HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hall, A. Articles by Girkin, J.M. Search for Related Content PubMed PubMed Citation Articles by Hall, A. Articles by Girkin, J.M.

A Review of Potential New Diagnostic Modalities for Caries Lesions

¹ Glasgow Dental Hospital & School, Adult Dental Care, Glasgow University, 378 Sauchiehall St., Glasgow G2 3JZ, UK; and
² Institute of Photonics, University of Strathclyde, Glasgow, G4 0NW, UK;

^* corresponding author FAX, +44-141-3112798, a.hall{at}dental.gla.ac.uk

	ABSTRACT

TOP ABSTRACT INTRODUCTION PHYSICAL PRINCIPLES BEHIND... MULTI-PHOTON IMAGING INFRARED THERMOGRAPHY INFRARED FLUORESCENCE OPTICAL COHERENCE TOMOGRAPHY ULTRASOUND TERAHERTZ IMAGING SUMMARY REFERENCES

 
This paper aims to present a simple overview of potential new diagnostic methods for dental caries. There are several novel methods of caries detection (with potential application to diagnosis) which have been proposed in the last few years, in addition to those that are gaining some commercial exposure and clinical acceptance. For the most part, these methods have been demonstrated in laboratories and are generally many years away from routine clinical application. They include multi-photon imaging, infrared thermography and infrared fluorescence, optical coherence tomography, ultrasound, and terahertz imaging.

KEY WORDS: caries diagnosis • optical imaging • terahertz imaging • ultrasound • thermography

	INTRODUCTION

TOP ABSTRACT INTRODUCTION PHYSICAL PRINCIPLES BEHIND... MULTI-PHOTON IMAGING INFRARED THERMOGRAPHY INFRARED FLUORESCENCE OPTICAL COHERENCE TOMOGRAPHY ULTRASOUND TERAHERTZ IMAGING SUMMARY REFERENCES

 
In this paper, we will provide a general background on the interaction of waves with dental tissue and then examine in some detail the methods of operation of these techniques and the potential advantages and disadvantages of each method. The techniques described are in their developmental stages. Some of them may never become practical applications for caries detection or diagnosis. However, the continual development of ways to monitor energy (wave) interactions with tooth tissue is currently the most promising route in which new and useful techniques may develop.

It is now recognized that the diagnosis of caries involves establishing the clinical presence of caries in addition to determining the activity of the lesion (Featherstone, 1996). The clinical presence of a lesion, however, is being continually redefined as advances are made in caries detection methodology. Descriptors such as the iceberg diagrams (Pitts, 1994) describe caries as a continuum from enamel, through dentin, to the pulp, where D₁ and D₂ represent caries limited to enamel only. In current clinical practice, this may be defined only in radiographic or histological terms, or inferred by visual methods of detection (Ekstrand et al., 1998). Other evolving methods of caries detection evaluate indirectly the quantity of mineral lost from a lesion relative to the surrounding sound tooth tissue. From this information, mineral loss may be inferred with various degrees of accuracy. The activity of a lesion may be determined by serial evaluation (Al-Khateeb et al., 1998). In some instances, activity may be inferred from a single observation. For example, recording the rate of fluid loss from a lesion has been suggested in relation to lesion activity. Techniques to measure the rate of fluid loss from a lesion have included Quantitative Light Fluorescence (QLF) (van der Veen et al., 1999) and infrared thermography (Kaneko et al., 1999).

If early detection and definition of lesion activity are our goal, then what methods are potentially useful in this quest? This conference will describe visual, tactile, radiographic, electrical, and two optical methods of caries detection. Without serial evaluation, none of these methods claims to be diagnostic for caries, with the exception, perhaps, of some of the visual methods in combination with clinical experience of the patient, the remainder of their dentition, and factors such as salivary flow rate and buffering capacity.

There are other methods potentially capable of detecting the demineralization which occurs as a result of the caries process. Some methods may not be applicable clinically, since they may damage the tooth in the search for caries. Some methods may be limited to particular teeth or particular surfaces of teeth. What is clear is that the usefulness and limitations of many methods may be defined by our understanding of how energy sources, which could be useful for detection purposes, interact with dental tissue.

	PHYSICAL PRINCIPLES BEHIND DIAGNOSTIC TECHNIQUES

TOP ABSTRACT INTRODUCTION PHYSICAL PRINCIPLES BEHIND... MULTI-PHOTON IMAGING INFRARED THERMOGRAPHY INFRARED FLUORESCENCE OPTICAL COHERENCE TOMOGRAPHY ULTRASOUND TERAHERTZ IMAGING SUMMARY REFERENCES

 
Most caries detection and diagnosis require observation of the interaction of energy which is applied to the tooth, or the observation of energy which is emitted from the tooth. Such energy is most often in the form of a wave. When waves interact with objects (including teeth), they can do so in the following ways (Fig. 1): scattering, reflection, absorption, or fluorescence. Scattering of waves can be divided depending on the wavelength of the light, relative to the size of the features within the sample. The scattered light will then emerge from the sample (transmission) or be backscattered toward the source. Fluorescence is mostly associated with light, although the phenomenon is applicable to other wavelengths. Fluorescence is observed in two ways. These are the measurement of the wavelength () of fluorescent energy (light) emitted or the time taken for the fluorescent energy (light) to disappear (so-called fluorescence decay or lifetime fluorescence). In truth, most interactions of waves are a combination of processes—for example, multiple scattering followed by absorption. Each of the ways in which waves interact with objects can be defined briefly.

View larger version (23K): [in this window] [in a new window]   Figure 1. A schematic diagram to show some of the methods by which waves may interact with objects.

Reflection
Reflection is a surface phenomenon resulting in the change in direction of a wave by a single interaction with a large object, the direction of the reflected wave often being opposite the incident wave. Here the analogy would be a single ‘bounce’ of the squash ball off the smooth floor of the court. In reflection, the wavelength, or energy, of the light or sound is not altered.

Absorption
Absorption requires the wave to be stopped by an object, the wave energy taken in by the object, and then converted into a different form, such as heat. An object can also stop the wave energy and then convert the energy into another wave which has less energy and hence a longer wavelength. When this happens to light waves, the emission of the longer wavelength light is known as fluorescence.

Wave energy can be of two forms, transverse waves (e.g., light waves) and longitudinal waves (e.g., sound waves). All electromagnetic waves (light, radio waves, x-rays) travel at the speed of light (c = 3 x 10⁸ m/sec in a vacuum), while for sound waves the speed is lower (~ 330 m/sec). For any wave, the speed of the wave is directly proportional to the wavelength and frequency:

The energy (E) of any wave is inversely proportional to its wavelength, which for electromagnetic waves is given by:

where h = 6.62 x 10^–34 J.s (Plank’s constant).

The wavelength of the wave is important for certain interactions, in particular for absorption and scattering. The probability of a scattering event depends on the relative size of the wavelength and the scattering site. Scattering probability decreases with increasing wavelength. Hence, longer wavelengths scatter less than shorter wavelengths and therefore can penetrate objects more deeply.

The use of longer wavelengths for diagnostic techniques may help with the penetration through the tissue, but the counterbalance to this is that the resolution of an image is directly proportional to the wavelength. This means that, as longer wavelengths are used, the ultimate resolution possible (the smallest feature that can be seen) falls. In an optical system, the collecting power of the lens in combination with the wavelength sets this resolution limit. As a rough guide, without resorting to sophisticated optical techniques, imaging with a resolution much less than the wavelength of the light is hard. For microscopic imaging systems (such as those used for histological studies), it is possible to achieve resolutions of around 0.3 µm using oil objective lenses and very short working distances that make such systems impractical for clinical dental diagnostic instrumentation. For sound-based systems, with their significantly longer wavelengths, the resolution limit is much lower (i.e., only larger features can be seen).

In relation to any diagnostic technique, in particular those that rely on imaging and diffraction-limited optics, the system’s ultimate resolution is not normally the factor limiting the size of feature that can be seen. There is no advantage in having the highest possible resolution optics if one has no contrast in the image. Contrast is defined as the difference between light and dark points, and if this difference cannot be seen, then the observer cannot resolve any features. Generally, the loss of contrast in optical imaging systems is caused by scattering events sending the light to the wrong point on the detector (which could be the eye or a camera).

Before we describe some novel techniques that are currently being evaluated for their diagnostic potential, it is worth considering the specific characteristics of some different types of waves. These are shown in the Table. It should be noted that electromagnetic irradiation is conventionally defined by wavelength, whereas ultrasound is conventionally defined by frequency.

In addition to those discussed at this conference, several methods using light are currently being proposed to detect caries. These include: multi-photon imaging, infrared fluorescence, optical coherence tomography, and terahertz imaging. Other wave energies—such as infrared radiation emitted by objects at room temperature, and sound waves—have also been proposed for the detection of caries.

Light is a particularly suitable tool for the study of teeth. The regular structure of teeth ensures good propagation of light through the crystalline enamel and the tubules of dentin. The size of the structures is comparable with the wavelength of visible and near-infrared light. Disruption to the ordered structure of a tooth increases the likelihood of scattering of light that passes into the tooth. The uptake of fluid into pores created by demineralization—in addition to the uptake of exogenous stain, bacterial breakdown products, and other contaminants present as a result of the caries process—will change the normal interaction of light with tooth structure. In addition to scattering, these changes will include absorption and fluorescence. Many of the techniques discussed use one or more of these interactions. Without realizing the actual physics of the processes, the skilled dental practitioner has been using these effects for years in a subjective manner.

	MULTI-PHOTON IMAGING

TOP ABSTRACT INTRODUCTION PHYSICAL PRINCIPLES BEHIND... MULTI-PHOTON IMAGING INFRARED THERMOGRAPHY INFRARED FLUORESCENCE OPTICAL COHERENCE TOMOGRAPHY ULTRASOUND TERAHERTZ IMAGING SUMMARY REFERENCES

 
Caries detection systems such as QLF rely on the fluorescence signal observed when teeth are exposed to blue light ( ~ 488–514 nm). This causes sound tooth structure to fluoresce. Carious tooth tissue may also fluoresce, but the disruption to the regular structure of the tooth at this point results in profound scattering, and no or little fluorescence is detected. Consequently, sound tooth structure fluoresces at > 520 nm, whereas carious tooth tissue appears dark. It is not possible to collect light specifically from different depths within the tooth. Other methods, such as confocal microscopy, can be used to collect light from different depths but only within the outer 100 microns of the tooth (Watson, 1997). However, information from deeper within the tooth is often required. Blue light tends to scatter substantially within caries lesions and therefore does not penetrate well through the lesion. At high intensity, blue light induces free-radical production and phototoxicity in live tissue, which could injure the pulp (Girkin et al., 1999).

The choice of a longer wavelength of light for imaging reduces the scattering, allowing the light to penetrate more deeply within the tooth. This may make any image of the tooth clearer and reduces the levels of phototoxicity. For multi-photon imaging of teeth, infrared light ( = 850 nm) has been used (Girkin et al, 1999).

In conventional fluorescence imaging (QLF), a single ‘blue’ photon is used to excite a fluorescent compound in the tooth. In the multi-photon technique, two infrared photons (with half the energy of the blue photon) are absorbed simultaneously. The probability of this happening is normally low, but by exposing the tooth to many more photons, it is possible to increase greatly the chances of two-photon absorption (the probability of two-photon absorption is proportional to the square of the light intensity). Generally, this means increasing the intensity of the light beam, which is also likely to generate heat within the tooth. To generate enough two-photon events, it has been calculated that a peak power of 2 kW would be required. Clearly, a tooth would not survive this substantial amount of power input for any length of time. It is possible to resolve this difficulty by using ultra-short pulses, measured in femto seconds (fs), (~ 100 fs = 100 x 10^–15 s) of laser light, to produce adequate peak laser power but low average power, to increase the chances of a two-photon event. Even with such high peak powers, the fluorescence is generated only in the focal plane, and hence one has a method of optically sectioning entire samples.

Ultra-short pulses (100 fs) of 850 nm laser light are generated at 200 MHz. The average beam power is in the milliwatt range. By scanning a focused beam, one can record, from the focal plane, the fluorescence resulting from two-photon excitation. If the focal plane is then changed, through the enamel toward the dentin, a series of optical sections can be created. With this technique, sound tooth tissue fluoresces strongly, whereas carious tooth tissue fluoresces to a much lesser extent. In practice, by using motors with micron accuracy, one can move the plane of focus through the tissue and record the sectional images from the tooth to form a three-dimensional image. Caries will appear as a dark form within a brightly fluorescing tooth. To highlight the diseased tissue, the image may be displayed in its negative form so that caries appears bright within a dark tooth (Fig. 2).

View larger version (119K): [in this window] [in a new window]   Figure 2. A composite image of a caries lesion made by a multi-photon technique superimposed on a scattered light image of the tooth.

	INFRARED THERMOGRAPHY

TOP ABSTRACT INTRODUCTION PHYSICAL PRINCIPLES BEHIND... MULTI-PHOTON IMAGING INFRARED THERMOGRAPHY INFRARED FLUORESCENCE OPTICAL COHERENCE TOMOGRAPHY ULTRASOUND TERAHERTZ IMAGING SUMMARY REFERENCES

 
Thermal radiation energy travels in the form of waves. It is possible to measure changes in thermal energy when fluid is lost from a lesion by evaporation. The thermal energy emitted by sound tooth structure is compared with that emitted by carious tooth structure. The technique has been described by Kaneko et al.(1999) and has been proposed as a method of determining lesion activity rather than a method of determining the presence or absence of a lesion.

The method described by Kaneko et al.(1999) uses indium/antimony thermal sensors, which can detect temperature changes in the order of 0.025°C. With a constant flow of air over the surface of the tooth, the change in temperature of the lesion is compared with that of the surrounding sound tooth structure. The source-to-sensor distance is 20 cm, and the time taken to capture the data for a lesion is up to 2 min. A study described by Matsuyama et al.(1998) found a reasonable correlation (0.67–0.79) between temperature changes and mineral loss and lesion depth, respectively.

The technique has not been used intra-orally. Problems will exist in relation to variations in the temperature of the mouth with respiration or fluid evaporation from other oral surfaces. The source-to-specimen distance is presently unsuitable for posterior teeth. Accessible smooth-surface lesions have been used in vitro, but there are no data on lesions which cannot be directly accessed. Additionally, the issue of lesion staining may also affect the heat transfer between the sound and carious tooth structure. To the authors’ knowledge, there is no evidence that the rate, or pattern, of fluid loss from a lesion is directly related to the subsequent reactivity of a lesion in vivo or in vitro.

	INFRARED FLUORESCENCE

TOP ABSTRACT INTRODUCTION PHYSICAL PRINCIPLES BEHIND... MULTI-PHOTON IMAGING INFRARED THERMOGRAPHY INFRARED FLUORESCENCE OPTICAL COHERENCE TOMOGRAPHY ULTRASOUND TERAHERTZ IMAGING SUMMARY REFERENCES

 
This technique has seldom been reported. In theory, the tooth is exposed to light (irradiation) with a wavelength of between 700 and 15,000 nm. Barrier filters are used to observe any resulting fluorescence. Studies by Alfano et al.(1984) mention exposure of teeth to wavelengths exceeding 700 nm, but the results were not presented. Unpublished reports commented upon by Longbottom (1999) suggest that the technique is able to discriminate between sound and carious enamel and dentin. Further work is required to determine if the fluorescence signal from exposure to infrared irradiation is greater than that from other wavelengths. Additionally, any heating effects from absorption of infrared irradiation may have potentially damaging effects on the dental pulp, given the increased penetration and decreased scattering of the longer wavelength. Specific coherent sources of such irradiation have been relatively difficult to acquire, and detection involves the use of infrared-sensitive detectors as charge-coupled devices (CCDs) or film.

	OPTICAL COHERENCE TOMOGRAPHY

TOP ABSTRACT INTRODUCTION PHYSICAL PRINCIPLES BEHIND... MULTI-PHOTON IMAGING INFRARED THERMOGRAPHY INFRARED FLUORESCENCE OPTICAL COHERENCE TOMOGRAPHY ULTRASOUND TERAHERTZ IMAGING SUMMARY REFERENCES

 
Optical Coherence Tomography (OCT) is a method of imaging that has been developed for transparent and semi-transparent structures. Teeth fall into the latter category. It was first developed in medicine for use in ophthalmology (Huang et al., 1991) before being developed for use with skin and tissues of the gastro-intestinal tract. In the past 6 years, interest in the uses of OCT for dental imaging has grown (Baumgartner et al., 2000).

OCT uses light, the wavelength of which dictates the scattering and therefore the depth of penetration of the imaging technique. Additionally, the wavelength of light also affects the resolution of the technique, which is in the order of 10 µm (Baumgartner et al., 2000). Most OCT techniques described for imaging dental tissue have used wavelengths of 840 to 1310 nm (Everett et al., 1999; Baumgartner et al., 2000). This has resulted in imaging depths of 0.6 to 2-0 mm, respectively. Colston et al.(2000) describe imaging depths of less than 4 mm (just how much less is not reported). The depth resolution of such systems varies between 10 µm and 17 µm.

OCT is based upon the interference of light. When a light beam is split into two and then recombined, interference produces a pattern, the intensity of which is determined by the level of light in each beam. OCT systems use Super Luminescent Diodes (SLDs) as a light source. This type of source produces light with a broad range of wavelengths, each of which will produce its own interference pattern. However, in certain circumstances, the merging of interference patterns will result in blurring of some signals and not others. The signals that are not blurred are the ones which can be detected and give the technique its ability to section the samples optically. The spectral bandwidth of the light (the difference between the shortest and longest wavelengths produced by the illumination source) determines the depth resolution of the technique (Colston et al., 2000).

