Medical Device Link.

Originally Published IVD Technology April 2002

Detection Technologies

Taking a fresh look at sensors

Advances in technology are changing how and where diagnostic sensors are used.

Gillian McMahon, Edel Minogue, and Dermot Diamond

Sensor technology is merging with communications and information technology to probe the world of the human body and report back using some of the most exciting inventions in recent times.

In terms of healthcare, this revolution is being spurred by two trends—increased economic pressures and reduced patient stays in hospitals (driven primarily by insurance companies). Keeping patients out of hospitals and reducing the length of time that they stay are putting increased strain on all tests being done, especially in emergency rooms, intensive care wards, and outpatient services.

As this pressure on hospital testing builds, some of these services have decentralized and moved into the community to the physician's offices, pharmacies, and medical-assistant-run laboratories where rapid diagnosis and monitoring can be done. With these centers, the patient has the benefit of easier access to testing, shorter travel times, and reduced waiting times for appointments and results. Also, an emerging trend is that tests are being carried out at home, either by a caregiver or by the patients themselves.

Overall, this situation has created a rapidly rising demand for new noninvasive and in vitro sensor technologies to speed up testing. Such new products are used anytime, anywhere, and are thriving in the climate of telemedicine and remote, wireless communications. They generate rapid results, making it possible for users to obtain real-time information about their health. Three types of sensors are contributing to this shift in clinical testing: noninvasive, or wearable, sensors; in vitro sensors, which are used in point-of-care (POC) or near-patient testing devices; and in vivo, or biomimetic, sensors.

Designing and Developing Sensors

New applications often require new technologies which do not come into being easily. It is not simply a case of making conventional laboratory instruments smaller or putting a sensor into a piece of clothing or into the human body. But despite these challenges, successful devices are already available on the market, and their quality is improving all the time. Areas to consider when developing a sensing device include those described in the following sections.

Medical Need. The device must satisfy a medical need by delivering new benefits to patients, offering a new way of monitoring a condition, developing a test that is cheaper, or creating a device that has significant advantages over currently available technology.

Market Potential. The new sensor to be developed must have commercial potential. If it is to be an improvement on already available technology, it must also be determined whether there is a demand for such a device in the market.

Intended Use. The sensing device's intended use must be clear from the outset. For example, most POC instruments utilize disposable test strips or cartridges. The cost of these strips and cartridges must be taken into account since the cost of ownership is a huge issue. Such is the case when considering not only the initial outlay for a sensor or sensing device but also the expense of consumables, maintenance, and possibly accessories. Wearable sensors and in vivo diagnostics usually monitor continuously, so calibration and self-cleaning are issues that should also be addressed. In addition, the barrier to use must be lower than the perceived benefits so that users find the technology user-friendly and will actually use it.

Sensing or Detection Element. A chemical, biological, or physical sensor produces a signal (e.g., voltage, absorbance, heat, or current) in response to an event, such as binding between two molecules. In the case of a biological or chemical sensor, this event typically involves a receptor (e.g., macrocyclic ligand, enzyme, or antibody) binding with a specific target molecule, known as the analyte, in a sample. On the other hand, physical sensors measure inherently physical parameters such as current or temperature, which can be due to reactions occurring. In any case, the signal is transduced by passing it to a circuit where it is digitized. The digital information can then be stored in memory, displayed visually on a monitor, or made accessible via a digital communications port.

Since it is essential that the sensor's measurement be detected, it is necessary that a mode of transduction such as electrochemical signals (electrochemical-based sensors), optical signals utilizing changes in fluorescence or absorbance (bulk optodes), or plasmon resonance be available. With most sensors, transduction is accomplished electrochemically or optically.

Instrumentation Platform. In the broadest sense, the instrumentation platform is an inherent part of the design of a particular device. The platform for a wearable sensor is self-explanatory (e.g., smart vests, a bra), while the platform for a POC device depends on the type of technology implemented to fabricate the device. Biomimetic sensors are essentially stand-alone systems that mimic their biological counterparts.

