Optical Fibers for Biomedical Applications
This is a continuation from the previous tutorial - Optical fibers for industrial laser application
1. INTRODUCTION
Optical fibers have revolutionized medicine in many ways. They have advanced both diagnostics and treatments capabilities. Their major advantages lie in the fact that they are thin and flexible so they can be introduced into the body to remotely sense, image, and treat.
This capability enabled the introduction of minimally invasive procedures, which became the preferred choice for surgery. Such procedures minimize the postoperative pain and discomfort, as well as shorten—and sometimes eliminate— hospitalization time, thus saving on costs and reducing the number of missed workdays.
They also reduce the risk of contamination from the hospital environment. And maybe, most importantly, they allow the patient return to the comfort of his or her home sooner.
People are taking it naturally to have endoscopes in the clinical practice. This instrument—and all its variations—is based on the existence of a coherent fiber bundle, which allows imaging of internal organs while working their way through the natural orifices of the body or in some cases through minimal incisions.
Fiber bundles were patented in the 1920s by Baird in England and Hansell in the United States in parallel, but most important progress and realization were made by Van Heel and Capany in the 1950s (also two parallel unconnected works). Now imaging bundles can have 100,000 fibers that are small enough and flexible to be used in various types of endoscopes.
In this chapter, we describe new progress in this exciting field of fiber optics into the infrared \(\text{(IR)}\) regimen and other new fiber-based sensing methods in the visible, ultraviolet \(\text{(UV)}\), and \(\text{IR}\).
We begin (Section MEDICAL LASER ARMS) with the development of fibers and waveguides that are capable to transmit high energies for tissue ablation, thus replacing cumbersome medical laser arms based on tubes, joints, and reflecting mirrors.
We continue (Section TRANSENDOSCOPIC SURGICAL APPLICATION) with an example of state-of-the-art operation of calculi fragmentation in the salivary gland ducts. The next two sections deal with two sensing methods that make use of fibers: absorption spectroscopy, which uses the hollow core to make spectral measurements of different biological gases, and evanescent wave sensing, which uses the penetration of waves outside the borders of the delivery medium.
Section FIBER OPTIC THERMAL SENSING deals with remote sensing of temperature, which can be applied through endoscopes as well.
The last section brings us into the imaging field again; however, it expends our capabilities into the mid-\(\text{IR}\) \(\text{(MIDIR)}\) range, thus allowing collection of additional important data that lie in this spectral range for temperature mapping for diagnostic and feedback mechanisms for treatments.
2. MEDICAL LASER ARMS
The most straightforward use of lasers in medicine is laser surgery such as benign lesion removal and tissue ablation. Although a variety of lasers are on the market, only a few might be suited for medical applications, and there is a need to optimize the laser for the required applications. The most common lasers used for medical applications are \(\text{CO}_2(\lambda=10.6\mu m)\), \(\text{Er}/\text{YAG}(\lambda=2.93\mu m)\), \(\text{Ho}/\text{YAG}(\lambda=2.07\mu m)\), \(\text{Nd}/\text{YAG}(\lambda=1.06\mu m)\), and some others in the visible range.
A \(\text{CO}_2\) laser is often referred to as the ‘‘surgical laser’’ because its action most resembles traditional surgery. Unlike that of any other medical laser, its action on tissue is directly visible as it is used.
The \(\text{CO}_2\) laser was the first laser widely used by surgeons and is still the most used of all the medical lasers. Strongly absorbed by water, which constitutes more than 80% of soft tissue, this laser emits continuous wave \(\text{(CW)}\) or pulsed far-\(\text{IR}\) light at 10.6 \(\mu m\), which can be focused into a thin beam and used to cut like a scalpel or defocused to vaporize, ablate, or shave soft tissue.
The \(\text{CO}_2\) laser may be operated in pulsed mode or used with scanning devices to precisely control the depth and area of ablation.
Its uses include
- Removal of benign skin lesion, such as moles, warts, keratoses
- As a ‘‘laser scalpel’’ in patients or body areas prone to bleeding
- ‘‘No-touch’’ removal of tumors, especially of the brain and spinal cord
- Laser surgery for snoring (LAUP)
- Shaving, dermabrading, and resurfacing scars, rhinophyma, skin irregularities
- Cosmetic laser resurfacing for wrinkles and acne
\(\text{Er}/\text{YAG}\) laser emits a MIDIR beam at 2.93, which coincides with the absorption peak for water. Its principal use is to ablate tissue for cosmetic laser resurfacing for wrinkles and skin irregularities.
The laser has been advertised to offer advantages of reduced redness, decreased side effects, and rapid healing compared to the pulsed or scanned \(\text{CO}_2\) laser but does so by its limited penetration into tissue, which limits the results compared to the more versatile \(\text{CO}_2\) laser. It has also been used as a dental drill substitute to prepare cavities for filling.
\(\text{Ho}/\text{YAG}\) Laser is relatively new to the medical/dental fields. It emits a \(\text{MIDIR}\) beam at \(2.07\mu m\). Its principal use is to precisely ablate bone and cartilage, with many applications in orthopedics for arthroscopy, urology for lithotripsy (removal of kidney stones), \(\text{ENT}\) for endoscopic sinus surgery, and spine surgery for endoscopic disc removal.
The \(\text{Ho}/\text{YAG}\) laser was approved for \(\text{TURP}\) (prostate removal).
To facilitate laser surgery and other applications, one has to direct the laser light from the laser to the application point. For the visible and near-\(\text{IR}\) region up to \(2\mu m\), this can be done by using standard silica fibers.