The intensity of the interference is a function of the scattering caused by the changes in tissue structure of the tooth. Variation in scattering measured in relation to depth from a single point on the tooth surface is called an "A-scan". Taking several A-scans along a line produces information from a ‘slice’ of tooth tissue, which is the tomogram. The movement along the line of A-scans is known as the "B-scan", and, according to Colston et al.(2000), it takes from 30 to 60 sec to acquire a 1-cm-long B-scan (Fig. 3).

View larger version (36K): [in this window] [in a new window]   Figure 3. A schematic diagram of the method by which OCT produces an A-scan and the way in which serial A-scans can be arranged to produce a tomographic slice of a tooth, known as a B-scan.

The method has been further enhanced by measurement of the changes in polarization of the beams of light as they pass through the tooth (Baumgartner et al., 2000; Colston et al., 2000).

Of clinical relevance is the development of prototype handpieces for intra-oral OCT, although no in vivo data have been reported. Analysis of caries lesions has been performed, and changes in signal are related to the degree of scattering and possibly the degree of mineralization. Further work has used OCT to assess the restoration-tooth interface for semi-transparent restorations. This could have implications for the non-invasive diagnosis of secondary caries. As with all optical methods, it is likely that uptake of any stain will confound the technique.

	ULTRASOUND

TOP ABSTRACT INTRODUCTION PHYSICAL PRINCIPLES BEHIND... MULTI-PHOTON IMAGING INFRARED THERMOGRAPHY INFRARED FLUORESCENCE OPTICAL COHERENCE TOMOGRAPHY ULTRASOUND TERAHERTZ IMAGING SUMMARY REFERENCES

 
The use of ultrasound to detect dental caries has been proposed for the past 30 years. The technique has received renewed interest at various times, particularly in the past 5 years.

Sound waves are longitudinal or pressure waves which travel through gases, liquids, and solids. They may also travel along the boundaries between and among gases, liquids, and solids. The frequency (number of oscillations per second) of sound generally detectable by the human ear ranges from about 20 Hz to almost 16,000 Hz for adults (1 Hz = 1 oscillation per sec). Ultrasound waves have a frequency of > 20,000 Hz. Sound waves have all the usual properties of waves, in that they may be reflected, scattered, refracted, or absorbed.

The speed at which a sound wave travels depends upon the medium through which it travels. In air, this is 330 m/sec. In liquids and solids, it is much higher. Different media have different abilities to transmit sound. The relative ability of a medium to transmit sound depends on its mechanical properties, such as elasticity, density, and the wavelength of the sound. Sound waves cause gases and liquids to undergo elastic deformation, which may result in compression or dilatation, whereas in solids, the elastic deformation due to transmission of sound waves may also include a shearing component. The relationship between the speed of sound within a given medium (v) and the elasticity (E) and density () of the medium is represented by the equation:

Additionally, a medium through which sound travels may also be defined by its ability to conduct sound. This is called specific acoustic impedance (Z) and is represented by represented by the formula:

When sound travels from one medium to another, some of the wave energy is reflected at the interface between the two media, while some of the wave energy is transmitted into the second medium. The transmitted energy can subsequently be reflected when it encounters a third medium and so on.

The intensity of the sound reflected (I_r) when traveling from one medium to another can be defined by the formula:

where I_i is the intensity of the original (incident) sound wave. The intensity of sound transmitted to the second medium (I_t) is represented by the formula I_t = (I_i - I_r) and has units of W/m².

It follows that the greater the difference in acoustic impedance between two media, the greater the amount of reflection of sound waves from the interface. Therefore, the amount of sound reflected provides information about the structure of the reflecting interface, whereas the time taken for sound to be reflected provides information about the position of the reflecting interface.

Ultrasound waves are usually produced by an alternating voltage applied to a piezo-electric crystal. Sound waves produced as a result of minute changes in the crystal dimension may be emitted continually, as a burst of waves (which crescendo then die away) or as a single pulse.

To reach the target medium (for dentistry, this is the tooth), sound waves have to travel through a coupling medium. The ideal coupling agent to link the source of the ultrasound to the specimen is one that has acoustic impedance similar to that of the specimen. This minimizes any reflection at the interface between the two media and maximizes the amount of ultrasound entering the specimen.

Various acoustic coupling agents have been used for ultrasound evaluation of teeth, including mercury (Reich et al., 1967) and aluminum rods bonded to the tooth surface (Barber et al., 1969). Of obviously greater clinical relevance is the use of water (Ng and Ferguson, 1988) and glycerine (Huysmans and Thijssen, 1998; Yanikolu et al., 1999).

