DOI: 10.1148/radiol.2241010511
(Radiology 2002;224:265-269.)
© RSNA, 2002
Microcalcifications in Breast Tissue Phantoms Visualized with Acoustic Resonance Coupled with Power Doppler US: Initial Observations1
Susan P. Weinstein, MD,
Chandra Seghal, PhD,
Emily F. Conant, MD and
Jill A. Patton, MS
1 From the Department of Radiology, University of Pennsylvania Medical Center, 1 Silverstein Bldg, 3400 Spruce St, Philadelphia, PA 19104. From the 1999 RSNA scientific assembly. Received February 23, 2001; revision requested April 12; revision received October 26; accepted December 12. Supported in part by DAMD17-00-1-0406. Address correspondence to S.P.W.
 |
ABSTRACT
|
|---|
Calcium carbonate particles embedded in gelatin and turkey breast tissues were visualized with acoustic resonance imaging and power Doppler ultrasonography. Sonography revealed that the region of color level detection corresponded to the location of the calcium carbonate particles. Correlation between color level detection and the location of the particles was confirmed on radiographs of the specimens obtained at core needle biopsy performed through the region of color level detection.
© RSNA, 2002
Index terms: Breast, calcification, 00.8119 • Breast, US, 00.12981 • Breast radiography, technology, 00.128
|
INTRODUCTION
|
|---|
It is estimated that ductal carcinoma in situ represents 20%–30% of breast carcinomas detected at screening mammography (1). Since the 1970s, the detection rate of ductal carcinoma in situ has steadily increased with the wider use of screening mammography. The detection rate for women younger than 50 years was 2.3 cases per 100,000 women in the 1970s and increased to 6.2 cases per 100,000 women by the 1990s (2). More dramatically, in the group of women older than 50 years, the rate has increased from 14.3 to 54.6 per 100,000 (2) in the same period.
Frequently, ductal carcinoma in situ manifests only as calcifications without an associated mass (3). Because ductal carcinoma in situ represents as much as half of mammographically depicted cancers (3), it would be efficacious to image microcalcifications with as many modalities as possible to allow flexibility in imaging-guided biopsy. Currently, even with all the technical advances in breast imaging, including magnetic resonance imaging, scintigraphy, and ultrasonography (US), mammography is the only reliable method to help detect, characterize, and localize microcalcifications for biopsy.
We have developed a technique in which acoustic resonance is used to visualize microcalcifications. Acoustic resonance is used on the basis of the size of the microcalcifications and the binding strength with the surrounding tissues in which they are embedded. When subjected to a wide frequency range, particles of different sizes will resonate at different frequencies given the same binding environment. "Tuning" to the appropriate frequency range would make it possible to selectively visualize microcalcifications of varying sizes. The purpose of our study was to evaluate US coupled with acoustic resonance to demonstrate small calcified particles for targeting and biopsy.
|
Materials and Methods
|
|---|
Phantoms
Gelatin phantoms were constructed by using a mixture of water, gelatin, and glycerol with uniformly dispersed oil-lecithin emulsion (particle diameter, 0.2–2.0 µm). The suspension was placed in a mold and refrigerated until a solid state was reached. Calcium carbonate particles 400–800 µm in diameter were suspended in the gelatin phantom in an area 1.0–1.5 cm in diameter. The phantom was kept refrigerated until use, at which time it was removed from the mold. Ten phantoms were made.
Twelve tissue phantoms were made from boneless turkey breast. The turkey breast was dissected along the tissue planes, and 400–800-µm-diameter calcium carbonate particles mixed in acoustic coupling gel were embedded carefully to avoid air trapping. The region of the calcium carbonate particles ranged from 1.5 to 2.0 cm in diameter
Imaging: US Coupled with Acoustic Resonance
Imaging was performed (S.P.W. or C.S., with the assistance of J.A.P.) by using gray-scale and power Doppler US coupled with acoustic resonance (Logic 700; GE Medical Systems, Milwaukee, Wis). A variable 10–13-MHz transducer was used. All the gelatin and tissue phantoms were scanned. The phantoms were imaged with B-mode US and power Doppler US, while the particles were resonated from 50 to 500 Hz. The frequency that corresponded to the maximum color level detection was noted.