Microelectromechanical systems (MEMS) are a technology that combines computers with tiny mechanical devices such as sensors, valves, gears, mirrors, and actuators embedded in semiconductor chips. As such, MEMS are their own instrumentation platform. While the term MEMS is most commonly used in the United States, other names are also used to describe this technology: micro systems technology (MST) in Europe, and micromachining in Japan. All of these involve making small-scale (1 µm–1 mm) devices that have functionality in physical domains outside of the integrated-circuit world, such as solid mechanics, fluidics, optics, acoustics, magnetics, and others. MEMS are already an established technology with microdevices that integrate sample handling, fluid transport, reagent mixing, analysis, detection, and output on the same tiny devices. Advantages of such devices include the integration of multiple steps in often-complex analytical procedures, diversity of application, submicroliter consumption of reagents and samples, and portability.¹

Adapting the MEMS technology for healthcare, BioMEMS apply microdevices to biological and medical problems. This technology has further expanded into two other areas: biomedical MEMS, which deals with in vivo applications such as biosensors, drug delivery, biotelemetry, and minimally invasive precision surgery; and biotechnological MEMS, which deals with in vitro applications.²

From MEMS emerged the concept of lab-on-a-chip. This term describes a chip which includes systems for pretreatment, mixing, and measuring microscopic liquid samples with reagents; moving the mixtures to an integrated reaction chamber; and determining the results with an onboard detector. Much research has been carried out on the fluid flow and mixing characteristics in these microchannels, resulting in the field of microfluidics.³ An example of the advances that have been made in this area is the recent agreement between Renal Plant (Southborough, MA) and Micronics (Redmond, WA) to develop a portable, disposable lab-on-a-chip–based artificial kidney for hemodialysis.⁴ Implementing the principle of microfluidics, the product is intended to replicate normal kidney functions and enhance medical outcomes for patients with end-stage renal disease.

Fabrication and Materials. The proper construction of a sensor device is critical for its success in a given application. For POC and wearable sensors, the construction must provide protection from the external environment while also facilitating the functioning of the device. Hence, a compromise must be reached between sealing the device from external elements, while still allowing sampling, analysis, and communications to take place.

In terms of materials, since the lab-on-a-chip concept emerged from the electronics industry, first-generation sensors based on this technology were made of silicon. However, it soon became clear that silicon was incompatible with sensors being used for biological purposes. A gradual shift from silicon to plastics occurred, and new fabrication technologies emerged around the use of plastics.⁵ The introduction of polymers simplifies the fabrication process for sensors, improving mass-production possibilities, while at the same time reducing costs and making the sterilization process easier.⁶ Common polymers such as polymethylmethacrylate, polystyrene, and polycarbonate all lend themselves well to the fabrication processes of hot embossing, injection molding, and micromilling.

Another factor is the development of nanotubes and self-assembling molecular structures and their applications in nanodevices. Nanosensors will make it possible for medical nanodevices to monitor environmental states at three different operational levels:

• Internal nanorobot states.
• Local and global somatic states (inside the human body).
• Extrasomatic states (sensory data originating outside the human body).⁷

Validation and Calibration. Some sensing devices need no calibration while some others require regular calibration. Wearable sensors and biomimetic sensors present huge challenges in the areas of validation and calibration, such as how to calibrate and how to obtain or create a suitable calibrant. These issues are related to obtaining reproducible and accurate readings, especially in the case of continuous monitoring and functioning.

For POC devices, the calibration issue is more straightforward. Many of these devices are similar to scaled-down laboratory instruments, so their calibration and validation can be based on protocols already in place. Some POC sensing devices have built-in electronic controls, while others have external quality control (QC) protocols in place. These external quality controls can be a separate test strip or a vial of QC sample, such as blood, plasma, or saliva. QC samples should be periodically processed in the same manner as patient samples to monitor the ongoing performance of the entire analytical process.

For quantitative tests, two levels of liquid QC are usually carried out. For hematology, coagulation, and blood gas tests, two different levels of control material are required during each 8-hour period of patient testing and each time there is a change of reagents.⁸

For qualitative tests, a positive and a negative control should be included with each run of patient specimens. Multiparameter urine-chemistry dipsticks can be an exception to this rule.

