Increasing healthcare costs, the prevalence of chronic diseases, an ageing ‘baby boomer’ community, and large emerging markets in countries such as China, India and Brazil, are creating tremendous demand for affordable, robust and reliable medical devices to improve the treatment and care of millions of patients worldwide, and to cure an increasing range of diseases. In turn, medical device designers are exploring new technologies from various industries to improve the diagnostic, monitoring and therapeutic capabilities of next-generation devices.
Whether talking about home-based testing, monitoring and diagnostics, clinical equipment or imaging applications, two trends have emerged to make medical devices more affordable and more accessible for patients: miniaturisation and portability. Today, medical manufacturers are moving entire systems—from home-based and clinical devices to imaging applications—into a portable unit the size of a cell phone or smaller. What was once huge equipment tethered to a wall has become available in mobile clinics, ambulances and in a doctor’s bag for house calls.
With the trends toward miniaturisation and portability come implications for reliability, form factor, power consumption and battery life when considering the semiconductor technologies used in medical applications. Today’s medical instruments, from home-based and clinical applications to imaging devices, are often very complex and highly use-specific. However, in addition to their core elements, medical devices also include unique functional blocks to complete their task. These changing features and requirements, complex functionality in a small footprint, low power, high accuracy and reliable operation make medical devices an excellent market for reprogrammable non-volatile semiconductor technologies like mixed-signal and low-power Flash-based FPGAs.
Semiconductors in medical applications
Traditionally, designers have leveraged standard products for their medical devices. Yet, as the pressure to deliver superior and less invasive healthcare grows, so too does the rate of medical device technology innovation. Advancements in the semiconductor industry are enabling this acceleration by delivering smaller, cheaper, more reliable solutions able to integrate critical functions such as analog circuits and embed microprocessors in a single device.
New packaging technologies and methodologies have also improved the cost and reliability of the assembly of packages with small footprints. Chip-scale packages, for example, have reduced semiconductor package sizes to the dimensions of the die inside. Other innovations, such as the use of a known good die or foldable printed circuit boards, have also provided considerable miniaturisation in the overall space occupied by the electronic portion of the medical device.
With low-power requirements and an allowance for the use of small batteries, small processors are helping to fuel the emergence of new wearable and implantable medical devices. Similarly, several advancements in FPGA technologies are furthering the development of portable, miniaturised medical applications. For example, the function of several standard chips can be integrated into one reprogrammable FPGA device, reducing footprint, cost and power while increasing reliability.
This integration has also addressed the obsolescence issue that plagues risk-averse medical product designers who use standard products. These customers need field-proven technology that will not become obsolete, enabling them to continue to meet stringent medical instrument guidelines. Certainly, with their inherent flexibility and re-programmability advantages, FPGAs have already played a significant role in the growth of medical electronic devices. Looking forward, the semiconductors with the greatest opportunities are those that can satisfy the more stringent demands of this marketplace – high reliability, small form factor and extremely low power.
Reliability
Complex and shrinking medical instruments must be reliable enough to withstand a wider range of operating conditions. It used to be enough for the machine to work in a spotless operating room, clinic or laboratory. Today, however, modern medical devices must deliver the same accuracy and reliability in a mobile clinic or in an ambulance. Ultimately, however, the medical device is only as reliable as the semiconductor devices used within. Therefore, it is critical for the component, such as the ASSP, microprocessor or FPGA, to be inherently reliable.
The reliability of the device and the system can be impacted by power-up requirements and firm-error immunity of the devices, as well as functional integration and the number of devices on the board. Well suited to medical applications due to their inherent high reliability, non-volatile Flash-based FPGAs are single-chip and live at power-up. In addition to the system operation benefits and the reduction in power consumption, the live-at-power-up feature offered by Flash-based FPGAs contributes to the reduction of the number of total components used on the board, increasing reliability.
In comparison, significant additional circuitry may be required when utilising SRAM-based FPGAs. In addition to a boot PROM and/or additional system memory for unsecure configuration code, a CPLD may be needed for system configuration and supervisory tasks. Clock and reset signal generation is also required upon power-up to help initialise components onboard. These issues reduce reliability, add complexity and cost to the system design, and slow down the development process.
Reliability can also be severely impacted by the susceptibility of the device to single-event errors where alpha and neutron radiation causes loss of configuration data. Programmable logic devices based on SRAM technology, for example, are susceptible to ‘soft errors,’ the transient corruption of a single bit of data, and ‘firm errors,’ the loss of the underlying FPGA configuration, which can cause system-level functional failure.
