The long-term prognosis for medical diagnostics, treatments, and therapies looks healthier than ever judging from the latest wave of microelectronic sensors and sensory implants. They give healthcare providers greater levels of understanding of a patient’s ailments and illnesses, as well as provide faster and higher-accuracy diagnoses and treatments for such conditions. Much of the technology behind these advances comes from progress made with microelectromechanical-systems (MEMS) devices.
The use of these microsystem elements is expected to rise rapidly over the next few years. According to the BioMEMS 2010 report from iSuppli Corp.’s Yole Développement, the microsystems technology market for healthcare applications will balloon from $1.2 billion in 2009 to $4.5 billion in 2015, representing more than 1 billion units per year by 2015 (Fig. 1).
Types of devices vary widely, including pressure sensors, silicon microphones, accelerometers, gyroscopes, optical MEMS and image sensors, microfluidic chips, microdispensers for drug delivery, flow meters, and IR temperature sensors, plus emerging MEMS devices like RFID, strain sensors, and energy harvesting devices.
Some of these microsystems are already commercially available or on the cusp of market introduction. Even those currently under development are expected to hit the market within a few years. In the meantime, existing MEMS IC products continue to find homes in more novel applications in the medical field.
For example, a miniaturized inertial management unit (IMU) developed by Movea uses MEMS three-axis accelerometer, gyroscope, and magnetometer sensors to help produce high-accuracy, wireless nine-degree-of-freedom measurements for rehabilitation and fitness activities. The company’s off-the-shelf, 2.4-GHz, wireless-transmitting MotionPod comes in a fully integrated printed-circuit board (PCB) module measuring 33 by 22 by 15 mm and weighing 14 g.
Basically the size of small wristwatch, it clips onto a strap for easy attachment to the body or can patch directly onto the body. Multiple MotionPods can be networked to gather information simultaneously from different parts of the body for applications such as performance analysis and full-body motion capture.
“Nine-axis sensing provides real-time and precise angular information with a dynamic accuracy of one degree,” says Movea CEO Sam Guilaumé.
Another interesting MEMS sensor development is Freescale Semiconductor’s MPL115A digital MEMS barometer. Worn by patients, the device essentially determines the altitude—wherein higher altitudes demand higher levels of oxygen and vice versa—to conserve oxygen and energy in ventilator systems. It can be used as a smart bandage for negative-pressure wound therapy, employing differential pressure measurements (Fig. 2).
Even conventional analog and mixed-signal ICs are being integrated upward in medical sensing applications. Texas Instruments’ low-power, eight-channel, 24-bit ADS 1298R analog front-end is designed specifically for bio-potential measurements generated by medical instrumentation sensors used in electrocardiograms (ECGs), electromyograms (EMGs), and electroencephalograms (EEGs). In essence, it’s an ECG-on-a-chip.
Looking further down the road, University of Michigan researchers created a piezoelectric MEMS device that generates 10 times more energy than conventional energy harvesters. It portends important implications for powering medical implants in the body as well as wireless sensor networks in automobiles.
The bulk micromachined device is packaged together with other tiny circuit elements, creating a complete vibration energy harvester in a tiny 27-mm3 package. It can harvest vibrational energy between 14 and 155 Hz to produce about 200 µW from 1.5 gs of vibration.
The device charges a supercapacitor to 1.5 V. The supercapacitor then powers up the wireless sensor that’s in place of the battery. The researchers estimate that the harvester can repeat this cycle for 10 to 20 years without any degradation.
The piezoelectric effect is also used in an ultrasonic pressure-sensing echo probe with an aluminum-nitride film that can non-invasively measure the tissues of living bodies. It was developed by Japan’s Industrial Technology Association. The 40-µm thick film measures contact pressure directly, with little effect on ultrasonic transmissions and receptions.
The sensor features mechanical strength and durability. This comes from the placement of a single internal electrode between a pair of thin-film external electrodes that have piezoelectric layers on the inner sides and by completely shielding the internal electrode from the outside between the external pair of electrodes (Fig. 3).
A tiny passive MEMS LC resonator lies at the core of CardioMEMS’s Champion implantable device for monitoring and treating aneurisms, a leading cause of heart failure (Fig. 4). The U.S. Food and Drug Administration (FDA) approved the device for monitoring, with treatment approval expected in the near future.
