IEEE SENSORS JOURNAL, VOL. 3, NO.

3, JUNE 2003

251

Emerging Biomedical Sensing Technologies and Their Applications

Gerard L. Cot, Senior Member, IEEE, Ryszard M. Lec, Member, IEEE, and Michael V. Pishko
AbstractRecent progress in biomedical sensing technologies has resulted in the development of several novel sensor products and new applications. Modern biomedical sensors developed with advanced microfabrication and signal processing techniques are becoming inexpensive, accurate, and reliable. A broad range of sensing mechanisms has significantly increased the number of possible target measurands that can be detected. The miniaturization of classical bulky measurement techniques has led to the realization of complex analytical systems, including such sensors as the BioChemLab-on-a-Chip. This rapid progress in miniature devices and instrumentation development will significantly impact the practice of medical care as well as future advances in the biomedical industry. Currently, electrochemical, optical, and acoustic wave sensing technologies have emerged as some of the most promising biomedical sensor technologies. In this paper, important features of these technologies, along with new developments and some of the applications, are presented.

I. INTRODUCTION HE ongoing mergence of the 20th century revolution in information technology with the 21st century revolution in biotechnology poses considerable demand for new sensors, in particular new biomedical sensors. Currently, the recent advances in the microelectronics industry, the availability of advanced microfabrication technologies down to the microand nanoscale, and inexpensive signal processing systems have made the development of a variety of novel biomedical sensors possible. A good indication of that demand is the growing use of personal monitoring devices such as glucose sensors for diabetics or the recently developed sensors for HIV detection. Biomedical sensors can also make medical care more personal and tailored to the individual needs of a patient. In the near future, a treatment procedure could be adjusted to address a patients unique metabolism and biological rhythms. For example, the dose of a drug could be correctly determined in order to optimize the healing process and minimize its side effects. In addition, biomedical sensors will enable a broad range of medical
Manuscript received May 23, 2001; revised June 10, 2002. The associate editor coordinating the review of this paper and approving it for publication was Prof. Henry Baltes. G. L. Cot is with the Department of Biomedical Engineering Program, Texas A&M University, College Station, TX 77843-3120 USA (e-mail: cote@tamu.edu). R. M. Lec is with the The School of Biomedical Engineering, Science and Health Systems, and the Department of Electrical and Computer Engineering, Drexel University, Philadelphia, PA 19104 USA (e-mail: rlec@cbis.ece.drexel.edu). M. V. Pishko is with the Department of Chemical Engineering, Pennsylvania State University, University Park, PA 16802 USA (e-mail: mpishko@engr.psu.edu). Digital Object Identifier 10.1109/JSEN.2003.814656

services at a patients home using a variety of systems that employ a personal computer and the Internet. One may envision a dedicated home-based analytical diagnostic system interfaced with a computer that could monitor and store medical data over the life time of the person. In such a case, specialized application software would be capable of recognizing incoming health problems and could notify a person in advance of her or his health conditions. A modern biomedical sensor is a device which consists of a biologically or biophysically-derived sensing element integrated with a physical transducer that transforms a measurand into an output signal. The requirements for any good biomedical sensor are specificity or the ability to pick out one parameter without interference of the other parameters, sensitivity or the capability to measure small changes in a given measurand, accuracy or closeness to the true measurement, time response, biocompatibility, aging characteristics, size, ruggedness and robustness, and low cost. In addition, the sensor must have compatibility with the chemical, optical, optoelectronic, or electronic integrated circuit (IC) technology. The above listed features have been researched comprehensively over the last two decades and critical knowledge has been accumulated and the challenges have been identified. One may claim that the biomedical sensor field has matured enough to be poised for commercial success. In this paper, an overview of three of the primary biomedical sensor technologies; electro-chemical, optical and acoustic are discussed along with many of the biomedical applications. II. MODERN BIOMEDICAL SENSORS A conceptual model of a biomedical sensor and its important design elements are shown in Fig. 1. This model presents a complete biomedical sensor scheme in which, in addition to a sensing section of a biomedical sensor, microfluidic, signal processing and packaging units are included. Simultaneous analysis and design of all these elements are essential for the development of marketable biomedical sensors. The principle of operation of such a biomedical sensor can be inferred by following its sensing path. A measurand is introduced to a biomedical sensor using sample delivery system or by bringing the sensor to the patient, as with implantable or indwelling biomedical sensor probes. Next, the measurand passes through a preprocessing section, such as semi-permeable membrane, which performs a initial selective screening of possible interfering factors. After that, the measurand is exposed to the sensing element, a biologically active substance which is selective to the measurand of interest (i.e., DNA, antibodies,

1530-437X/03$17.00 2003 IEEE

252

IEEE SENSORS JOURNAL, VOL. 3, NO. 3, JUNE 2003

Fig. 1. General diagram of a biomedical sensor system. It should be noted that the biomedical sensor could be an array that would allow for simultaneous detection of multiple measurands.

enzymes, or cellular components). When a measurand interacts with the sensing element, microscopic physical, chemical, and/or biochemical changes are produced. These microscopic changes cause the macroscopic physical changes in the sensing element, which are converted by the physical transducer into an output electric signal. The electric signal is conditioned, processed, and displayed. The processing can include such important sensor features as self-calibration, self-diagnostics, and advanced pattern recognition analyses. All these functional design elements can be enclosed in the sensor package that provides measurement integrity to the device. In recent years significant progress has been made in all areas listed above. However, the widest researched area has been on the sensing elements and sensing mechanisms. Below is a short overview of the current state in these areas with an emphasis on the optical, electrochemical and acoustic biomedical sensing technologies. III. BIOMEDICAL TRANSDUCER TECHNOLOGIES AND APPLICATIONS Biomedical sensor development has in a large part depended upon technologies primarily developed for other purposes. The silicon-based microfabrication (IC) and micromechanical (MEMS) techniques have been successfully applied for the fabrication of a large range of miniature electrochemical biomedical sensors. Similarly, the progress in optical biomedical sensors has been founded on fibers and devices purposed for fiberoptic communication. Also, acoustic/piezoelectric biomedical sensors capitalized on the decades-long growth of RF telecommunication technologies. Piezoelectric elements

utilized in radar, cellular phones, electronic watches, etc., have been well applied to biomedical sensors. Other types of biomedical sensors based on calorimetric, magnetic techniques, etc. have also been effected significantly by modern IC and MEMS techniques. Currently, electrochemical, optical, and acoustic wave transducer technologies have emerged as some of the most promising biomedical sensor technologies. In the following sections, important features of these technologies, some applications and new trends and developments are presented. A. Electrochemical Biomedical Sensing Technologies Electrochemical biomedical sensors have been the subject of research for a number of decades, with the vast majority of devices consisting of enzymes coupled to an electrode. These devices typically can be operated in two modes: potentiometric or amperometric. Amperometric devices are by far the most prevalent and have seen commercialization for measurands such as glucose and lactic acid. In an amperometric enzyme electrode, an enzyme is immobilized on the surface of an electrode and the product of the enzymatic reaction detected at the electrode surface anodically or cathodically. For example, the commercially available Yellow Springs Instruments glucose electrode uses the enzyme glucose oxidase immobilized on a membrane placed over a platinum electrode. Glucose oxidase catalyzes the reaction shown at the bottom of the page. One product of this reaction, hydrogen peroxide, is oxidized at 700 mV (versus a saturated calomel reference electrode) on the platinum electrode surface, producing a current that is directly proportional to the amount of glucose in sample. Amperometric enzyme electrodes for other measurands such as lactate operate in a sim-

COT et al.: EMERGING BIOMEDICAL SENSING TECHNOLOGIES

253

ilar fashion. Amperometric glucose sensors have been by far the greater commercial success and can be found on most pharmacy shelves in the form of home glucose test meters, such as the Freestyle meter from Therasense and the One Touch meter from Lifescan. In addition to in vitro enzyme electrodes based on this detection scheme, the FDA has recently approved a subcutaneously implantable glucose sensor [1], [2] based on the same principles. Miniature needle-type amperometric glucose sensors have been commercialized by MiniMed Corporation (Slymar, CA) for the short-term monitoring of interstitial fluid glucose (Fig. 2). Despite considerable efforts to develop a membrane and enzyme system that is minimally affected by the foreign body response, most subcutaneous sensors develop significant drift immediately following implantation. A rapid and progressive loss of sensitivity can be attributed to protein/cellular fouling of the membrane, enzyme dysfunction due to heavy metal toxicity, and unstable levels of oxygen in the subcutaneous tissues. Accurate real-time monitoring in the clinical setting requires frequent recalibration of the in vivo sensor using an external glucose measurement reference method [3]. Markwell Medical Corporation (Madison, WI) has developed a miniature implantable enzyme based amperometric sensor with telemetry capable of measuring interstitial fluid glucose long-term. Unique to this technology is a multilayer protective membrane that promotes angiogenesis and prevents the formation of a diffusion limiting fibrous capsule [4]. An iontophoretic transdermal glucose sensor [5][7] developed by Cygnus, Inc. (Redwood City, CA) also used similar glucose detection chemistry, but uses an electric field to extract interstitial fluid from the skin and uses the sensor to measure glucose in this field. Thus, the issue of a foreign body response is removed, but significant extraction times (great than 7 min) are required to remove enough interstitial fluid for a measurement. Medical Research Group Corporation (Slymar, CA) has developed a glucose sensor for long-term implantation within the blood stream [8]. This device is also based on glucose oxidase immobilized on a Pt electrode. However, the consumption of oxygen is measured rather than the production of peroxide. The miniature system looks like a pacemaker generator with two flexible intravascular leads. An external patient module provides a data display, visual and audible alarms for hypo- and hyperglycemia, data storage, and telemetry for recalibration. The protective membranes are reported to be relatively unaffected by protein deposition and thrombosis by placing the catheter-based sensor within the large vein of the chest (vena cava). Oxygen sensors are used to measure the decrease in oxygen that occurs in proportion to the oxidation of glucose. Venous blood oxygen levels are measured by a second reference oxygen sensor to provide enhanced specificity for glucose. Excellent long-term stability has been demonstrated in human testing, requiring infrequent recalibration. Eventual fouling of the protective membranes and enzyme depletion limits the current system to six months of continuous use. Methods to safely remove the old system and surgically implant a new intravenous system are beingdeveloped.Anintegratedsensor-controlledinsulindelivery system (artificial endocrine pancreas) is being developed by MRG consisting of the vena cava glucose sensor, an implantable insulin pump and an adaptive computer control algorithm.

Fig. 2. Example of a typical bio-chemical sensing system (minimed continuous glucose monitoring system; http://www.minimed.com/files/ cgms/patient_pg2.htm).

