Home Synthetic Aperture Imaging in Breast Ultrasound
Post
Cancel

Synthetic Aperture Imaging in Breast Ultrasound

Posted Jul 31, 2012 Updated Apr 10, 2023
By Won Hwa Kim MD , Jung Min Chang MD PhD , Choye Kim BS , Jongho Park MS , Yangmo Yoo PhD , Woo Kyung Moon MD PhD , Nariya Cho MD PhD , Byung Ihn Choi MD PhD
9 min read

Rationale and Objectives

To compare the image quality between conventional and synthetic aperture (SA) imaging in breast ultrasound (US).

Materials and Methods

Twenty-four patients with 31 breast lesions were included in our study. The US data were processed with SA algorithm. For quantitative analysis, contrast-to-noise ratio (CNR) and signal-to-noise ratio (SNR) were calculated. For qualitative analysis, conventional and SA images were reviewed by three radiologists and diagnostic preference (conspicuity, margin sharpness, and contrast) was assessed. The radiologists also determined whether artifacts were present. Parameters were analyzed using a paired t -test, Wilcoxon signed-rank test, and chi-square test.

Results

The mean CNRs were higher in SA images compared with conventional images (mean, 2.56 versus 2.28, P = .004). The mean SNRs were higher in SA images compared with conventional images (31.62 versus 25.26, P < .0001). SA images were considered as being “better” or “much better” in 16–23 (51.6–74.2%) lesions of total 31 lesions for conspicuity, in 17–24 (45.2%–77.4%) for margin sharpness, and in 13–23 (41.9%–74.2%) for contrast. Significant preferences in SA images were demonstrated (conspicuity, P < .05 for all radiologists; margin sharpness and contrast in two radiologists). Refraction and speckle artifacts were less frequently observed in SA images (refraction, P < .05 for all radiologists; speckle, P < .05 for two radiologists), whereas reflection artifacts were more frequent in SA images ( P < .05 in two radiologists).

Conclusion

SA imaging provides better image quality than conventional imaging in patients with focal breast lesions in breast US.

Introduction

Ultrasonography (US) of the breast plays a critical role in the diagnosis and management of breast diseases. US with a high-frequency transducer is essential for noninvasive diagnosis of screening-detected or palpable masses, and it allows radiologists to differentiate between benign or malignant masses using descriptive parameters . The limitations of US, however, are that it is operator-dependent and has considerable interobserver variability in terms of radiologists’ descriptions and assessments of breast lesions . Because inherent artifacts, such as speckle, acoustic shadowing, and acoustic enhancement, can compromise image quality and contribute to the inconsistent and inaccurate interpretation of breast images, several US techniques had been developed and are commercially used to overcome these problems . Nevertheless, conventional US still has its own set of problems (eg, fixed transmit focus and limitation of penetration) that is responsible for compromising image quality.

Synthetic aperture (SA) imaging is a US algorithm technique that can potentially provide higher image quality than conventional imaging by using the superposition of the acoustic fields The SA technique was originally developed from synthetic aperture radar for geological and sonar applications, but it has been modified for use in medical imaging for several decades . The major difference between SA and conventional imaging is that SA imaging simultaneously produces the full range of images at each transmission, whereas a single image line is built sequentially to form the full range of an image in conventional imaging. In SA imaging, the simultaneously obtained full range of the images yields low-resolution images, which are synthesized to create a high-resolution image ( Fig 1 ) . During this process, the SA imaging system calculates the transmit focus in all points of an image with information from each emission, which provides more uniform resolution beyond the fixed focus depth.

Figure 1, Schematic illustrations of conventional and synthetic aperture (SA) imaging. In conventional imaging, an image is formed by sequentially building up each image line achieved at each transmission. In SA imaging, a full line of low-resolution image is produced from each transmission, and the low-resolution images are synthesized coherently to provide a high-resolution image.

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Materials and methods

Patients

Get Radiology Tree app to read full this article<

US Unit and Imaging Acquisition

Get Radiology Tree app to read full this article<

Quantitative Image Analysis

Get Radiology Tree app to read full this article<

CNR=|μ1−μ2|σ21+σ22√ CNR

=

|

μ

1

−

μ

2

|

σ

1

2

+

σ

2

2

where μ 1 and μ 2 are the mean gray intensities within a rectangular ROI drawn within the lesion and in the background, respectively, and σ 1 and σ 2 are the variances (standard deviation) of the mean gray intensities within the rectangular ROI in the lesion and the background, respectively. Similarly, the SNR is given by the following equation:

SNRdB=10log(PsignalPnoise), SNR

dB

=

10

log

(

P

signal

P

noise

)

,

where P signal and P noise are the powers of the signal and the noise, respectively.