With all such POC tests, it is also important to verify reagent performance because reagents may be stored under a variety of uncontrolled conditions. Several methods may be used, such as comparing the reagents with reference materials or checking against routine controls. Where individually packaged reagents or kits are used, criteria should be established for monitoring reagent quality and stability. Running wet controls is a typical way of validating the reagent quality and operator technique.

Communications. In order to address the issue of connectivity among POC devices, the Connectivity Industry Consortium (CIC) was formed to set up a standard communications platform for all devices.⁹ The CIC identified five user requirements: bidirectionality, device connection commonality, commercial software interoperability, security, and QC and regulatory compliance. After 14 months the CIC finalized these standards, which now permit POC instruments that are made by different manufacturers to connect to laboratory and hospital information systems.¹⁰ Under these standards, new devices should seamlessly link into an existing data management system without additional expense. Some software vendors, in fact, are already advertising their products as being "CIC-compliant."

Traditionally, monitoring units consisted of a sensor for a particular analyte, some circuitry to convert the signal into a digital result, and a cable to communicate with a base station (e.g., a laptop). Advances in technology now enable sensors to be integrated with wireless communications technology that frees the sensors from being physically attached to a base station. Interest in such freestanding monitoring units is growing rapidly in healthcare, since they offer the potential for developing integrated networks of sensing devices that can detect, diagnose, and monitor various health problems. The merging of computing with wireless communications systems and sensors has led to increased accessibility to real-time information in digital form. The communications network that has assembled over the past decade and continues to attract huge investment will fuel demand for more sources of health-related information and data.

One global industrial alliance that has explored the potential applications of wireless communications in healthcare is Bluetooth. With the participation of such companies as Nokia, Ericsson, Microsoft, Intel, IBM, and 3Com, Bluetooth has developed a technology that incorporates an inexpensive, radio-frequency-based system onto a small microchip using low power for short-range communications.¹¹ The chips can be integrated into electronic devices, allowing seamless data transmission among them and eliminating the need for cumbersome and costly peripheral cabling. With this technology, wireless downloading of results can occur from all types of sensing devices, including wearable, POC diagnostics, and biomimetic sensors.

Figure 1. The integration of wireless technology with sensing technology, showing selected examples of physical, chemical, biological, and optical targets, as well as sensor parameters in these areas. Photo courtesy Graviton Inc. (click to enlarge)

Companies such as Graviton (San Diego) have explored other applications of wireless communications and developed wireless distributed sensor networks. By integrating sensors and code division multiple-access spread-spectrum wireless transmitters, Graviton has created custom wireless-sensor networking configurations that allow for continuous, uninterrupted data transmission (see Figure 1). Graviton has also developed self-organizing sensor networks that permit seamless interconnection, filtering, and routing of data to the appropriate data hub. Applying this communications technology to an autonomous medical device could result in managing and filtering results that are obtained from a medical sensor and passing specific data to other electronic devices.

Regulatory Framework. Many regulations must be considered when developing sensors for medical applications, especially if the device will be used at home or for self-testing, or will be available over the counter (OTC). While the regulations related to wearable and biomimetic sensors remain unclear, regulatory requirements for POC devices are quite definite. The main barriers to market for POC devices are premarket review regulations established by FDA and under the Clinical Laboratory Improvement Amendments of 1988 (CLIA).

Wearable Sensors

Wearable sensors incorporate wiring and electronics into clothing, eyeglasses, and even shoes. Although wearable sensors are noninvasive, when they are capable of analyzing human sweat and tears, they are functioning like in vitro diagnostics.

Remarkable progress has been made in the area of intelligent fabrics during the past few years, and these materials have now been developed such that they can be employed to solve real biomonitoring problems. Embedding sensors into fabrics for measuring parameters such as stress, strain, temperature, and pH increases the potential use of this technology.

One example is the intelligent knee sleeve that is currently being used by Australian Rules football players to monitor knee strain or injury during training.¹² Strapped to the knees, the sleeve provides feedback to players by emitting an audio tone.

