Development of a focal lesion phantom with clinically relevant lesion characteristics for image quality evaluation of breast ultrasound scanners

Phys Med. 2022 Feb:94:65-74. doi: 10.1016/j.ejmp.2021.12.019. Epub 2022 Jan 5.

Abstract

Purpose: Contrast-detail (C-D) and anechoic-target (A-T) detectability are measures of an ultrasound scanner's ability to image lesions of varying contrast and size from background tissue and, as such, they are important tools for grading the imaging capabilities of ultrasound scanners. The objective of this study was to develop a range of contrast- and anechoic-detail phantoms with clinically relevant lesions, of various contrast and sizes, for performance testing of breast ultrasound equipment.

Methods: Tissue mimicking materials that represent the acoustic properties of breast fibroglandular tissue were produced and moulded to construct a range of C-D and A-T phantoms. Two phantom designs were produced, containing cylindrical and spherical targets. Both phantom types were constructed with contrast targets covering the range anechoic, -1, -2, -3 and -4 dB, with lesion diameters of 1-4 mm, positioned at four clinically relevant depths (10, 25, 40 and 55 mm). An image analysis program was developed to objectively analyse the lesion images and to determine the lesion-signal-noise-ratio (LSNR).

Results: Both phantoms were used to evaluate the performance of a breast ultrasound scanner. The use of cylindrical phantoms led to an artificially higher image quality performance compared with the more clinically relevant spherical lesion phantom, thus indicating the importance of using the appropriate targets in ultrasound phantoms.

Conclusion: The spherical lesion phantoms, coupled with the quantitative metric of LSNR, provides a comprehensive approach for performance and quality control testing, as well as the evaluation of advanced ultrasound imaging modes and technologies.

Keywords: And clinically relevant phantoms; Anechoic target detection; Breast ultrasound quality assurance; Contrast detail.

MeSH terms

  • Image Processing, Computer-Assisted*
  • Phantoms, Imaging
  • Ultrasonography