MRI Brain Perfusion and Arterial Spin Labeling (ASL)
The MR Perfusion & Arterial Spin Labeling web based training focuses on perfusion dynamic susceptibility contrast (DSC) and arterial spin labeling 2D and 3D techniques.
Welcome to the web-based training for Perfusion Dynamic Susceptibility Contrast and Arterial Spin Labeling. Our focus in module one is Perfusion Dynamic Susceptibility Contrast (DSC). MR Perfusion -- Dynamic Susceptibility Contrast (DSC) Diagnostic & research tool Evaluate regional variations in cerebral microvasculature MR Perfusion Techniques Dynamic Susceptibility Contrast or (DSC) Dynamic Contrast Enhanced Imaging (DCE) Arterial Spin Labeling (A S L) In this course, we covered the following concepts: MRI Perfusion (DSC) Arterial Spin Labeling (ASL) Dynamic Susceptibility Contrast Perfusion Recommendations for patient set up and the bolus injection Inline Perfusion Maps Perfusion parameter cards Arterial Input Function (AIF) and the two techniques (Global and Local) Define Arterial Spin Labeling (ASL) Principle and application for 2D & 3D ASL ASL Parameter card 2D and 3D ASL image results Advantages and applications of 2D & 3D ASL You must pass the final assessment with a minimum score of 80% to complete this course successfully. When you have completed this section, you will be able to: Define Dynamic Susceptibility Contrast Perfusion List the recommendations for patient set up and the bolus injection Identify the inline Perfusion Maps Define the Perfusion parameter cards Explain Arterial Input Function (AIF) & the two techniques (Global & Local) Dynamic Imaging Dynamic measurements ≥ 50 measurement Observe signal drop during first pass of contrast injection Perfusion Sequence ep2d_fid FID – Single Shot Echo Planar Imaging Readout T2* (susceptibility) weighted images Contrast agent bolus Changes in tissue T2* following the injection of a paramagnetic MR contrast agent are mainly due to the de-phasing of the extracellular spins in the spatially non-uniform field created by the magnetic susceptibility differences between vascular and extravascular compartments. This susceptibility difference causes magnetic field distortions in the vicinity of blood vessels, resulting in a decrease of T2* in the extracellular compartment. In a T2* weighted image the decrease of T2* is visible as an attenuation of signal intensity Power injector recommended to infuse contrast media 18 to 20 gauge needle Ante-cubital Contrast agent dose in ml per Radiologist (e.g., Gd-DTPA) Rate of contrast injection per Radiologist discretion 20 ml saline flush following contrast injection Perfusion images – same number of slices, position, thickness and gap as routine images acquired. Phase encoding direction Anterior to Posterior (A>P) 50 or > measurements Start scan Wait 7-10 seconds Baseline established for more accurate perfusion evaluation Start bolus injection – follow Radiologists recommendations Inline Perfusion maps are calculated Inline Perfusion Maps – Grey Scale GBP – Global Bolus Plot PBP – Percentage of Baseline at Peak TTP – Time to Peak Map relCBV – relative Cerebral Blood Volume Map relCBF – relative Cerebral Blood Flow Map relMTT – relative Mean Transit Time Map relCBVCorr – correct T1 relCBV Measurements: Number of times the sequence is repeated within a dynamic measurement at regular intervals ART – Advanced Retrospective Technique Removes patient motion artifacts Rigid body model motion correction Datasets are acquired and retrospectively realigned in 3D using k-space interpolation On – Patient motions in image plane corrected Off – Motion correction not performed Low Pass Filter for smoothing images Increases signal-to-nose Decreased spatial resolution Spatial Filter > Filter Width Filter width is applied in the image domain Range 0.1 – 20.0 mm Filter Width 1.No Filtering 2.Weak (2.0) 3.Medium (1.0) 4.Strong (0.5) Global time intensity curve Inline Map – automatically calculated Assess bolus passage that occurs over time Displays time course of average signal in a central slice ≥ 15 measurements required Validates bolus passage Used to monitor time course of contrast uptake by (mainly) normal tissue for quality assurance purposes Signal is plotted as a percentage of the maximum value across all time points Dip in curve not seen (less than 30%) → Poor results Average amount of time it takes for the contrast media (or) blood to pass through the voxel vasculature Unit – seconds (s) Volume of blood per unit of time passing through a given region of brain tissue Unit – ml (of blood) per minute per 100 gram (of tissue) Color map post-processed in Perfusion Task Card Total volume