The smart bra is an example of how wearable sensors can be useful in these applications.¹³ This garment can tighten and loosen its straps, or stiffen and relax its cups, by responding to real-time data from sensors in the fabric. These sensors monitor parameters such as breast bounce, heart rate, and breathing. Overall, the smart bra restricts breast motion and increases support, thereby reducing breast pain and strain, especially during exercise.

Figure 2. The cystic fibrosis wristwatch: the sensor wells for sodium and chloride measurement (a); the sweat collector collects the sweat sample in a coiled tube before it moves into the sensor (b); and the complete wristwatch (c). Illustration courtesy South of England Cochlear Implant Centre (click to enlarge)

Another step in the evolution of wearable biochemical sensors is the development of a test device for cystic fibrosis.¹⁴ A small portable detector that is worn like a wristwatch gives a result in minutes, rather than the full day that is typically needed for a laboratory test. The wristwatch device uses an electric field to push pilocarpine nitrate into the skin, thereby dilating the pores. Sweat is then sucked up and stored in a duct in the watch. The sample is analyzed by a sensor that is commonly used for blood analysis and records the levels of sodium, chloride, and potassium ions. A laptop computer analyzes this data and gives an immediate diagnosis. (see Figure 2).

The GlucoWatch by Cygnus (Redwood City, CA) is also a wristwatchlike device that measures glucose in the interstitial fluid as a low electric current pulls glucose through the skin.¹⁵ Another wearable sensor that takes a similar approach measures blood oxygen transcutaneously.¹⁶

Other sensor inventions that have been developed can conduct tests in the human eye. Last year, CIBA Vision (Duluth, GA) developed a disposable contact lens that measures the glucose levels in tears.¹⁷ While holding a power pack close to the eye that is wearing the lens, a button is pushed, and a light flashes. The wavelength of the light that bounces back to the power pack indicates the level of glucose. Data from studies showed that the results obtained in this way were consistent with standard blood measurements. In addition to correcting the vision of diabetics, this contact lens will soon be able to carry out glucose monitoring as well.

Point-of-Care Devices

Some of the latest POC devices are in the areas of pregnancy testing and ovulation prediction, diabetes monitoring, coagulation testing, HIV and infectious-disease testing, tumor markers, cardiac markers, and drugs of abuse.

Figure 3. The OneTouch Ultra glucose test strip from Lifescan. The electrode is assembled in several layers on a plastic substrate using a screen-printing technique. Photo courtesy Lifescan (click to enlarge)

In the diabetes management market, many POC diagnostics are available for glucose self-testing. The OneTouch Ultra blood glucose monitoring system from Lifescan (Milpitas, CA) features a 5-second test time, advanced electrochemical biosensor test strips that require only a 1-µl blood sample, and easy downloading of data to a personal computer. Its detection system is based on an enzyme (glucose oxidase) biosensor, and the electrode test strip is manufactured using a screen-printing technique. This system brings about immobilization of the active components via the controlled deposition of inks onto the solid, planar electrode surface. The materials such as enzymes, mediators, and conducting and insulating polymers are incorporated into the inks, and these multiple ink layers are patterned on top of one another to build up the complex electrode structure (see Figure 3).

With so many people taking anticoagulants such as warfarin and requiring regular monitoring of their hemostasis using a prothrombin time (PT) test, a market has emerged for coagulation monitors that can be used at home. In the POC coagulation testing industry, many new devices are coming to the market, often smaller than previously developed devices and offering technical advantages and new features.

Figure 4. The Protime microcoagulation system from International Technidyne Corp. Photo courtesy International Technidyne Corp.

At least four such devices have gained FDA approval since 1997, one of which is the Protime Microcoagulation system by International Technidyne Corp. (see Figure 4). After blood is drawn from the finger, detection is conducted by an optical method, and the PT result is available in about 5 minutes. The lightest PT home testing monitor is the Avosure by Avocet (San Jose) which weighs only 6 oz. This system gives faster results but costs more than twice the other machine.

