves were generated. The in-stent increase in temperature was measured in vitro using a fiberoptic thermometry system. The SNR was 5.0 +/- 0.6 outside the stent and 23.2 +/- 14.1 within the stent ( < 0.0 5) in plain MRA, 19.5 +/- 5.0 outside and 30.7 +/- 8.2 within the stent ( < 0.05) in contrast enhanced MRA, and 5.8 +/- 1.6 and 13.9 +/- 5.9, respectively ( < 0.05) in the magnitude images of the flow measurements. Flow volume curves within and distal to the stent were comparable. The expandable active MRI stent produces local signal enhancement in MRA and MR flow measurements after catheter placement and thus may improve assessment of the stented vessel segment by MR imaging.

Correlations between local conversion and hydrodynamics in a 3-D fixed-bed esterification process: An MRI and lattice-Boltzmann study

In situ magnetic resonance (MR) visualisation techniques are used to study chemical conversion within a fixed-bed reactor. The reaction chosen for study is the esterification of acetic acid by methanol, catalysed by an acidic ion exchange catalyst. MR visualisation shows that fractional variations in steady-state conversion within slice sections of the bed, perpendicular to the direction of superficial flow, are of order 20%. The origin of this heterogeneity in conversion lies in the heterogeneity in the fluid flow field within the bed as is seen directly from MR flow visualisation. A lattice-Boltzmann code is validated for the prediction of the flow field within this fixed bed of porous resin packing. The lattice-Boltzmann predictions of the flow field are then used to explore any dependence of local conversion within the bed with locally averaged fluid velocity, over varying length-scales within the bed. These data report a first demonstration that MR visualisation techniques can be used to gain insight into the true rate-limiting processes influencing the performance of fixed-bed reactors.

Tetralogy of Fallot: postoperative delayed recovery of left ventricular stroke volume after physical exercise assessment with fast MR imaging.

In six asymptomatic patients with corrected tetralogy of Fallot and nine healthy control subjects, the authors assessed left ventricular (LV) function during recovery from supine bicycle exercise by performing fast magnetic resonance (MR) flow mapping in the ascending aorta. Abnormal recovery of LV function after exercise was observed in the patients. MR flow mapping allows assessment of cardiac recovery after exercise.

NMR study of the dissolution of laser-polarized xenon

NMR of laser-polarized xenon is used to probe the dissolution behaviour of the noble gas in different liquids. The dissolution and self-relaxation rates are extracted via a macroscopic model, and comparison of the decay rate of the xenon magnetization in deuterated and non-deuterated solvent pairs allows the determination of the pure dipole-dipole contribution to relaxation. A transient convective effect, tentatively assigned to the xenon concentration gradient, is observed and characterized by diffusion encoding MRI experiments. The flow of xenon penetrates inside the solvent near the walls of the NMR tube, the longitudinal images showing a “ ” shape, the transverse ones a “O” shape. This convection effect has implications for delivery conditions of laser-polarized xenon in continuous flow experiments and magnetic resonance imaging.

Blood Flow Modeling in Carotid Arteries with Computational Fluid Dynamics and MR Imaging

Rationale and Objectives The authors' goal was to develop a noninvasive method for detailed assessment of blood flow patterns from direct in vivo measurements of vessel anatomy and flow rates. Materials and Methods The authors developed a method to construct realistic patientspecific finite element models of blood flow in carotid arteries. Anatomic models are reconstructed from contrast material–enhanced magnetic resonance (MR)angiographic images with a tubular deformable model along each arterial branch. A surfacemerging algorithm is used to create a watertight model of the carotid bifurcation for subsequent finite element grid generation, and a fully implicit scheme is used to solve the incompressible NavierStokes equations on unstructured grids. Physiologic boundary conditions are derived from cine phasecontrast MR flow velocity measurements at two locations below and above the bifurcation. Vessel wall compliance is incorporated by means of fluid-solid interaction algorithms. Results The method was tested on imaging data from a healthy subject and a patient with mild stenosis. Finite element grids were successfully generated, and pulsatile blood flow calculations were performed. Computed and measured velocity profiles show good agreement. Flow patterns and wall shear stress distributions were visualized. Conclusions Patientspecific computational fluid dynamics modeling based on MR images can be performed robustly and efficiently. Preliminary validation studies in a physical flowthrough model suggest that the model is accurate. This method can be used to characterize blood flow patterns in healthy and diseased arteries and may eventually help physicians to supplement imaging-based diagnosis and predictand evaluate the outcome of interventional procedures.

