Program Information
Tolerance Levels and Trends in Ultrasound Transducer Depth of Penetration Performance: A Retrospective Study
C Brunnquell*, Z Li , J Zagzebski , University of Wisconsin, Madison, WI
Presentations
SU-F-708-5 (Sunday, July 30, 2017) 2:05 PM - 3:00 PM Room: 708
Purpose: The maximum depth of penetration (DOP) for visualizing background scatterers in a tissue phantom is used to assess system sensitivity in ultrasound QA. We aim to establish action levels for transducer DOP measurements based on past system tests and compare manual and automated measurements.
Methods: DOPs measured by four trained observers were extracted from yearly tests of 20 scanners over 4 years. Aggregated results were used to establish action levels. Manual measurements were compared to automated measurements in UltraIQ (termed “SNR” in UltraIQ, the depth where the signal in a phantom is equal to 1.4xnoise from an “air-image”). Temporal trends in transducer performance were evaluated.
Results: For 24 combinations of transducer type, frequency, and mode (fundamental/harmonic), DOP tolerances were established based on 4-year measurement distributions with average=μ and standard deviation=σ. The “watch” action level was set to the depth μ-σ and “consider replacement” set to μ-2σ. Of 558 DOP measurements, 58 were flagged; 20 flagged measurements were repeats within the same transducer over time. DOP consistency among transducers and settings varied (0.3cm<σ<0.8cm), supporting establishment of consistent measurement techniques with case-specific tolerance criteria. Automated measurements were significantly greater than manual estimates (p<0.001). Automatic analysis often failed on low-frequency transducers because the phantom far edge interfered with automated measurement. No significant trends in DOP performance were observed in individual transducers over time. The slope of DOP vs. time varied considerably at -0.036±0.558 cm/year, averaged over all transducers.
Conclusion: This study provides valuable baseline data for comparing systems and probes. It reveals that consistent techniques and case-specific tolerances can improve sensitivity to transducer DOP performance trends in clinical system tests. Automated software is in most cases an appropriate replacement for subjective DOP measurements, but should be incorporated with caution due to possible changes in action levels and applicability to specific transducer/phantom combinations.
Funding Support, Disclosures, and Conflict of Interest: An early version of UltraIQ software was provided by Cablon Medical B.V. for initial evaluation and feedback.
Contact Email: