Engineering advances over the last decade have resulted in clinical technology that may permit physicians to use their ultrasound scanners in ways they have previously only dreamed about.
Myocardial perfusion imaging with ultrasound can sometimes replace traditional techniques, with harmonic and subharmonic imaging generating better images of the coronaries and other parts of the body. While 3-D has yet to reach its potential in real-time image presentation, many mainstream systems generate volumetric images in less than a minute, opening the door to new diagnostic opportunities.
In achieving these capabilities, no single technological advance stands alone; several techniques and technologies are combined in novel ways to address previously intractable problems.
No better example arises than the evolution of microbubble contrast agents. Several years ago, these agents were designated as the best approach for visualizing myocardial perfusion. Microbubbles improve image quality, partly because their shells and the gas inside strongly reflect ultrasound waves. This reflectivity vastly increased signal strength, but clinical tests found the agents lacking-until ultrasound technology was adapted especially for use with microbubbles. This innovation changed the picture, most notably in myocardial perfusion.
Harmonic imaging and microbubble contrast agents, possibly in combination with pulse inversion and power Doppler, have produced clinical results comparable to those of nuclear medicine. At Northwick Park Hospital in Harrow, U.K., researchers studied power Doppler harmonic imaging in the assessment of myocardial perfusion. They compared ultrasound using microbubble contrast with single-photon emission computed tomography (SPECT) with technetium-99m sestamibi in 15 patients.1
Using a five-segment model by two blinded observers, perfusion was determined to be present or absent for each modality. The research team reported a high concordance (93%) between the two modalities for the presence as well as the location of the defects: 96% for the anteroseptal region, 93% for the inferoposterior region, and 87% for the apical region.
"Potentially, contrast echo has similar accuracies to SPECT, although it is certainly not a clinically proven technique," said Dr. Avijit Lahiri, director of cardiac research at the hospital.
A larger study comparing the two modalities was performed at the University of Texas Southwestern,2 where harmonic power Doppler imaging (HPDI) and SPECT were performed during rest and pharmacological stress in 123 patients known or suspected to have coronary artery disease. Myocardial perfusion by HPDI was graded for each coronary territory as absent, patchy, or full.
Persistently absent or patchy myocardial perfusion by HPDI between rest and adenosine was interpreted as a fixed defect, whereas any decrease in perfusion grade was interpreted as a reversible defect. Ultrasound and SPECT produced the same conclusions in 83 of 103 (81%) for normal versus abnormal perfusion. Discrepancies occurred mostly in imaging the circumflex, where HPDI identified fixed defects in 33% of patients but SPECT identified only 14%.
"The next stage of development is to assess real-time perfusion, and that is even more exciting because we can watch the contrast agent come and then wash out," said Dr. Sheila K. Heinle, an assistant professor of medicine at the university. "That will possibly give us additional information about coronary flow reserve in conjunction with pharmacologic agents."
Harmonic imaging alone offers substantial advantages over fundamental imaging, with or without microbubble contrast. It increases contrast between vessels and background tissues, has spatial and temporal resolution similar to that of gray-scale imaging, and is unaffected by flash artifact. Harmonic imaging also allows increased penetration without loss of detail. The advantage of artifact reduction and greater intrinsic contrast sensitivity is especially apparent when heightening low-contrast lesions.3
"The increase in contrast allows a better interpretation of images and an increased penetration with fewer artifacts," said Francois Tranquart, chief of ultrasound at Hospital Bretonneau in Tours, France.
But harmonic imaging may be less useful in some circumstances than others. Research conducted at Hutzel Hospital/Wayne State University in Detroit found that harmonic imaging improved resolution in at least one fetal structure in only 51.4% of patients studied. Differences were most evident for four-chamber views of the heart, but in those cases improved visualization was seen in only 30.5% of patients. Patient weight and gestational age were found to be major factors, as the researchers concluded that obese patients during the second trimester benefited the most.
"Women with a normal body mass index did not have improvement in visualization rates with harmonic tissue imaging, and in fact there seemed to be some image deterioration in normal-weight patients when we used harmonic tissue imaging instead of traditional fundamental imaging," said Dr. Marjorie C. Treadwell, director of obstetric ultrasound at Hutzel.
The improved resolution that came with deeper penetration worked, but only in patients who needed the deeper look.
The nuances, exceptions, and niche applications that are accumulating in ultrasound indicate the care clinicians will have to take in applying this modality to realize its benefits. The modality is already famous for its dependence on user skills. Diagnostic accuracy and reproducibility of results are linked exactly to the sonographer's ability to visualize, recognize, and capture data that indicate disease or trauma. And the silver bullet that solves the personnel problem, in the eyes of many, is real-time volumetric imaging.