A number of research programs have sought to minimize the oxygen dependency and improve the signal to noise level of enzyme electrodes by replacing oxygen in the enzyme catalyzed reaction with an electron acceptor/donor also called a mediator [9]. In the case of glucose oxidase, the reaction becomes the following:

where is the mediator in its oxidized form and is the mediator in its reduced form. These mediators are typically in the form of organometallic complexes such as ferrocene and its derivatives and Os(imidazole) (bis-bipyridine) . The mediators have lower oxidation potentials than hydrogen peroxide and thus can be operated at lower potentials (200500 mV versus saturated calomel). Thus they are less suspectible to interference from electroactive compounds such as ascorbic acid and uric acid that are pervasive in vivo and in bodily fluids. This type of chemistry has lead to commercial home glucose test meters such as those manufactured by Abbott Laboratories, Therasense, Roche, and Bayer. A number of biomedical sensors based upon redox polymers coupled to oxidoreductases have also been studied. In these studies, the polymer served to immobilize the enzyme via formation of an insoluble protein/polymer complex, through the physical entrapment of the enzyme in a polymer film and/or through the covalent crosslinking of the enzyme and polymer. These polymers are suitable as mediators for implantable sensors because they are less susceptible to mediator leaching from the sensor as compared to small molecule mediators. Amperometric biomedical sensors based on redox polymer/enzyme complexes were shown to be miniaturizable and could measure desired species either intravenously or subcutaneously when implanted in rats, primates and human volunteers [10][12]. Because of the high current density these sensors exhibit, they can be calibrated in vivo using a one-point calibration, i.e., the device is calibrated from a blood glucose measurement and the sensor current at one point in time. A significant area recently has been in the development of biomedical sensor arrays using nanostructured materials, for both redundant sensing of a single analyte or multianalyte sensing with a single device. Recent research on patterning biomolecules on surfaces has focused primarily on self-assembled monolayers (SAMs) and tethered biomolecules on surfaces that may potentially form addressable patterned arrays.

254

IEEE SENSORS JOURNAL, VOL. 3, NO. 3, JUNE 2003

Amperometric patterned enzyme electrodes were formed using SAMs patterned by scanning electrochemical microscopy [13] or via microcontact printing followed by the electrostatic assembly of a multilayer enzyme/redox polymer thin film [14]. Willner and colleagues reported amperometric glucose sensors based on pyrroloquinoline quinone/glucose dehydrogenase SAMs that exhibited high current densities [15][19]. These surfaces are easily formed, particularly using alkane thiols and their derivatives on gold-coated surfaces. SAMs also permit the site specific immobilization and orientation of biomolecules on a surface. However, two-dimensional approaches such as SAMs may limit the number of biomolecule recognition sites on the sensor surface and thus may have low signal levels and require shielding or other measures to reduce noise. The structure of self-assembled molecules on a surface can also result in defects or pinholes in the monolayer and contribute to instability, particularly at applied potentials. The current adhesion chemistry used in the fabrication of SAMs also permits monolayer formation only on a limited number of surfaces, most commonly gold. In addition to monolayers, photolithography and other photoinduced patterning chemistries were highlighted in a few studies, demonstrating the formation of patterned biomolecule surfaces and micropatterned polymers for optical chemical sensing. Electrode arrays can be fabricated using various methods, however, thin film technology adopted from microelectronics has been the predominant approach due to ease, quality, reproducibility and low cost of manufacturing. Electrode arrays have been photolithographically microfabricated on thermally oxidized silicon, polyimide, and other insulating substrates. Methods of immobilizing sensing components for glucose and lactate detection onto electrode arrays, include electrodeposition of proteins such as glucose oxidase and the casting of proteins over the array surface. These studies demonstrated the potential for fabricating successful sensor arrays, however, there is still a need for a reproducible and simple way of immobilization of sensing elements which also allows for oxygen-free, mediated glucose, and lactate monitoring. One approach which affords excellent reproducibility and control over architecture of the deposited species is based upon multilayer electrostatic assembly [20]. This scheme has been used to deposit oppositely charged redox polymers and enzymes on electrode surfaces. Specifically, ferrocene and osmium based cationic redox polymers were stacked alternately with with anionic oxidoreductases such as pyruvate oxidase (for pyruvate biosensors) and lactate oxidase, both of which as polyanions at neutral pH. Resultant sensors yielded reproducible current responses and showed no saturation for physiologically relevant concentrations of analytes such as glucose lactate and pyruvate. Examination of the nanostructure of these thin films suggest that there is significant intercalation between the cationic and anionic layers leading to a very compact and stable nanocomposite structure. B. Optical Biosensing Technologies Recent advances in optic and electronic technologies have lead to the research and development of many optic and fiber

Fig. 3. Michelson interferometer using bulk optics. The beam splitter is often replaced with fiber optics.

optic biomedical devices. Indeed, there has been much enthusiasm as well as a strong effort by newly formed medical device companies, established medical device industry leaders, and universities to investigate optical technologies for a host of diagnostic and sensing procedures, including both bio-chemical and bio-physical based sensors. Examples of some optical bio-physical sensors include the measurement of body temperature, blood velocity and intracranial pressure measurements while the optical bio-chemical sensor examples include blood chemical detection, blood micronutrient monitoring, and cancer detection. In terms of optical biomedical sensors for monitoring these parameters remotely into the body, fiberoptic probes can be used. The utility of fiberoptic probes is that they offer the potential for miniaturization, good biocompatability for the visible and near-infrared wavelengths, fast speed since light is used, and safety, since no electrical connections to the body are required. In this section, we will briefly outline the latest mechanisms being explored for biomedical optical sensing today including interferometry, infrared absorption, scattering, luminescence and polarimetry. 1) Interferometric Biomedical Sensing: The phenomenon of interference is a method by which physical light interaction takes place and it depends on the superposition of two or more individual waves, typically originating from the same source. There are several variations for producing light interference using both bulk optics and fiber optics but the Michelson interferometer is the most common instrument used today (Fig. 3), especially in the Fourier transform infrared (FTIR) machines used for absorption spectroscopy and the relatively new field of optical coherence tomography [21]. One example of a fiber optic based interferometric sensor is the commercially available intracranial fluid pressure monitoring system for patients with severe head trauma or a condition known as hydrocephalus, which is an increased amount of cerebral spinal fluid in the ventricles and/or subarachnoid spaces of the brain [22]. In addition, the Michelson interferometer is currently being investigated for sensing other biophysical parameters such as tissue thickness, particularly for corneal tissue as feedback for the radial keratectomy procedure, which is laser removal or shaving of the cornea to correct vision [23], [24]. As mentioned, this interferometric approach, when used with

COT et al.: EMERGING BIOMEDICAL SENSING TECHNOLOGIES

255

a low coherent light source and scanning or imaging array technology, has been shown to produce morphological data in vivo by a process known as optical coherence tomographic (OCT) imaging [21], [25], [26]. Optical coherence tomography is a fundamentally new sensing technology that allows visualization through tissue at very high resolution. It measures the intensity of backreflected infrared light and allows resolutions of 1020 m, but with a very limited depth of field, typically 12 mm. This approach has been used to investigate several ocular diseases including macular disease, genetic retinal disease, retinal detachment and retinoschisis, choroidal tumors, optic nerve disorders, and glaucoma [25]. In addition, OCT is useful over approximately the same distance of a biopsy at high resolution and in real time. Consequently, the most attractive applications for OCT are those where conventional biopsies cannot be performed or are ineffective [21]. This interferometric-based imaging approach could also be used in cardiology for guiding coronary procedures [27], [28]. 2) Absorption-Based Biomedical Sensing: Research in the area of bio-chemically based optical sensing and, in particular, infrared (IR) absorption-based sensing has been fueled in a large part by the glucose sensing market. However, the optical monitoring approaches currently clinically available are the pulse oximeter, IR ear thermometer, hemoglobin and hematocrit meter, and the new IR bilirubin sensor. The pulse oximeter is based on the detection of changes in the strong optical absorption peaks of oxygenated and de-oxygenated hemoglobin and are available from vendors such as Criticare, Datex-Ohmeda, Invacare, Novametrix, and Nellcor, to name a few [29], [30]. The ear thermometer is based on receiving infrared light from the tympanic membrane and is available from vendors such as Becton Dickinson, Braun, Omron, and Safety 1st [31]. Through the work of Groner, Winkelman, et al. [32], and recently commercialized by Cytometrics, Inc. as the Hemoscan 1000 [33], a noninvasive optical approach has become available which has the potential to monitor hemoglobin and hematocrit as an indicator of iron deficiency in a clinical or field setting. The approach is based on imaging the near-infrared absorption spectral peaks of hemoglobin but using a polarized light input and been referred to as orthogonal polarization spectral (OPS) imaging. The images are striking but the key to this technology becoming truly useful as a field device for blood monitoring is in the software development to provide quantifiable data. The new bilirubin sensor is a near infrared approach pioneered through the early work of Jacques et al. [34], [35] and commercially developed by scientists at SpectRx Inc. In general, light absorption in a sample is governed by the Beer-Lambert law in which the transmitted light is dependent on the wavelength and intensity of the incident light, the path length and the absorption coefficient or rather the sum of the multiplication of the molar absorptivity times the concentration of all the different components in the measurand [36]. The primary means for varying the wavelength of the light include dispersive and nondispersive methods. The optical filter-based nondispersive NIR systems are more commonly being developed by several companies [37], [38], in particular for glucose sensing, because they can be configured inexpensively and generally can have

better throughput if all the light is passed through the sample. However, they have had limited success in producing repeatable and quantifiable results in vivo. The lack of repeatability of the NIR signal in vivo both within and especially between patients is because the signal variations in most instances are not fully understood or accounted for in the system. The primary known drawbacks to taking this technology from an in vitro to an in vivo monitoring device include the pathlength variability of a pliable tissue, temperature variability of the peripheral site such as the finger or earlobe and the presence of other chemically confounding substances (protein, urea, cholesterol, alcohols, etc.). The latest technology in the IR biosensing field is in the development of clinical and chemical laboratories, in particular at the Oak Ridge National Lab (ONRL), in which researchers have miniaturized an infrared microspectrometer to the size of a sugar cube [39]. Carved out of a solid block of plastic, the device measures 1.5 cm on a side and has no moving parts. It can be used for blood chemistry analysis, as well as a number of nonmedical processes. The plastic device uses a light source to excite certain types of compounds in gases, liquids and solids. These excited compounds give off infrared light of various wavelengths. The measured emissions wavelengths are fed into a microchip, which determines the concentrations of chemicals in a sample. 3) Biomedical Sensing Using Scattered Light: There are fundamentally two types of optical scattering for diagnostics and monitoring, elastic and inelastic. The elastic scattering can be described using Mie theory (or Rayleigh scattering for particles small compared to the wavelength), in which the intensity of the scattered radiation can be related to the concentration, size, and shape of the scattering particles. This type of scatter is broadband and not typically specific enough for biosensing [36], [40]. However, as a diagnostic screening tool for cancer detection, measurement of the scatter in thin tissues or cells may hold promise [41]. Many of the changes in tissue due to cancer are morphologic rather than chemical and thus occur with changes in the size and shape of the cellular and subcellular components. Thus, the changes in elastic light scatter should occur with morphologic tissue differences. If the wavelength of the elastic light scattering is carefully selected so as to be outside the major absorption areas due to water and hemoglobin and if the diffusely scattered light is measured as a function of angle of incidence, there is potential for this approach to aid in pathologic diagnosis of disease [36]. One inelastic scattering approach in which the polarization of the particle is not constant is known as Raman scattering, which is observed when monochromatic (single wavelength) radiation is incident upon the media. The shift is associated with transitions between rotational, vibrational, and electronic levels. As with infrared spectroscopic techniques, Raman spectra can be utilized to identify molecules since these spectra are characteristic of variations in the molecular polarizability and dipole moments. The Raman signal in general is weak but the technology has advanced, with the replacement of slow photomultiplier tubes with faster CCD arrays, as well as the manufacture of higher power near infrared laser diodes, to allow for biomedical sensing. However, this technology is still bulky and expensive for most biomedical sensor applications and thus cur-