Barber et al.(1969) used ultrasound to detect the enamel/dentin junction and the dentin/pulp interface by bonding an aluminum rod with epoxy resin to non-carious extracted teeth. The use of water as a coupling agent (Ng and Ferguson, 1988) did not permit ultrasound imaging in extracted teeth as deep as the dentin/pulp interface but was sufficient to detect artificially produced enamel caries lesions in addition to the enamel/dentin junction. Additionally, a correlation between the mineral loss from the lesion body measured by transverse microradiography and ultrasound measurements was reported (Ng and Ferguson, 1988). However, measurements were made on only 6 teeth. Huysmans and Thijssen (1998) used ultrasound to determine the thickness of enamel overlying the enamel/dentin junction as a method of measuring mineral loss by erosion. Their results indicated that ultrasound could potentially measure enamel thickness, but further work demonstrated that changes of less than 0.4 mm could not be detected reliably (Louwerse and Huysmans, 2001). Yanikolu et al.(2000) examined 20 natural proximal lesions in extracted teeth and demonstrated a sensitivity of 0.88 and specificity of 0.86 when histology was used as the gold standard.

One group of researchers has taken a slightly different approach to the use of ultrasound to detect caries. They have used ultrasound waves which travel across the surface of the tooth along the interface between enamel and air, rather than through the tooth structure. In this way, ultrasound detects surface discontinuity present as a result of cavitated proximal lesions. Bab et al.(1997) compared in vitro the radiographic and clinical evaluations of 6 lesions with ultrasound measurements. Using a flexible probe tip which would fit into the wedge-shaped interproximal contours and conform to the shape of the tooth, they demonstrated much stronger ultrasound reflections from cavitated lesions compared with non-cavitated lesions. A later in vitro study by Ziv et al.(1998) compared ultrasonic measurements at 70 approximal sites with radiographic and histologic findings. Using histology as a gold standard, they obtained a sensitivity of 1.0 and a specificity of 0.92 for ultrasound measurements. Bab et al.(1998) described an in vivo study involving 253 restoration-free approximal sites. The authors suggested that their Ultrasonic Caries Detector (UCD) could discriminate dentin interproximal caries from an intact site judged with bitewing radiography as the standard. No further information is available from this study. It is not known how many such lesions were cavitated and if the surface area of the cavity was related to the strength of the ultrasound reflection.

Ultrasound may be a quick and reliable tool for the detection of dental caries in enamel. The use of longitudinal waves to measure demineralization in relation to the ADJ is very useful, as is the potential for surface sound waves to detect cavitation. Further work is required in vivo to relate ultrasound findings to other clinical measurements.

	TERAHERTZ IMAGING

TOP ABSTRACT INTRODUCTION PHYSICAL PRINCIPLES BEHIND... MULTI-PHOTON IMAGING INFRARED THERMOGRAPHY INFRARED FLUORESCENCE OPTICAL COHERENCE TOMOGRAPHY ULTRASOUND TERAHERTZ IMAGING SUMMARY REFERENCES

 
This method of imaging uses waves with terahertz frequency (= 10¹² Hz or a wavelength of approximately 30 µm). This wave-form is short enough to provide reasonable resolution but long enough to prevent serious loss of signal due to scattering. A good overview of the technique is provided by Arnone et al.(2000) and Ciesla et al.(2000).

For many years, no practical sources or detectors of terahertz radiation were known. In the early 1980s, it was discovered that photoconductive emitters or certain crystals (e.g., zinc-telluride) exposed to short pulses (< 10^–12 s) of visible or infrared light would emit electromagnetic waves with a frequency in the terahertz range. To detect terahertz irradiation, photoconductive detectors can be used in addition to a technique called "free-space electro-optical sampling" (EOS).

The advantages of terahertz imaging include: the relative transparency of human tissue to terahertz rays, low powers used for imaging (~ 1 µW), the use of non-ionizing radiation, and no alteration of electrical charge of the tissues examined. Any adverse thermal effects are thought to be unlikely, and the cost of the system is currently similar to that of magnetic resonance imaging. However, the price of the expensive, ultra-fast laser technology required is expected to fall. Additionally, both average and spectroscopic absorption and transmission data can be recorded, as well as refractive index. The use of coherent detection methods permits the simultaneous collection of such data. The low signal-to-noise ratio for terahertz imaging facilitates extremely clear imaging, but with low spatial resolution (250 µm) due to the long wavelength of the source. Finally, care is required in image interpretation, since terahertz waves are strongly absorbed by water, a potential complication in the mouth.