However, once we choose to use lasers in the \(\text{MIDIR}\) \((>2\text{mm)}\), other means are needed. There are two means to deliver \(\text{MIDIR}\) laser power to the application point: articulated arms and \(\text{MIDIR}\) optical fibers.
Each of the solutions could be coupled to a hand-piece that shapes the laser beam to a desired energy distribution.
At the beginning of laser use in medicine, energy was transmitted by articulated arms, which was a set of tubes connected by joints and reflecting mirrors to three freedom ranks (Fig. 1). At first, this setup suffered from power drifts and required a lot of maintenance. In time, it was improved and the laser power drifts were reduced, but still the device was cumbersome (Fig. 2) and limited to external use.
Although many \(\text{MIDIR}\) lasers still use articulated arms, a better choice is to use an optical fiber. The main advantages of optical fibers are their size and flexibility. These features enable the operator more maneuverability and flexibility (Fig. 3) but most important may allow operating inside the body in minimally invasive procedures.
Whether the laser beam is guided from the laser using an articulated arm or an optical fiber, in most of the cases there is a hand-piece at their distal end (Fig. 4). A hand-piece is a device that serves a few purposes.
First, it enables the operator to easily hold the distal end of the optical fiber or articulated arm and may force the operator to work at a predetermined working distance. Moreover, it serves as a beam shaper, which reshapes the laser beam to a
predefined spot size. Last, it may allow the use of cooling gas flow to the treated tissue in order to cool it quickly and avoid thermal damage.
3. TRANSENDOSCOPIC SURGICAL APPLICATION
Quite a few fiber-based applications have been developed and used for the past years. We have chosen to present one new application in which we took part. This demonstrates the advantages of this method in minimally invasive surgery.
Sialolithiasis (salivary gland stones) is the most frequently occurring of the diseases affecting the salivary glands. In postmortem studies, incidence was found to be 1.2%, while in hospital admissions, it is estimated to be 1 per
15,000. Sialolithiasis results in a mechanical obstruction of the salivary duct causing repetitive swelling during meals, which can remain transitory or be complicated by bacterial infections. The introduction of sialoendoscopy in the mid 1990s has significantly improved the diagnosis and treatment of salivary gland inflammatory diseases, yet the endoscopic removal of larger and impacted salivary gland stones remains challenging, often requiring sialoadenectomy, which is associated with the risk of facial paralysis and requires general anesthesia and hospitalization.
Although the patient can function without one of the glands, it is involved with decreased secretion of saliva, which creates dry mouth and potentially more dental caries. Extracorporeal shockwave lithotripsy is also used for treatment. This technique requires several sessions at intervals of a few weeks.
The remaining stone debris may function as the ideal nidus for further calcification and sialolithiasis recurrence. This technique creates potential risk to bones and teeth.
The success rate is also rather low and does not pass 50% in the best cases. It is much less effective for stones of 7-\(\text{mm}\) diameter. Interventional sialoendoscopy with a wire basket through a miniaturized endoscope was also carried out in the 1990s. It has good success for the 1- to 3-\(\text{mm}\) diameter stones, but most stones are 3–7\(\text{mm}\).
Endoscopic laser lithotripsy can potentially treat most cases of salivary gland stones with minimal complications while preserving a functional salivary gland.
In urology, the holmium laser has become the gold standard for endoscopic lithotripsy with delivery of the energy through highly reliable, low-\(\text{OH}\) silica fibers with core diameters as small as 200 microns.
Some successful attempts have been reported on the use of this laser for sialolithiasis, but the smaller anatomy of the salivary ducts increases the risk to surrounding soft tissues.
Metal hollow waveguides were used because they are extremely durable, inexpensive, and biocompatible and can withstand high energies and powers. Hollow metal waveguides can be produced with low attenuation \((<0.5\) \(\text{dB}/\text{m})\) and can be bent to a radius of curvature as low as 5 \(\text{cm}\).
Hollow waveguides must be hermetically end-sealed for use in a water environment because even a small amount of water in the waveguide will completely attenuate the beam. Silver hollow waveguides of 0.9 \(\text{mm}\) outer diameter and 0.13 \(\text{mm}\) wall thickness were prepared with an \(\text{AgI}\) internal dielectric coating.
A total of 99.9% pure silver tubes with a 20 \(\text{RMS}\) inner surface finish were used. The tube’s inner surface underwent chemical reaction with a solution of iodine dissolved in ethanol, resulting in a thin dielectric layer of \(\text{AgI}\).
Coating deposition time and iodine concentrations were adjusted to optimize transmission at around 3 microns. Polished sapphire rods of 0.63-\(\text{mm}\) diameter and 5-\(\text{mm}\) length were cemented at the tip of the waveguide using a biocompatible adhesive.
A plano-convex \(\text{ZnSe}\) lens was designed to focus the 1-\(\text{mm}\) output aperture of the OpusDent fiber into the 0.63-\(\text{mm}\) waveguide, which was mounted in a \(\text{Luer}/\text{SMA}\) connector to facilitate quick connection to both the erbium laser and the sialoendoscope (Fig.5).
Clinical Tests
Helsinki committee approval was obtained to conduct human clinical trials at the outpatient facility of the Oral & Maxillofacial Department, Barzilai Hospital, Ashkelon, Israel, and performed by Professor \(\text{O}\). Nahlieli.
The trial was conducted on 17 patients (9 females, 8 males) aged 11–72 years, with salivary stones of 1–15 \(\text{mm}\) in diameter located in the posterior part of the salivary ducts.
Altogether, 21 stones located in 18 glands, 16 submandibular and 2 parotid, were involved in the trial. Stone size and location were documented by plain radiographs, sialography, and high-resolution ultrasound before the procedure, immediately after the procedure (radiography only), and at 12 weeks postoperative.