256

IEEE SENSORS JOURNAL, VOL. 3, NO. 3, JUNE 2003

rently in the bench top study phase. The technology has allowed researchers to consider the possibility of distinguishing normal and abnormal tissue types as well as studying a variety of biological molecules including proteins, enzymes and immunoglobulins, nucleic acids, nucleoproteins, lipids and biological membranes, and carbohydrates quantifying blood chemicals in near-real time [42][49]. The diagnostic approaches look for the presence of different spectral peaks and/or intensity differences in the peaks due to different chemicals present in, for instance, cancerous tissue. For monitoring, investigators have applied statistical methods such as partial least squares (PLS) to aid in the estimation of biochemical concentrations from Raman spectra [45], [46]. In addition to normal Raman scattering, biomedical sensors have been developed using surface-enhanced Raman spectroscopy (SERS) that allows for orders of magnitude increase in sensitivity. In particular, the ONRL group has developed surface-enhanced Raman gene (SERG) probes that can locate free DNA molecules that have hybridized to other DNAs fixed on a surface [39]. This group uses a hybridization method in which DNA is transferred from the nylon membrane to a glass strip coated with tiny silver spheres. The dye labels attached to the DNA bases have unique Raman infrared spectra, but the normally weak Raman lines are greatly enhanced by the presence of the silver spheres. This enhancement allows DNA bases to be detected at sufficient sensitivity to be useful for DNA sequencing studies. Using a silver colloid, SERS has also been used to detect elevated levels of amino acids, in particular glutamate, in cerebral spinal fluid after head trauma [50]. These elevated levels could cause further damage and thus need to be quantifiably measured in order to provide for pharmaceutical intervention. 4) Use of the Luminescence Property of Light for Biomedical Sensing: Luminescence is the absorption of photons of electromagnetic radiation (light) at one wavelength and re-emission of photons at another wavelength. The luminscent effect can be referred to as fluorescence or phosphorescence. Fluorescence is luminescence that has energy transitions that do not involve a change in electron spin and therefore the re-emission occurs much faster. Consequently, fluorescence occurs only during excitation while phosphorescence can continue after excitation. The measurement of fluorescence has been used for both diagnostic and monitoring purposes. Obtaining diagnostic information, in particular with respect to cancer diagnosis or the total plaque in arteries, has been attempted using the intrinsic fluorescence of tissue [51], [52]. The intrinsic fluorescence is due to the naturally occurring proteins, nucleic acids, and nucleotide coenzymes while extrinsic fluorescence is induced by several mechanisms. For instance, antibody-based extrinsic fluorescent biomedical sensors have been developed for number of biomedical applications including cancer diagnosis [53]. In fact, such fluorescent biomedical sensor probes can be made at the nanometer scale in which a laser beam (blue) that penetrates a living cell is used with monoclonal antibodies that recognize and bind to benzo(a)pyrene tetrol (BPT), indicating that the cell has been exposed to a cancer-causing substance [54]. Extrinsic fluorescence has also been investigated to monitor such measurands as glucose, intracellular calcium, proteins, and nucleotide coenzymes [36]. Unlike the use of fluorescence in dilute solu-

tions, the intrinsic or autofluorescence of tissue as well as the scattering and absorption of the tissue act as a noise source for the extrinsic approach. Most extrinsic fluorescent biomedical sensors are based on either the measurement of intensity or lifetime, in which the lifetime can be measured in the time or frequency domain [55][58]. Bench-top systems typically are bulky and include dual monochromators (grating based wavelength separation devices) are used with either a photomuliplier tube as the detector or a CCD array detector. However, once the optimal configuration for a particular biomedical application, such as cervical cancer detection or glucose sensing, has been investigated using the bench-top machine, an intensity or phase lifetime measurement system can be designed with a simpler, more robust, configuration. Such a system can be designed with wavelength specific filters instead of monochromators and made to work at two or more discrete wavelengths. In addition, optical fibers can be used for delivery and collection of the light to the remote area [58]. Since the excitation wavelength and fluorescent emission wavelengths are different, the same fiber or fibers can be used to both deliver and collect the light. As with many of the optical approaches discussed, a number of novel fluorescence-based techniques seem to have evolved from the glucose sensing application. Unlike many of the other, noninvasive, optical approaches being investigated, because the fluorescent glucose sensing approaches have to be in contact with the sample, they have the advantage of being highly specific to the glucose. Those approaches that seem to have demonstrated the most promise generally fall into two categories: the glucose-oxidase based sensors and the affinity-binding sensors [43]. In the first category the sensors use the enzyme catalyzed oxidation of glucose by glucose-oxidase (GOX), similar to the electrochemical sensors but in this case they generate an optically detectable glucose-dependent signal. Several methods for optically detecting the products of this reaction and hence the concentration of glucose driving the reaction have been devised [59], [60]. The primary drawback to GOX based sensors is that their response depends not only on glucose concentration but on local oxygen tension as well [43]. The affinity-based sensors do not depend on local oxygen; however, many of the earlier affinity-binding techniques were investigated for short term use since they required indwelling probes [57], [58], [61][63]. In more recent work, these and other investigators have exploited the phenomenon of fluorescence resonance energy transfer (FRET) whereby an acceptor in close proximity to a fluorescent donor can induce fluorescence quenching in the latter. Most recently, Russell et al. has reported the use of poly(ethylene glycol) or PEG particles to encapsulate the FRET assay [55]. In the work of Russell, it was reported that it is possible to create a microparticle-based fluorescent glucose assay system potentially suitable for subcutaneous implantation. In addition to the above application, high sensitivity, short collection time requirements, lack of sample contact, and capability of scanning large areas/volume render fluorescence methods an attractive alternative for microbial detection. One fluorescence-based approach uses the intrinsic fluorescence of tryptophan, other amino acids, and DNA which are excited and fluoresce in the UV (200300 nm excitation, 300400 nm

COT et al.: EMERGING BIOMEDICAL SENSING TECHNOLOGIES

257

Fig. 4. Schematic diagram of a biochip system (adapted from the virtual poster presentation of ORNL [http://www.ornl.gov/virtual/biomedical sensors/]).

emission). Observation of these markers is a sensitive indicator of the presence of biological materials. However, all biological materials contain these building blocks and, in addition, particles, such as dust, pollen, smoke, etc., preferentially scatter UV/blue radiation (Raleigh scattering). These interferences significantly reduce the utility of these markers. However, a variety of microbial cell components exhibit intrinsic fluorescence: NAD[P]H, tryptophan, tyrosine, DNA, lumazines, pterins, flavoproteins, and other secondary metabolites [64]. Of these, NAD[P]H (360 nm excitation, 450/475 nm emission) has been the most extensively studied. Other DPNH metabolites have excitation and emission energies near that of NAD[P]H and these metabolic signals can be obtained by integration of the 425500-nm region fluorescence. This fluorescence is directly proportional to the concentration of the metabolite, e.g., NAD[P]H and related to the number of live (metabolizing) cells. Other components, such as flavoproteins and cytochromes, have some absorption in this region, but their concentrations are factors of 10100 less in living cells. In fact, there are no interferences of biological origin in this region and this fact has made redox fluorimetry based on NAD[P]H/DPNH a vital tool in the investigation of cellular metabolism and tissue oxygenation [65]. One method and prototype device for the detection of microbial life on surfaces, such as foods, glass, plastics, cloth, stainless steel, etc., developed by Estes et al., in environmental has resulted in a sensitivity of conditions [66].

Last, the latest research in luminescent sensing is a new generation of reagents that report on specific molecular events in living cells, called fluorescent protein biomedical sensors or rather bioluminescent chips. These biomedical sensors have evolved from in vitro fluorescence spectroscopy and fluorescent analogue cytochemistry. The various probe designs measure the molecular dynamics of macromolecules, metabolites, and ions in single cells emerge from the integrative use of contemporary synthetic organic chemistry, biochemistry, and molecular biology [67]. For instance, a simple device consisting of a coating of bioluminescent bacteria on top of a light-sensitive integrated circuit can be used to emit light in direct correlation to an measurand concentration [68]. The overall biochip design can be represented as shown Fig. 4, as adapted from the virtual poster presentation of ORNL [69]. 5) Biomedical Sensing Using Light Polarization Properties: Two of the emerging applications of polarized light are for biochemical quantification such as glucose and tissue characterization, in particular, to aid in cancer identification [70][73]. The concept behind these devices for measurand quantification is that the amount of rotation of polarized light by an optically active substance depends on the thickness of the layer traversed by the light, the wavelength of the light used for the measurement, the temperature, the pH of the solvent, and the concentration of the optically active material [43]. For polarimetry to be used as a noninvasive technique, for instance in blood glu-

