MR QA Flashcards
What is Health Technology Management?
Operates independently of vendors and departments
Procurement of appropriate equipment, technology and maintenance agreements
Includes:
* Acceptance testing and clinical user acceptance of what was agreed to be purchased
- Routine QC testing and annual performance testing
- Appropriate transport or disposal of equipment
Responsibilities of Radiation Practitioners (as per AHPRA)
- Daily QC and maintenance of imaging systems
- Identify and arrange equipment repairs when there is an equipment fault
- Clinical decision making whether equipment is still appropriate for use on patients or stop until equipment issue is resolved
- Following vendor specific instructions of operation and use of equipment
How to ensure Safety and Quality Management
- Medical physicists, biomedical engineers and imaging technicians provide supports in clinics
- Utilise established pathway to escalate and notify TGA of issues
- Investigation of equipment faults or failures, patient and staff safety, and other clinical sites
- Keep documented evidence
- Record equipment history: (acceptance, reports, servicing, repairs, testing)
- Requires MDT collaboration
Australian Standards and Guidelines for Practice
RANZCR :
* Standards of Practice for Clinical Radiology (2020)
- MRI Safety Guidelines (2017)
Australasian College of Physical Scientists and Engineers in Medicine (ACPESEM) (AUS equivalent of AAPM)
- AAPM Report 100
- ACR MRI Quality Control Manual
- IPEM Report 112 Quality Control and Artefacts in MRI
Checks required to assess MRI System Performance (MR QA)
- Setup and Table Position Accuracy
- Slice Position Accuracy
- Geometric Accuracy
- Slice Thickness Accuracy
- High Contrast Spatial Resolution
- Low Contrast Object Detectability
- Percent Image Uniformity
- Percent Signal Ghosting
- Magnetic Field Homogeneity
MRI QA: Set-up (General System Checks)
Assess table docking and movement
* Table Docking –> process of ensuring couch aligns and locks into desired place within MRI scanner
Assess RF Coil Integrity and connections
Assess room temperature
Assess laser operation
MRI QA: Table Position Accuracy Test
Performed using the Distance Accuracy Test Method
- Place a phantom with known dimensions on the table
- Drive the table to specific positions (10cm, 20cm and 30cm) along the longitudinal axis with the MR bore lasers on (these positions are predetermined)
- Compare object dimensions with the change in displayed table position
- Can be verified with a physical measuring device (e.g., ruler) to measure the distance travelled (in relation to the laser on the phantom)
MRI QA: Importance of Table Position Accuracy
Ensures accurate patient alignment
* Proper table positioning ensures that the patient or phantom is correctly aligned within the centre of the magnetic field
- Crucial for maintaining image quality
Allows for consistent measurements
Contributes to a reduction in Artifacts
MRI QA: Importance of Slice Position Accuracy
Helps verify that an MRI system can accurately position imaging slices relative to a selected region
Ensures that the targeted anatomical area is correctly imaged
Ensures consistency across differing MRI scan protocols
MRI QA: Slice Position Accuracy Test
Setup:
1. Start with quick, low resolution scan (localiser) to choose the central slice of the region you want to scan
- MRI system calculates exact location of the selected slice. Adjusts magnetic gradients (to fine tune position of slice within the phantom) and table position to ensure alignment relative to the isocentre
ACR Phantom:
* Placed on the table
* Internally has two equal length bars
Scan and Measure
* Take a central scan through the phantom
- If MRI slice is accurate, both bars should appear the same length
- Scan adjacent slices. The lengths of the bars in these slices may change slightly
Check for accuracy:
* Compare the bar lengths
* Uneven bars or larger differences indicate a problem with slice positioning
Direction of Error:
* Direction of the shorter bar in adjacent slices shows which way the slice position may be off
MR QA: Geometric Accuracy Definition
Measure of the difference between the actual spatial location of an object and its position in the MR image
In Theory:
* Static magnetic field is homogeneous
*Linearly varying gradient fields are used to encode spatial positions
In practice:
* Static field is often inhomogeneous, so manufacturers apply correction factors to improve accuracy
MR QA: Geometric Accuracy Test Method
- Use a phantom with a uniform grid or hole pattern, with known geometry and dimensions
- Scan the phantom across the MRI’s FOV (ensure that all parts of the phantom are included in the image)
* can evaluate geometric accuracy across the full extent of the FOV - Use a Spin Echo (SE) sequence without distortion correction to test spatial linearity and gradient calibration
Spatial Linearity: Definition
Refers to how accurately the MRI system maps spatial locations in the scanner to the corresponding positions in the MR image
Gradient Calibration: Definition
Ensures that the magnetic field gradients applied during an MRI scan are accurate
Tests whether the gradients vary in a linear and predictable manner over the scanner’s FOV