Videoendoscopy of the procedure was recorded and included a postoperative view of the lumen in the area (Fig. 6).
before the procedure, immediately after the procedure (radiography only), and at 12 weeks postoperative. Videoendoscopy of the procedure was recorded and included a postoperative view of the lumen in the area (Fig. 6).
Patients were followed up for lumen patency, symptoms of sialoadenitis, and any potential complications, immediately, at 4 weeks, and at 12 weeks after the procedure.
All procedures were performed under local anesthesia in an ambulatory environment, using standard video-sialoendoscopy protocol. Under this protocol, the orifice of Wharton or Stensen duct is first identified and a lacrimal probe gently inserted. Papillotomy is preferably performed with a \(\text{CO}_2\) laser.
Lacrimal dilators from 1 to 3 \(\text{mm}\) in diameter are then sequentially used for duct dilation. The rigid scope, connected to a videocamera and monitor, is introduced with the help of isotonic saline irrigation. The salivary ducts are diagnosed and the stone is located.
If no contraindications are detected, the fiber is inserted through the operating channel of the endoscope until its distal end is within the field of view of the endoscope.
Actual laser irradiation is performed only under clear vision, when the fiber is seen to be in direct contact with the stone and tangential to the duct. Adequate irrigation is maintained throughout the procedure using an intravenous bag connected to the irrigation port on the sheath of the endoscope.
The fiber was connected to a standard Lumenis Opus 20, dental erbium laser. Fragmentation usually began at a setting of 150 \(\text{mJ}\)/pulse and a pulse rate of 10 \(\text{Hz}\).
While the initial plan was to attempt full fragmentation on all stones, it was soon realized that because of anatomical and operational constraints, such as tortuous ducts or poor visibility, alternative techniques to assist in stone removal can be applied.
Consequently, three alternative methods were used depending on each case: total fragmentation, mostly for stones under 5 \(\text{mm}\); creation of a traction point for miniforceps; and separation of surrounding soft tissues in cases of impacted stones.
Of the 21 stones treated, 5 were fully fragmented, 7 stones of 5–7 \(\text{mm}\) were prepared for extraction by miniforceps, and 9 stones with diameters up to 15 mm were released from surrounding soft tissues for subsequent endoscopic or surgical removal.
A total of 15 of the 18 treated glands returned to normal function without any symptoms, whereas 2 nonfunctional glands remained nonfunctional but were completely asymptomatic. No other complications have been noted during the 2–12 months of follow-up so far.
4. ABSORPTION SPECTROSCOPY
Introduction
Spectroscopy studies the way electromagnetic radiation interacts with matter. When radiation meets matter, the radiation is either scattered, emitted, or absorbed. This gives rise to three principal branches of spectroscopy. Emission spectroscopy observes light emitted by atoms excited by radiation–matter interactions. Raman spectroscopy monitors light scattered from molecules. Absorption spectroscopy studies radiation absorbed at various wavelengths.
In absorption spectroscopy, light illuminates a sample of material to be analyzed. The sample, which can be liquid, solid, or gas, is usually enclosed in an absorption cell. Each element or compound in the sample absorbs particular wavelengths of light, resulting in one or more dark lines on its spectrum.
These lines are a ‘‘fingerprint’’ identifying which chemical substances are present in the sample and their quantities, as well as other information about detailed structure and activity.
When an atom or molecule absorbs energy, electrons are promoted from their ground state to an excited state. In a molecule, the atoms can rotate and vibrate with respect to each other. These vibrations and rotations also have discrete energy levels that can be considered as being packed on top of each electronic level.
Absorption spectroscopy is not limited to a specific wavelength region and can be observed at a wide range of wavelengths. However, the absorption in each wavelength range can be accounted to light interaction with different types of energy levels in the examined sample.
The absorption in the \(\text{UV}\) and visible spectrum is due to electronic absorption, whereas that in the IR region is due to vibrational and rotational energy levels of the molecules (Fig. 7).
The absorption of \(\text{UV}\) or visible radiation corresponds to the excitation of outer electrons. The absorption in organic molecules is restricted to certain functional groups (chromophores) that contain valence electrons of low excitation energy. The spectrum of a molecule containing these chromophores is complex.
This is because the superposition of rotational and vibrational transitions on the electronic transitions gives a combination of overlapping lines. This appears as a continuous absorption band.
Many inorganic species (such as semiconductors) show charge–transfer absorption and are called charge–transfer complexes. For a complex to demonstrate charge–transfer behavior, one of its components must have electron-donating properties and another component must be able to accept electrons.
Absorption of radiation then involves the transfer of an electron from the donor to an orbital associated with the acceptor.
\(\text{IR}\) radiation does not have enough energy to induce electronic transitions as seen with \(\text{UV}\). Absorption of \(\text{IR}\) is restricted to compounds with small energy differences in the possible vibrational and rotational states. For a molecule to absorb IR radiation, the vibrations or rotations within a molecule must cause a net change in the dipole moment of the molecule.
The alternating electrical field of the radiation interacts with fluctuations in the dipole moment of the molecule. If the frequency of the radiation matches the vibrational frequency of the molecule, then radiation will be absorbed, causing a change in the amplitude of molecular vibration.
Medical Applications of Absorption Spectroscopy
Absorption spectroscopy is difficult to perform in vivo because the human tissue scatters light considerably, so a quantitative analysis is difficult to obtain. However, it can be performed in situ. In situ studies had been conducted for several applications: glucose and other blood analytes analysis, \(\text{NO}\) and \(\text{CO}_2\) measurements in breath and molecular tissue mapping.