Get Radiology Tree app to read full this article<

Qualitative Image Analysis

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Statistical Analysis

Get Radiology Tree app to read full this article<

Results

Quantitative Image Analysis

Get Radiology Tree app to read full this article<

Figure 2, The box plots show the distribution of contrast-to-noise ratios (CNRs) and signal-to-noise ratios (SNRs) in conventional and synthetic aperture imaging. (a) CNR values were significantly higher in synthetic aperture (SA) imaging ( P = .0004). (b) SNR values were significantly higher in SA imaging ( P < .0001). Box values are from lower to upper quartiles, and the central lines represent the median values. The whiskers extend from minimal to maximal values, and the dots represent outliers.

Get Radiology Tree app to read full this article<

Qualitative Image Analysis

Get Radiology Tree app to read full this article<

Table 1

Subjective Preference Scores of Synthetic Aperture Imaging Compared with Conventional Imaging

Parameters Conventional Images Much Better (Score = −2) Conventional Images Better (Score = −1) Same (Score = 0) SA Images Better (Score = +1) SA Images Much Better (Score = +2) Average Score_P_ Value † Conspicuity Radiologist 1 0 (0) 6 (19.4) 9 (29) 13 (41.9) 3 (9.7) 0.42 .019 ∗ Radiologist 2 0 (0) 4 (12.9) 9 (29) 18 (58.1) 0 (0) 0.45 .003 ∗ Radiologist 3 0 (0) 5 (16.1) 3 (9.7) 12 (38.7) 11 (35.5) 0.94 <.0001 ∗ Margin sharpness Radiologist 1 0 (0) 6 (19.4) 8 (25.8) 15 (48.4) 2 (6.5) 0.42 .016 ∗ Radiologist 2 0 (0) 6 (19.4) 11 (35.5) 14 (45.2) 0 (0) 0.26 .074 Radiologist 3 0 (0) 6 (19.4) 1 (3.2) 16 (51.6) 8 (25.8) 0.84 <.0001 ∗ Contrast Radiologist 1 0 (0) 6 (19.4) 9 (29) 13 (41.9) 3 (9.7) 0.42 .019 ∗ Radiologist 2 0 (0) 6 (19.4) 12 (38.7) 13 (41.9) 0 (0) 0.23 .074 Radiologist 3 0 (0) 3 (9.7) 5 (16.1) 16 (51.6) 7 (22.6) 0.87 <.0001 ∗

Data are presented as the mean values, and the numbers in parentheses are percentages.

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Figure 3, The bar graphs show the frequency distribution of preference scores of synthetic aperture imaging compared with conventional imaging for (a) lesion conspicuity, (b) margin sharpness, and (c) contrast.

Figure 4, A 45-year-old woman with invasive ductal carcinoma in her left upper breast. (a) Conventional ultrasound imaging demonstrates an ill-defined hypoechoic mass ( arrow ). (b) Synthetic aperture imaging shows improvements in margin sharpness and lesion contrast of the mass ( arrow ) and connective tissue in the subcutaneous fat ( arrowhead ). For lesion conspicuity and margin sharpness, two of the three radiologists scored (b) as “better,” and the other radiologist scored (b) as “much better.” For the contrast measurement, all radiologists scored (b) as “better.” The contrast-to-noise ratio (CNR) and signal-to-noise ratio (SNR) were higher in synthetic aperture imaging (CNR, 2.69 vs. 2.47; SNR, 36.46 vs. 34.93).

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Table 2

Artifacts of Synthetic Aperture Imaging versus Conventional Imaging

Artifacts Conventional ( n = 31) Synthetic Aperture ( n = 31)P Value Refraction Radiologist 1 12 7 .025 ∗ Radiologist 2 12 5 .020 ∗ Radiologist 3 19 7 .001 ∗ Speckle Radiologist 1 14 5 .029 ∗ Radiologist 2 10 3 .052 Radiologist 3 15 7 .033 ∗ Reflection Radiologist 1 2 14 .003 ∗ Radiologist 2 4 24 <.0001 ∗ Radiologist 3 5 12 .071 Posterior shadowing Radiologist 1 3 3 1.000 Radiologist 2 14 6 .083 Radiologist 3 2 0 .317 Posterior enhancement Radiologist 1 1 0 .317 Radiologist 2 5 3 .157 Radiologist 3 0 0 1.000

Numbers are frequencies of artifacts in the 31 image pairs.

Get Radiology Tree app to read full this article<

Figure 5, A 68-year-old woman with invasive ducal carcinoma in her left upper outer breast. (a) Conventional ultrasound imaging demonstrates a hypoechoic round mass ( arrow ), which is difficult to distinguish from the surrounding breast parenchyma. The refraction artifacts ( arrowhead ) are severe. (b) In synthetic aperture imaging, the mass and the surrounding normal tissues are more conspicuous and there is a sharper margin and higher contrast of the mass. In addition, a minimal degree of refraction artifact is visible. For lesion conspicuity, margin sharpness and contrast, two of the three radiologists scored (b) as “much better,” and the other radiologist scored (b) as “better.” The contrast-to-noise ratio (CNR) and signal-to-noise ratio (SNR) were higher in synthetic aperture imaging (CNR, 2.53 vs. 2.19; SNR, 34.57 vs. 23.98).