Influence of blood-pool contrast media on MR imaging and flow measurements in the presence of pulmonary arterial stents in swine.

To compare the effects of various stents on magnetic resonance (MR) imaging flow volume measurements and to determine the value of a blood-pool MR imaging contrast medium in assessment of vascular stents. In 11 pigs, six nitinol stents (Memotherm), four platinum stents (NuMed), and one elgiloy stent (Wallstent) were placed in the main pulmonary artery under x-ray fluoroscopic guidance. MR imaging was performed 3 months after stent placement before and after injection of NC100150 contrast medium. Blood flow volumes were assessed with velocity-encoded cine MR imaging through and next to the stent. The signal-to-noise ratio and width of susceptibility artifacts of the stents also were determined. Measurements were analyzed with the paired Student t test and Bland-Altman test, where appropriate. Blood flow volumes measured through the nitinol and platinum stents disclosed no significant difference between velocity-encoded cine MR imaging measurements through and next to the stent. On cine MR images, small susceptibility artifacts were observed around the nitinol and platinum stents. Signal-to-noise ratio in the stent lumen was reduced in nitinol and platinum stents when compared with that next to the stent. The elgiloy stent produced severe susceptibility artifacts, making measurement of flow volumes impossible. NC100150 injection caused no significant effect on flow volume measurements. It improved the signal-to-noise ratio of the pulmonary arterial lumen outside and, to a lesser extent, inside the stent. Assessment of morphology and flow volumes through nitinol and platinum stents is feasible with MR imaging. Blood-pool contrast media provide persistent signal enhancement in the pulmonary artery and, to a lesser extent, in the lumina of nitinol and platinum stents.

The active magnetic resonance imaging stent (AMRIS): initial experimental in vivo results with locally amplified MR angiography and flow measurements.

Magnetic resonance (MR) is limited by artifacts in vessels after stenting. An active MR imaging stent (AMRIS) allows for artifact-free imaging with local improvement in signal-to-noise ratio (SNR). In a rabbit model, we evaluated the imaging properties by MR angiography (MRA) and flow measurements. The AMRIS was placed in the abdominal aorta of five rabbits. At 1.5 T, MRA (three-dimensional fast low-angle shot) was performed before and after intravenous injection of an iron oxide-based, blood-pool contrast medium (dose, 50 micromol Fe/kg), and flow measurements were performed (electrocardiographically triggered phase-contrast cine gradient-echo sequence). Mean SNRs were calculated and flow volume curves were generated. The SNR was 6.0 +/- 0.6 (outside the stent) versus 12.3 +/- 1.1 (inside the stent, P < 0.05) for plain MRA, 21.2 +/- 0.6 versus 40.6 +/- 5.2 (P < 0.05) for contrast-enhanced MRA, and 5.4 +/- 0.4 versus 13.7 +/- 2.1 (P < 0.05) for the magnitude images of flow measurements. Flow volume curves within and distal to the stent were comparable. By using the AMRIS as a vascular stent, the stented vessel segment can be examined with enhanced signal intensity on MRI.