taken up by the capillary bed within a voxel, based on the mass of tissue supplied Volume of blood in a voxel divided by the mass of the voxel Units of ml (of blood) per 100 gram (of tissue) Color map post-processed in Perfusion Task Card Inline T1 Corrected relative CBV map reconstruction Based on the local AIF method Blood Brain Barrier breakdown – extravasation of contrast agent yields T1 effects and misestimating the relCBV Local AIF calculation with T1 correction Result – Calculated corrected CBV map Arbitrary units Most important hemodynamic maps Cerebral Blood Volume (CBV) Cerebral Blood Flow (CBF) Mean Transit Time (MTT) Offline perfusion evaluation Inline perfusion evaluation Hemodynamic maps not quantifiable Time to peak (TTP) Percentage of Baseline at Peak (PBP) One slice across time Optimal Bolus PBP Map TTP Map GBP Map PBP Map TTP Map Converting signal to concentration Signal – peak of curve Concentration – area beneath curve Signal intensity of an image not only TR, TE, T1, T2 susceptibility Complicated by Cross-talk Ghosting Noise Artifacts Distortions Blurring Everything making up contrast & image quality AIF – required for calculation of Perfusion Maps Two techniques available: Local AIF & Global AIF Global AIF – Manual selection Offline Calculation in the Perfusion Task Card Local AIF – Automatic Offline Calculation in the Perfusion Task Card Inline Calculation @ scanner Global AIF -- Manual Selection Determines incoming flow of contrast agent for a reference volume around every voxel Calculates an AIF for each voxel in the volume (3cm x 3 cm x 3cm) Minimized bolus delay and dispersion between the artery and the tissue of interest Avoids the manual selection of the arterial input function Supports the automated calculation of these inline and offline maps relCBF relCBV relMTT relCBVCorr Global (Manual) AIF post-processing Local (Automated) AIF post-processing Now that you have completed the MR Perfusion DSC content, you should be able to: Define Dynamic Susceptibility Contrast Perfusion List the recommendations for patient set up and the bolus injection Identify the inline Perfusion Maps Define the Perfusion parameter cards Explain Arterial Input Function (AIF) and the two techniques (Global and Local) When you have completed this section, you will be able to: Define Arterial Spin Labeling (ASL). Describe the principle and application for 2D and 3D ASL. Define the ASL Parameter card. Discuss 2D and 3D data acquired. Explain the advantages and applications of 2D and 3D ASL. Explain the advantages and applications of 3D ASL Method to measure Cerebral Blood Flow (Perfusion) Endogenous arterial blood water is magnetically labeled instead of exogenously administered tracer, and the magnetic label decays with T1 Longitudinal magnetization of arterial blood water must be manipulated (labeling) so it differs from tissue magnetization ASL key parameters required include assumptions about T1 values in blood and tissue Labeling efficiency Arterial transit time ASL methods particularly benefit from high magnetic field strengths due to Higher sensitivity T1 lengthens Allows more of the label to accumulate Inversion Array Size 3D PACE –– Volume based Prospective Motion Correction Real-time matching of patient and MR system coordinates Motion is detected and position is updated in real-time before next acquisition Motion correction via PACE is performed during measurement – Means that it is also retained in the uncorrected series (Control Series) Advantages of prospective correction Removes orientation dependency of ghost and distortions Ensures accurate spin history and slice selection Perfusion mode – default selection cannot be modified PICORE – label scheme Q2 TIPS – definition of bolus Results Better inversion profile at distal edges Better accuracy – improved perfusion measurements ASL Schematic ASL Schematic Base ImageHotspotsText BlocksImage FileDefinition of bolus Periodic saturation pulses – Periodic train of thin-slice saturation pulses at distal end of the tagged region (i.e., train of thin slice saturation pulses) Thin-slice saturation pulses are less sensitive to B1 inhomogeneity Profiles of these thin-slice saturation pulses are sharper Label Scheme Quality Check Rejects images with excessive motion On Off On-Extended Note: See Quality Check Highlights pop-up for more information on On, Off, and On-Extended. Based on difference, “control-label” Average signal in the different images are processed Images without motion – Average intensity is very small In case of motion, label and control images are shifted and the different images have a much higher average intensity