Quick and accurate measurements of blood analytes without sample preparation may enable a better treatment. Kim et al. showed that it is possible to measure the blood glucose levels in whole blood without sample preparation. They used absorption spectroscopy in the \(\text{MIDIR}\) region 6.5–11 \(\mu m\).
In this wavelength region, there are distinct glucose absorption lines, along with several hemoglobin absorption lines.
The hemoglobin absorption lines are used for calibration. The measured prediction error ranged between 10 and 50 \(\text{mg}/\text{dl}\).
Gas analysis of human breath may provide valuable biomedical and clinical information. The detection of gas traces such as ammonia, \(\text{NO}\) (Table 1), may be used for noninvasive medical diagnosis and monitoring the success of a medical treatment.
As for today, the most common method for exhaled gas analysis is mass spectroscopy with different kinds of ionization. This method provides good results, but it cannot be realized as a turnkey instrument.
Table 1. Exhaled gas traces and the illness they indicate
Spectroscopic analysis of exhaled gas has been investigated for several years. Most of the groups have been investigating exhaled gas in the near-\(\text{IR}\) \(\text{(NIR)}\) and \(\text{MIDIR}\) region. However, there is also research on spectroscopic response in the \(\text{UV}\). Eckhardta et al. investigated gas spectroscopy in the \(\text{UV}\) using hollow waveguides.
The main advantage of conducting spectroscopic analysis in the \(\text{UV}\) is that the absorption of oxygen in that region is small, thus enabling measurement in air. The main advantage of using a fiber instead of a flow cell is the ability to achieve a longer path length and, therefore, better sensitivity. They showed theoretically that it is possible to measure several gas traces simultaneously and measured experimentally \(\text{NO}\).
The use of \(\text{MIDIR}\) spectroscopy for exhaled gases has been investigated thoroughly by a number of research teams around the world. Roller et al. used a tunable laser in the 5.2-\(\mu m\) region and an absorption cell to measure the amount of \(\text{NO}\) in human exhaled breath.
They showed that measuring \(\text{NO}\) and \(\text{CO}_2\) simultaneously eliminates the need for system calibration. Using their system design, \(\text{NO}\) levels of 44 parts per \(10^9\) to 20 parts per \(10^9\) were measured, which is the near-normal level.
Ganser et al. developed an on-line sensing measurement of nitric oxide \(\text{(NO)}\) traces based on \(\text{CW}\) quantum cascade laser, operating near 5.2 \(\text{m}\).
They used a Faraday modulation technique that provides unique selectivity because it exploits the magnetic moment of \(\text{NO}\) molecules. The minimum detectable \(\text{NO}\) concentration was found to be 4 \(\text{ppb}\) (sampling time: 10 seconds).
Absorption spectroscopy can be used to detect abnormalities in tissues and organs. Tumors or other irregularities in tissue and organs might be detected directly by measuring their spectral response.
Another method is monitoring the organ metabolism, blood flow in the tissue, and blood oxygenation.
Hynes et al. investigated the use of microspectroscopy for dental applications. Chronic periodontitis is an inflammatory disease of the supporting structures of the teeth.
\(\text{IR}\) microspectroscopy has the potential to simultaneously monitor multiple disease markers, including cellular infiltration and collagen catabolism and, hence, differentiate diseased and healthy tissues.
They showed that connective tissue contains lower densities of \(\text{DNA}\), protein, and lipids compared to higher densities in epithelial tissue.
Collagen-specific tissue mapping by \(\text{IR}\) microspectroscopy revealed much higher levels of collagen deposition in the connective tissues compared to that in the epithelium.
Thus, inflammatory events such as cellular infiltration and collagen deposition and catabolism can be identified by \(\text{IR}\) microspectroscopy.
Neurological complications during critical illness remain a common cause of morbidity and mortality. The current methods used to monitor cerebral function including electroencephalography, jugular bulb mixed venous oxygen saturation, and transcranial Doppler either require an invasive procedure or are not sensitive enough to effectively identify patients at risk for cerebral hypoxia.
Tobias et al. investigated \(\text{NIR}\) spectroscopy as a method to measure oxygen in tissue. This method is noninvasive, which works similar to pulse oximetry, allows for the penetration of living tissue, and provides an estimate of brain tissue oxygenation by measuring the absorption of \(\text{IR}\) light by tissue chromophores.
5. EVANESCENT WAVE SPECTROSCOPY
Introduction
\(\text{NIR}\) and \(\text{MIDIR}\) spectroscopy is a reliable method for obtaining the fingerprint of solids, liquids, and gases. It can detect small amounts of materials and quantify the amount of material inside a given sample.
One of the methods of \(\text{NIR}\) and \(\text{MIDIR}\) spectroscopy is evanescent wave spectroscopy \(\text{(EWS)}\), which is also known as attenuated total reflectance spectroscopy \(\text{(ATRS)}\).
Evanescent waves are the attenuated waves that ‘‘leak’’ from an optical waveguide, guiding crystal or fiber core to its cladding or surroundings in the case of a waveguide, crystal, or optical fiber without a cladding (Fig. 8).
Because the evanescent waves are highly attenuated, their interaction with the waveguide’s surroundings is limited to a small distance from the guiding material. The amount of attenuation depends on the wavelength that is transmitted through the guiding material and the type of material in the surroundings.
The penetration depth \(d_p\) is given by Eq. (1), where \(\lambda\) is the wavelength, \(\theta\) is the angle of incidence, \(n_1\) and \(n_2\) are the index of refraction of the surrounding and the core, respectively. \(d_p\) is very small, usually in the range of 0.5 \(\mu m\):
\[\tag{1}d_p=\frac{\lambda}{2\pi\sqrt{n_2^2\sin^2\theta-n^2_1}}\]
Experimental Setups
There are several optical setups for performing \(\text{EWS}\). Each setup contains a tunable light source (tunable laser, monochromator), a detector, and depending on the examined object, some kind of a sample chamber. One of the most used systems for \(\text{EWS}\) is an \(\text{FTIR}\).