Prolonged cardiac recovery from exercise in asymptomatic adults late after atrial correction of transposition of the great arteries: evaluation with magnetic resonance flow mapping

After atrial correction of transposition of the great arteries (TGA), dysfunction of the systemic right ventricle at rest and during exercise has been reported. Information on changes in systemic right ventricular function during recovery from exercise is lacking. This study evaluates cardiac recovery from supine exercise using magnetic resonance (MR) imaging in patients with asymptomatic TGA after atrial correction. Flow in the ascending aorta, representing stroke volume of the systemic ventricle, was assessed with MR flow mapping in 10 asymptomatic patients with atrially corrected TGA and in 12 controls at rest during exercise and an 8-minute recovery period. In response to exercise, the patients had a smaller increase in heart rate, stroke volume, and cardiac output than did controls. After exercise, no significant difference in halftime of heart rate recovery was observed (patients, 48 ± 7 seconds; controls, 39 ± 4 seconds [p >0.05]). In the patients, the time course of stroke volume recovery was significantly different (p <0.001). Stroke volume in the patients, as a percent difference from rest, remained significantly elevated, from 2.5 minutes (+16 ± 5% vs +7 ± 6%; p <0.05) to 8 minutes (+4 ± 7% vs −3 ± 5%; p <0.05) after exercise. Subsequently, cardiac output remained significantly elevated, from 4.5 minutes (+27 ± 13% vs +15 ± 11%; p <0.05) to 7 minutes (+22 ± 11% vs +12 ± 12%; p <0.05) after exercise. We conclude that heart rate recovery is within normal limits in patients with atrially corrected TGA. Furthermore, cardiac recovery from exercise, assessed with MR flow mapping, is prolonged in patients with asymptomatic TGA after atrial correction. Abnormal recovery may reflect dysfunction of the systemic right ventricle and an altered metabolic response to exercise.

Biophysical properties of the normal-sized aorta in patients with Marfan syndrome: evaluation with MR flow mapping.

To investigate the feasibility of magnetic resonance (MR) flow mapping in the assessment of aortic biophysical properties in patients with Marfan syndrome and to detect differences in biophysical properties in the normal-sized aorta distal to the aortic root between these patients and matched control subjects. Seventy-eight patients with Marfan syndrome with aortic root dilatation and 23 matched control subjects underwent MR flow mapping in four locations in the normal-sized aorta (1, ascending aorta; 2, thoracic descending aorta; 3, descending aorta at the level of the diaphragm; and 4, abdominal descending aorta). Distensibility at each location and flow wave velocity between locations were calculated. Compared with the control subjects, patients with Marfan syndrome had decreased aortic distensibility at three of the four locations (levels 1, 2, and 4; P <.05) and increased flow wave velocity between all locations (P <.05) in the aorta. In patients with Marfan syndrome, flow wave velocity was also significantly increased along the entire aortic tract beyond the aortic root (from level 1 to level 4). MR imaging reveals abnormal biophysical properties of the normal-sized aorta in patients with Marfan syndrome. Monitoring of these properties is relevant for evaluating disease progression and treatment options.

Improved MR flow mapping in coronary artery bypass grafts during adenosine-induced stress.

To validate a recently developed fast high-temporal-resolution magnetic resonance (MR) flow sequence and use it to assess coronary artery bypass graft function during pharmacologic stress. Aortic and internal mammary artery flow was measured in 11 healthy volunteers by using conventional cine gradient-echo imaging as a reference standard method and turbo-field echo-planar imaging (TFEPI). By using TFEPI, breath-hold flow mapping with a spatial and temporal resolution of 0.8 mm(2) and 23 msec, respectively, can be performed. This sequence was applied in 20 angiographically normal grafts, and total blood flow at rest and during adenosine infusion (140 microgram/kg/min) was measured. Good agreement in aortic and internal mammary artery flow values between conventional fast-field echo and TFEPI techniques was found. The mean bypass graft total flow (+/- SD), as assessed with TFEPI, increased from 30.8 mL/min +/- 13.5 to 76.7 mL/min +/- 36.5 (P <.05) to yield a flow reserve of 2.7. Furthermore, this sequence revealed a difference in total flow between single and sequential grafts at rest (25.4 mL/min vs 40.9 mL/min; P <.05) and during stress (65.2 mL/min vs 98.3 mL/min; P <.05). Breath-hold TFEPI provides fast accurate flow measurements with high temporal resolution and allows motion-compensated flow quantification in multiple coronary artery bypass grafts during one 6-minute adenosine infusion.