258

IEEE SENSORS JOURNAL, VOL. 3, NO. 3, JUNE 2003

cose monitoring, the signal must be able to pass from the source, through the body and to a detector without total depolarization of the beam. Since the skin possesses high scattering coefficients, which causes depolarization of the light, maintaining polarization information in a beam passing through a thick piece of tissue (i.e., 1 cm), which includes skin, would be very difficult, if not impossible, although at least one company is attempting this approach [74]. Tissue thickness of less than 4 mm that include skin have been tried [75] and may potentially be used but the polarimetric sensing device must be able to measure millidegree rotations in the presence of greater than 95% depolarization of the light due to scattering from the tissue. As an alternative to transmitting light through the skin, the aqueous humor of the eye has been investigated as a site for detection of in vivo glucose concentrations since this sensing site is a clear biological optical media [70], [71]. It is also known that glucose concentration of the aqueous humor of the eye correlates well with blood glucose levels, with a minor time delay (on the order of minutes), in rabbit models [71]. The eye as a sensing site, however, is not without its share of potential problems. For instance, potential problems with using the eye include corneal birefringence and eye motion artifact [70]. In most tissues, including the eye, the change in rotation due to other chiral molecules such as proteins needs to be accommodated in any final instrument. In addition, most other tissues also have a birefringence associated with them that would need to be accounted for in a final polarimetric glucose sensor. It is the birefringence and retardation of the polarized light, as well as polarized scattering of the cells and tissue, that is the signal rather than the noise when using polarized light for tissue characterization. It has been shown that scanning laser polarimetry provides statistically significant higher retardation for normal eyes in certain regions over eyes with glaucoma. In addition to retardation in the retinal nerve fiber layer, images generated from the scatteringofvariousformsofpolarizedlighthavealsobeenshown to be able differentiate between cancer versus normal fibroblast cells in vitro [76]. Using a simplified, cross-polarized system, the potential of polarized light to not only distinguish between pigmented nevus and a freckle but also to retrospectively identify the true border around a patient with a sclerosing basal cell carcinoma has been tested in vivo [73], [77]. C. Acoustic Biomedical Sensors Acoustic waves have been used to study physico-chemical properties of gases, liquids, and solids for decades [78], [79]. In recent years, there has been an increase in efforts to utilize acoustic waves for the development of biomedical sensors [80][83]. Acoustic wave transducer technology provides a wide range of devices that are sensitive, accurate, small, portable, robust, and have excellent aging characteristics. In addition, such transducers can be produced using standard photolithography and hence are inexpensive. Acoustic waves can be generated and received by a variety of means including piezoelectric, magnetostrictive, optical, and thermal techniques [84]. The piezoelectric transduction effect almost exclusively has been utilized for generation and reception acoustic waves in sensor applications.

Acoustic biomedical sensors are typically designed to operate in a resonant type sensor configuration implemented as an oscillator. In this case, the output sensor signal is the resonant frequency shift, which is a function of the magnitude of the measurand. This is an important feature, since one can measure frequency relatively easily. In addition, the frequency can be considered as a quasidigital signal, which facilitates the subsequent signal processing operations. Sensors configured as an oscillator can be equipped with an antenna for remote sensing and control. Several recent oscillator circuits include an option for measuring sensor losses which significantly expands measurement capabilities of portable acoustic sensors [80], [85]. Another advantageous feature of using piezoelectric materials is that the same electro-mechanical transduction mechanism can be used not only for sensing but also for actuation. This property is essential for the design of piezoelectric surgical microcutters [86] or liquid microflow systems [87]. Therefore, it is possible to develop smart biostructures in which both sensing and actuating are realized within the same piezoelectric technology platform. Recently, a piezoelectrically-based Chem-Lab-on-a Chip capable of detecting a variety of hazardous chemicals has been reported [88]. In summary, the piezoelectric sensing platform offers a very versatile technology base for the development of sensors, actuators and smart structures. 1) Acoustic Waves, Piezoelectric Transducers and Acoustic Sensing Mechanisms: There are different types of acoustic wavesthatcanbeusedforbiomedicalsensing.Knowledgeof their properties is important for the selection of the optimal acoustic wave for a given measurand. Acoustic waves can be considered as a source of distributed force acting on a medium. The resultant deformation of a medium can be compressional or shear, or consists of a combination of both. The type of the deformation accompanying the wave is important because it determines resultant acoustic sensing processes.Compressional deformation is associated with the structural relaxational processes in the medium, while shear is coupled with medium viscoelastic properties and therefore they are sensitive to different molecular processes [78], [79]. Compressional deformations are easily transmitted through any gaseous, liquid or solid media. Shear deformation, on the other hand, propagates only through solids and penetrates only into liquids and gases. This last feature is very favorable because it makes shear waves sensitive to a numerous interfacial phenomena and this configuration is utilized in most acoustic wave biomedical sensors. The wave penetration depth, which depends on the frequency of the wave and the density and viscoelastic properties of the medium, ranges from microns to nanometers. Therefore, required sample volumes for sensing are small and the sensitivity of the sensor is high. In addition to mechanical phenomena, acoustic wave sensors can also sense electrical properties of a medium. Electric field probing of a medium is either generated by acoustic wave displacement (via the piezoelectric effect) or is provided by a sensor electrode structure. As a result, acoustic wave sensing mechanisms are very broad and acoustic sensors are capable of measuring the changes in mass/density, elastic modulus, viscosity, electrical conductivity, and dielectric constant. There are several types of waves that can be excited by piezoelectric transducers and they can be classified as bulk and

COT et al.: EMERGING BIOMEDICAL SENSING TECHNOLOGIES

259

Fig. 5. Classification of acoustic waves (bold-marked acoustic waves already have been used in biomedical sensors).

surface generated waves. The bulk generated waves are usually excited by metalized bulk piezoelectric elements such as disks or rods, whereas the surface generated acoustic waves are excited by an interdigital system of metallic electrodes (IDT) placed on the surface of piezoelectric materials. The electrodes are used to connect the transducer with electronic circuitry for the excitation and/or reception of acoustic waves. Two of the most common configurations of piezoelectric transducers include a thin metalized disk used for the excitation of bulk waves which excite thickness shear modes (TSM) and an interdigital transducer (IDT), which is applied for the excitation of a surface Rayleigh wave (SRW), a surface transverse wave (STW), a shear-horizontal acoustic plate mode (SH-APM), and a flexural plate wave (FPW). The STWs form a large family of various waves and include shear horizontal SAW (SH-SAW), surface skimming bulk wave (SSBW), and Love wave modes [83], [90]. In Fig. 5, a general acoustic wave classification diagram is presented. Knowledge of the properties of acoustic waves is important for the optimal design of a biomedical sensor [80], [89], [91]. Specific properties of these waves, such as the type of accompanying mechanical displacement, spatial distribution of mechanical and electrical fields, susceptibility to spurious mode coupling and the sensitivity of the wave to ambient conditions such as temperature, pressure, etc., depend on the cut of piezoelectric materials from which transducers are made. There are many available materials for manufacturing piezoelectric transducers including crystals, composites, and hybrid structures, which provide a wide range of possible sensing material design options. The most often used material is quartz, which is chemically inert, has superior mechanical properties and is temperature compensated [80], [81]. ATquartz cut orientation is routinely utilized for fabrication of TSM-based resonator biomedical sensors. In Fig. 6, a schematic presentation of a distribution of shear mechanical displacement generated by a disk-shaped AT-cut transducer immersed in water is given when the transducer operates at the fundamental (a) and harmonic frequencies (b). As a general rule, acoustic transducers can slice a biological interface at

Fig. 6. Conceptual model for a TSM transducer exposed on one side to water. (Top) Sensor operating at the fundamental frequency. (Bottom) Sensor operating at the fundamental frequency and higher harmonics.

different depths [Fig. 6(b)], hence providing important spatial , information. For SAW based devices various cuts of and crystals are used [80], [83]. Here, commonly [82], ZX [83] and ST-90 a 36Y rotated [86] materials are utilized. In the case of the APF sensors a silicon-ZnO hybrid structure is employed [80]. In a typical biomedical sensor, piezoelectric transducers are integrated with biological sensing films in order to obtain a required specificity (Fig. 1). Many biological materials such as proteins (enzymes, antibodies, receptors), organelles, cells, and tissue (microorganizms, animal and plant cells, and tissues) have been used as molecular recognition elements in piezoelectric biomedical sensors. An important issue impacting the

260

IEEE SENSORS JOURNAL, VOL. 3, NO. 3, JUNE 2003

Fig. 7.

Typical acoustic sensing process.

sensor performance is a proper attachment of a biological film to the sensor surface. In addition to classical immobilization techniques [90], several novel bio-deposition techniques have been recently developed and are successfully used with piezoelectric sensors. Examples include SAMs [92], molecularly imprinted polymers (MIPS) techniques for immobilization of antibodies [93], and soft lithography patterning for proteins and cells [94]. A typical acoustic sensing process is schematically presented in Fig. 7 and the corresponding measurement systems are depicted in Fig. 8. When a measurand interacts with the sensing element, microscopic physical, chemical and/or biochemical changes are produced. These microscopic changes cause the macroscopic acoustical/mechanical or electrical changes in the sensing element. Specifically, the density, viscosity, elasticity, electric conductivity, or dielectric constant of the sensing element undergo changes, which in turn modify the acoustic field quantities of the acoustic wave transducer. The acoustic wave transducer, which consists of a piezoelectric element with an array of metal electrodes, acts as a converter which transducts measurands into an output electric signal. For example, in the acoustic immunosensors, the antibodies are immobilized in the form of a thin film at the surface of the acoustic wave transducer. When the target antigen is introduced into the sensor environment, the elasticity, density, and viscosity of the film vary and these variations modify the acoustic parameters of the sensor, which finally leads to the changes in the output sensor signal. Molecular interpretation of the sensor response could be very simple or very complicated. When the sensing film is rigid and acoustically thin, then the sensor response can be related directly to mass accumulation of the measurand at the interface [80], [96]. Examples include gas absorption by

thin rigid polymer films [83] and thin metal oxide films [97], or some immunological reactions [98]. However, when the film is acoustically thick then, in addition to mass effects, the film viscous and elastic properties make significant contributions to the sensor response and the relationship between the sensor response and molecular events could be very complex [80], [99], [100]. Also, other factors such as the type of the boundary conditions or the topography of a sensor surface come into importance and influence sensor response, as well [101], [102]. Therefore, different biological systems and measurement conditions should be carefully studied in order to select or to develop the correct sensor model representation. Modeling of the sensor response is an area of ongoing, very active research, and though significant progress has been made in the last several years, many issues still need to be addressed [103], [104]. Among them are the molecular nature of the electro-bio-mechanical boundary conditions, interfacial viscoelasticity, and the development of dedicated computational techniques. Acoustic biomedical sensors, from an electronic measurement point of view, can be referred to as electrical radio-frequency (RF) components such as resonators, filters, or delay lines. Therefore, standard laboratory microwave measurement techniques based on network analyzers, vector voltmeters, and impedance meters are routinely used for the characterization of biomedical sensors (Fig. 8). However, the resolution of these techniques is limited and they are not applicable for real-life applications because of the cost and size. Therefore, in recent years, a lot of effort has been placed on the development of small portable biomedical sensor measurement systems. Several designs allowing simultaneous measurement of the transducer frequency and dissipation change have been proposed that are based on the frequency and amplitude measurements (oscillators) [85], the decay time and amplitude [87] or their combinations [91], but the overall performance of these systems is only satisfactory [80]. Finally, a very difficult and rather neglected issue is that of an enclosure for a sensor. The separation of a fluid-based biological environment from the electronic circuitry poses a very difficult task. Though several ingenious mechanical measurement cell designs have been devised [104] and various electrical passivating techniques layers [96] and deposited polymer utilizing sputtered thin films [97] have been successfully tested, the complete design that is commercially viable with an appropriate fluidic delivery system still needs to be developed. 2) New Acoustic Sensing Devices and Applications: A novel monolithic piezoelectric sensor (MPS) has been recently presented for the detection of chemical and biochemical measurands [105]. This new sensor overcomes several deficiencies associated with the typical two electrode system of a bulk TSM (QCM) sensor, while still employing a well-characterized, temperature-stable thickness-shear mode (TSM) response. This new three-electrode structure, depicted in Fig. 9, is applicable to both gaseous and liquid phase measurements; however, the principal benefit of the MPS is in liquid phase measurements. In these applications, it offers the ability to operate simple, yet stable, oscillator circuits in relatively viscous media. This novel MPS structure should accelerate the commercialization of piezoelectric sensor technology, particularly in such areas as biomedical, biochemical and environmental testing.

