- Contact Us
- |
- TI Worldwide: United States
- |
- my.TI Login
Block Diagram
Design ConsiderationsUltrasound Systems, both medical and industrial, use focal imaging techniques to achieve imaging performance far beyond what can be achieved through a single-channel approach. Using an array of receivers, a high-definition image can be built by time shifting, scaling, and intelligently summing echo energy. The concept of time shifting and scaling, which is based on receiving signals from a transducer array, provides the ability to "focus" on a single point in the scan region. By subsequently focusing at different points, an image is assembled. When initiating a scan, a pulse is generated and transmitted from each of the 8 to 512 transducer elements. The pulses are timed and scaled to "illuminate" specific regions of the body. After transmit, the transducers immediately switch into receive mode. The pulse, now in the form of mechanical energy, propagates through the body as high-frequency sound waves, typically in the range of 1 to 15MHz. The signal weakens rapidly, falling off as the square of the distance traveled. As the signal travels, portions of the wave front energy are reflected. The reflections are the echoes that the receive electronics detect. Signals reflected immediately will be very strong, as they are from reflections close to the surface, while reflections that occur long after the transmit pulse will be very weak, reflecting from deep in the body.
Because of limits on the amount of energy that can be put into the body, the industry developed extremely sensitive receive electronics. Receive echoes from focal points close to the surface, require little if any amplification. This region is referred to as the near field. But receive echoes from focal points deep in the body, are extremely weak and must be amplified by a factors of 1,000 or more. This region is referred to as the far field. In the high-gain (far field) mode, the limit of performance is the sum of all noise sources in the receive chain. The two largest contributors of receive noise are transducer/cable assembly and receive low noise amplifier (LNA). In low gain (near field), the limit of performance is defined by the magnitude of the input signal. The ratio between these two signals defines the dynamic range of the system. Many receive chains integrate the LNA with a variable gain amplifier. Low-pass filtering is needed between the VCA and ADC as an anti-aliasing filter and to limit the noise bandwidth. Often two- to five-pole filter, linear phase topologies are used here. Primary considerations for op amps used include signal swing, minimum and maximum input frequencies, harmonic distortion and gain requirements. Analog-to-digital converters (ADCs) are typically 10- and 12-bit. SNR and power consumption are important issues, followed by channel integration. Another ADC trend is the implementation of an LVDS interface between the ADC and the beam former. By serializing the data coming out of the ADC, the number of interface lines can be reduced from 6144 to 1024 for a 512-channel system. This reduction translates to smaller and lower-cost PC boards. DSPs are used in the imaging system for Doppler processing, 2D, 3D and even 4D imaging as well as a host of post processing algorithms to increase functionality and performance. The key requirements of the imaging system are high performance and high bandwidth. DSPs that run at 1GHz or above can handle the intensive processing needs of ultrasound and the SerialRapidIO peripheral provides 10 Gbps full duplex bandwidth. Some ultrasound systems require a high dynamic range or have functions requiring several cycles. Examples of these functions are spectral reduction and square root functions. When an ultrasound solution requires an operating system, the TMS320DM6446 may fill the need. In addition to having a powerful core and video accelerators to handle the imaging needs, the DM6446 also has an ARM9™ core capable of handling the OS requirements. Signal assembly is accomplished with a digital beam former. This is typically a custom-designed ASIC, but this function has been implemented in different forms of programmable logic. Within the beam former the digitized signal is scaled and time delayed to create the focusing effect in the receive chain. The properly adjusted signals are then summed together across all receive channels and passed to the imaging system. The imaging system can be developed as a separate ASIC, can be a programmable processor such as a DSP. Transmit elements require the control of 100V to 200V of signal swing. This is almost always accomplished with the use of high-voltage FETs. Control of the FETs can take one of two forms: on-off (push-pull) or class-AB linear control. The most popular is the push-pull approach, as it requires a much simpler and lower-cost interface to the FETs. The class-AB approach dramatically improves harmonic distortion but requires more complex drivers and consumes more power. A wide variety of TI products have been chosen by system and equipment manufacturers for their ultrasound imaging applications, including op amps; single, dual and octal ADCs (all with fast-input overload recovery and excellent dynamic performance); digital signal processors; and the VCA8617, an integrated 8-channel, low-power ultrasound front-end IC. TI is also offering the ADS5270, an advanced 8-channel, 12-bit data converter with serialized LVDS Interface. Application Notes
|
All TI End-Equipment Solutions
Products | Applications | Design Support | Sample & Buy 
TI Worldwide | About TI | Contact Us | Investor Relations | Press Center | Corporate Social Responsibility | Careers | Tags | my.TI Login | All Searches | Site Map
© Copyright 1995-2008 Texas Instruments Incorporated. All rights reserved. Trademarks | Privacy Policy | Terms of Use