MR QA: Geometric Accuracy Test - Tolerance Criteria
< 2% error to perform treatment planning (AAPM Report 100)
For testing larger FOV or tighter tolerances, ACR phantom may not be suitable:
< 2mm error presence using ACR phantom
MR QA: Geometric Distortion (%GD) Equation
(%) GD = 100 x ((actual phantom dimensions) - (measured phantom dimensions on image)) / (measured phantom dimensions on image)
MR QA: Slice Thickness Accuracy Importance
Checks the performance of the RF and gradient subsystems
Affects spatial resolution, SNR, and minimal slice gaps
Can be affected by:
* RF excitation bandwidth and gradient field amplitude
MR QA: Slice Thickness Accuracy Test
- Use a phantom with crossed wedges (e.g., NEMA)
* Wedges have a slope related to slice thickness - Choose an MRI protocol with a known slice thickness (e.g., 5mm)
* Ensure no slice thickness modifications are applied for an accurate test - Scan the phantom capture an image of the wedges
* Bars in the image will represent the axial slice thickness - Measure Bar Lengths
* Lengths should correspond to the expected slice thickness
* Signal profile across the bar can be used to assess the thickness - Evaluate the results
* If slice thickness was incorrectly set, bar lengths will vary from the expected size
MR QA: Slice Thickness Accuracy - NEMA Phantom Tolerance
Applies for slice thickness of 5mm slices or more:
<10% deviation from intended thickness (NEMA) (-1mm to +1mm margin of error in the actual slice thickness)
MR QA: High Contrast Spatial Resolution Accuracy: Definition
Ability to distinguish between two nearby objects with minimal noise
Depends on the acquisition matrix size (which determines pixel size)
Influenced by image processing and the resolution of the display monitor used to view the images
MR QA: High Contrast Spatial Resolution Accuracy: Test Method
- Use a phantom with high-contrast objects of varying sizes (e.g., small circles, lines, or patterns designed for resolution testing)
- Choose an appropriate FOV for the scan (e.g., 250 mm)
- Set the acquisition matric (e.g., 256 x 256) to define the resolution and pixel size
- Conduct the MRI scan using a standard high-contrast sequence to capture the image of the phantom
- Ensure no distortion correction methods are applied, to accurately measure the spatial resolution
- Visually inspect the smallest resolvable objects in the phantom (e.g., 1mm wide objects spaced 1mm)
- Compare the image with known phantom features to determine if the system can resolve objects at the expected resolution
- Ensure the display monitor’s resolution matches or exceeds the acquisition resolution to prevent loss of perceived spatial resolution
MR QA: High Contrast Spatial Resolution Accuracy Test: Tolerances
Must be able to distinguish between objects that are at lease on pixel width in size, and separated by at least one pixel width
Modulation Transfer Function (MTF): Definition
Measure of an imaging systems ability to reproduce fine details from the object being imaged
Quantifies how well different spatial frequencies (fine vs coarse) details are captured in the image
MTF of 1 = Perfect Resolution (object detail is perfectly preserved)
MTF close to 0 = Poor Resolution (detail is lost, and objects blur together)
Often used as an objective test for high-contrast spatial resolution in imaging systems
MR QA: High Contrast Spatial Resolution: MTF Test Method
- Capture an image of a high-contrast phantom
- Measure Input and Output
* Input = Known Contrast of Phantom
* Output = Measured Contrast in image - Calculate Contrast
- Calculate MTF
- Plot MTF curve
* MTF values vs spatial frequency (detail size)
MR QA: Low Contrast Object Detectability: Definition
Ability to distinguish between objects that have low contrast and are surrounded by significant noise
Dependent on field strength and scan protocol used
* Higher field strength = improved detectability
MR QA: Low Contrast Object Detectability: Test Method
- Place an ACR phantom with 30 low-contrast circles of varying sizes inside the MRI scanner
- Use scan protocol that tests low-contrast resolution (often T1-weighted sequence)
- Perform an MRI scan
- Visually inspect the image to determine how many of the low-contrast circles are detectable
- Compare the results with known standards
MR QA: Low Contrast Object Detectability: Importance
Essential for detecting subtle differences in soft tissue (e.g., distinguishing between normal and abnormal tissue)
Important for imaging structures with low inherent contrast (e.g., brain, liver and muscles)
Ensures SNR is sufficient for identifying small, low-contrast objects
MR QA: Image Intensity Uniformity: Definition
Regions with uniform tissue should show the same MRI signal intensity across the image
Affected by RF coil design, subsystem performance, and field inhomogeneity
MR QA: Image Intensity Uniformity: Test Method
- Place a homogenous phantom (e.g., water phantom) inside the MRI scanner
- Ensure phantom is centred in the FOV
(e.g., If testing head coil uniformity, centre phantom within coil) - Use a slice thickness and FOV that fully covers the phantom to assess the uniformity across the entire image
- After acquiring the image, identify two ROIs.