Two typical EWS systems are shown in Fig. 9. The first system (Fig. 9a) is known as fiber optic EWS (FEWS). The experimental setup uses an \(\text{FTIR}\), along with a partially uncladded optical fiber made of \(\text{AgBrCl}\) or chalcogenide glass (if the measurement is done in the \(\text{MIDIR}\) region) or silica (if the measurement is done in the \(\text{UV}\), \(\text{VIS}\), or \(\text{NIR}\) region).
These fibers are inserted into a sample chamber or are placed in contact with the sample. The fiber’s surface may be sensitized to bond a certain material to increase the sensor sensitivity.
The second method (Fig. 9b) is to couple the light into a crystal that serves as a waveguide. The crystal is then placed in contact with the sample. Here, as in the previous setup, it is possible to use different light sources. The light source choice depends on the examined material.
To obtain good sensitivity, it is recommended that the contact length between the fiber–crystal be as long as possible. Each reflection from the fiber– crystal wall increases the interaction between the light and the sample. As the number of reflections increases, the light attenuation decreases and a clearer spectrum is being recorded. The number of reflections can be obtained using geometrical optical and is given by Eq. (2), where \(\text{L}\) is the interaction length, \(\text{d}\) is the fiber diameter, and \(\theta\) is the angle of incidence:
\[\tag{2}N(\theta,d,L)=L\frac{\tan(90-\theta)}{d}\]
Equation (2) shows that it is beneficial to use long fibers/crystals. However, one has to take into account the size limitation of the sample itself. Although long fibers are usually applicable to liquid or gas samples, shorter fibers/crystals would be needed for solid samples and in \(\text{vivo}\) measurements.
Lately a new type of \(\text{EWS}\) sensors was suggested. These sensors are based on two dimensional photonic crystals. It was shown that solutions and biological fluid that enters the air holes and gaps of the photonic crystal causes a shift in their photonic band-gap.
The amount of shift in the photonic band-gap is proportional to the amount of biomolecule in the solution. \(\text{EWS}\) might be applicable to many spectroscopic applications. Remote analysis is required for many industrial applications in chemistry and biochemistry.
A few examples are in situ monitoring of chemical reactions, on-site pollution control, in \(\text{vivo}\) monitoring of biological processes, and biomedical sensing such as noninvasive glucose measurements and cancer diagnosis.
Chemical Sensing
Chemical Analysis and Monitoring
All organic species exhibit vibrational modes within the 2- to 16-\(\mu m\) wavelength region. This makes \(\text{FEWS}\) an excellent tool for remote detection of a wide variety of chemicals and biochemicals ranging from proteins and enzymes to simple molecules such as ethanol and benzene.
Chemical detection of butanone by means of \(\text{FEWS}\) was reported during the late 1980s. Chemical analyses were then performed on acetone, ethanol, and sulfuric acid using chalcogenide fibers. It was shown that the detection versus concentration curve follows a linear Beer–Lambert law, allowing quantitative analysis of solute samples.
A wider range of organic species, including carcinogens such as benzene, toluene, and trichloroethylene, were later detected. Sensitivity of several parts per million levels was achieved.
Spectral monitoring performed continuously and in real time allows one to observe and follow changes in molecular structures throughout a chemical reaction. For example, conversion of fructose and glucose into ethanol can be monitored quantitatively during the fermentation process in cider production.
Similarly, variation in the primary amine signal can be used to follow the curing stage in epoxy resins, and \(\text{C}\)–\(\text{H}\) vibrations can be used to characterize the polymerization of styrene films.
Finally, remote fiber sensing is particularly useful for remotely characterizing reactions that take place in hazardous environments such as microwave ovens. For example, the microwave-assisted synthesis of diethoxypenthane can be monitored using the shift in the \(\text{C}\)–\(\text{O}\) band.
Ouyang et al. have shown that it is possible to detect large biomolecules using two-dimensional photonic crystals based on macroporous silicon microcavities. They were able to detect rabbit immunoglobulin \(\text{G}\) \(\text{(IgG)}\) \(\text{(150}\) \(\text{kDa}\), \(1\;\text{Da}=1\;\text{gmol}^{-1}\)).
The sensing performance of the device was tested and it was shown that concentrations down to 1–2 \(\mu M\) could be detected with sensitivity of 1–2% of a protein monolayer.
Another study by Jensen et al. demonstrated highly efficient evanescent wave detection of fluorophore-labeled biomolecules in aqueous solutions positioned in the air holes of the microstructured part of a photonic crystal fiber. The air-suspended silica structures located between three neighboring air holes in the cladding crystal guide light with a large fraction of the optical field penetrating into the sample even at wavelengths in the visible range.
An effective interaction length of several centimeters is obtained when a sample volume of less than 1 \(\mu l\) is used.
Pollution Control
A portable \(\text{FTIR}\) and long optical fibers may enable one to conduct environmental on-site pollution measurements in water reservoirs, soil, and marine environments with very high sensitivity. Fuel contamination was detected in soil, and several researches were conducted on different liquids, especially in water.
It was shown that pollutants such as benzene could be detected in water. Lately real-time monitoring during a pilot scale test of a natural aquifer in Munich, Germany, took place.
