Design Article
New components offer flexibility in ultrasound system design
Rob Reeder, Converter Applications Engineer, Analog Devices, Inc.
1/13/2009 9:00 AM EST
This article will review the up-front considerations that must be addressed to offer portable, performance-based products that provide the ultrasound system designer with the flexibility required to bring new imaging products to this global market.
System trade-offs
Ultrasound systems remain complicated, even though years of research and development have produced significant technological advancements. As with other complex systems, many approaches to system partitioning exist.
For many years, manufacturers implemented these complicated systems by designing their own custom application specific integrated circuits (ASICs). This solution usually consisted of two ASICs which implemented most of the time-gain compression (TGC) and receive/transmit (Rx/Tx) paths, as shown in Figure 1.
Figure 1: ASIC approach to ultrasound signal-processing channel
(Click on image to enlarge)
This approach was common before multichannel voltage-controlled amplifiers (VGAs), analog/digital converters (ADCs), and digital/analog converters (DACs) became widely available. The custom circuits, which allowed designers to incorporate low-cost, flexible functions, provided cost savings over time, because incorporating much of the signal chain in them minimized the number of external components.
Unfortunately, as time went on, the technology available via these lithographies showed its age in both scale and power consumption. ASICs have a high number of gates, but their digital technology is not optimized to successfully implement analog functions such as high performance ADCs. ASICs also limit the system designer, due to the limited number of suppliers.
Although high-performance imaging can be achieved using this partitioning method, it is not optimal in terms of portability, size, and power consumption. The advent of quad-channel and octal-channel TGCs, ADCs, and DACs has allowed both the size and power to be reduced, but without compromising performance, thus bringing new system approaches and new players into these markets.
Multichannel components allow the designer to put devices closer together, increasing the number of channels in the system. They also allow the designer to divide the sensitive circuits between two or more boards to complete the system. This allows for effective reuse of the electronic circuits over many platform developments.
Note that as channel counts increase, this leads to dynamic range improvement. Effectively, noise can be treated as uncorrelated in the system. By doubling the number of channels in a system, the noise is halved and the dynamic range increases by 3 dB. Therefore, a 64-channel system can add as much as 12-dB improvement in dynamic range over a 16-channel system. But there are disadvantages: an increase in channel count can make printed-circuit board (PCB) trace routing a nightmare, forcing the designer to use lower channel-count devices in some cases. This also poses new thermal challenges for mechanical designers, increasing both the system cost and the fan noise.
Today, IC manufacturers are able to integrate complete multichannel TGC paths, Figure 2. Multichannel, multi-component integration makes the design approach easier, reducing PCB size and power dissipation without sacrificing performance. As higher level integration schemes become more predominant, advantages once again follow in cost, size, and power reduction, leading to less heat in the system and longer battery life.
Figure 2: Commonly sought integration paths for multichannel designs
(Click on image to enlarge)
An ultrasound subsystem, such as Analog Devices' AD9272/AD9273 which integrates the low noise amplifier (LNA), VGA, antialiasing filter (AAF), ADC and crosspoint switch, implements the complete time-gain compression path, the most common receiver path found in an ultrasound system. The two devices offer the system designer the flexibility to trade performance for power consumption: the high-performance AD9272 features low noise (0.75nV/rt-Hz), while the low-power AD9273 consumes only 100 mW per entire TGC channel at 40 MSPS. Since pin count and housing size are critical, the pin-compatible devices employ serial I/O. The compact, 14-mm - 14-mm - 1.2-mm packages reduce per-channel area and power dissipation by more than 33% as compared to multi-chip solutions.
Most ultrasound companies acknowledge that their critical intellectual property (IP) lies within the probe and beamformer technology. Multichannel ICs are quickly becoming commodity devices, putting an end to high-cost components, and the endless tweaking and optimization of individual TGC path, needed to complete the system and get that extra bit of performance or savings in power. Other portions of the ultrasound system are being considered for further integration; studies have shown that if the front-end electronics were closer to the probe, there would be less probe loss and better signal sensitivity, allowing system designers to relax the requirements on the front-end components (LNA/VGA). Integration of these portions of the signal chain might prove beneficial.