Fortunately, neutron and alpha radiation do not have adverse effects on true non-volatile Flash-based FPGAs at ground and sea levels or at high altitudes, making them more suitable for medical applications where failure is not an option. Further, the integration of diverse functions into a single chip can increase reliability through the elimination of discrete devices. For example, the integration of analog functions (sensing) and non-volatile memory for data-logging can significantly improve reliability.
Small form factor
For a large portion of medical applications, the enabling factor is getting just enough processing power into a given space. Portable glucose meters, for example, are about the size and shape of a PDA. In the case of implanted devices, such as cardiac pacemakers, neurostimulators for treating central nervous system disorders, and hearing aids, advances in semiconductors have led to significant size reductions, enabling less intrusive placement of these devices.
For example, advances in packaging technologies and methodologies for small footprint devices have improved such that miniaturisation is not only possible, but cost effective while maintaining reliability. Chip scale packaging, for example, combines the size and electrical performance of bare die assembly with the reliability of encapsulated devices. With some chip-scale alternatives, the semiconductor package has been reduced to the dimensions of the die inside, making it ideal for size-sensitive medical applications.
As mentioned previously, single-chip, Flash-based FPGAs offer increased reliability due to a reduction in the number of components used on the board. Similarly, the integration of diverse functions into a single mixed-signal FPGA also reduces component count. Certainly, the removal of these devices also minimises board space and, ultimately, the form factor for the end application.
Low power
In an implanted device, the benefits of minimal power consumption, minimal heat dissipation and extended battery life are clear. However, as electronic components get smaller, medical devices that were once fixed are portable, making them sensitive to power consumption.
Ultrasonic imagers, once stationary due to their size and weight, are now available in laptop-like form factor, and handheld sizes that weigh from three to seven pounds. Automatic external defibrillators are now commonplace. Portable oxygen concentrators extract oxygen from the air and can be carried over the shoulder like a purse. All of these devices are enabled by ultra-low power consumption and extended battery life.
Today, the design considerations regarding power consumption can be complicated. For designers of portable medical electronics, selecting chips on advanced process geometries means not only the ability to achieve higher levels of integration with a smaller overall die size, but also increased leakage and increased static power. Because of the complexity of medical devices, power consumption in standby mode becomes more critical, making low leakage a design requirement. As a result, a mainstream process technology, such as 130 nm, with lower leakage is often more desirable than an aggressive technology node that increases standby or static power consumption.
Today, FPGA technology is increasingly utilised in low-power applications, which makes achieving lower system power an increasingly important challenge. The various FPGA technologies have significantly different power profiles, and these differences can have a profound impact on the overall system design and power budget. The power profile of an FPGA is determined by the base technology of the interconnect element used. For example, non-volatile, Flash-based FPGAs utilise a single Flash cell to form their efficient interconnect.
In comparison, SRAM-based FPGAs utilise a six-transistor SRAM cell to perform the interconnection between routing lines and logic cells, resulting in higher static and dynamic power. When evaluating the respective programmable technologies, there are five components to the FPGA power profile that must be considered: inrush current, configuration current, static power, dynamic power and power during sleep/low-power mode.
Unlike SRAM-based FPGAs, Flash-based FPGAs have no power-up or configuration power components. They offer lower static and dynamic power consumption over a large temperature range and support low-power modes that allow further minimisation of power consumption when the system is idle.
Home-based applications
Traditionally, home-based applications have been used for testing and monitoring only. Today, however, home-based applications are expected to do much more. Next-generation medical applications are now communicating with each other or are integrating several different medical devices into a single monolithic unit. Most mainstream glucose meters, for example, now communicate wirelessly with the insulin pump to adjust the insulin.
However, advances in insulin pump devices have allowed periodic intelligent measurement of the patient’s glucose in addition to insulin injections based upon the glucose results. Often battery-powered, all of these feature-rich, home-based medical devices, such as home infusion pumps, digital ear thermometers, respiratory therapy products, and portable blood glucose, cholesterol, and blood pressure monitors, require low-power operation, small form factor, and high reliability.
Several functional blocks are common to most portable home-based applications, including bio-sensors; amplification and analog-to-digital conversion of the sensor input; power management, such as system power control and power sequencing; microcontrollers for low-power operation and control; and user interfaces and control for human machine interface (HMI) and display. Additional requirements may drive the need for interfaces to multiple storage standards, wired and wireless interfaces, and audio feedback or notification.
These devices may also require integrated functionality, such as data-logging and wireless communications, while maintaining the same or lower power, footprint and cost. For example, blood pressure meters are now benefiting from more extensive data-logging features as well as communication ports for real-time sharing of information with the healthcare provider.
| Tel: | +27 11 315 8316 |
| Email: | [email protected] |
| www: | www.asic.co.za |
| Articles: | More information and articles about ASIC Design Services |
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version