During that study, variations in pollutant concentration profiles were recorded following injection of test quantities of chlorinated hydrocarbons in the aquifer inlet. Control measurements performed simultaneously by chromatographic analysis were in good agreement with the optical detection. This study and other studies demonstrated the potential of \(\text{FEWS}\) for permanent pollution-monitoring devices in water wells.
Biochemical Sensing
Biochemical Analysis
Biological molecules, microorganisms, and tissues have an IR fingerprint, which permits differentiation among different kinds of bacteria and species. Moreover, it may enable the identification of changes in the metabolic process such as structural changes in tissue and protein unfolding.
Several animal studies had been conducted on body fluids drawn from animals, especially blood serum. Keirsse et al. showed that it is possible to differentiate between a fat mouse blood serum to the one from a healthy mouse. Shimony et al. investigated the amount of bovine albumin in water suspension.
Kishen et al. monitored mutans streptococci activity in human saliva. They used FEWS to monitor a bacterial-mediated biochemical reaction. To achieve this, a short length of the cladding is removed; the fiber core surface is treated and coated with a thin film of porous glass medium using the sol-gel technique. The mutants streptococci-mediated reaction with sucrose is monitored using a photosensitive indicator, which is immobilized within the porous glass coating.
Spectroscopic analysis shows that the transmitted intensity at 597 \(\text{nm}\) increases conspicuously when monitored for 120 minutes. Two distinct phases are observed, one from 0 to 60 minutes and the other from 60 to 120 min. A negative correlation coefficient between the rate of increase in absorption peak intensity was seen \(R=-0.994)\).
Cytron et al. have measured the amount of salt in human urine in real time and without any sample preparation to quantitatively assess the urine composition for the diagnosis of urolithiasis in patients.
Urine samples were obtained from two groups of patients: 24 patients with stone formation after shock-wave lithotripsy and 24 healthy subjects of similar age. \(\text{IR}\) absorption measurements were performed in real time, using \(\text{IR}\)-transmitting silver halide fibers.
The absorption data were compared with the \(\text{IR}\) absorption spectra of aqueous solutions prepared in our laboratory with known concentrations of known urinary salts. The results were used for the study of the chemical composition of these salts in the urine samples and for a quantitative analysis of the concentration of the salts.
They determined the composition of the stones in 20 of the 24 patients on the basis of the characteristic absorption peaks for the oxalates, carbonates, urates, and phosphates observed in their urinary samples.
Using the method mentioned earlier, they found the concentration of different salts in urine with an average error of 20%.
In Vivo Measurements and Live Organisms Monitoring
One of the advantages of \(\text{FEWS}\) is the ability to collect the signal far from the working place. This feature allows one to perform measurements on live subjects and inside the patient body in minimally invasive procedures. Such measurements might replace conventional biopsies.
Moreover, \(\text{FEWS}\) enables one to examine cell cultures, live microorganisms, and tissue explants directly within an experimental setup designed to provide appropriate environmental conditions.
In \(\text{vivo}\) spectroscopic analyses have been applied to medical diagnostics and to metabolic monitoring. As an example, noninvasive glucose measurements were performed through the mucous membrane of the lips of human patients.
Clinical tests on diabetic patients showed that the \(\text{FEWS}\) results closely matched the control measurements obtained by a glucose oxidase method. Glucose levels were monitored in real time after patients had been subjected to pulsatile intravenous glucose injections.
Similar in \(\text{vivo}\) measurements were applied to cancer diagnostics on anesthetized live animals. Human breast tumors grown near the surface of mice skin showed noticeable spectral variations, and in \(\text{vivo}\) tests demonstrate the feasibility of quantitative measurement of dye clearance in the gastroesophageal tract.
Sukuta et al. investigated skin cancer using \(\text{FEWS}\) using chemical factor analysis. They isolated the eigenspectra of biochemical species and some of the eigenspectra have been preliminarily identified as due to protein peptide bond and lipid carbonyl vibrations.
Cluster analysis was used for classification, and good agreement with prior pathological classifications, specifically for normal skin tissue and melanoma tumors, has been found.
Live cell monitoring can be used to evaluate metabolic activities on various types of cell cultures. The dynamics of bacterial cell colonies were recorded during the swarming of bacterial biofilms.
The synthesis of lubricant slime during active migration of Proteus mirabilis bacteria was observed in the exo-polysaccharide features of the second derivative spectra. Another study showed that it is possible to detect bacterial activity in human saliva. The sensor determines the specific concentration of Streptococcus mutans in saliva, which is a major causative factor in dental caries.
Several groups had investigated human metabolism. Lucas et al. studied the metabolic responses of human lung cells exposed to various toxicants. A monolayer of a lung epithelial cell line \(\text{(A549)}\) attached to a chalcogenide fiber surface was exposed to micromolar quantities of Triton \(X\)-100 surfactant.
A rapid alteration of the cell membrane was observed within minutes of exposure. This study showed that because of shallow penetration depth of the evanescent wave, the \(\text{FEWS}\) technique collects strong signals from the cell itself while minimizing signal from the surrounding fluid. Another study examined the water diffusion into the human skin.
6. FIBER OPTIC THERMAL SENSING
Changes in local temperature of blood, body organs, tissue, and skin in biological systems depend on the thermal energies received and lost, as well as on chemical reactions as metabolisms. Increase or decrease in temperature might indicate some kind of disease or an abnormal function of the body organ.
Temperature measurements can be done using standard thermometers, thermocouples, thermistors, and radiometric methods and optical methods.
Although measurements of the body core temperature indicate some kind of illness, they do not indicate what the illness is and whether it is just a symptom of it.
However, measurement of a local temperature of an organ might give more information about its condition. Moreover, a thermal image of an organ might reveal structural changes in the organ.