For an image to be obtained by terahertz irradiation, the object is placed in the path of the terahertz beam. Alternatively, the terahertz beam can be scanned over the surface of an object. It is also possible to record terahertz images using a CCD detector. Some of the first images were reported by Hu and Nuss (1995). They demonstrated images of the inside of a silicon chip and the change in water content of a leaf over time. The application to diseased human tissue followed.

Dental applications for this technique have been limited but promising. A longitudinally hemisected sound human premolar tooth has been imaged from the intact surface. Images have demonstrated the outline of the enamel-dentin junction as well as the dentin-pulp interface. Longitudinal sections through 3 teeth have demonstrated increased terahertz absorption by early occlusal caries and, intriguingly, an apparent ability to discriminate dental caries from idiopathic enamel hypomineralization. Work is in progress to image intact teeth with early caries lesions.

No reports indicate the time required to acquire such images. Additionally, the cost of the equipment, the complexity of the laser source, and the requirements for precise specimen manipulation, with a computer-controlled X-Y stage, mean that it will probably take a long time for the technique to be used clinically.

	SUMMARY

TOP ABSTRACT INTRODUCTION PHYSICAL PRINCIPLES BEHIND... MULTI-PHOTON IMAGING INFRARED THERMOGRAPHY INFRARED FLUORESCENCE OPTICAL COHERENCE TOMOGRAPHY ULTRASOUND TERAHERTZ IMAGING SUMMARY REFERENCES

 
We have described a selection of techniques, each of which has the potential to measure changes in teeth occurring as a result of mineral loss due to dental caries. Each of the methods measures the result of the caries process. Some methods may have the potential to determine lesion activity indirectly. Many of the techniques have been applied to extracted teeth and require extensive testing on vital teeth to ensure that there are no adverse reactions. However, the large array of potential methods gives cause for optimism that rapid, non-invasive quantifiable methods of caries detection, and possibly diagnosis, will afford future dentists the ability to provide optimal preventive advice with minimal early intervention.

	FOOTNOTES

 
Presented at the International Consensus Workshop on Caries Clinical Trials, Glasgow, Scotland, January 7–10, 2002

	REFERENCES

TOP ABSTRACT INTRODUCTION PHYSICAL PRINCIPLES BEHIND... MULTI-PHOTON IMAGING INFRARED THERMOGRAPHY INFRARED FLUORESCENCE OPTICAL COHERENCE TOMOGRAPHY ULTRASOUND TERAHERTZ IMAGING SUMMARY REFERENCES

 
Al-Khateeb S, Forsberg C-M, de Josselin de Jong E, Angmar-Månsson B (1998). A longitudinal laser fluorescence study of white-spot lesions in orthodontic patients. Am J Orthod Dentofacial Orthop 113:595–602.[ISI][Medline]

Alfano RR, Lam W, Zarrabi HJ, Alfano MA, Cordero J, Tata DB, et al. (1984). Human teeth with and without caries studied by laser scattering, fluorescence and absorption spectroscopy. IEEE J Quantum Electr 20:1512–1516.

Arnone D, Ciesla C, Pepper M (2000). Terahertz imaging comes into view. Physics World 13:35–40.

Bab IA, Fuerstein O, Gazit D (1997). Ultrasonic detector of proximal caries. Caries Res 31:322.

Bab I, Ziv V, Gazit D, Fuerstein O, Findler M, Barak S, et al. (1998). Diagnosis of approximal caries in adult patients using ultrasonic surface waves (abstract). J Dent Res 77(Spec Iss A):255.

Barber FE, Lees S, Lobene RR (1969). Ultrasonic pulse-echo measurements in teeth. Arch Oral Biol 14:745–760.[ISI][Medline]

Baumgartner A, Dichtl S, Hitzenberger CK, Sattmann H, Robl B, Moritz A, et al. (2000). Polarization-sensitive optical coherence tomography of dental structures. Caries Res 34:59–69.[ISI][Medline]

Ciesla CM, Arnone DD, Corchia A, Crawley D, Longbottom C, Linfield EH, et al. (2000). Biomedical applications of terahertz imaging. Proc SPIE 3934:73–81.

Colston BW, Everett MJ, Sathyam US, DaSilva LB, Otis LL (2000). Imaging of the oral cavity using optical coherence tomography. In: Monogr Oral Sci. Assessment of oral health—diagnostic techniques and validation criteria. Faller RV, Whitford GM, editors. Basel: Karger, pp. 32–55.