Correlation of hemodynamic impact and morphologic degree of renal artery stenosis in a canine model.

In a noninvasive comprehensive magnetic resonance (MR) examination, the morphologic degree of renal artery stenosis was correlated to corresponding changes in renal artery flow dynamics. Different degrees of stenosis were created with the use of a chronically implanted inflatable arterial cuff in seven dogs. For each degree of stenosis, an ultrafast three-dimensional gadolinium MR angiography with high spatial resolution was performed, followed by cardiac-gated MR flow measurements with high temporal resolution for determination of pulsatile flow profiles and mean flow. Flow was also measured by a chronically implanted flow probe. In three of the dogs, trans-stenotic pressure gradients (DeltaP) also were measured via implanted catheters. Five different degrees of stenosis could be differentiated in the MR angiograms (0%, 30%, 50%, 80%, >90%). The MR flow data agreed with the flow probe within +/-20%. Stenoses between 30 and 80% gradually reduced the early systolic peak (Max(1)) of the flow profile but only minimally affected the midsystolic peak (Max(2)) or mean flow. Stenoses of more than 90% significantly depressed mean flow by more than 50%. The ratio between Max(1) and Max(2) (Rmax(1/2)) gradually fell with the degree of stenosis. The onset of significant mean flow reduction and DeltaP was indicated by a drop of Rmax(1/2) below 1 to 1.2. Thus, the analysis of high-resolution flow profiles allows detection of early hemodynamic changes even at degrees of stenoses not associated with a reduction of mean flow. Rmax(1/2) allows differentiation of the grade of hemodynamic compromise for a given morphologic stenosis independent of mean flow in a single comprehensive MR examination.

Endoscopic Third Ventriculocisternostomies In The Infant: Pre- & Post-Operative Magnetic Resonance Imaging EvaluationElective Project: Undergraduate Prize 2000

PURPOSE: To determine whether it is possible to select patients with obstructive hydrocephalus, in the under 1 age group for endoscopic third ventriculocisternostomy (ETV) using pre-operative T2 weighted turbo spin echo (T2W-TSE) sagittal sequence Magnetic Resonance (MR) imaging; and to assess ventriculocisternostomy patency using post-operative T2W-TSE MR.PATIENTS AND METHODS: A retrospective review of MR examinations and clinical notes of 11 patients under 1 year of age who had ETV, was performed. The post-operative flow MR images were divided into the presence or absence of flow-related signal changes.RESULTS: In 6 of the 11 patients, ETV was successful (54.5%) i.e. no VP shunt or revision of the ETV was required. 9 patients had post-operative T2W-TSE MR examinations — 8 of these 9 MR studies correlated to the clinical situation (89%). The remaining MR examination showed a CSF flow void but the ETV failed at 3 weeks.CONCLUSION: Pre-operative MR using T2W-TSE to select suitable candidates for ETV improves the success rate from < 40% to 54.5%. Post-operatively MRI is a good predictor of whether the ETV has been successful or not.Sharman, A. (2000). Clinical Radiology55, 989.

Coronary sinus flow measurement by means of velocity-encoded cine MR imaging: validation by using flow probes in dogs.