Optical fibers play a major role in temperature sensing. Optical fiber can be used as a temperature sensor; for example, it is possible to embed fluorescent material inside the fiber and monitor the fluorescence parameter changes caused by temperature changes.Moreover, they can transmit thermal radiation to a radiometer that converts it to thermal readings.
Finally, optical fiber bundles can be used to deliver, for example, thermal images for breast cancer imaging.
The use of optical fiber for temperature sensing and imaging has many advantages. They are very small (Fig.10) and could be used during minimally invasive procedures. Furthermore, optical signals are not affected by \(\text{EMI}\) and \(\text{RFI}\) interferences, so these sensors can be used during magnetic resonance imaging \(\text{(MRI)}\) procedures.
Fiber Optic Thermal Sensor
Fiber optic thermal sensors are based on two major technologies: changes in the relaxation time of fluorescent material due to temperature changes, and changes in the optical path of a Fabry–Perot interferometer.
In fiber optic thermal sensors based on fluorescence, the fiber probe communicates optically with a temperature-sensitive fluorescence material such as phosphor (Fig.11). The fluorescence signal is activated by pulses laser diode and is transmitted back through the fiber to a detector.
The phosphor tip, at the far end, is either embedded in the medium to be measured or placed in contact with its surface. The probe is composed of silica fiber and various jacketing layers, all of which are stable over the full temperature measurement range of the instrument.
The phosphor, which is typically stable to much higher temperatures than the glass itself, can respond to temperature in various ways: change of quantum efficiency, spectral shift of emission and/or excitation bands, and alteration of the fluorescence lifetime (decay time). Of these temperature-dependent mechanisms, change in decay time provides the most robust approach to measuring temperature.
The reason is that lifetime changes can be quantified in a manner independent of instrumental and environmental variables.
The Fabry–Perot fiber optic temperature sensor measures the changes in the interference pattern, which are caused by changes in the optical path of one of the interferometer legs. The changes of the optical path are caused by changes in the optical properties (index of refraction, Bragg grating reflectance, etc.) of the material when changes in temperature occur.
Two types of Fabry–Perot fiber optic temperature sensor could be found: intrinsic Fabry–Perot in which the perturbation of interest affects the fiber itself and extrinsic Fabry–Perot in which the fiber is used only to guide the light. The structure of an extrinsic Fabry–Perot temperature sensor is shown in Fig. 12. The cavity can be formed by encapsulating two separate fibers inside a glass capillary tube (Fig.12a) or by mounting a miniature cavity—formed by glass or silicon micromachining techniques—on the tip of a fiber (Fig.12b), with a deformable membrane having a thin film reflector.
The power source could be either a laser (usually a laser diode) or an \(\text{LED}\). When the sensor is exposed to the perturbation of interest (temperature, pressure, etc.) and the reference \(\text{FPI}\) is not, a difference in the optical path length is created. This change causes a change in the interference pattern at the interferometer output.
This change is analyzed and is being converted into a temperature reading.
Fiber optic temperature sensors could be used for a wide range of medical applications. The main advantage of these sensors is that they are immune to \(\text{EMI}\) and \(\text{RFI}\).
Therefore, they could be used during other procedures such as \(\text{MRI}\) examinations. Other applications could be temperature measurements of internal organs during minimally invasive procedures.
In the field of minimally invasive surgery, radiofrequency \(\text{(RF)}\) waves are used locally via a probe or over a region of the body introduced in a tunnel. This method permits resection of masses or tumor treatment with minimal bleeding and minimal damage to surrounding tissue.
However, it is also involved in close temperature monitoring. The small size and intrinsic immunity of optical sensors to electromagnetic fields permit continuous in situ temperature measurements without adding harm to the tissue. Integration of ultraminiature temperature sensor from Fiso, Inc. into minimally invasive surgery involving \(\text{RF}\) waves could help the optimal control of these type of therapies.
Optical Fiber Radiometry
A warm body at room temperature \(\thicksim 300\;^\circ K)\) emits radiation according to Planck’s law at the wavelength region of the mid-\(\text{IR}\) (3–20 \(\mu m)\). The radiation intensity is given by Stephan Boltzmann law:
\[\tag{3}I=\varepsilon\sigma T^4,\]
where \(\text{I}\) is the radiation intensity, \(\varepsilon\) is the emissivity, and \(\sigma\) is the Stephan Boltzmann constant. The emissivity of the human body is unity \(\varepsilon=1\) and the radiation peak wavelength is around 10 \(\mu m\).
As can be seen from Eq. (1), by measuring the radiation intensity, one can calculate the body’s temperature. A device that measures the radiation intensity and converts it into a temperature reading is called radiometer. A typical radiometer usually consists of three parts: (1) optics that collects the \(\text{IR}\) radiation, (2) an \(\text{IR}\) detector that converts the photons into an electrical signal, and (3) an electronic system for signal processing.
A fiber optic radiometer utilizes an optical fiber, which is suitable for \(\text{MIDIR}\) radiation transmission, as the collection optics. A typical fiber optic radiometer system is shown in Fig.13.
Fiber optic radiometers could be used for many medical applications. The most straightforward is measurement of body core temperature. Another application is temperature measurements during \(\text{MRI}\) procedures. There are many cases when it is necessary to measure the temperature of a sample while it is being imaged in an \(\text{MRI}\) system.
For example, there are surgical procedures in the brain, where the treatment of a tumor involves heating with a laser beam or cooling with a cold finger. Medical treatment depends on the accurate measurement of the temperature inside the tumor. In the space between the magnets poles, there are strong magnetic fields, strong gradients of magnetic fields and \(\text{RF}\) fields. The measurement of temperature in such a hostile environment with conventional instruments (e.g., thermocouples) is not easy.