* One ROI in the area with the highest pixel intensity
* One ROI in the area with the lowest pixel intensity - Calculate the mean pixel intensity for each ROI
- Calculate Percent Image Uniformity
* PIU = (mean of lowest ROI / mean of highest ROI) x 100
MR QA: Image Intensity Uniformity: Importance
Ensures uniform signal across the entire image –> essential for accurate diagnosis
Assess the performance of RF coil to deliver a uniform signal
Evaluates the magnetic field homogeneity –> can impact the image quality if the field is not consistent across the scan area
MR QA: Image Intensity Uniformity: Test Method Tolerances
Water-filled Phantoms
* 1.5T MRI systems (PIU > 90%)
* 3T MRI systems (PIU 80-85%)
MR QA: Percent Signal Ghosting: Definition
Ghosting artifact occurs in the phase encoding direction
Can be caused by motion or system imperfections (e.g., hardware, coil design, or pulse sequence issues) –> leads to repetitive signals appearing as ‘ghosts’ in the image
MR QA: Percent Signal Ghosting: Ghosting Artifact: Definition
A ghosting artifact is a repetitive, blurred signal in an MRI image that appears as multiple copies of the object, typically along the phase encoding direction
Causes
* Patient Movement (e.g., breathing, heat beat)
* System imperfections
Appearance:
* usually appear as faint, repeating versions of the object
MR QA: Percent Signal Ghosting: Phase Encoding Direction: Definition
Phase encoding direction is one of the two directions (along with the frequency encoding direction) in MRI where spatial information is encoded during the scan
Typically the axis along which ghosting artifacts appear
MR QA: Percent Signal Ghosting: Test Method
- Place a homogenous phantom in the MRI scanner
- Acquire the image with a standard MRI sequence (e.g., spin-echo, or gradient-echo sequence)
- Choose three ROIs for analysis
* Background ROI (region away from the phantom, in the background with no object signal)
- Ghosting ROI (region where ghosting artifacts are visible)
- Phantom ROI (region within the phantom where the mean signal is uniform)
- Calculate PSG
* PSG = ((ghosting signal + background signal) / mean phantom signal) x 100 - Compare PSG value to the acceptable tolerance (<1 % of the mean phantom signal)
* Higher value = excessive ghosting
MR QA: Magnetic Field Homogeneity: Definition
Refers to the variations in the magnetic field across a defined area, typically within a Diameter Spherical Volume (DSV)
Affected by:
* MRI hardware
* Coil design
* Nearby ferromagnetic structures
Optimised by:
* Magnetic shimming (adjusting the magnetic field to make it as uniform as possible)
MR QA: Magnetic Field Homogeneity: Test Method
- Use a spherical phantom to measure homogeneity
- Acquire scan using two different bandwidths (e.g., low and high)
* Uniformity maps or bandwidth variations are used to highlight areas of magnetic field non-uniformity - Compare the change in diameter or distance of the affected areas between the two bandwidths
* Results are usually reported in parts per million (ppm) - Perform the test in multiple orientations (e.g., axial, sagittal, coronal) to ensure homogeneity across all axes
- Tolerance:
* Depends on the dimensions and features of the phantom used
MR QA: Magnetic Field Homogeneity: Importance
Ensures the MR system maintains a uniform magnetic field, which is critical for image quality and accurate spatial encoding
MR QA: Magnetic Field Homogeneity: Bandwidth
Refers to the range of frequencies (or spectrum) that the MRI system uses to encode the signal from the patient’s body
Determines the range of frequencies the system can process for each pixel or voxel in the image
Wider Bandwidth
* Reduces the likelihood of artifacts (like chemical shift) but can increase noise, lowering signal-to-noise (SNR)
Narrower Bandwidth
* Increases SNR but may lead to more artifacts, especially near interfaces between different tissues (like fat and water)
Changing Bandwidth
* Helps highlight variations in the magnetic field, since the degree of shift or distortion in the image can change depending on the bandwidth used
MR QA: Magnetic Field Homogeneity: Uniformity Maps
Visual representations that show how consistent or uniform the magnetic field or signal intensity is across the scanned area
Used to assess magnetic field homogeneity or image intensity uniformity
Checks required to assess MRI System Performance (MR SIM QA)
- Check for RF Coils (PIU, PSG, SNR)