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Table 3

Levels of Agreement for Interobserver Variability for Conventional and Synthetic Aperture Images

Parameters Radiologists 1 and 2 Radiologists 1, 2, and 3 Radiologists 2 and 3 Conspicuity 0.59 ± 0.13 0.65 ± 0.1 0.56 ± 0.08 Margin sharpness 0.46 ± 0.15 0.44 ± 0.01 0.44 ± 0.1 Contrast 0.68 ± 0.07 0.57 ± 0.11 0.44 ± 0.09 Artifact 0.32 ± 0.06 0.50 ± 0.06 0.27 ± 0.06

Data are presented as the mean ± standard deviation. A kappa value of 0 to 0.20 was considered to be slight agreement, 0.21–0.40 was considered fair agreement, 0.41–0.60 was considered moderate agreement 0.61–0.80 was considered substantial agreement, and 0.81–1.00 was considered almost perfect agreement.

Get Radiology Tree app to read full this article<

Discussion

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

Get Radiology Tree app to read full this article<

References

  • 1. Stavros A.T., Thickman D., Rapp C.L., et. al.: Solid breast nodules: use of sonography to distinguish between benign and malignant lesions. Radiology 1995; 196: pp. 123.

  • 2. Baker J.A., Kornguth P., Soo M.S., et. al.: Sonography of solid breast lesions: observer variability of lesion description and assessment. AJR Am J Roentgenol 1999; 172: pp. 1621.

  • 3. Szopinski K.T., Pajk A.M., Wysocki M., et. al.: Tissue harmonic imaging: utility in breast sonography. J Ultrasound Med 2003; 22: pp. 479.

  • 4. Mesurolle B., Helou T., El-Khoury M., et. al.: Tissue harmonic imaging, frequency compound imaging, and conventional imaging: use and benefit in breast sonography. J Ultrasound Med 2007; 26: pp. 1041.

  • 5. Huber S., Wagner M., Medl M., et. al.: Real-time spatial compound imaging in breast ultrasound. Ultrasound Med Biol 2002; 28: pp. 155-163.

  • 6. Pedersen M.H., Gammelmark K.L., Jensen J.A.: In-vivo evaluation of convex array synthetic aperture imaging. Ultrasound Med Biol 2007; 33: pp. 37-47.

  • 7. O’Donnell M.: Coded excitation system for improving the penetration of real-time phased-array imaging systems. IEEE Trans Ultrason Ferroelectrics Frequency Control 1992; 39: pp. 341-351.

  • 8. Jensen J.A., Nikolov S.I., Gammelmark K.L., et. al.: Synthetic aperture ultrasound imaging. Ultrasonics 2006; 44: pp. e5-e15.

  • 9. Bae M.H., Jeong M.K.: A study of synthetic-aperture imaging with virtual source elements in B-mode ultrasound imaging systems. IEEE Trans Ultrason Ferroelectrics Frequency Control 2000; 47: pp. 1510-1519.

  • 10. Gammelmark K.L., Jensen J.A.: Multielement synthetic transmit aperture imaging using temporal encoding. IEEE Trans Med Imaging 2003; 22: pp. 552-563.

  • 11. Kim K., Hwang J., Jeong J., et. al.: An efficient motion estimation and compensation method for ultrasound synthetic aperture imaging. Ultrason Imaging 2002; 24: pp. 81-99.

  • 12. Nikolov S., Jensen J.A.: Synthetic aperture tissue and flow ultrasound imaging.Synthetic aperture tissue and flow ultrasound imaging.2001.Technical University of Denmark, Department of Electrical Engineering, Biomedical EngineeringDenmark:

  • 13. Baker J.A., Soo M.S., Rosen E.L.: Artifacts and pitfalls in sonographic imaging of the breast. AJR Am J Roentgenol 2001; 176: pp. 1261.

  • 14. Pedersen M.H., Misaridis T.X., Jensen J.A.: Clinical evaluation of chirp-coded excitation in medical ultrasound. Ultrasound Med Biol 2003; 29: pp. 895-905.

  • 15. Trahey G.E., Nock L.F.: Synthetic receive aperture imaging with phase correction for motion and for tissue inhomogeneities. II. Effects of and correction for motion. IEEE Trans Ultrason Ferroelectrics Frequency Control 1992; 39: pp. 496-501.

Academic Radiology, Volume 19, Issue 8
Journals General Radiology
This post is licensed under CC BY 4.0 by the author.
Recently Updated
Contents