COT et al.: EMERGING BIOMEDICAL SENSING TECHNOLOGIES

261

(a)

(b) Fig. 9. Physical structure of the MPS. (a) The two active electrodes (upper surface) each have area, A and are seperated by a gap width g. (b) The MPS equivalent circuit consists of two TSM equivalent circuits with mechanical coupling (modeled as mutual inductance) and a parasitic capacitance. (c) Experimental amplitude (solid) and phase (square) response of the prototype MPS with various viscous water-glycerol mixture loading [105].

Fig. 8. Laboratory and portable electronic measurement systems utilized with acoustic biomedical sensors.

As well as the acoustic transducer technology, novel interfaces are also being developed. Biological recognition elements require appropriate immobilization on the surface of a transducer in order to maintain their bioactivity and specificity of biointeraction. Several immobilization techniques, surface modification methods, and a variety of novel intermediate materials supporting the attachment have been studied extensively [106]. In particular, several polymer-based interfaces have been proposed [107], [108]. One technique is a new very promising polymer nanofiber structure for cell immobilization [109]. Nanofiber architecture closely resembles a natural collagen structure which supports and promotes growth and proliferation of tissue cells. In addition to the aforementioned novel acoustic sensor device and the artificial biomedical sensor interfaces, there have been several new applications for acoustic sensors. These include antibody-based, nucleic acid-based, enzyme-based, and cell-based acoustic biomedical sensors. Antibody-based or immunosensors have been the most explored and the most advanced class of acoustic biomedical sensors. The interest results from the fact that antibodies can be obtained against almost any substance, antibody immobilization techniques are numerous and well developed and the pertinent acoustic sensing mechanism is simple and mostly determined by mass accumulation [110]. The detection process of antigens does not require labeling, which is the obvious advantage. However, in a case when nonspecific binding may interfere with detection, special care needs to be taken to eliminate it [111]. Other advantages

of piezoelectric immunosensors are that they are reusable with no noticeable degradation in performance [112]. In addition, piezoelectric immunosensors can operate in optically opaque media. Furthermore, piezoelectric sensors are inexpensive, easy-to-use, and feature rapid response; hence, they may allow for wide screenings and the development of effective preventive strategies for a broad range of diseases. Such sensors are used to determine the concentration of immunoglobulin IgM and C-reactive protein [112]; human cells such as T-lymphocytes [113], erythrocytes [114], herbicides in drinking water [115]; bacteria such as E.coli [116], Stalphylococcus aureus [117]; drugs (methamphetamine) in human urine [118]; and viruses such as human herpes [119], hepatitis [120], African swine fever [121], and HIV [122]. In particular, piezoelectric sensors facilitated the detection of antibodies that are specific against HIV within a few minutes in human serum sample. It is worth noting that the selectivity and sensitivity was on a par with that of a licensed HIV ELISA and a typical HIV sensor response is given in Fig. 10 [122]. Exploiting advances in genetic sequence information technology and utilizing an inherent strong affinity interaction between complementary nucleic acid strands, nucleic acid, or DNA sensors are emerging as a very important class of biomedical sensors [123], [124]. Unlike immunosensors, which are susceptible to nonspecific binding, DNA sensors provide higher selectivity and reliability. Similar to acoustic immunosensors, no label is required for the detection of DNA. However, much more complex immobilization procedures challenge development of DNA sensors. DNA biomedical sensors typically employ an immobilized single-stranded DNA molecule for hybridizing with the complementary strand in a given sample. Recently, a TSM quartz sensor was used to monitor hybridization and affinity of several synthesized antisense oligonucleotides [125]. Antisense oligonucleotides are potential therapeutic agents used against cancer and viral

262

IEEE SENSORS JOURNAL, VOL. 3, NO. 3, JUNE 2003

(a)

(b) Fig. 10. (a) Resonance frequency change versus the time of a TSM HIV immunosensor and (b) comparison of the results measured with the TSM HIV sensor and with licensed ELISA [122].

in the concentration range between 30200 uM. Chang and Shih [134] developed a fullerene-cryptan-coated piezoelectric membrane glucose enzyme sensor by measuring gluconic acid, a product of glucose oxidase in glucose aqueous solutions. They found that the interference from various common species found in the human blood was negligible. Piezoelectric glucose meters are in a position to pose an alternative solution to well known electrochemical portable glucometers. Last, acoustic biomedical sensors incorporating living cells are capable of delivering functional information in contrast to previously discussed protein-based sensors, which provide analytical data. Functional information, i.e., information about the physiological effect of a measurand on a living system, is often desired in many important applications in pharmacology, toxicology, cell biology, and environmental measurements [135]. Important properties of cultured animal cells such as attachment, proliferation and cell-substrate interaction under different conditions were successfully monitored using TSM sensors. In particular, extracellular matrix deposition, the integrity of the actin cytoskeleton, cell-substrate separation distance, and mechanical properties of the narrow cleft between cell and substrate were determined [136]. Steinem et al. [137] utilized acoustic sensors for analysis of the interactions ganglioside-lectin and ganglioside-toxin (cholera, tetanus, pertusis). Because a cell-based sensing process is physiologically relevant to natural cellular machinery, these sensor types will experience growing significance in the near future. IV. CONCLUSION About $20 billion per year are spent for analytical testing worldwide. Specialized laboratories located away from a patient, doctor, or hospital perform nearly all of the testing causing significant time delays in reporting results. Modern biomedical sensors developed with advanced microfabrication and signal processing techniques are becoming inexpensive, accurate, and reliable and, with an average detection time on the order of a few minutes, can significantly reduce the delay time as well as bring the testing to doctors offices and patients homes. As a result, the wide use of biomedical sensors may lead to more individualized health care services that will be tailored for the needs of a patient and will match a specific genotype. Indeed, one may envision using biomedical sensors for optimizing drug doses, monitoring the effectiveness of treatments, and monitoring health conditions over persons life time. In addition to decreased time delays, miniaturization of biomedical sensors and integration with microfluidic devices is yielding advanced analytical microsystems such as a BioChemLab-on-a-Chip. Integration of several sensors on a single substrate produces transducer arrays. A few recent examples include electronic noses and tongues that are capable of performing multimeasurand detection in a few minutes. For a more complete listing of various biomedical, biological, and biosensing applications, the authors refer the reader to the following references [138][140]. Overall, these exciting developments forecast a biomedical sensor revolution that could dramatically change the way in which the medical, pharmaceutical, and environmental industries practice their fields in the future.

diseases. A combination of a piezoelectric biomedical sensor with PCR-amplified real Aeromonas bacteria samples allowed detection of hybridization using only 205 bp fragments of extracted DNA [126]. Genetic defects were identified with a TSM biomedical sensor by measuring a single mismatch in a 15mer single-strand target of the p53 suppresser gene [127]. Other work reports a successful study on denaturation of DNA [128], the detection of DNA polymorphisms [129] and the real-time monitoring of enzymatic cleavage of nucleic acids [130]. In comparison to traditional DNA analysis methods [123], acoustic sensors offer a rapid and marker-free detection of nucleic acid hybridization. Acoustic DNA sensors should find a broad application in medical area for fast and inexpensive screening of variety of diseases, in the pharmaceutical industry during drug development process and also address many sensing needs in monitoring environmental conditions. Enzymes provide another important interface capable of highly specific reactions with a variety of biological substances. Though enzymes are biocatalysts, they participate in biological transduction processes and have been extensively used for biomedical sensor development. Enzymes catalyze bioreactions at a very high reaction rate, have well-characterized mechanisms of action, are easily immobilized, are available for broad range of applications and are inexpensive [131]. Several AT-cut TSM quartz glucose sensors using immobilized hexokinaze [132] and glucose oxidase [133] have been studied. In the last case, glucose levels were measured in situ and the sensor responded in 80100 s with a linear relation to glucose