The frequency used to excite the particles into resonance was emitted from a thin (5-mm) lightweight (13-g) piezoelectric speaker element with the capability to transmit low-frequency vibrations from 50 Hz to 2 kHz. The device was housed in a fiberglass case. The piezoelectric vibrator was designed for 8 
of impedance to be driven by an audio amplifier. The vibrator was specially designed by the biomedical medical instrumentation group at our university; they specialize in building custom instruments, and this vibrator was approved for human use by the university institutional review board. The disk was coupled with US gel (E-Z-Gel; E-Z-Em, Westbury, NY) and placed adjacent to the tissue to be scanned. The images were videotaped for analysis.
Image Analysis
The videotaped images were subsequently analyzed. Imaging time for each scan was approximately 2 minutes. At the video frame rate of 30 images per second, approximately 120 x 30, or 3,600, images were generated. An algorithm was developed to reduce the data by sorting the images on the basis of the frequency of the vibration. These images were subsequently used in a computer program developed in our laboratory to analyze the vibrational response of the microcalcifications.
Color was analyzed for each image: Mean color level, percentage of fractional area of color, and color-weighted fractional area were computed. To determine the mean color level, the color palate on the image was read by the computer and divided equally on a scale of 0–100. With this scaling system, the computer constructed a lookup table for hue, saturation, and brightness values for the colors in the palette bar. Next, the computer identified colored pixels on the image and used the lookup table to assign a color value to each pixel in the region of interest. The color level of the pixels in the region of interest was summed and divided by the number of color pixels to calculate the mean color level. The percentage of fractional area of color was defined as the area covered by colored pixels divided by the area of the region of interest, multiplied by 100%. Color-weighted fractional area was defined as the mean color level multiplied by the percentage of fractional area of color, all divided by 100, and indicated the presence of net motion in the region of interest. Each parameter was plotted with respect to the frequency range.
Biopsy
All the gelatin and tissue phantoms were scanned from 50 to 500 Hz. The location where the maximum color level was detected and the corresponding frequency were noted. To demonstrate that the area of color level detection in the phantoms corresponded to the calcium carbonate particles, biopsy was performed through the region of color detection level. A 14-gauge disposable core biopsy needle (Monopty Biopsy Instrument; Bard, Covington, Ga) was used to perform biopsy in the region of interest during in real-time scanning by using power Doppler US coupled with acoustic resonance imaging. Approximately seven to 10 samples were obtained for each phantom. The core biopsy samples were placed in a core needle biopsy specimen container (Beekley, Bristol, Conn). Radiography of the specimens (alpha RT; Instrumentarium, Milwaukee, Wis) was performed at 22 kVp and 6 mAs, with magnification of 1.8, to confirm that the biopsy target, the region of depiction at power Doppler US, corresponded to the calcium carbonate particles.
|
Results
|
|---|
Gelatin Phantom
The calcium carbonate particles were readily visible as white particles in all the gelatin phantoms at B-mode US. Figure 1a shows one of the phantoms. The calcium carbonate particles were readily visible in all 10 specimens. With power Doppler US evaluation, a gradual increase in color level detection was seen in the region of the calcium carbonate particles (Fig 1b). The maximum detection level was reached and then plateaued at frequencies of 200–380 Hz (Fig 1c). This plateau was followed by a gradual decrease in the color level detection (Fig 1d). This trend in color level detection was seen in all the gelatin phantoms evaluated. As seen in Figure 1, there was a minimal color level detected outside the region of the calcium carbonate particles. Graphs of mean color level, percentage of fractional area of color, and color-weighted fractional area versus frequency demonstrate the results seen on the static images in Figure 1, that is, a gradual increase in depiction level, followed by a decrease (Fig 2).
