Current Design Trends in Medical Electronics

Electromedical manufacturers and component suppliers are taking steps to improve both performance and cost-effectiveness of increasingly miniaturized, high-precision, portable devices.
Fernando Lynch
One of the staple concepts of futuristic and science-fiction literature and films is the notion that the boundary between humans and machines is dissolving. These scenarios depict us advancing toward an age in which people are partly or mostly robots—or controlled by computers. Though it is unlikely that this will occur during our lifetimes, it is entirely possible that many of us may find ourselves with one or more sensitive electronic devices implanted into our bodies. In fact, products derived from nanotechnology and microelectromechanical systems (MEMS)—with machines fabricated at the millimeter or molecular level—have already become a reality.
For example, the idea of an electronic device stimulating the heart to beat was considered lunacy in 1930, when the first external pacemakers required ac power and penetration of the chest cavity with a probe. By 1958, battery-powered pacemakers were being implanted. Implantable pacers are now considered a mature product (approximately 150,000 annually are installed), with implantable cardioverter defibrillators (ICD) not far behind.
To date, more than one million people on the planet have been implanted with some kind of electronic device. Implantable electronics are now competing with or complementing pharmaceutical and other treatments for such ailments as brachycardia, tachycardia, Lou Gehrig's disease, Huntington's disease, Parkinson's disease, intractable and chronic pain, muscle spasticity, irregular breathing, urge incontinence, diabetes, and deafness. Implantable electronic products include drug pumps, monitors and delivery systems, cochlear implants, and neurostimulators. Experimentation is already under way on electronic retinal implants that could lead to at least a partial cure of blindness.
Although today the average age of a pacemaker recipient is 70, demographic statistics indicate that an enormous number of now-middle-aged baby boomers will be prime candidates for implantable products in the not-too-distant future. The pressure on manufacturers is to produce devices that are smaller and lighter, with lower total system costs. In addition to these ongoing developments in implantable devices, the trend toward portability and delivery of care at the bedside is accelerating the development of a range of next-generation monitoring, display, and testing equipment designed to be more compact, accurate, and versatile. Products that can meet these objectives while providing lower power consumption, superior functionality, or ease of manufacture will fill a profitable niche in this burgeoning industry. This article outlines several areas of electromedical product design in which the limits of conventional semiconductor technology are being extended through the use of existing circuits in new ways or through the combination of several cell or block functions into a single electronic system (see Table I).
Application COMPONENT DESCRIPTION
Implantable Pacers, Defibrillators, and Neurostimulators
Input protection TVS (transient voltage suppressor) die, 5–20 V, variety of sizes and metal contacts.
TVS, 3–12 die array, monolithic or "chip-on-strap."
TSPD (thyristor surge-protection device), 3–12 monolithic die array, provides smaller footprint than conventional TVS.
ASICs ASIC, analog or mixed signal ultra-low power.
Diode/bridge Schottky die, 20–100 V, variety of sizes and metal contacts.
Schottky die, single and dual, 40–70 V, variety of sizes and metal contacts.
Input protection,
blanking/tip switch MOSFET die, 0.110 sq in., 1 KV, 13.5 ohm.
MOSFET, 1 KV, 13.5 ohm.
MCM, 6-array MOSFET (MSAFA1N100D).
High-voltage
switching bridge IGBT die, 0.160 sq in., 1200 V, 55-A surge.
MCM, half-bridge, capacitive-coupled, IC-driven IGBT.
Thyristor-based (SCR and Triac) up to 1200 V.
Schottky die, single and dual, 40–70 V, variety of sizes and metal contacts.
Charging circuit Rectifier, monolithic-microwave surface-mount (MMSM) package, flip-chipable, up to 70 V, 20 mA.
Rectifier die, up to 1200 V, 55-A surge, standard and ultrafast recovery.
Schottky, 500 V, 1 A, on silicon-carbide substrate.
Rectifier, up to 600 V, ultrafast recovery.
Voltage regulation Zener die, 1.8–300 V, variety of sizes and metal contacts.
ASICs ASIC, analog or mixed signal, ultra-low power.
Diagnostic Imaging and MRI
MR surface coils PIN diode, axial and stud mount for receipt and transmit.
MR transmitters PIN diode, 1–3 KV, 13 W, stud mount for high-power transmit.
MR receivers PIN diode, 1 KV, 10 W, axial and stud mount.
Hearing Aids
Class D amplifier Ultra-low-power, low quiescent current, true 1-V operation, thin die.
Portable Diagnostic Meters (Glucose, Oximetry, Pulse Analyzers)
Analog power management Analog IC interfaces with the microprocessor for analog functions such as measuring current, temperature, or time. Very low quiescent/standby current (~1–2 µA) and operating current (a few hundred µA).
ESD protection Polymer-based bidirectional transient-voltage suppressor. Reacts almost instantly to the transient voltage and effectively clamps it below 60 V, resulting in less voltage stress during the clamp period and greater IC protection.
Silicon-based bidirectional transient-voltage suppressor. Low clamping voltages at 1.7 and 3.3-V levels.