COT et al.: EMERGING BIOMEDICAL SENSING TECHNOLOGIES

263

REFERENCES
[1] J. J. Mastrototaro et al., An Electroenzymatic Glucose Sensor Fabricated on a Flexible Substrate, pp. 300302, 1990. , An electroenzymatic glucose sensors fabricated on a flexible sub[2] strate, Sens. Actuators B, vol. 5, pp. 139144, 1991. [3] W. K. Ward and J. E. Troupe, Assessment of chronically implanted subcutaneous glucose sensors in dogs: The effect of surrounding fluid masses, ASAIO 45, pp. 555561, 1999. [4] S. J. Updike, B. J. Gilligan, M. C. Shults, and R. K. Rhodes, A subcutaneous glucose sensor with improved longevity, dynamic range and stability of calibration, Diabetes Care, vol. 23, no. 2, pp. 208214, 2000. [5] G. Rao et al., Reverse iontophoresis: Noninvasive glucose monitoring in vivo in humans, Pharm. Res., vol. 12, no. 12, pp. 18691873, 1995. [6] J. Tamada, N. Bohannon, and R. Potts, Measurement of glucose in diabetic subjects using noninvasive transdermal extraction, Nature Med., vol. 1, no. 11, pp. 11981201, 1995. [7] J. Tamada et al., Noninvasive glucose monitoring. Comprehensive clinical results, J. Amer. Med. Assoc., vol. 282, no. 19, pp. 18391844, 1999. [8] J. C. Armour, J. Y. Lucisano, B. D. McKean, and D. A. Gough, Application of chronic intravascular blood glucose sensor in dogs, Diabetes, vol. 39, pp. 15191526, 1990. [9] A. E. G. Cass et al., Ferrocene-mediated enzyme electrode for amperometric determination of glucose, Anal. Chem., vol. 56, pp. 66771, 1984. [10] E. Csoregi et al., Design, characterization and one-point in vivo calibration of a subcutaneously implanted glucose electrode, Anal. Chem., vol. 66, pp. 31313138, 1994. [11] E. Csoregi, D. Schmidtke, and A. Heller, Design and optimization of a selective subcutaneously implantable glucose electrode based on wired glucose oxidase, Anal. Chem., vol. 67, pp. 12401244, 1995. [12] C. P. Quinn et al., Kinetics of glucose delivery to subcutaneous tissue in rats: A study utilizing amperometric biosensors, Amer. J. Physiol., vol. 269, no. 32, p. E155, 1995. [13] G. Wittstock and W. Schuhmann, Formation and imaging of microscopic enzymatically active spots on an alkanethiolate-covered gold electrode by scanning electrochemical microscopy, Anal. Chem., vol. 69, pp. 50595066, 1997. [14] K. Sirkar and M. Pishko, Amperometric biosensors based on oxidoreductases immobilized in photopolymerized poly(ethylene glycol) redox polymer hydrogels, Anal. Chem., vol. 70, pp. 28882894, 1998. [15] A. N. J. Moore, E. Katz, and I. Willner, Immobilization of pyrroloquinoline quinone on gold by carbodiimide coupling to adsorbed polyamines: Stability comparison to attachment via a chemisorbed thiol monolayer, Electroanal., vol. 8, no. 12, p. 1092, 1996. [16] A. Riklin et al., Improving enzyme-electrode contacts by redox modification of cofactors, Nature, vol. 376, pp. 672675, 1995. [17] I. Willner et al., Electrical wiring of glucose oxidase by reconstitution of FAD-modified monolayers assembled onto au-electrodes, J. Amer. Chem. Soc., vol. 118, pp. 10 32110 322, 1996. [18] I. Willner and B. Willner, Electrical communication of redox proteins by means of electron relay-tethered polymers in photochemical, electrochemical and photoelectrochemical systems, React. Polym., vol. 22, pp. 267279, 1994. [19] I. Willner, E. Katz, and B. Willner, Electrical contact of redox enzyme layers associated with electrodes: Routes to amperometric biosensors, Electroanal., vol. 9, no. 13, pp. 965977, 1997. [20] K. Sirkar, A. Revzin, and M. Pishko, Glucose and lactate biosensors based on redox polymer/oxidoreductase nanocomposite thin films, Anal. Chem., vol. 72, no. 13, pp. 29302936, 2000. [21] D. Huang, E. A. Swanson, C. P. Lin, J. S. Schumann, W. G. Stinson, W. Chang, M. R. Hee, T. Flotte, K. Gregory, C. A. Puliafito, and J. G. Fujimoto, Optimal coherence tomography, Science, vol. 254, pp. 11781181, 1991. [22] R. A. Wolthuis, G. L. Mitchell, E. Saaski, J. C. Hartl, and M. A. Afromowitz, Development of medical pressure and temperature sensors employing optical spectrum modulation, IEEE Trans. Biomed. Eng., vol. 38, pp. 97481, Oct. 1991. [23] C. K. Hitzenberger, A. Baumgartner, W. Drexler, and A. F. Fercher, Interferometric measurement of Corneal thickness with micrometer precision, Amer. J. Ophthal., vol. 118, no. 4, pp. 468476, 1994. [24] J. Merchant, Optical Coherence Reflectometry for the Measurement of Collagen Thickness, Masters thesis, Dept. Biomed. Eng., Texas A&M Univ., College Station, 1998.

[25] P. Hrynchak and T. Simpson, Optical coherence tomography: An introduction to the technique and its use, Optom. Vis. Sci., vol. 77, no. 7, pp. 34756, July 2000. [26] Light Lab Imaging, Westford, MA. [Online]. Available: http://www.lightlabimaging.com/oct.html [27] M. E. Brezinski, G. J. Tearney, B. E. Bouma, J. A. Izatt, M. R. Hee, E. A. Swanson, J. F. Southern, and J. F. Fujimoto, Optical coherence tomography for optical biopsy, Properties Demonstration Vasc. Pathol., Circ., vol. 93, no. 6, pp. 120613, Mar. 1996. [28] M. E. Brezinski, G. J. Tearney, N. J. Weissman, S. A. Boppart, B. E. Bouma, M. R. Hee, A. E. Weyman, E. A. Swanson, J. F. Southern, and J. G. Fujimoto, Assessing atherosclerotic plaque morphology: Comparison of optical coherence tomography and high frequency intravascular ultrasound, Heart, vol. 77, no. 5, pp. 397403, May 1997. [29] Profox Oximetry Software. PROFOX Associates, Inc., Escondido, CA. [Online]. Available: http://www.profox.net/pages/2compdevs.html [30] Pulse Oximetry. Novametrix Medical Systems Inc., Wallingford, CT. [Online]. Available: http://www.novametrix.com/ix_ox.html [31] Ear Thermometer Review From Epinions Inc., Brisbane, CA. [Online]. Available: http://www.epinions.com/well-Supplies-Home_DiagnosticsThermometers-All-Ear [32] W. Groner, J. W. Winkelman, A. G. Harris, C. Ince, G. J. Bouma, K. Messmer, and R. G. Nadeau, Orthogonal polarization spectral imaging: A new method for study of the microcirculation, Nat. Med., vol. 5, no. 10, pp. 120912, Oct. 1999. [33] Hemoscan 1000 Noninvasive Optical Monitor for Hemoglobin and Hematocrit. Cytometrics Inc., Philadelphia, PA. [Online]. Available: http://www.cytometrics.com/ [34] I. S. Saidi, S. L. Jacques, and F. K. Tittel, Monitoring neonatal bilirubinemia using an optical patch, in Proc. SPIE Opt. Fibers Med. V, vol. 1201, A. Katzir, Ed., 1990, pp. 569578. [35] I. S. Saidi, S. L. Jacques, M. Keijzer, and F. K. Tittel, Optical fiber probe monitor for neonatal bilirubinemia, in Proc. 11th Int. Conf. IEEE Eng. Med. Biol. Soc. 2, 1989, pp. 120102. [36] J. Enderle, S. Blanchard, and J. Bronzino, Introduction to biomedical engineering, in Biomed. Optics and Lasers Contributed, G. L. Cot, S. Rastegar, and L. Wang, Eds. New York: Academic, 2000, ch. 17, pp. 843903. [37] K. Day, SEC accuses Futrex Incorporated of fraud, Washington, DC, 1996. [38] P. Sabatini, Biocontrols Diabetes monitor still faces hurdles, Burlington, MA, 1996. [39] B. K. Jacobson. Biosensors and Other Medical and Environmental Probes. ONRL Center for Biotechnology. [Online]. Available: http://www.ornl.gov/ORNLReview/rev29_3/text/biosens.htm [40] G. L. Cot, Noninvasive optical glucose sensingAn overview, J. Clinical Eng., vol. 22, no. 4, pp. 2539, 1997. [41] J. R. Mourant, M. Canpolat, C. Brocker, O. Esponda-Ramos, T. M. Johnson, A. Matanock, K. Stetter, and J. P. Freyer, Light Scattering From Cells: The Contribution of the Nucleus and the Effects of Proliferative Status, vol. 5, no. 2, pp. 13137, Apr. 2000. [42] A. Mahadevan-Jansen and R. Richards-Kortum, Raman spectroscopy for the detection of cancers and precancers, J. Biomed. Opt., vol. 1, no. 1, pp. 3170, 1996. [43] R. J. McNichols and G. L. Cot, Optical glucose sensing in biological fluids: An overview , J. Biomed. Opt., vol. 5, no. 1, pp. 516, Jan. 2000. [44] A. J. Berger, I. Itzkan, and M. S. Feld, Feasibioity of measuring blood glucose concentration by near infrared Raman spectroscopy, Spectrochim. Acta A., vol. 53, no. 2, pp. 28792, 1997. [45] M. J. Goetz Jr., G. L. Cot, W. F. March, R. Erckens, and M. Motamedi, Application of a multivariate technique to Raman spectra for quantification of body chemicals, IEEE Trans. Biomed. Eng., vol. 42, pp. 72831, July 1995. [46] J. Lambert, M. Storrie-Lombardi, and M. Borchert, Measurement of physiological glucose levels using Raman spectroscopy in a rabbit aqueous humor model, IEEE-LEOS Newslett., vol. 12, no. 2, pp. 1922, 1998. [47] R. V. Tarr, The non-invasive measure of D-glucose in the ocular aqueous humor using stimulated Raman spectroscopy, Ph.D. dissertation, Georgia Inst. Technol., Atlanta, 1991. [48] S. Y. Wang et al., Analysis of metabolites in aqueous solutions by using laser Raman spectroscopy, Appl. Opt., vol. 32, no. 6, pp. 925929, 1993. [49] J. P. Wicksted, R. J. Erkens, M. Motamedi, and W. F. March, Monitoring of aqueous humor metabolites using Raman spectroscopy, in Proc. SPIE, vol. 2135, 1994, pp. 264274.