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. Color Doppler US images of gelatin phantoms evaluated with acoustic resonance and power Doppler US. Color is represented in gray scale. (a) Echogenic calcium carbonate particles are clearly visible in the gelatin phantom. (b) As the frequency is gradually increased to 100 Hz, the color pixels are seen as gray areas (arrows) in the region of the calcium carbonate particles. The pixels in this region were seen in color on the original images. (c) At 210 Hz, the entire region of the calcium carbonate particles is filled in with color pixels. (d) At 400 Hz, only a few color pixels (arrows) are seen.
|
|
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. Color Doppler US images of gelatin phantoms evaluated with acoustic resonance and power Doppler US. Color is represented in gray scale. (a) Echogenic calcium carbonate particles are clearly visible in the gelatin phantom. (b) As the frequency is gradually increased to 100 Hz, the color pixels are seen as gray areas (arrows) in the region of the calcium carbonate particles. The pixels in this region were seen in color on the original images. (c) At 210 Hz, the entire region of the calcium carbonate particles is filled in with color pixels. (d) At 400 Hz, only a few color pixels (arrows) are seen.
|
|
View larger version (128K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1c. Color Doppler US images of gelatin phantoms evaluated with acoustic resonance and power Doppler US. Color is represented in gray scale. (a) Echogenic calcium carbonate particles are clearly visible in the gelatin phantom. (b) As the frequency is gradually increased to 100 Hz, the color pixels are seen as gray areas (arrows) in the region of the calcium carbonate particles. The pixels in this region were seen in color on the original images. (c) At 210 Hz, the entire region of the calcium carbonate particles is filled in with color pixels. (d) At 400 Hz, only a few color pixels (arrows) are seen.
|
|
View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1d. Color Doppler US images of gelatin phantoms evaluated with acoustic resonance and power Doppler US. Color is represented in gray scale. (a) Echogenic calcium carbonate particles are clearly visible in the gelatin phantom. (b) As the frequency is gradually increased to 100 Hz, the color pixels are seen as gray areas (arrows) in the region of the calcium carbonate particles. The pixels in this region were seen in color on the original images. (c) At 210 Hz, the entire region of the calcium carbonate particles is filled in with color pixels. (d) At 400 Hz, only a few color pixels (arrows) are seen.
|
|
View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. Graphs plot (a) color-weighted fractional area (CWFA), (b) percentage of fractional area, and (c) mean color level versus frequency in the gelatin phantom. Detection levels at power Doppler US increase, then plateau, and then rapidly decrease relative to frequency.
|
|
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. Graphs plot (a) color-weighted fractional area (CWFA), (b) percentage of fractional area, and (c) mean color level versus frequency in the gelatin phantom. Detection levels at power Doppler US increase, then plateau, and then rapidly decrease relative to frequency.
|
|
View larger version (23K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2c. Graphs plot (a) color-weighted fractional area (CWFA), (b) percentage of fractional area, and (c) mean color level versus frequency in the gelatin phantom. Detection levels at power Doppler US increase, then plateau, and then rapidly decrease relative to frequency.
|
|
Turkey Phantom
The turkey phantom was also imaged with B-mode US and power Doppler US. In all 12 turkey phantoms, unlike in the gelatin phantoms, the calcium carbonate particles were not as readily visible at B-mode US before use of acoustic resonance imaging coupled with power Doppler US (Fig 3a). When the turkey phantom was imaged with power Doppler US coupled with acoustic resonance imaging, a gradual increase in color level detection was seen, a peak was reached, and a decrease followed (Fig 3b–3d). Similar findings in power Doppler detection were seen in all the turkey phantoms evaluated.
View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. (a) US scan depicts a turkey breast phantom. Calcium carbonate particles are not as obvious as they are in the gelatin phantom. Color is represented in gray scale. (b-d) Power Doppler US scans coupled with acoustic resonance depict some color pixels in the region of the calcium carbonate particles (arrows) at 155 Hz (b); the number of color pixels continues to increase at 270 Hz (c); followed by a decrease at 355 Hz (d).
|
|
View larger version (158K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. (a) US scan depicts a turkey breast phantom. Calcium carbonate particles are not as obvious as they are in the gelatin phantom. Color is represented in gray scale. (b-d) Power Doppler US scans coupled with acoustic resonance depict some color pixels in the region of the calcium carbonate particles (arrows) at 155 Hz (b); the number of color pixels continues to increase at 270 Hz (c); followed by a decrease at 355 Hz (d).
|
|
View larger version (165K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c. (a) US scan depicts a turkey breast phantom. Calcium carbonate particles are not as obvious as they are in the gelatin phantom. Color is represented in gray scale. (b-d) Power Doppler US scans coupled with acoustic resonance depict some color pixels in the region of the calcium carbonate particles (arrows) at 155 Hz (b); the number of color pixels continues to increase at 270 Hz (c); followed by a decrease at 355 Hz (d).
|
|
View larger version (162K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3d. (a) US scan depicts a turkey breast phantom. Calcium carbonate particles are not as obvious as they are in the gelatin phantom. Color is represented in gray scale. (b-d) Power Doppler US scans coupled with acoustic resonance depict some color pixels in the region of the calcium carbonate particles (arrows) at 155 Hz (b); the number of color pixels continues to increase at 270 Hz (c); followed by a decrease at 355 Hz (d).
|
|
Biopsy Results
The radiograph of a specimen is shown in Figure 4. All radiographs of the specimens depicted the calcium carbonate particles in the specimens.
|
Discussion
|
|---|
The concept of acoustic resonance imaging is relatively new. There is a large body of literature that reports various modes of US imaging to characterize motion and elasticity of soft tissues induced by external vibrations or due to natural motion (4–8). None of these authors deal with the issue of detection of microcalcifications, nor do they use the concept of acoustic resonance as a means to identify the properties of inhomogeneities in the tissue. Our goal was to be able to visualize the resonance motion of inhomogeneities relative to that of the surrounding tissue.
The 400–800-µm calcium carbonate particles were excited into resonance when exposed to low-frequency vibrations from 50 to 500 Hz. The vibrations were of low amplitude, with no known harmful side effects. When the particles were in resonance and combined with power Doppler US, the calcium carbonate particles were expected to demonstrate a color level relative to the degree of resonance. At maximal resonance, there would be maximal color level detection. The frequency at which maximal resonance is demonstrated depends on the size of the microcalcifications and the binding properties between the tissue and the microcalcifications, which is demonstrated with the phantom models as shown in Figures 1 and 3. Graphs of the mean color level, percentage of fractional area of color, and color-weighted fractional area versus frequency demonstrate an increase in power Doppler US depiction level, followed by a plateau, then a decrease. The plateau, seen in the curves, may represent the particles of different sizes reaching resonance over a different range of frequencies.
In the simple gelatin phantom, the microcalcifications were visible without power Doppler US, which allowed direct correlation between the location of the particles and the area of color-level depiction. Unlike the homogeneous background of the gelatin phantom, the turkey phantom more closely approximates human breast tissue, with muscle striations and specular interfaces similar to the appearance of Cooper ligaments and other specular reflectors in human glandular tissue. Therefore, the location of the calcium carbonate particles is not obvious until power Doppler US coupled with acoustic resonance imaging is used. Core biopsy was performed in the area where color level was detected in both types of phantoms, and radiography of the specimens was performed. Findings on the radiographs of the specimens confirmed that the region of color level detection corresponded to the calcium carbonate particles.
There are multiple potential applications for this method. Once the microcalcifications are detected at US, biopsy could be performed with this dynamic imaging guidance. Results of various studies have demonstrated the cost-effectiveness of core needle biopsy compared with excisional biopsy (9–12) or US-guided core needle biopsy versus stereotactic core needle biopsy (13). With the ability to perform biopsy with vacuum assistance and US guidance, this method would allow performance of biopsy in calcifications by using US and possibly dynamic visualization of the extraction of calcifications, which would allow confirmation of targeting and, conceivably, a procedure time shorter than that for stereotactic core needle biopsy. Currently, sonographically guided biopsy is not always feasible because calcifications cannot always be definitively visualized with US alone.