Step-up dc-dc converter High-efficiency (>90%) boost converter IC. Low (typically 16 µA) quiescent/standby current, low (<1 µA) shutdown current, and adjustment via analog reference or direct PWM input.
Power regulation Low-cost Schottky rectifiers. Applications include battery charge/discharge regulation and general-purpose/low VF rectification.
LED output detection Visible enhanced photo detector diode. Low-cost die with 1-sq-mm active area or clear SMD package for low dark current and low noise.
CCFL backlight inverter Direct-drive, high-efficiency IC or complete module with 100:1 wide-range dimming for extended battery life.
Table I. A sampling of electromedical design application areas and selected available components.
SYSTEM SOLUTIONS FOR IMPLANTABLES
Devices such as implantable pacers or defibrillators are really miniature computers that employ sensitive, low-voltage, low-power, application-specific integrated circuits (ASICs) to monitor, regulate, and control the delivery of electrical impulses to the heart. Implantable cardioverter defibrillators (ICDs) have been in common use for a number of years. When it detects a potentially life-threatening cardiac fibrillation, the ICD applies a high-voltage pulse between two electrodes connected to the heart. The pulse can be as high as 800 V, with the resulting current (during a few milliseconds) reaching several tens of amperes.
The high voltage is generated and stored on a large capacitor through the use of a charge pump. Normally, the shock is delivered to the heart via a two-phase pulse. Figure 1 shows a principal block diagram of a two-phase defibrillator system that features a typical high-voltage bridge required to generate the biphasic pulse. The application consists of two identical half bridges, each having two switches—one to ground and the other to the high voltage. Insulated-gate bipolar transistors (IGBTs) are very often used as the switch element, since they offer minimum on-resistance relative to silicon area. The high-side IGBT requires a gate voltage that is approximately 10 to 15 volts higher than the voltage to be switched. Normally, a transformer is used for level shifting between the high-voltage controller and the switch. Figure 2 shows a principal block diagram of the components required for one half bridge.

Figure 1. Principle diagram of a defibrillator system. Notice the two bridges (see Figures 2 and 3).
The use of a transformer in such a design, however, can pose significant disadvantages. It often involves separate discrete components from different vendors, which can complicate manufacturing and increase costs. Transformers also tend to be bulky and difficult to handle in production, decreasing ultimate device reliability.
Half-Bridge Module Technology. To overcome some of these limitations, electromedical design engineers are currently developing high-voltage, half-bridge modules for ultralow-power, low-voltage applications in which space and quiescent current are of primary concern (Figure 3). These designs seek to provide solutions for low-speed and low-duty-cycle, high-voltage switching applications.
Figure 2. Half-bridge with transformer as isolation for the high-side driver.
With a low-voltage interface, the bridge can be controlled directly from a CMOS-level controller chip (logic-level inputs), with complete integration of all necessary components for a half bridge. Features include ultralow quiescent current as well as a small form factor in a ball-grid-array (BGA), multichip-module (MCM) package. The BGA packaging techniques allow for complete circuits to be packaged at considerable size and cost savings. Their flip-chip construction simplifies assembly, as wire bonds and associated testing are eliminated.
Figure 3. High-voltage half-bridge module with capacitor as isolation for the high-side driver.
In order to attain the smallest possible form factor and quiescent current, the half-bridge design calls for high-side-level shifting and charge transfer via capacitive coupling achieved by two specialized integrated circuits (IC1 and IC2 in Figure 3). Low-gate-charge IGBTs enable the use of lower capacitance on the isolation/charge-transfer capacitor.
Half-Bridge Circuit Function. Operation of the half-bridge technology occurs through the two logic-level inputs, LO_EN and HI_EN, which respectively control the low- and the high-side switch turn-on (Figure 3). The LO_EN signal drives the gate of the lower-side IGBT via a level shifter and driver in IC1. The HI_EN pin enables a high-frequency oscillator in IC1. The output of this oscillator is level-shifted to the IGBT gate-drive voltage (VDD) and drives the isolation and charge-transfer capacitor. The capacitor's other side is connected to IC2, where the clock signal is rectified to produce the gate emitter voltage. IC2 also has a switch that shorts the IGBT's gate to the emitter when no clock signal is present (Figure 4).

Figure 4. Detailed block diagram of IC2, part of half-bridge module.
Protective Components. Another important design aspect with critical implanted devices is that the sensitive electronics be adequately protected, as when, for example, a patient with an implanted pacer has a heart attack and requires external defibrillation that entails being shocked with thousands of volts. The portfolio of protective components includes products such as zeners, high-voltage diodes, transient-voltage suppressors (TVSs), and thyristor surge-protection devices (TSPDs). It can be confusing to the circuit designer to know when to use each type of protective element, which depends on such variables as the power dissipated or the current generated by clamping external or internal surges. For instance, the TSPD is a crowbar device, whereas the diode-based suppressors (zeners and TVSs) are clippers. Therefore, TSPD-based suppressors can be used in an application if a particular voltage drop to on-state voltage level is permissible and if the impedance of the protected line will allow current drop below the holding current value.