The RF-addressed wireless pressure sensor needs no batteries, as it’s powered by external inductive coupling. Pressure changes deflect the transducer’s diaphragm and change the LC circuit’s resonant frequency, which can be monitored externally.
The pressure sensor and its wireless antenna are inserted near the heart with a catheter, a procedure that takes only a few minutes. Blood-pressure readings are sent to a wireless scanner. When blood-pressure readings taken over several days remain outside a desired range, doctors can be notified by phone for further action.
CardioMEMS (an off-shoot of the Georgia Institute of Technology, or Georgia Tech) manufactures the electronics reader, signal-processing circuitry, and the transmission circuitry. MEMSCAP supplies the device’s sensor, antenna, and packaging. Results thus far have been very encouraging.
“Patients monitored with the Champion had 38% fewer hospitalizations than the current gold standard of care,” says Mark Allen, a professor at Georgia Tech and a cofounder of CardioMEMS.
Many procedures such as endoscopies and robotic surgeries are becoming simpler and easier to perform thanks to new device developments. Portuguese firm Awaiba Lda developed a wafer-level digital CMOS image sensor that’s customizable to low-power medical needs. The Nan Eye camera measures a mere 0.5 by 0.5 mm—roughly the size of a matchstick tip—and features a resolution of 140 by 140 pixels at a 40-frame/s rate (Fig. 5).
The camera’s lens, based on B33 (Borofloat) glass, is designed so the surface toward the viewed object is flat, minimizing the influence caused by the medium between the lens and the viewed object. Therefore, only the lens’ opening angle is reduced when the system operates in contact with body fluids.
Replete with a 3-µm pitch 250- by 250-pixel rolling shutter, the camera provides clear and sharp color images utilizing Bayer pattern filters. A low-power, 1.8-V battery-operated version is available that dissipates only 600 µA.
Eye Is For Implant
Ophthalmic implants are lately garnering lots of attention. For patients with ailments such as glaucoma, retinitis pigmentosa, and age-related macular degeneration, this hopeful news may soon bear fruit.
For example, STMicroelectronics teamed up with Switzerland’s Sensimed AG to develop a smart contact lens called Triggerfish. It can measure, monitor, and control intra-ocular pressure levels for patients as well as catch early cases of glaucoma. It monitors this pressure for 24 hours and then provides a record to the attending physician. The pressure sensor, a MEMS strain gauge developed by STMicroelectronics, is made on a flexible substrate (Fig. 6).
Circumferential fluctuations in the area of the corneoscleral junction, directly correlated in intra-ocular pressure readings, are measured. This information is subsequently transmitted from a recorder via wireless communications.
Cumbersome glaucoma tests, which require a visit to an ophthalmologist, could soon be history thanks to a test that provides earlier and more accurate detection of the malady. An easy to use self-test probe devised by Eniko Enokov, professor of aerospace and mechanical engineering at the University of Alabama, allows patients to gently rub the eyelid in the comfort of their own home.
“The system detects the stiffness, and therefore infers the interocular pressure,” says Enokov.
Although the probe concept appears simple, the technology behind it is rather complex. It involves a system of microforce sensors, specially designed microchips, and math-based procedures programmed into the probe.
“We went through several years of refinement and modifications to arrive at the current design,” says Enokov. “The innovation in our device is that it is non-invasive, simpler to use, and applies to a variety of situations that are either difficult to address or impossible to test using current procedures.”
Progress has already been made in the use of a retinal prosthesis that can be turned on by light. This non-invasive retina, the product of research at the Imperial College of London, allows the light to control neurons, opening up the possibility for significantly more powerful brain-machine interfaces. A gallium-nitride LED array on a sapphire substrate is used to fire 1-mW/mm2 pulses to activate the neurons.
This development allows biomedical engineers to activate chosen sets of neurons, not simply whatever cells happen to be near the stimulation site, which is the case with stimulation probes. Light can also be used to inhibit a neuron’s firing, whereas probes can only stimulate. Perhaps most intriguing is the engineering of light-triggered brain cells that could begin to pave the way to a hybrid computer that uses an optical link to unite biological and silicon components.
How does the brain function? It’s a question that perpetually drives researchers to seek answers. Based on the some of the latest developments, they’re making deeper inroads toward solutions for a number of disorders.