We have shown that power Doppler US coupled with acoustic resonance imaging has the ability to demonstrate microcalcifications in phantom models. Future studies are needed to develop this technique for use in patients. US evaluation may be performed in women prior to conventional stereotactic or excisional biopsy. Correlation between particle size, peak resonance, binding properties, and pathologic results may provide beneficial information. Analysis of acoustic resonance properties may help determine the strength of adhesion between the calcium deposits and the surrounding tissues. Such forces are likely to be related to the molecular characteristics of the deposits and the mechanisms that cause calcifications to develop at specific sites in the tissues. Measurement of adhesive forces with acoustic resonance, in conjunction with morphologic parameters detected at mammography, may help characterize calcifications.
|
FOOTNOTES
|
|---|
Author contributions: Guarantors of integrity of entire study, all authors; study concepts and design, all authors; literature research, S.P.W.; experimental studies, all authors; data acquisition and analysis/interpretation, all authors; manuscript preparation, S.P.W.; manuscript definition of intellectual content, editing, revision/review, and final version approval, all authors.
|
REFERENCES
|
|---|
- Kopans D. Pathologic, mammographic, and sonographic correlation. In: Kopans D, eds. Breast imaging. 2nd ed. Philadelphia, Pa: Lippincott-Raven, 1998; 511- 615.
- Morrow M, Schnitt SJ, Harris JR. In situ carcinomas. In: Harris JR, Lippman ME, Morrow M, Hellman S, eds. Diseases of the breast. Philadelphia, Pa: Lippincott-Raven, 1996; 355-373.
- Ernster VL, Barclay J, Kerlikowske K, et al. Mortality among women with ductal carcinoma in situ of the breast in the population-based surveillance, epidemiology and end results program. Arch Intern Med 2000; 160:953-958.[Abstract/Free Full Text]
- Dickenson RJ, Hill CR. Measurement of soft tissue motion using correlation between A-scans. Ultrasound Med Biol 1982; 8:263-271.[CrossRef][Medline]
- Wilson LS, Robinson DE. Ultrasonic measurement of small displacements and deformations of tissue. Ultrason Imaging 1982; 4:71-82.[CrossRef][Medline]
- Lerner RM, Huang SR, Parker KJ. Sonoelasticity images derived from ultrasound signals in mechanically vibrated tissues. Ultrasound Med Biol 1990; 16:231-239.[CrossRef][Medline]
- Adler RS, Rubin JM, Bland PH, et al. Quantitative tissue motion analysis of digitized M-mode images: gestational differences of fetal lung. Ultrasound Med Biol 1990; 16:561-566.[CrossRef][Medline]
- Ophir J, Cespedes I, Ponnekanti H, et al. Elastography: a quantitative method for imaging the elasticity of biological tissues. Ultrason Imaging 1991; 13:111-134.[CrossRef][Medline]
- Hillner BE, Bear HD, Fajardo LL. Estimating the cost-effectiveness of stereotaxic biopsy for nonpalpable breast abnormalities: a decision analysis model. Acad Radiol 1996; 3:351-360.[CrossRef][Medline]
- Liberman L, Fahs MC, Dershaw DD, et al. Impact of stereotaxic core breast biopsy on cost of diagnosis. Radiology 1995; 195:633-637.[Abstract/Free Full Text]
- Lee CH, Egglin TK, Philpotts L, et al. Cost-effectiveness of stereotactic core needle biopsy: analysis by means of mammographic findings. Radiology 1997; 202:849-854.[Abstract/Free Full Text]
- Yim JH, Barton P, Weber B, et al. Mammographically detected breast cancer: benefits of stereotactic core versus wire localization biopsy. Ann Surg 1996; 223:688-697.[CrossRef][Medline]
- Liberman L, Feng TL, Dershaw DD, et al. US-guided core breast biopsy: use and cost-effectiveness. Radiology 1998; 208:717-723.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