IMPROVING AMPLIFIER TECHNOLOGY
Among the most prominent applications for electromedical devices is the hearing aid. The production of hearing aids is particularly challenging for integrated circuit design engineers because of the constraints inherent in comfortably fitting ever-smaller, ever-more-sophisticated devices for individuals with hearing loss.
As in most consumer electronics products, the drive to decrease size and power consumption while increasing performance dominates hearing aid integrated-circuit requirements. This imperative has resulted in hearing aids changing from large behind-the-ear devices to today's completely-in-the-canal (CIC) aids, which are so small that they generally cannot be seen at all. As hearing aids move further into the ear, the size of the ear canal determines the overall size requirements of the aid, and therefore of the integrated circuit.
An additional challenge for design engineers is the power requirement. As hearing aid sizes have decreased, battery manufacturers have been obliged to decrease battery sizes, thereby reducing the total battery power available. Current models typically run off a single one-volt battery and demand even smaller high-performance ASICs that also operate at extremely low power levels. The creation of ultra-low power blocks such as compression amplifiers and Class D amplifiers designed specifically for the hearing aid market have resulted in ASICs with maximum performance and minimum power consumption and size.
Although state-of-the art analog hearing aid ASICs are a mainstay for many circuit manufacturers, there is a growing demand from the industry for both digital hearing aids and extended-life analog models. Each of these requirements presents its own set of difficulties. Digital hearing aids can match the fidelity of analog versions, but with added complexity in chip design and relatively high current consumption. The challenge is to design 16-bit sigma-delta converters with extremely low power consumption and to create efficient digital-signal-processing engines.
Extended-life analog hearing aids push the limit of ultra-low-power analog design expertise, since power consumption must be radically reduced. The goal is to attain anywhere from several weeks to months of device use without battery replacement—compared with today's aids that require a new battery approximately once a week. Engineers are designing new integrated circuits to address these challenges for both digital and extended-life analog hearing aids.
OPTIMIZING LCD DISPLAYS
As with implantable devices, many of the portable medical instrumentation and information systems coming on the market today are designed to meet the objective of enhancing patient care while at the same time lowering healthcare costs. Applications with tremendous promise in this area include monitoring equipment that could link caregivers and patients at home via the Internet as well as hand-carried ultrasound and other imaging devices that could open up new ways to improve treatment, emergency response, and service to remote locations.
Achieving high efficiency in powering the display systems used in these devices—a major consumer of battery power—is extremely important in sustaining reliable operation and extending battery life. One approach to solving this problem is evident in recent efforts to design an effective cold-cathode fluorescent lamp (CCFL) backlight inverter for portable medical applications.
Direct-Drive Efficiency. The direct-drive topology of the CCFL inverter eliminates the inductor and resonant capacitors necessary to implement a conventional Royer oscillator–based inverter solution. The design uses a fixed-frequency pulse-width modulator (PWM) control circuit connected directly to a high-voltage transformer. By removing cost- and power-hungry components, efficiency can be improved and component size reduced. As a liquid-crystal display (LCD) with a CCFL backlight can consume from 70 to 80% of the total system power of the unit, use of a direct-drive inverter can gain 30-60% in efficiency and translate directly into longer battery life.
Wide-Range Dimming. Many medical applications require low brightness settings from the LCD display. For example, patient monitoring equipment requires low brightness levels at night so that patients are not kept awake, but the display must be bright enough for the attendant to read. This specific functionality can be attained via wide-range dimming technology.
The light emitted by a CCFL bulb is proportional to the current flowing through it. There are two ways to control a continuous ac current: either by adjusting the amplitude (analog dimming) or by varying the amount of time that a burst of the full current is present (digital or wide-range dimming). This latter approach not only lowers the power consumption compared with conventional amplitude control methods, but also achieves much lower dimming levels (by ratios of 100:1) without causing lamp flicker or partial one-sided extinction of the bulb at low levels.
Lamp Life. Another major benefit of digital dimming technology is enhanced lamp life. Extensive testing has demonstrated that, at 2-mA average current, bulb degradation with digital dimming is approximately 23%, compared with 38% for analog dimming. The slowest aging conditions of the lamp occurred at a range of between 2 and 40% duty cycle for digital dimming.
CONCLUSION
With the increasing role of sophisticated electronics in advanced medical devices and instrumentation, the forces of continuous improvement are following a path not unlike that of consumer electronics, with an emphasis on size reduction, portability, and extended battery operation. The life-sustaining role of many electromedical devices lends a particular urgency to efforts to improve battery performance, protect critical electronic circuitry, or provide higher-quality displays. The examples discussed in this article represent merely a selection of the areas in which designers are extending the frontiers of system-engineered solutions to create safer and more effective medical electronics.
Fernando Lynch is corporate medical market segment manager at Microsemi Corp. (Garden Grove, CA). In addition to projects in diverse industries such as military, aerospace, and telecommunications, the company specializes in developing innovative semiconductor solutions for manufacturers of medical products.