Ekstrand KR, Ricketts DN, Kidd EA, Qvist V, Schou S (1998). Detection, diagnosing, monitoring and logical treatment of occlusal caries in relation to lesion activity and severity: an in vivo examination with histological validation. Caries Res 32:247–254.[ISI][Medline]

Everett MJ, Sathyam US, Colston BW, Da Silva LB, Fried D, Ragadio JN, et al. (1999). Polarization sensitive optical coherence tomography. In: Early detection of dental caries II. Proceedings of the 4th Annual Indiana Conference. Stookey GK, editor. Indianapolis: Indiana University School of Dentistry, pp. 143–154.

Featherstone JDB (1996). Clinical implications: new strategies for caries prevention. In: Early detection of dental caries. Proceedings of the 1st Annual Indiana Conference. Stookey GK, editor. Indianapolis: Indiana University School of Dentistry, pp. 287–295.

Girkin JM, Hall AF, Creanor SL (1999). Multi-photon imaging of intact dental tissue. In: Early detection of dental caries II. Proceedings of the 4th Annual Indiana Conference. Stookey GK, editor. Indianapolis: Indiana University School of Dentistry, pp. 155–168.

Hu BB, Nuss MC (1995). Imaging with terahertz waves. Optics Lett 20:1716–1718.

Huang D, Swanson EA, Lin Cp, Schuman JS, Stinson WG, Chang W, et al. (1991). Optical coherence tomography. Science 254:1178–1181.[Abstract/Free Full Text]

Huysmans MCDNJM, Thijssen JM (1998). Ultrasonic measurement of enamel thickness: a tool for monitoring erosion? (abstract) Caries Res 32:292.

Kaneko K, Matsuyama K, Nakashima S (1999). Quantification of early carious enamel lesions by using an infrared camera in vitro. In: Early detection of dental caries II. Proceedings of the 4th Annual Indiana Conference. Stookey GK, editor. Indianapolis: Indiana University School of Dentistry, pp. 83–100.

Longbottom C (1999). Quantification of red/infrared laser fluorescence for early caries detection. In: Early detection of dental caries II. Proceedings of the 4th Annual Indiana Conference. Stookey GK, editor. Indianapolis: Indiana University School of Dentistry, pp. 75–82.

Louwerse C, Huysmans MCDNJM (2001). Reproducibility of ultrasonic enamel thickness measurements: an in vitro study (abstract). Caries Res 35:295.

Matsuyama K, Nakashima S, Kaneko K (1998). An in vitro study on the detection of early enamel carious lesions by use of an infra-red camera (abstract). Caries Res 32:294.

Ng SY, Ferguson MWJ (1988). Ultrasonic studies of unblemished and artificially demineralised enamel in extracted human teeth: a new method for detecting early caries. J Dent 16:201–209.[ISI][Medline]

Pitts NB (1994). Discovering dental public health: from Fisher to the future. Community Dental Health 11:172–178.[Medline]

Reich FR, Brenden BB, Porter NS (1967), Ultrasonic imaging of teeth. Report of Battelle Memorial Institute, Pacific Northwest Laboratory, Richland, WA, USA (privately issued). Comment in: Arch Oral Biol 14:745–760, 1969.

van der Veen MH, de Josselin de Jong E, Al-Khateeb S (1999). Caries activity detection by dehydration with quantitative light fluorescence. In: Early detection of dental caries II. Proceedings of the 4th Annual Indiana Conference. Stookey GK, editor. Indianapolis: Indiana University School of Dentistry, pp. 251–260.

Watson TF (1997). Fact and artefact in confocal microscopy. Adv Dent Res 11:433–441.[Abstract]

Yanikolu FC, Analoui M (1999). Ultrasonic methods for early caries detection. In: Early detection of dental caries II. Proceedings of the 4th Annual Indiana Conference. Stookey GK, editor. Indianapolis: Indiana University School of Dentistry, pp. 101–122.

Yanikolu FC, Öztürk F, Hayran O, Analoui M, Stookey GK (2000). Detection of natural white spot caries lesions by an ultrasonic system. Caries Res 34:225–232.[ISI][Medline]

Ziv V, Gazit D, Beris D, Fuerstein O, Aharonov L, Bab I (1998). Correlative ultrasonic histologic and Roentgenographic assessment of approximal caries (abstract). Caries Res 322:294.

This article has been cited by other articles:

				K. S. Kornman Diagnostic and Prognostic Tests for Oral Diseases: Practical Applications J Dent Educ., May 1, 2005; 69(5): 498 - 508. [Abstract] [Full Text] [PDF]

				N.B. Pitts and J.W. Stamm Preface J. Dent. Res., July 1, 2004; 83(suppl_1): C4 - C5. [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hall, A. Articles by Girkin, J.M. Search for Related Content PubMed PubMed Citation Articles by Hall, A. Articles by Girkin, J.M.