Last year, NeuroPace Inc. submitted a request to the FDA for approval of a brain implant to treat epilepsy. The company expects to get such approval for its RNS system soon. RNS is a novel investigational device that utilizes responsive brain neurostimulation, significantly reducing the frequency of seizures among people who have a common form of epilepsy that’s otherwise difficult to treat with medication.
“Over the next decade, I believe a variety of closed- and open-loop brain stimulation devices will replace destructive surgical procedures,” says Martha Morrell, chief medical officer at NeuroPace.
The device is one of many neural surgical implants under investigation that serve to alleviate and treat conditions ranging from pain management and depression to Parkinson’s and Alzheimer’s diseases. NeoStim and Trifectas Medical Corp. are just two of many other companies also investigating this field.
The CSI (Central Nervous System Imaging) European project launched last year aims to improve the diagnosis and therapy of brain diseases and lower their associated costs. It hopes to achieve substantial advances in state-of-the-art medical 3D imaging platforms for sensing, computing, and equipment platforms. Members include leading European electronics companies, universities, and scientific research centers.
At the University of South Florida, researchers are using deep brain stimulation surgery to treat essential tremor. Essential tremor, which affects the hands, head, and voice, is three times more prevalent than Parkinson’s disease. This largely hereditary neurological condition can cause uncontrollable shaking that interferes with normal daily activities.
The researchers recently reported that the technique allowed 77% of patients undergoing this stimulation to stop using medications afterward, one year after treatment. The therapy uses an implanted pacemaker-like device to stimulate a targeted region of the brain with electrical impulses, blocking or correcting abnormal nerve signals that cause the tremor. The FDA approved the procedure in 1997.
Microfluidic technology is steadily making significant gains in implantable devices and for lab-on-a-chip technology. Many lab-on-a-chip developments focus on producing low-cost, high-accuracy, and rapid diagnosis of the blood for cancer detection. In fact, that’s the goal of the Miracle (Magnetic Isolation and Molecular Analysis of Single Circulating and Disseminated Tumor Cells) project for detecting cancer in the blood, launched last year by Belgium’s IMEC and its partners.
In another development, microfluidic technology forms the basis of a novel device created by the University of Tokyo that simulates the journey taken by food and oral medications when flowing through the body. Its developers believe the device could be useful for applications such as drug screening and risk assessment of chemicals.
The developers designed a three-stop organ journey in which a micro-intestine and micro-liver absorb and metabolize the chemical before passing it to breast cancer cells—the target tissue. They bundled cells of these three organs onto a glass and plastic microfluidic chip that measures 7.5 by 2.5 cm. Samples are entered into the cells via an inlet that leads sequentially into the three chambers. Results are measured at the output (Fig. 7).
One of the more notable microfluidic-based drug-delivery mechanisms is the Jewel insulin pump from Debiotech, co-developed with STMicroelectronics using Debiotech’s microfluidic MEMS technology. (It’s pending FDA approval.) The pump can be mounted on a disposable skin patch to provide continuous insulin infusion. It promises substantial improvement in the treatment efficiency and quality of life of diabetic patients.
Smart infusion pumps are complicated machines that require careful design considerations. A recent FDA analysis of such pumps found that of the 56,000 medical device reports it received relating to the use of infusion pumps (during a five-year period), more than half of the problems were caused by user errors, where software errors were found to be common.
The FDA uncovered shortcomings in patient education on proper settings and other matters. However, the FDA also gave high marks to the technology behind such pumps and stated that problems were more likely to stem from user errors rather than device defects.
Researchers at the Imperial College of London’s Centre for Bio-Inspired Technology decided to use a biomimetic approach to simplify insulin delivery via a bionic pancreas. Just like a natural pancreas organ, the biomimetic approach relies on two populations of hormone cells to work: beta cells that secrete insulin when the blood glucose is high, and alpha cells that release a hormone called glucagon when the glucose level is low. Both were simulated in chip form.
The organ the researchers devised consists of an electrochemical glucose sensor that penetrates the skin, a microchip, and two small pumps worn on the body (one for each hormone). Every five minutes, the sensor detects the blood glucose level, activating each appropriate pump with a signal that drives a motor. The motor will push a dispensing syringe when needed.
Smelling, breathing, touching, hearing, and seeing all are undergoing intense scrutiny using electronic device technology as basic diagnostic and therapeutic building-block platforms. Implications for better human healthcare are bound to be historic as we await the results of these developments that are just around the corner.