Images with an average intensity above a certain threshold are rejected ASL Tab – Quality Check Quality Check Highlights Base ImageHotspotsText BlocksImage FileOff: relCBF maps and PW images without image rejectionOn: relCBF maps and PW images with image rejection (default)On-Extended: relCBF and PW images with and without image rejection, difference maps Perfusion Mode – FAIR Q2TIPS FAIR – Flow-sensitive Alternating Inversion Recovery Control image – Pulse played out with the slice-selection gradient Inversion of spins within imaging slab Spins outside slab left unaffected Label image – Without an accompanying slice selection gradient, inversion pulse affects spins in the entire volume Slice-selection gradient – Played with zero amplitude or at a different time Works as a spoiler, helps cancel Eddy Current Compensation (ECC) effects between tagged and control image Oblique angle – Rotation Transverse to Coronal should not exceed 30 (T>C ≤ 30°) Labeling slab is locked to imaging slab Minimize TE 9 slices (determined by readout time) Keep total readout time of slices short Resolution 64 x 64 ASL = (1% of signal), large voxels maximize SNR iPAT – Possible but not recommended for rCBF maps iPAT on – No separate M0 scan – M0 scan is calculated from coil reference scans that result in a slightly decreased image quality, affects resulting rCBF maps Patient Browser 2D ASL – Advantages Better comparisons among different patients Better comparisons for same patient obtained on different days 3D ASL – Advantages Higher SNR Better Resolution Better Background Suppression Perfusion Weighted Images Left: 2D ASL; Right: 3D ASL Now that you have completed this section, you should be able to: Define Arterial Spin Labeling (ASL). Describe the principle and application for 2D and 3D ASL. Define the ASL Parameter Card. Discuss 2D and 3D data acquired. Explain the advantages and applications of 2D and 3D ASL. PBP quantifies the amount of signal loss at peak time relative to the baseline How much does the signal drop at the peak ≥ 15 measurements required Influenced by: Relaxivity of contrast agent Flow rate of injection Cardiac output Entry point of injection Dilution of contrast agent Color map post-processed in Perfusion Task Card Duration from arterial injection of contrast media to bolus peak ≥ 15 measurements required Influenced by Delay of contrast media injection relative to the start of the measurement Flow rate of injection Cardiac output Entry point of injection Color map post-processed in Perfusion Task Card Tag Image (or) Label Image Perturbation of magnetization induced by inversion pulse (labeling or tagging) Proximal to slice location After a time delay labeled arterial blood reaches the tissue capillary bed → Changes in MR signal occur Pulse sequence used to acquire image data at the slice location Control Image Same pulse sequence repeated without tagging Subtraction Two sequences subtracted inherently [Tag (Label) image and Control image] relCBF is estimated 2D ASL Principle 2D ASL Principle Quality Check Images Open the popup window to view images that result from a volunteer's head movement during the scan. Quality Check Images Quality Check Images Volunteer moves head during scan Quality Check Off: PWI without rejection of moved measurements Quality Check On: PWI with rejection of moved measurements Inversion Time Diagram Specifies the array of inversion times to be measured in the sequence Related to which blood will be labeled Inversion time should be long enough to transport bolus into the image slices Should be equal or larger than arterial transit time – Typical values: 1400-2000 ms Adjusting TI – Might be useful to consider Patient age, pathology, and expected blood flow rates If low blood flow rates are expected – Might be beneficial to extend Inversion Time If short arterial transit time is expected – TI can be lowered to increase SNR and improve image quality Pediatric Disclaimer - MR scanning has not been established as safe for imaging fetuses and infants less than two years of age. The responsible physician must evaluate the benefits of the MR examination compared to those of other imaging procedures. Bolus Duration Diagram Bolus Duration Diagram Bolus Duration Related to saturation pulses – Saturation pulses define the duration of the bolus independent of the Bolus Arrival Time Typical values 500 ms – 700 ms Less SNR can be expected with short Bolus Duration (e.g., <500 ms) Bolus Durations (e.g., 500 ms, can improve left to right symmetry of the perfusion signal in some cases Bolus Duration Diagram Bolus Duration Diagram Flow Limit Suppresses signal from large vessels with