Capturing a volume of data enables the diagnostician to see all relevant data at once and in context. Pathologies are frozen in time and place, and reproducibility is absolute, since the same data set could be reconfigured into volumes or slices from any perspective or angle. Faster scans and less operator error boosts productivity and patient throughput; lower cost and increased revenue will follow. And patient recalls would become a thing of the past.
Real-time imaging, while preferred, may not be necessary. Delays in image processing, which may be no more than a few seconds, allow computing engines built into scanners to do some powerful processing. The Voluson 530 D Digital Volume Sonography system from Medison acquires a volume of data that can be sliced in planes, which is otherwise difficult if not impossible to achieve. The system, which requires a six-second scan to obtain the data for a 3-D image, has been used in ob/gyn, abdominal, pelvic, vascular, and endocavitary applications. Harmonic imaging is offered as an option.
Siemens Medical Systems offers 3-Scape, which compiles both color Doppler and gray-scale images into volumes that can be examined interactively. The system, built into the company's flagship Sonoline Elegra, also measures tissue structures. Other vendors, including Toshiba, GE, and ATL, have introduced 3-D technologies, first on their flagship systems and then on mid-tier units. In such systems, studies point toward potential use of 3-D imaging in obstetrics to identify malformations, in gynecology to indicate uterine and ovarian disease, in gastroenterology for the characterization of pancreatic and hepatobiliary tumors, in uronephrology to detect stones and prostatic tumors, and in the assessment of rectal carcinomas.4
Echocardiography presents the ultimate challenge and promises the greatest patient benefits for volumetric sonography. Instantaneously capturing the beating heart at stop-action speeds would dramatically reduce exam time and improve the diagnostic value of the data. The Model One-and still experimental-V360 Real-time Volumetric Ultrasound system from Volumetrics Medical Imaging incorporate many features found in conventional 2-D imaging, as well as real-time volumetric acquisition. Interactive capabilities allow retrospective multiview analysis of the heart. A work-in-progress shown at the 1999 annual meeting of the American Heart Association featured on-the-fly viewing of 3-D volumes of the heart.
"Based on our market research, within three to five years every company will have to offer 3-D capability," said Jim Mundell, general manager at Volumetrics.
Three-D echocardiography is especially suited to imaging the fetal heart, which for obvious reasons is not a candidate for nuclear medicine or cardiac cath. Researchers at the University of California, San Diego, evaluated 10 human fetuses in utero, four of whom had congenital heart disease.5 They performed freehand transabdominal scanning, using a real-time, 3-D echocardiography system from Volumetrics.
Four scans, lasting just 1.5 seconds and acquiring 20 volumes per second, were obtained for each fetal heart and stored for off-line analysis. Two-dimensional images displayed simultaneously in four planes were reconstructed in real-time during the acquisition, providing the operator with enough feedback to ensure that an adequate 3-D volume was being captured. Off-line reconstructions allowed the heart to be slowed, stopped, or viewed at its original speed.
The researchers consistently visualized most structures and views, as well as cardiac function, and they could readily detect abnormal structures.
"A lot of progress is being made, in scanners as well as in display and visualization technology and approaches," said Thomas Nelson, Ph.D., a physics professor in UCSD's department of radiology. "But the goal here is not really technology-it's to obtain better visualization of particular entities."
Nelson predicts a merger of 3-D technology with harmonic and pulse inversion approaches for contrast media. Pulse inversion harmonics with contrast produces spectacular results when coupled with 3-D, he said.
Yet the current of clinical medicine is hampered by clinical and economic realities. While the technologies necessary to make volumetric imaging a reality are almost at hand, the perceived clinical value of volumetric imaging has not resulted in widespread adoption of the technology. Some kinks remain to be worked out, such as exactly demonstrating blood flow.
Because flow consists of motion, it can occur in any direction, depending on the vascular tree. Flow, therefore, must be plotted according to its velocity. Success at this promises enormous benefits, as the information can be translated into images, as well as into quantifying a host of cardiovascular indicators. Most appealing is the possibility of characterizing complex flow events, such as regurgitant flow jets that are hard to visualize in two dimensions. Research has shown it's feasible in both animal6 and in-vitro models.7
Despite the perception of 3-D as leading-edge technology, 2-D imaging is serving as the pathfinder for new types of data acquisition and processing, both qualitative and quantitative. Researchers at Thomas Jefferson University in Philadelphia have been exploring ways to get around some of the limitations of harmonic imaging, especially the problems caused by second harmonic generation and harmonic frequency accumulation within the tissue itself. They are attempting to create subharmonic images by transmitting at the fundamental frequency and receiving at half that frequency.8 This technique brings better lateral resolution as well as improved scanning of deep-lying structures. The latter are better visualized with the higher transmit frequency and smaller attenuation of scattered subharmonic signals.
