ACR MR - Large Phantom

Image Quality Parameters Evaluated;
  • High contrast spatial resolution
  • Slice thickness accuracy
  • Slice position accuracy
  • Geometric distortion
  • Image intensity uniformity
  • Ghosting
  • Signal-to-noise ratio (SNR)
  • Low contrast resolution (automatic and visual detection)
  • Geometric distortion – Localizer

Output Available As;
  • HTML
  • PDF
  • NoSQL database
  • Export database query results to spreadsheet application
  • Image quality parameter trend analysis links embedded in HTML file
  • Export trend data to spreadsheet application
  • Export trend plot to graphics file
  • User enabled Pass/Fail evaluation
  • User defined Pass/Fail limits
  • Sample report in PDF format
AutoQA MAX MR ACR Trending
Application Settings
Application Settings
  • License supports multiple users (5)
  • Supports local and regional number and date formats
  • Chinese, Japanese and Spanish translation supported (other language translations to be added)
  • Shared Database location user defined
  • PDF report storage folder location user defined
  • Self-service ‘Off-line’ license activation supported, use mobile phone to scan QR code
  • Easy license deactivation if license is to be moved or used by another users
DICOM Image Selection
  • Create DICOM directory of local image folders or online storage locations
  • Intuitive image folder processing with Study and Series selection screens
  • Study and Series DICOM fields order can be re-arranged by drag and drop
Image Display Tools
  • DICOM image header dump
  • Report pixel value user the cursor
  • Thumbnail image display supports window center (w/c) controls
  • ‘Apply All’ w/c levels to series images
  • Series image statistics view
Image Processing
  • High Contrast Spatial Resolution
    • User input is required for this measurement. ‘High Contrast Spatial Resolution’ screen displays slice #1 with window/center (w/c) and magnification controls to support proper viewing for resolution determination.  User needs to assess the upper left (UL) and lower right (LR) resolution for the 1.1 mm, 1.0 mm, 0.9 mm sections, and identify the minimum hole size with a row (UL) or column (LR) of holes distinguishable from adjacent holes.  Final w/c will be saved and used for the BMP image stored in the final report.
  • Slice Thickness Accuracy-
    • Horizontal count density profiles are sampled and integrated for the top ramp (TR) and bottom ramp (BR) measurements.
    • Sample size is matrix dependent, with sampling width between approximately 2.5 – 3.1 mm’s.
    • The integrated profile is first processed with a 5 pt smoothing kernel. The peak pixel value is determined for the TR and BR profiles and the FWHM is calculated with a baseline correction.
    • Slice Thickness Accuracy reported is the average of the TR and BR FWHM measurements divided by 10.
    • Slice Offset – is the difference between the TR and BR ramp centers multiplied by the ramp angle magnification factor.
  • Slice Position Accuracy- the difference between the endpoints of left and right wedge profiles are computed and reported according to the instructions provided in the ACR Large Phantom Guidance documentation.
  • Geometric Accuracy – reported distance measurements from single pixel profile samples.
  • Image Uniformity Intensity- reported measurements calculated according to the instructions provided in the ACR Large Phantom Guidance documentation.
  • Ghosting- reported measurements calculated according to the instructions provided in the ACR Large Phantom Guidance documentation
  • Signal-to-Noise (SNR) – reported measurement uses the large mean ROI calculated from the Image Intensity Uniformity measurement and the same two background noise (SD) measurements from ghosting measurement.
  • Low Contrast Resolution –
    • Automated low contrast target detection is based on maximizing the target means using ROI diameters approximately equal to the target diameters.
    • Optimization methods make adjustments in centering and rotation angles of expected target locations.
    • Target visualization threshold is >1.0 SD of the target and background mean difference. The mean and standard deviation for background ROI’s were averaged for each target; ROI’s tangential to target for outer two rings and radial to the target for the inner ring.
    • Contrast to noise ratio (CNR) is calculated and reported for each Low Contrast image (#8- #11) using 3 targets found in spoke #1 ( 7mm targets).

Follow QA Benchmark for product information and updates on

© 2019 QA Benchmark.