264

IEEE SENSORS JOURNAL, VOL. 3, NO. 3, JUNE 2003

[50] D. P. ONeal, M. Motamedi, J. Chen, and G. L. Cot, Surface-enhanced Raman spectroscopy for the near real-time diagnosis of brain trauma in rats, in Proc. SPIE Int. Symp. Biomed. Opt., Optical Diagnostics of Biological Fluids V3918, San Jose, CA, Jan. 2228, 2000. [51] U. Utzinger, E. V. Trujillo, E. N. Atkinson, M. F. Mitchell, S. B. Cantor, and R. Richards-Kortum, Performance estimation of diagnostic tests for cervical precancer based on fluorescence spectroscopy: Effects of tissue type, sample size, population and signal-to-noise ratio, IEEE Trans. Biomed Eng., vol. 46, pp. 1293303, Nov. 1999. [52] A. L. Bartorelli, M. B. Leon, Y. Almagor, L. G. Prevosti, J. A. Swain, C. L. McIntosh, R. F. Neville, M. D. House, and R. F. Bonner, In vivo human atherosclerotic plaque recognition by laser-excited fluorescence spectroscopy, J. Amer. Coll. Cardiol., vol. 17, no. 6 Suppl B, pp. 160B168B, May 1991. [53] M.-C. Moreno-Bondi, J. Mobley, J.-P. Alarie, and T. Vo-Dinh, Antibody-based biosensor for breast cancer with ultrasonic regeneration, J. Biomed. Opt., vol. 5, no. 3, pp. 35054, July 2000. [54] Nanosensor Probes Single Living Cells. ONRL Center for Biotechnology. [Online]. Available: http://www.ornl.gov/ORNLReview/rev32_3/nanosens.htm [55] R. Russell, C. Gefrides, M. McShane, G. Cote, and M. Pishko, A fluorescence-based glucose biosensor based on concanavalin A and dextran encapsulated in a Poly(ethylene glycol) hydrogel, Anal. Chem., vol. 71, no. 15, pp. 31263132, 1999. [56] E. M. Sevick-Muraca, J. S. Reynolds, T. L. Troy, G. Lopez, and D. Y. Paithankar, Fluorescence lifetime spectroscopic imaging with measurements of photon migration, Ann. N.Y. Acad. Sci., vol. 838, pp. 4657, Feb. 1998. [57] J. R. Lakowicz and B. Maliwal, Optical sensing of glucose using phasemodulation fluorimetry, Anal. Chim. Acta, vol. 271, pp. 15564, 1993. [58] J. S. Schultz, S. Mansouri, and I. J. Goldstein, Affinity sensor: A new technique for developing implantable sensors for glucose and other metabolites, Diabetes Care, vol. 5, no. 3, pp. 245253, 1982. [59] Z. Rosenzweig and R. Kopelman, Analytical properties and sensor size effects of a micrometer-sized optical fiber glucose biosensor, Anal. Chem., vol. 68, no. 8, pp. 140813, 1996. [60] B. P. Schaffar and O. S. Wolfbeis, A fast responding fiber optic glucose biosensor based on an oxygen optrode, Biosens. Bioelectron., vol. 5, no. 2, pp. 13748, 1990. [61] R. Ballerstadt and J. S. Schultz, Competitive-binding asay method based on fluorescence quenching of ligands held in close proximity by a multivalent receptor, Anal. Chim. Acta, vol. 345, pp. 20312, 1997. [62] S. Mansouri and J. S. Schultz, A miniature optical glucose sensor based on affinity binding, Biotechnol., pp. 88590, 1984. [63] D. L. Meadows and J. S. Schultz, Design, manufacture and characterization of an optical fiber glucose affinity sensor based on an homogeneous fluorescence energy transfer assay system, Anal. Chim. Acta, vol. 280, pp. 2130, 1993. [64] R. Dalterio, W. Nelson, D. Britt, J. Sperry, J. Tanguay, and S. Suib, The steady-state and decay characteristics of primary fluorescence from live bacteria, Appl. Spectrosc., vol. 41, p. 234, 1987. [65] C. Barlow and B. Chance, Ischemic areas in perfused rat hearts: Measurement by NADH fluorescence photography, Science, vol. 193, p. 909, 1976. [66] C. Estes, A. Duncan, B. Wade, and L. Powers, Real-time multiwavelength microbe detection instrument, in Abstract for Poster Presentation at the EMBS/BMES Joint Conf., Atlanta, GA, Oct. 1316, 1999. [67] K. A. Giuliano, P. L. Post, K. M. Hahn, and D. L. Taylor, Fluorescent protein biosensors: Measurement of molecular dynamics in living cells, Annu. Rev. Biophys. Biomol. Struct., vol. 24, pp. 40534, 1995. [68] G. Spera, Biosensor research targets medical diagnostics, MDDI, Nov. 1997. [69] T. Vo-Dinh. Virtual Poster Session: Biosensors and Biochips for Environmental and Biomedical Applications. Advanced Monitoring Development Group, Life Sciences Division, Oak Ridge Nat. Lab., Oak Ridge, TN. [Online]. Available: http://www.ornl.gov/virtual/biosensors/ [70] B. D. Cameron, H. Gorde, B. Satheesan, and G. L. Cot, The use of polarized laser light through the eye for noninvasive glucose monitoring, Diabetes Technol. Therapeutics, vol. 1, no. 2, pp. 13543, 1999. [71] W. F. March, B. Rabinovitch, and R. L. Adams, Noninvasive glucose monitoring of the aqueous humor of the eye: Part II animal studies and the scleral lens, Diabetes Care, vol. 5, no. 3, pp. 259265, 1982. [72] M. J. Rakovic, G. Kattawar, M. Mehrubeoglu, B. D. Cameron, L. V. Wang, S. Rastegar, and G. L. Cote, Light backscattering polarization patterns from turbid media: Theory and experiment, Appl. Opt., vol. 38, no. 15, pp. 33993408, May 1999.

[73] S. L. Jacques, J. R. Roman, and K. Lee, Imaging superficial tissues with polarized light, lasers in surgery and medicine, vol. 26, pp. 119129, 2000. [74] F. Madarasz. The Photonic Molecular Probe, a Noninvasive Blood Glucose Meter. International Diagnostic Technologies, Inc., Madison, AL. [Online]. Available: http://www.idtscience.com/Default.htm [75] V. Kupershmidt, Method and Apparatus for Non-Invasive Phase Sensitive Measurement of Blood Glucose Concentration, U.S. Patent, 5 448 992, Sept. 1995. [76] A. H. Hielscher, J. R. Mourant, and I. J. Bigio, Influence of particle size and concentration on the diffuse backscattering of polarized light from tissue phantoms and biological cell suspensions, Appl. Opt., vol. 36, pp. 12535, 1997. [77] S. L. Jacques. (1998) Video Imaging With Polarized Light Finds Skin Cancer Margins not Visible to Dermatologists. Oregon Medical Laser Center NewsEtc.. [Online]. Available: http://omlc.ogi.edu/news/feb98/polarization/index.html [78] A. J. Matheson, Molecular Acoustic. New York: Wiley-Interscience, 1971. [79] A. B. Bathia, Ultrasonics Absorption. London, U.K.: Clarendon, 1967. [80] D. S. Ballantaine, R. M. White, S. J. Martin, A. J. Ricco, E. T. Zellers, G. C. Frye, and Wohltjen, Acoustic Wave Sensors. New York: Academic, 1997. [81] R. L. Bunde, E. J. Jarvi, and J. Rosentreter, Piezoelectric quartz crystal biosensors, Talanta, vol. 46, pp. 12231236, 1998. [82] R. M. Lec and P. A. Lewin, Acoustic wave biosensors, Proc. 20th Annu. IEEE Eng. Med. Biol. Soc., vol. 20, no. 6, pp. 27852790, 1998. [83] J. C. Andle and J. F. Vetelino, Acostic wave biosensors, Sens. Actuators A, vol. 44, pp. 167176, 1994. [84] M. G. Silk, Ultrasonic Transducers for Nondestructive Testing. New Yorkl: Adam Hilger, 1984. [85] F. Eichelbaum, R. Borngraber, J. Schroder, and R. Lucklum, Interface circuits for quartz-crystal microbalance sensors, Rev. Sci. Instrum., vol. 70, no. 5, pp. 211, 1999. [86] H. Takeuchi, H. Abe, and K. Yamanouchi, Ultrasonic micromanipulation of small particles in liquid using VHF-range leaky wave transducer, in Proc. Ultrason. Symp., Cannes, France, 1994, pp. 607611. [87] A. Lal, Silicon-Based ultrasonic surgical actuators, Proc. 20th Annu. IEEE Eng. Med. Biol. Soc., vol. 20, no. 6, pp. 27852790, 1998. [88] Sandias Tiny Acoustic Wave Sensors Will Detect Minute Traces of Dangerous Chemicals, Sandia Nat. Lab. News Releases, Albuquerque, NM, 1999. [89] G. McHale, M. I. Newton, M. K. Banerjee, and J. A. Cowen, Acoustic wave-liquid interactions, Mater. Sci. Eng. C, vol. 12, pp. 1722, 2000. [90] M. Vellekoop, Acosutic wave sensors and their technology, Ultrason., vol. 36, pp. 714, 1998. [91] F. Josse, R. Dahint, J. G. Schmacher, J. C. Andale, and J. F. Vetelino, On the mass sensitivity of acoustic-plate-mode sensors, Sens. Actuators A, vol. 53, pp. 243248, 1996. [92] X. Su, F. Chew, and S. F. Y. Li, Self-assembled monolayer-based piezoelectric crystal immonosensor for the quantification of total human immunoglobulin E, Anal. Biochem., vol. 273, pp. 6672, 1999. [93] J. Rickert, T. Weiss, W. Kraas, G. Jung, and W. Gopel, A new affinity biosensor: Self-assembled thiols as selective monolayers coatingd of quartz crystal microbalances, Biosens. Bioelectron., vol. 11, pp. 591598, 1996. [94] L. I. Andersson, Molecular imprinting: Developments and applications in the analytical chemistry field, J. Chromatography B, vol. 745, pp. 313, 2000. [95] R. S. Kane, S. Takayama, E. Ostuni, D. E. Ingber, and G. M. Whitesides, Patterning proteins and cells using soft lithography, Biomater., vol. 20, pp. 23632376, 1999. [96] R. Lucklum, C. Behling, and P. Hauptmann, Role of mass accumulation and viscoelastic film properties for the response of acoustic-wave-based chemical sensors, vol. 71, pp. 24882496, 1999. [97] C. Behling, R. Lucklum, and P. Hauptmann, Response of quartz-crystal resonators to gas and liquid measurand exposure, Sens. Actuators A, vol. 68, pp. 388398, 1998. [98] R. Lucklum, C. Behling, and P. Hauptmann, Gravimetric and nongravimetric chemical quartz resonators, Sens. Actuators B, vol. 65, pp. 277283, 2000. [99] B. Knig and M. Grtzel, A piezoelectric immunosensor for hepatitus viruses, Anal. Chim. Acta, vol. 309, pp. 1925, 1995. [100] S. J. Martin and G. C. Frye, Dynamics and response of polymer-coated acoustic devices, solid-state sensor and actuator workshop, in 5th Tech. Dig., 1992, pp. 2731.