velocities greater than the defined value Flow compensation is recognized as a part of flow encoding gradients along the slice selection axis Played out during the imaging module between the 3 reference scans (phase correction scans for EPI sequence) and the echo train Flow limit is switched off when set to Maximum (100 cm/s) Recommended values to see some improvements in the image quality are 10-20 cm/s Shorter inversion times (1200-1500 ms), lower flow limits of 1-10 cm/s are required GRASE – GRAdient Spin Echo implemented as TGSE – Turbo Gradient Spin Echo Fast Spin Echo based on 3D volume sequence Combination of Spin Echo and Gradient Echoes with an RF refocusing pulse applied before every phase-encoding step in the slice direction After each refocusing pulse – k-space is sampled using an EPI-like Gradient Echo train (centered about the Spin Echo) Refocusing pulses are re-phased and the spin de-phasing is caused by static inhomogeneity in the B0 field Imaging areas affected by background gradients caused by susceptibility differences will benefit (air-tissue or bone-tissue interfaces 3D ASL Contrast Parameter Card ASL Tab 3D ASL Contrast Parameter Card ASL Tab Source Images and (Motion Corrected Images) M0, tag, control, tag, control….. Perfusion Weighted Images (PWI) Derived from relative Cerebral Blood Flow equation (relCBF) Relative Maps (abitrary units) Relative cerebral Blood Flow (relCBF maps) Derived from relative Cerebral Blood Flow equation (relCBF) Relative Maps (Arbitrary units) Anatomical mask is applied – Movements can be masked → always better to check motion artifacts in PWI 2D ASL Princple Control Tag 2D ASL Princple Control Tag FAIR Perfusion Mode for 3D ASL Arterial blood feeding tissue from both proximal and distal sides of the imaging slice is tagged Flow direction unknown or the feeding arteries have tortuous paths, FAIR reduces underestimation of perfusion Full brain coverage Better resolution Improved SNR Q2TIPS – definition of the bolus Periodic saturation pulses – periodic train of thin slice saturation pulses at distal end of tagged region Results Better inversion profile at distal edges Better accuracy – improved perfusion measurements Bolus Duration Related to saturation pulses Saturation pulses define the duration of the bolus independent of the Bolus Arrival Time Typical values 500 ms – 700 ms Less SNR can be expected with short Bolus Duration (e.g., <500 ms) Shorter Bolus Durations (e.g., 500 ms) Can improve left to right symmetry of the perfusion signal in some cases Pediatric Disclaimer: MR scanning has not been established as safe for imaging fetuses and infants less than two years of age. The responsible physician must evaluate the benefits of the MR examination compared to those of other imaging procedures. Inversion Time Specifies the array of inversion times measured in the sequence Related to labeled blood Inversion times should be long enough to transport bolus into the image slices Should be equal or larger than arterial transit time – Typical values: 1400-2000 ms Adjusting TI – Might be useful to consider Patient age, pathology, and expected blood flow rates If low blood flow rates are expected – Might be beneficial to extend Inversion Time If short arterial transit time is expected – TI can be lowered to increase SNR and improve image quality Pediatric Disclaimer: MR scanning has not been established as safe for imaging fetuses and infants less than two years of age. The responsible physician must evaluate the benefits of the MR examination compared to those of other imaging procedures. Averaging Mode (Default) Sets mode for distributing averages over the specified TI array Inversion Array Size Number of Inversion Times used Inversion Array Size > 1 – several “Inversion Times” used during acquisition Only first two “Inversion Times” can be selected All following TIs are spaced equidistantly in time Advantages Provides data for a better estimation of Bolus Arrival Time (BAT) Provides data for calculation of corrected perfusion maps Suppression Mode: enables background suppression of multiple tissues Gray and white matter ASSIST: Attenuating the Static Signal In Arterial Spin Tagging Signal from static brain water protons can be reduced by multiple-inversion background suppression approaches Two nonselective inversion pulses then invert water protons in the entire sample (background suppression) Timing of pulses is optimized to attenuate signal from Gray and White Matter Source Images Perfusion Weighted Images (PWI) Only if “Inversion Array Size” = 1 Derived from regional Cerebral Blood Flow equation (rCBF) Relative maps – Arbitrary Units