- Accuracy of External Lasers and MRI Bore Lasers
- Geometric Distortions for RT Planning
- RT Positioning Equipment
- Patient-related Artefacts
- Standardised Display Monitors
MR SIM QA: RF Coil Tests: Importance
RF coils are critical for image quality in MRI –> affecting signal reception, image uniformity and SNR
Regular testing ensures coil integrity and performance over time
MRI SIM QA: RF Coils: Test Method
Vendor Specific
* Method of testing RF coils depends on the specific vendor’s design and instructions for routine maintenance
Phantom Usage
* Homogenous phantom is used for all tests
* Phantom is scanned using the coil configuration (e.g., head, body) for the desired clinical application (e.g., 2x 4 channel, flexible coils –> if to be QA’d for head and neck patient)
Key Checks Performed:
1. Image Uniformity
* Measured as PIU
* Compares pixel intensity across the image to ensure the RF coil provides consistent reception through the FOV
- Signal Ghosting
* Measured as PSG
* Evaluates how much ghosting artifacts appears in the image (caused by coil or system malfunctions) - SNR
* Assessed with NEMA method
MR QA SIM: RT Positioning Equipment Check
Assess the suitability of RT positioning equipment within an MR environment
MR Safe
* Contains no metal or conducting materials
MR Compatible
* May contain metal or conducting materials but have minimal impact on image quality and MR conditions for safe use
Vendor can assist the team in setting up the machine to ensure accurate testing and alignment with RT protocols
MR QA SIM: End to End Testing: Definition
Simulates clinical conditions as closely as possible, testing all steps of the RT process
Ensures reproducibility of the entire clinical workflow from image acquisition to treatment planning and delivery
MR QA SIM: End to End Testing: Examples
- Employ CT and MRI-compatible phantoms that simulate human anatomy
- After acquiring image, import them into the TPS
* Ensure images are accurately registered and can be used for dose planning and positioning
MR QA SIM: Display Monitor Standards
Primary Monitor
* Used for interpreting medical images
* Applies to diagnostic imaging and image-guided therapy
Secondary Monitor
* Used at operators console
* Used to adjust images before being sent to PACS
* Used in managing patient’s care by other clinical staff
* Do not need back light and luminance range as great as primary monitors
MR SIM QA: Display System Specifications and Performance
- Image Bit-Depth vs. Display:
Medical images are acquired at higher bit-depth than most display monitors can show, leading to potential loss of detail. - Image Compression:
Compressed image files can cause loss of information, potentially reducing image quality. - Display Resolution:
The display resolution limits the perceived spatial resolution of medical images. - Luminance Intensity:
The monitor’s backlight affects perceived contrast. Lower backlight intensity reduces visible contrast in images. - Reading Environment:
Factors like reflections and ambient lighting in the viewing environment can reduce perceived image quality. - Monitor Performance Over Time:
Backlight intensity and luminance calibration degrade over time, reducing the monitor’s performance. - Consistency in Follow-Up:
The monitor performance and luminance calibration should remain consistent when reviewing follow-up images of the same patient to ensure accurate comparison.
MR QA SIM: Definition and Intrinsic/Extrinsic Artifacts
- Definition:
Patient-related distortion occurs when signals are shifted from their original positions due to patient characteristics. - Intrinsic Factors:
* Magnetic susceptibility (variation in tissues’ magnetic properties).
* Chemical shift: signal differences between water and fat.
* Tissue boundaries (e.g., organs with different magnetic properties). - Extrinsic Factors:
* Metallic or conductive implants: create strong distortions in the magnetic field. - Patient Motion:
* Movement or breathing can cause signal ghosting and distortions.
MR QA SIM: Artifacts Effects on Imaging
- Signal Localization:
Distortion affects accurate delineation of structures like organs-at-risk (OARs). - Water and Fat Interface:
The interface between water and fat can be problematic; scan protocols can suppress fat signals to reduce distortion. - Image Quality:
If artifact overlap is too large, the image may be unacceptable for clinical use.