COT et al.: EMERGING BIOMEDICAL SENSING TECHNOLOGIES

265

[101] T. W. Schneider and S. J. Martin, Influence of compressional wave generation on thickness-shear mode resonator responses in a fluid, Anal. Chem., vol. 65, no. 18, pp. 33243335, 1995. [102] M. Urbakh and L. Daikhin, Surface Morphology and The Quartz Crystal microbalance response in Liquids, Coll. Surf. A, vol. 134, pp. 7584, 1998. [103] W. C. Duncan-Hewitt, Four-layer theory for the acoustic shear wave sensor in liquids incorporating interfacial slip and liquid structure, Anal. Chem., vol. 64, pp. 94105, 1992. [104] N. M. McHale, M. Banerje, and S. Rowan, Interaction of surface acoustic waves with viscous liquids, Faraday Discuss, vol. 107, pp. 1526, 1997. [105] M. G. Schwyer, J. A. Hilton, J. E. Munson, J. C. Andle, J. M. Hammond, and R. M. Lec, A novel monolithic piezoelectric sensor, in Proc. IEEE Int. Freq. Contr. Symp., Orlando, FL, May 2931, 1997, pp. 3241. [106] T. Wessa, M. Rapp, and H. J. Ache, New immobilization method for SAW-biomedical sensors: Covalent attatchment of antibodies via CNBr, Biomed. Sens. Bioelectron., vol. 14, pp. 9398, 1999. [107] N. Bari, M. Rapp, H. Sigrist, and H. J. Ache, Covalent photolinker-mediated immobilization of an intermediate dextran layer to polymer-caoted surfaces for biosensing applications, Biomed. Sens. Bioelectron., vol. 13, pp. 855860, 1998. [108] B. Ratner, Surface modification of polymers: Chemical, biological and surface analytical challenges, Biosens. Bioelectron., vol. 10, pp. 797804, 1995. [109] S. J. Kwoun, R. M. Lec, B. Hun, and F. Ko, A novel polymer nanofiber interface for chemical sensor applications, in Proc. IEEE/EIA Int. Freq. Contr. Symp., Kansas City, MO, 2000, pp. 5257. [110] S. P. Sakti, S. Rosler, L. Lucklum, P. Hauptmann, F. Bhling, and S. Ansorge, Thick polystylene-coated quartz crystal microbalanece as a basis of a cost effective immunosensor, Sens. Actuators A, vol. 76, pp. 98102, 1999. [111] N. Barie, H. Sigrist, and M. Rapp, Development of immunosensors based on commercially available surface acoustic wave (SAW) devices, Analusis, vol. 27, pp. 622629, 1999. [112] X. Chiu, J. Jiang, G. Shen, and R. Yu, Simultaneous immunoasssay using piezoelectric immunosensor array and robust method, Anal. Chim., vol. 336, pp. 185193, 1996. [113] B. Konig and M. Gratzel, Detection of human T-lymphocytes with a piezoelectric immunosensor, Anal. Chim. Acta, vol. 281, pp. 1318, 1993. , Development of piezoelectric immunosensor for the detection of [114] human erythrocytes, Anal. Chim. Acta, vol. 276, pp. 329233, 1993. [115] K. Yokoyama, Ikebukuro, E. Tamiya, I. Karube, N. Ichiki, and Y. Arikawa, Highly sensitive quartz immunosesnors for multisample detection of herbicides, Anal. Chim. Acta, vol. 304, pp. 139145, 1995. [116] H. Maramatsu, Y. Watanabe, M. Hikuma, I. Ataka, E. Tamiya, and I. Karube, Piezoelectric crystal biosensor system for detection of E. coli., Anal. Lett., vol. 229, pp. 21552166, 1989. [117] D. Le, H. Feng, J. Tai, N. Lihua, and S. Yao, A goat-anti-human modified piezoimmunosesnor for staphylococcus aureus detection, J. Microbiol. Methods, vol. 23, pp. 229234, 1996. [118] N. Miura, H. Higobashi, G. Sakai, A. Takeyashu, T. Uda, and N. Yamazoe, Piezoelectric crystal immunosensor for sensitive detection of methamphetamine in (stimulant drug) in human urine, Sens. Actuators B, vol. 13, pp. 188191, 1993. [119] B. Konig and M. Gratzel, A novel immunosensor for herpes viruses, Anal. Chem., vol. 68, pp. 341344, 1994. , A piezoelectric immunosensor for hepatitis viruses, Anal. Chim. [120] Acta, vol. 309, pp. 1925, 1995. [121] E. Uttenthaler, C. Kolinger, and S. Drost, Characterization of immobilization methods for African swine fever virus protein and antibodies with a piezoelectric immunosensor, Biosens. Bioelectron., vol. 13, pp. 12791286, 1998. [122] F. Arberl and H. Wolf, HIV serology using piezoelectric immunosensors, Sens. Actuators B, vol. 1819, pp. 271275, 1994. [123] C. M. Roth and M. L. Yarmush, Nucleic acid biotechnology, Annu. Rev. Biomed. Eng., vol. 01, pp. 265297, 1999. [124] J. Wang, DNA biosensors based on peptide nucleic acid (PNA) recognition layers, Biosens. Bioelectron., vol. 13, pp. 752762, 1998. [125] T. Ketterer, H. Stadler, E. Rickert, E. Bayer, and W. Gpel, Detection of oligonucleotide sequences with quartz crystal oscillators, Sens. Actuators B, vol. 65, pp. 7375, 2000.

[126] T. Sara, M. Marco, S. Cristina, and A. P. F. Turner, A DNA piezoelectric biosensor assay coupled with a polymerase chain reaction for bacterial toxicity determination in environmental samples, Anal. Chim. Acta, vol. 418, pp. 19, 2000. [127] P. Wittung-Stafshede, M. Rodahl, B. Kasemo, P. Nielson, and B. Norden, Detection of point mutations in DNA by PNA-based quartz-crystal biosensor, Coll. Surf., vol. 174, pp. 269273, 2000. [128] Y. Wu, A. Zhou, Q. Xie, and S. Yao, A piezoelectric quartz crystal impedance study on denaturation of herring DNA, Microchem. J., pp. 6774, 2000. [129] S. Tombelli, M. Mascini, L. Braccini, M. Anichini, C. Anichini, and A. P. Turner, Coupling of a DNA piezoelectric biosensor and polymerase chain reaction to detect apolipoprotein E polymorphism, Biosens. Bioelelcron., vol. 15, pp. 363370, 2000. [130] J. Wang, M. Jiang, and E. Palecek, Real-time monitoring of enzymatic clevage of nucleic acids using a quartz crystal microbalance, vol. 48, pp. 477480, 1999. [131] R. Freitag, Utilization of enzyme-substrate interactions in analtical chemistry, J. Chromatography B, vol. 722, pp. 279301, 1999. [132] S. J. Lasky and D. A. Butry, Chemical sensors and instrumentation, in ACS Symposium Series 43. [133] S. M. Reddy, J. P. Jones, T. J. Lewis, and P. M. Vadgama, Development of an oxidase-based glucose sensor using thicknerss-shear-mode quartz crystals, Anal. Chim. Acta, vol. 363, pp. 203213, 1998. [134] M. Chang and J. Shih, Fullerence-cryptand-coated piezoleectric crystal membrane glucose enzyme sensor, Sens. Actuators B, vol. 67, pp. 275281, 2000. [135] J. J. Pancrazio, J. P. Whelan, D. A. Borkholder, W. Ma, and D. A. Stenger, Development and application of cell-based biosensors, Ann. Biomed. Eng., vol. 27, pp. 697711, 1999. [136] C. Steinem, A. W. Janshoff, W. Ulrich, W. Willenbrink, and G. H. Sieber, Impedance and shear wave resonance analysis of ligand-receptor interactions at functionalized syrfaces and of cell monolayers, Biosens. Bioelectron., vol. 12, no. 8, pp. 787808, 1997. [137] A. Janshoff, C. Steinem, M. Sieber, A. Bay, M. Schmidt, and H. Galla, Quartz crystal microbalance investigation of the interaction of bacterial toxins with ganglioside containing solid supported membranes, Eur. Biophys. J., vol. 26, pp. 261270, 1997. [138] J. Biosens. Bioelectron. (formerly Biosens.), Elsevier. [139] Europtrodes Conf. Proc., Anal. Chem., Biosens. Bioelectron., Sens. Actuators B. [140] SPIE Photonics West BiOS Conf. Proc.

Gerard L. Cot (SM98) received the B.S.E.E. degree from the Rochester Institute of Technology, Rochester, NY, in 1986 and the M.S. and Ph.D. degrees in bioengineering from the University of Connecticut, Storrs, in 1987 and 1990, respectively. He is a Professor and faculty member of the Department of Biomedical Engineering and the Director of the Optical Biosensing Laboratory, Texas A&M University, College Station. His primary research interests focus on the use of optics for medical diagnostics such as cancer and optical biosensing including various analytes (glucose, excitatory amino acids, lactate, nitric oxide, oxygen, etc.). He has over 120 publications. Dr. Cot has been the Chair of several conference tracks and sessions and has served as a reviewer on several government panels. He is also an Associate Editor of the IEEE SENSORS JOURNAL. He is a member of the SPIE Biomedical Optics Group, the Biomedical Engineering Society (BMES), the American Society for Engineering Education (ASEE), and Tau Beta Pi Engineering Honor Society. He is the recipient of several awards, including the ASA Statistics in Chemistry Award, the JDRFI Mary Jane Kugel Award, the Faculty Distinguished Achievement Award in Teaching, and the TEES Senior Fellow Award.

266

IEEE SENSORS JOURNAL, VOL. 3, NO. 3, JUNE 2003

Ryszard M. Lec (M88) received the Ph.D. degree from the Warsaw University of Technology, Warsaw, Poland, in 1978. He is a Professor at the School of Biomedical Engineering, Science and Health Systems, Drexel University, Philadelphia, PA. From 1980 to 1981, he was Visiting Scientist at the Department of Electrical Engineering at the University of Marine, Poland. He then joined the Institute of Fundamental Technological Research, Warsaw, where he was Head of the Ultrasonic Spectroscopy Laboratory from 1981 to 1986. From 1986 to 1990, he was Visiting Professor in the Electrical and Computer Engineering Department, University of Marine. From 1991 to 1998, he was Research Professor of electrical and computer engineering, University of Marine, and Technical Leader of the Industrial Process Control Sensor System Program, College of Engineering, University of Marine. He has been active in the areas of material science and instrumentation for more than 20 years. Specifically, his research efforts have been devoted to the study of viscoelastic, acoustooptic (AO), ultrasonic properties, and liquid and solid media, with a focus on biomedical applications. In addition, he has developed several associated electronic instrumentations, including ultrasonic spectrometers, AO Q-switches and filters, and acoustic resonant systems. Since the mid-1980s, his interests have been in the application of acoustic, piezoelectric, ultrasonic, and optical technologies for the development of sensors. In particular, he has designed and fabricated a variety of sensors for the medical, biochemical, chemical, and automotive industries, including immunosensors, micro-viscometers, engine oil quality sensors, gas sensors, and sensors for monitoring the kinetics of chemical reactions. Dr. Lec has been a Vice Chairman of the Technical Program Committee of the IEEE International Frequency Control Symposium and a co-Chairman of the IEEE UFFC Standard Sub-Committee on Sensors since 1997. He was a member of the IEEE Sensor Council from 1998 to 2000, and he has authored and co-authored over 50 publications and one book and is co-inventor of three issued U.S. patents.

Michael V. Pishko received the B.S. and M.S. degrees in chemical engineering from the University of Missouri, Columbia, and the Ph.D. degree in chemical engineering from the University of Texas, Austin. He was a postdoctoral associate at the Massachusetts Institute of Technology, Cambridge. He is an Associate Professor in the Departments of Chemical Engineering and Materials Science and Engineering, The Pennsylvania State University, University Park. His research interests include biosensing, biomaterials, and drug delivery. He has over 50 publications and 18 issued U.S. patents. Dr. Pishko is a member of the editorial board and an Associate Editor of the IEEE SENSORS JOURNAL. He has received several awards, including an Alfred P. Sloan Research Fellowship and the JDRFI Mary Jane Kugel Award.