Trends in ultrasound
With so many applications requiring ultrasound technologies, the demand for both performance and portability is high. Performance-driven applications, such as cardiology and 4-D image processing, contain the largest number of channels, features, and options. Power is not a driving factor because these systems are used at the patient's bedside, operating room suite, or nurses' triage, but performance is key as these systems are used for human diagnoses.
Portable ultrasound offers a different array of application opportunities, especially in locations where reliable electrical power is scarce or does not exist, such as remote village clinics, emergency medical services, animal farming, bridges, and large machinery inspections, Figure 3.
Figure 3: Portable ultrasound systems used for large-animal farming and breeding (System photo courtesy of E.I. Medical Imaging)
(Click on image to enlarge)
Ultrasound systems can generally be divided into three classes:
- High-end ultrasound systems are at the forefront of the most recent technology and market features, produce the best images, and are more expensive.
- Mid-range systems generally have a sub-set of the features that a high-end ultrasound system offers without much sacrifice in image quality.
- Low-end systems are scaled down even further and, in some cases, serve a particular application, which can be clinical or other.
A new tool in ultrasound design
Ultrasound covers a wide range of varying applications, so the tradeoffs a system designer must make have increased. Each modality has limitations that are defined generally by performance vs. power. Today these challenges have been met with components that allow the designer to scale the performance vs. power ratio within the IC, thus cutting down on time to market. The AD927x from Analog Devices offers a host of configurability within the IC to scale input range, bias, sample rate, and gain. Depending on which imaging modality or probe type is required, the system designer can literally system-scale the design appropriately in real-time, offering the maximum performance at minimum power.
In a typical application, the system needs to acquire images from a 5-MHz probe producing a 0.5-Vpp signal. If the LNA that has 0.86 nV/rt-Hz of noise, or 1.4 nV/rt-Hz for the entire channel, and is source terminated with 50 ohms, the input dynamic range of the system would be 92 dB, thus producing a noise figure (NF) of 3.8 dB. This translates to an output dynamic range of 66.3 dB, allowing the system designer to maintain optimum performance while only dissipating 191 mW/channel at 40 MSPS.
If the system performance exceeds expectations, the system designer may decide that a 2-dB reduction in input dynamic range could be justified in the system. if it saves 50 mW channel. The system designer could experiment with gain, bias, termination, and other parameters to see if this is feasible, but it can be difficult to understand the system tradeoffs when changing all these parameters.
A graphical configuration tool, such as the one offered by Analog Devices for its AD927x family, Figure 4, makes it convenient for the system designer to evaluate performance. In this tool, all the system features have been put in place, allowing the system designer to make these tradeoffs quickly and system scaling to be pushed down into the IC-level directly. As a result, the designer no longer needs to change real hardware, and produce cumbersome image-processing tests to validate these tradeoffs. Furthermore, the configuration tool will translate the optimized configuration parameters to digital settings and generate a file that can replicate the part's final configuration setup for the system.
Figure 4: AD9272/9273 Graphical configuration tool
(Click on image to enlarge)
Conclusion
Advances in integrated multichannel devices are pushing system flexibility even further today. New innovative products and configuration tools make the system designer's life easier without a doubt. This provides a means of developing diversified ultrasound products that are configurable and scalable, depending on the imaging modality, when it comes to performance versus power.
About the author
Rob Reeder is a senior converter applications engineer working in the high-speed signal-processing group at Analog Devices, Inc. in Greensboro, NC. He was formerly a designer for ADI's Multi-Chip Products group, designing converter modules for space and military applications. Rob received his MSEE and BSEE from Northern Illinois University in DeKalb, IL He can be reached at rob.reeder@analog.com.
Special thanks
The author would like to give a special thanks to product development engineers Gina Kelso and Ramya Ramachandran, for creating the AD927x family configuration tool and GUI.