Figure 5. The mode of action of the Rapidpoint Coag POC coagulation monitor. Illustration courtesy Pharmanetics Inc. (click to enlarge)

The Rapidpoint Coag POC coagulometer by Bayer Diagnostics (Tarrytown, NY) is a multitest analyzer for monitoring anticoagulation therapy and determining a patient's coagulation status. By using a test-card format, the system provides a full menu of POC coagulation tests, including prothrombin time, activated partial thromboplastin time, and a heparin management test. Other benefits of this device include electronic quality control, which helps reduce the costs of regulatory compliance, and flexibility in sample origin (e.g., fingerstick, citrated whole blood, or plasma).

The Rapidpoint Coag detection system photomechanically monitors the formation (clotting) and dissolution (fibrinolysis) of blood clots in a flat capillary chamber on the surface of a credit-card-sized disposable card. The system's technology impedes the movement of small paramagnetic iron-oxide particles (PIOPs) on the card in response to an oscillating magnetic field in the analyzer (see Figure 5). The PIOPs combine with the appropriate chemicals and formulate into dry reagents.

Biomimetic Sensors

Figure 6. A cochlear implant inside a human ear.

In vivo monitoring of biochemical parameters has been most developed for the area of acute treatment, where devices only need to function for a few days at most. There are still considerable barriers to the use of implantable sensors for more-extended periods of time, despite the tremendous success of devices such as hearing aids (or cochlear implants) and pacemakers, which can be left in situ for years (see Figure 6). An artificial pacemaker consists of a lithium battery, a pulse generator, and a wire that connects the pacemaker to the heart. However, if the pulse generator and connecting wire were fabricated using a biomimetic material, the battery could be eliminated as the pacemaker's environment would influence its functioning. Research on biomimetic materials is still very much at the molecular level, but major advancements have been made.¹⁸

Figure 7. Diagram of a biomimetic sensor in which a molecular imprinted polymer (MIP) selective for the fluorescently labeled amino acid dansyl-L-phenylalanine was applied as a layer on the tip of a fiber-optic sensing device. Illustration courtesy Analytical Chemistry (click to enlarge)

The ultimate goal is to develop smart implantable systems. Most chemical sensors and biosensors require intimate contact with the sample at a sensitive membrane or film in order to provide information about the target species. The surface characteristics of these membranes and films rapidly change when exposed to body fluids, such as blood, that are rich in diagnostic information. Hence, the response characteristics of the sensors also rapidly change, leading to a need for regular recalibration, which in turn complicates how they function and how they are packaged.

Biocompatibility, calibration, and power requirements, along with the materials' long-term properties, remain serious issues to be addressed. One possible answer to some of these issues could be the development of molecular-imprinting-based biomimetic sensors for medical applications, which could provide an alternative to biosensors (see Figure 7).

Future Trends

It has been predicted that the trend in clinical diagnostics lies in the technology of autonomous sensing with next-generation handheld, portable POC diagnostic devices, wearable sensing technology, and in vivo biomimetics. The only limitation to progress is the fact that the sensors—especially the chemical and biological ones—lag behind the electronics, so there is still a long way to go.

In the future, the evolution of an integrated healthcare network incorporating these sensor types may emerge, much like a healthcare "nervous system," which would comprise multitudes of sensors and sensing technologies. Such systems could provide information-gathering nodes for healthcare applications everywhere by sharing more of the healthcare load.

Due to developments in communications and connectivity, data from these sensors can even now be easily accessed via personal digital assistants, PCs, mobile phones, and networks. Telemedicine is quickly becoming a part of our lives, making possible real-time communication of clinical results to a physician, as well as immediate feedback on action to be taken by the patient.

As the gradual shift occurs from the conventional healthcare system to a decentralized one incorporating and embracing these emerging sensing technologies, it will ultimately be a balance of the two that will supply our diagnostic and monitoring needs for a long time to come.

---

References

1. LJ Kricka, "Microchips, Microarrays, Biochips and Nanochips: Personal Laboratories for the 21st Century," Clinica Chemica Acta 307, no. 1–2 (2001): 219–223.