To validate coronary sinus flow measurements for quantification of global left ventricular (LV) perfusion by means of velocity-encoded cine (VEC) magnetic resonance (MR) imaging and flow probes. Measurements of coronary sinus flow were performed in seven dogs by using VEC MR imaging at baseline, single coronary arterial stenosis, dipyridamole stress, and reactive hyperemia. These measurements were compared with flow probe measurements of coronary blood flow (CBF) in the left anterior descending coronary (LAD) and circumflex (CFX) arteries (CBF(LAD+CFX)) and coronary sinus. LV blood perfusion was calculated in milliliters per minute per gram from coronary sinus flow, and LV mass was obtained by using VEC and cine MR imaging. LV mass was validated at autopsy. CBF(LAD+CFX) and coronary sinus flow at VEC MR imaging showed close correlation (r = 0.98, P: <.001). The difference between CBF(LAD+CFX) and MR coronary sinus flow was 3.1 mL/min +/- 8.5 (SD). LV mass at cine MR imaging was not significantly different from that at autopsy (73.2 g +/- 12.8 vs 69. 4 g +/- 12.8). At baseline, myocardial perfusion was 0.40 mL/min/g +/- 0.09 at VEC MR imaging, and CBF(LAD+CFX) was 0.44 mL/min/g +/- 0. 08 (not significant). Reactive hyperemia resulted in 2.7- and 2. 3-fold increases in coronary sinus flow at VEC MR imaging and flow probe CBF(LAD+CFX), respectively. VEC MR imaging has the potential to measure coronary sinus flow during different physiologic conditions and can serve as a noninvasive modality to quantify global LV perfusion in patients.

Combined diagnostic imaging of uterine carcinosarcoma: A case report.

The diagnostic imaging characteristics of uterine carcinosarcoma have not been established because the tumor is very rare. We studied magnetic resonance (MR) imaging, power Doppler ultrasonography, and positron emission tomography (PET) in one such carcinosarcoma to define its imaging characteristics. On T1-weighted MR images, the tumor mass in the uterus was low-signal intensity and had a slight inhomogeneous appearance. On T2-weighted images, it had a heterogeneous appearance and was of medium- or high-signal intensity; marked flow void was apparent from the myometrium to the tumor mass. The MR flow void was also detected by power Doppler ultrasonography. Increased glucose metabolism was detected on PET images, suggesting a malignant tumor in the endometrial cavity. Combined diagnostic images with MR, power Doppler ultrasonography, and PET is recommended as the diagnostic method for uterine carcinosarcoma.

Reconstruction of blood flow patterns in a human carotid bifurcation: a combined CFD and MRI study.

The carotid bifurcation is a common site for clinically significant atherosclerosis, and the development of this disease may be influenced by the local hemodynamic environment. It has been shown that vessel geometry and pulsatile flow conditions are the predominant factors that determine the detailed blood flow patterns at the carotid bifurcation. This study was initiated to quantify the velocity profiles and wall shear stress (WSS) distributions in an anatomically true model of the human carotid bifurcation using data acquired from magnetic resonance (MR) imaging scans of an individual subject. A numerical simulation approach combining the image processing and computational fluid dynamics (CFD) techniques was developed. Individual vascular anatomy and pulsatile flow conditions were all incorporated into the computer model. It was found that the geometry of the carotid bifurcation was highly complex, involving helical curvature and out-of-plane branching. These geometrical features resulted in patterns of flow and wall shear stress significantly different from those found in simplified planar carotid bifurcation models. Comparisons between the predicted flow patterns and MR measurement demonstrated good quantitative agreement.

Transient streamlines: texture synthesis for in vivo flow visualisation.

Magnetic resonance (MR) imaging is a versatile technique for providing detailed information on blood vessel morphology and function. With its ability to acquire multi-dimensional cine flow data, MR is also an important tool for providing insight into blood flow patterns in vivo. The purpose of this paper is to describe the application of texture synthesis for flow visualisation. Two related issues are addressed, one is the removal of image noise from the acquired velocity data to ensure a correct representation of the underlying flow structure, and the other is the formation of transient streamlines through flow texture synthesis. The process of noise removal is achieved by using a convex projection algorithm based on the principle of mass conservation, whereas transient streamlines are formed via an iterative orientated pattern formation and enhancement procedure. The method described provides realistic visualisation of the flow patterns and avoids distortions caused by integration errors associated with conventional streamline tracking techniques. Effectiveness of the method applied to MR flow data acquired in healthy volunteers and patients is demonstrated.