Flemming Forsberg, Ph.D., an associate professor of radiology at Thomas Jefferson, and his colleagues are also looking at ways to optimize data acquisition through quantitative assessment. They have documented that varying intracavitary pressure can affect the size of the bubbles and therefore their reflectivity. Differences in reflectivity might be used to noninvasively measure changes in the condition of the patient, thereby indicating cardiovascular health.
Each new capability, however, adds another layer of complexity to a modality that already depends heavily on operator skill. Achieving the potential of ultrasound in the future, therefore, may depend as much on the user as on the technology.
"There's always been that element in ultrasound; it's always been a choice of using B mode or color flow; of choosing the tools you need to get the information," Forsberg said. "That's one of the things that make it such an interesting field."
SMALLER, BETTER, CHEAPER
While early engineering advances in ultrasound were limited primarily by what could be done, economic factors are now in control. Iterative improvements that reduce vendors' costs are sought after, with no technical achievement more prized than the one that cuts cost while at the same time offering design and upgrade flexibility, without any sacrifice in image quality. The miniaturization of electronics is one of these improvements, and the replacement of hardware with software is another.
The impact of miniature, high-performance electronics is apparent in the appearance of the handheld ultrasound scanner. Manufacturers of these products, including SonoSite and Terason, want to put ultrasound in the hands of primary-care physicians and emergency medical personnel to enable quick screening or assessment of patients. It's an ambitious goal that could change the practice of medicine. Meanwhile, conventional practitioners will benefit from the trend toward smaller, lighter, and easier-to-operate diagnostic equipment.
The microchip is the source of power here. Toshiba America points to its new generation of transducers, featuring a "chip-in-the-tip"-an integrated circuit that captures and amplifies raw echo signals. The chip is designed to minimize information loss when data move through the cabling between transducer and system.
"Largely due to miniaturization of electronic circuitry, we can pack more computing power into a smaller footprint," said Scott Yarde, director of ultrasound marketing for Toshiba.
The other side of the technological coin is the digital signal processor (DSP). Linked in parallel, DSPs promise to deliver similar speed with markedly greater flexibility than application-specific integrated circuits. Whereas ASICs perform repetitive tasks because they are hard-wired, DSPs are controlled by software instructions. Because these chips are driven by algorithms, new clinical developments can be added to scanners in the form of software upgrades. Siemens was among the first to take advantage of DSPs, incorporating them into the Sonoline Elegra.
"Using the same hardware and by reprogramming the image processor, we were able to do panoramic viewing with Siescape and then evolve that into 3-D," said Heike Seck, product manager for the Elegra.
With the time between concept and commercialization compressed, the imaging community can look forward to a changing landscape in the clinic. The evolution could either tax the skill and knowledge of practitioners, or technology could come to the rescue.
Increased spatial resolution from more sensitive transducer arrays and improved contrast agents could make diagnostic conclusions easier. Real-time volumetric scanning could reduce the need for operator skill, making ultrasound more reproducible and productive. Advanced computing platforms might even make automatic feature recognition possible, as lesions and tissues are outlined, measured, and compared to databases containing normal and abnormal ranges.
If the past is any indicator, these advances are only a sampling of what's to come, as technologies are combined in new and creative ways. But it will always be the clinician's skill and art that count most.
References for this article are available at diagnosticimaging.com.
1. Senior R, Kaul S, Soman P, Lahiri A. Power Doppler harmonic imaging: a feasibility study of a new technique for the assessment of myocardial perfusion. Am Heart J 2000;139(2 Pt 1):245-51
2. Heinle SK, Noblin J, Goree-Best P, et al. Assessment of myocardial perfusion by harmonic power Doppler imaging at rest and during adenosine stress: comparison with (99m)Tc-sestamibi SPECT imaging. Circulation 2000;102(1):55-60.
3. Tranquart F, Grenier N, Eder V, Pourcelot L. Clinical use of ultrasound tissue harmonic imaging. Ultrasound Med Biol 1999;25(6):889-94.
4. Candiani F. The latest in ultrasound: three-dimensional imaging. Eur J Radiol 1998;27 Suppl 2:179-82.
5. Sklansky MS, Nelson T, Strachan M, Pretorius D. Real-time three-dimensional fetal echocardiography: initial feasibility study. J Ultrasound Med 1999;18(11):745-52.
6. Mori Y, Shiota T, Jones M, et al. Three-dimensional reconstruction of the color Doppler-imaged vena contracta for quantifying aortic regurgitation: studies in a chronic animal model. Circulation 1999;99(12):1611-1617.
7. Li X, Shiota T, Delabays A, et al. Flow convergence flow rates from 3-dimensional reconstruction of color Doppler flow maps for computing transvalvular regurgitant flows without geometric assumptions: An in-vitro quantitative flow study. J Am Soc Echocardiogr 1999;12:1035-1044.
8. Forsberg F, Shi WT, Goldberg BB. Subharmonic imaging of contrast agents. Ultrasonics 2000;38(1-8):93-98.
Back to top