2. L Cao, S Mantell, D Polla, "Design and Simulation of an Implantable Medical Drug Delivery System Using Microelectromechanical Systems Technology," Sensors and Actuators A: Physical 94, no. 1–2 (2001): 117–125.

3. NA Polson and MA Hayes, "Microfluidics: Controlling Fluid in Small Places," Analytical Chemistry 73, no. 11 (2001): 312A–319A.

4. "Micronics and Renal Plant Enter Agreement to Develop Artificial Kidney Microfluidic Lab-On-A-Chip Platform for Kidney Hemodialysis" in Micronics Inc. Home Page [on-line] (Redmond, WA: Micronics Inc., October 1, 2001 [cited 11 March 2002]); available from Internet: http://www.micronics.net/news/news_release_detail.php?nr_id+30.

5. H Becker, LE Locascio, "Polymeric Microfluidic Devices," Talanta 56, no. 2 (2002): 267–287.

6. WA Whitaker III, "Acrylic Polymers: A Clear Focus," in Medical Plastics and Biomaterials [on-line] January 1996 [cited 11 March 2002]; available from Internet: http://www.devicelink.com/mpb/archive/96/01/001.html.

7. RA Freitas Jr., Nanomedicine Volume I: Basic Capabilities (Georgetown, TX: Landes Bioscience, 1999).

8. "Commission on Laboratory Accreditation, Laboratory Accreditation Program, Point-of-Care Testing Checklist," [on-line] (Northfield, IL: College of American Pathologists, October 2001 [cited 11 March 2002]); available from Internet: http://www.cap.org/html/checklist_html/cklst_poc.html.

9. Connectivity Industry Consortium (CIC) Home Page [on-line] (CIC, 2000 [cited 8 March 2002]); available from Internet: http://www.poccic.org/index.shtml.

10. A Reder, "Regulating the Point of Care: The IVD Connectivity Industry Consortium," in Medical Device & Diagnostic Industry [on-line] April 2001 [cited 11 March 2002]; available from Internet: http://www.devicelink.com/mddi/archive/01/04/001.html.

11. "About the SIG," [on-line] (Bluetooth Technology SIG Inc., 2001 [cited 11 March 2002]); available from Internet: http://www.bluetooth.com/sig/about.asp.

12. "Intelligent Knee to Save Costly Sporting Injuries," [on-line] (Wollongong, NSW, Australia: University of Wollongong, 2001 [cited 11 March 2002]); available from Internet: http://www.uow.edu.au/science/research/ipri/kneesleeve.html.

13. "Smart Bra Heralds New Age of Intelligent Fabrics," [on-line] (Wollongong, NSW, Australia: University of Wollongong, 2000 [cited 11 March 2002]); available from Internet: http://www.uow.edu.au/science/research/ipri/smartbra.html.

14. A Lynch et al., "Point-of-Need Diagnosis of Cystic Fibrosis Using a Potentiometric Ion-Selective Electrode Array," Analyst 125, no. 12 (2000): 2264–2267.

15. JA Tamada et al., "Non-Invasive Glucose Monitoring: Comprehensive Clinical Results," Journal of the American Medical Association 282, no. 19 (199): 1839–1844.

16. JA Wahr and KK Tremper, "Non-Invasive Oxygen Monitoring Techniques," Critical Care Clinics 11, no. 1 (1995): 199–217.

17. J Stenson, "Contact Lens Measures Glucose," in MSNBC.com Home Page [on-line] June 25, 2001 [cited 11 March 2002]; available from Internet: http://www.msnbc.com/news/591718.asp.

18. D Kriz et al., "Molecular Imprinting-Based Biomimetic Sensors Could Provide an Alternative to Often Unstable Biosensors for Industry, Medicine, and Environmental Analysis," in Analytical Chemistry [on-line] June 1997 [cited 11 March 2002]; available from Internet: http://pubs.acs.org/hotartcl/ac/97/jun/mol.html.

Gillian McMahon, PhD, is a senior research officer, Edel Minogue, PhD, is a senior research fellow, and Dermot Diamond, PhD, is associate director at the National Centre for Sensor Research (Dublin, Ireland).

Copyright ©2002 IVD Technology