Figure 13. A typical fiber optic radiometric system (used, with permission.)
Thermal measurement of the skin during different procedures could be facilitated using a fiber optic radiometer. Temperature measurement of skin and tissue during laser heating may help prevent the doctor from damaging the surrounding tissue. This is important in applications such as skin resurfacing, laser tissue soldering, and burn generation.
Moreover, the temperature measurement could be used as a feedback to the procedure and could automate some of the procedures by stopping the laser when the temperature reaches a certain value.
7. THERMAL IMAGING
The use of thermal imaging is increasing mainly because of the improvement in thermal cameras. In vivo studies have increased the clinical value of this equipment in many disciplines such as breast cancer (risk assessment, detection, prognosis, and therapeutic monitoring), burn trauma (staging), diabetes, cardiology, neurology, urology, pulmonary, dermatology, ophthalmology, neonatology, pain management, and anxiety detection.
Fewer clinical studies have reported the use of \(\text{IR}\) imaging of internal tissue surfaces intraoperatively. Neurosurgeons have exhibited the \(\text{IR}\) imaging capability to detect vascular occlusion and reperfusion, as well as its potential to explore cerebrovascular disease, epilepsy, functional cortical activation, and brain tumors.
Unlike direct \(\text{IR}\) measurements, utilizing \(\text{IR}\) imaging by minimally invasive means is a challenge as the \(\text{IR}\) images need to be transferred from the internal tissue surface to the \(\text{IR}\) camera trans-endoscopically.
Because minimally invasive surgeries have been extensively replacing conventional surgeries during the last decade, resolving this problem has become essential.
Transmitting a thermal image through an endoscope could be done using a scanning fiber optic radiometer or \(\text{IR}\) fiber bundles coupled to an \(\text{IR}\) camera. The first method is quite complicated because scanning an image with a single fiber might take a long time and controlling the position of the fiber tip is challenging.
The second method is to use \(\text{IR}\) optical fiber bundles for transmitting the thermal radiation to an \(\text{IR}\) camera. Over the last 2 decades, several groups had been trying to manufacture \(\text{IR}\) bundles, unfortunately with limited success. The first ideas for \(\text{IR}\) imaging cones were reported by Kapany. The first group who reported an \(\text{IR}\)-transmitting coherent bundle was Saito et al. in 1985. They used \(\text{As}\)-\(\text{S}\) fibers, which transmit radiation in the 2- to 6-\(\mu m\) wavelength range.
Their aim was to be able to map temperatures in hard-to-reach areas of engines, furnaces, and nuclear reactors. They have created 200–1000 fiber bundles with \(\text{As}\)-\(\text{S}\) core and \(\text{FEP}\) Teflon cladding. The bundle was coupled to an \(\text{InSb}\) camera (with 3.0–5.4 \(\mu m\) operating spectral range) to collect various thermal images.
At first, a blackbody with uniform distribution of \(773\;^\circ\text C\) was imaged, although different temperatures were measured at each spot in relation to each fiber in the bundle. This was caused by different transmittance of each fiber. After calibration for these differences, images of an iron and a candle flame were successfully taken.
In 1987, Klocek et al. reported a more flexible bundle made from GeSeSb fibers that were fused together only at both ends to overcome the rigidity of Saito’s bundle, which was fused along the entire length of the bundle. In 1991, Nishii et al. reported using an \(\text{As2S3}\)-based imaging bundle. Nishii et al. created three types of chalcogenide glasses.
Using a method suggested by Kapany, 8400 fibers, 100 cm long, were stacked in a rectangle. The ends of the bundle were ground and polished for an optical flat surface. The bundle was inserted into a Teflon tube and coupled to an \(\text{IR}\) camera. Images of a human face were taken and different temperature areas were observed. The high \(\text{NA}\) of the fibers caused some resolution lowering. Anti-reflecting coatings on both surfaces improved the image as well, and temperatures as low as \(25\;^\circ\text C\) could be captured by the camera through the bundle.
Ray Hilton, from Amorphous Materials in Garland, Texas, is also using chalcogenide glass to create thermal imaging bundles. The bundle created was 3600 elements of 68-\(\mu m\) diameter each with 52% active area.
\(\text{A}\) \(\text{USAF}\) target with a heat source (human palm behind) was used to test resolution, which was found to be 7 \(\text{lp/mm}\) (Fig.14). The dark spots in the picture are due to a few fibers that were damaged in the fabrication process.
The human subject was also imaged by a 3- to 5-\(\mu m\) camera with reasonable results.
The chalcogenide glass was, in fact, the first used in endoscopic setting. Naghavi et al. reported the first prototype of a flexible \(\text{IR}\) fiber optic catheter for the \(\text{IR}\) imaging of atherosclerotic plaques.
The system had thermal resolution of \(0.01\;^\circ\text C\) and spatial resolution of \(100\;\mu m\). Temperature heterogeneity was successfully detected in a phantom model, simulating blood vessels and hot plaques, as well as in \(\text{in}\) \(\text{vivo}\) animal study. They concluded that the technique is feasible and can be used for thermal detection of vulnerable atherosclerotic plaques.
Katzir et al. reported on an imaging bundle made of silver halide fibers. A 900-element bundle was used to transfer images of heated tungsten wire at \(40\;^\circ\text C\) into a thermal camera with reasonable resolution. These types of bundles were used by Gannot’s group as a feedback mechanism to optimize laser-based tissue ablation. Figure 15 shows an \(\text{in}\) \(\text{vivo}\) setup of an \(\text{IR}\) laser interaction on mouse tissue. Figure 16 shows the temperature mapping.
Gopal et al. reported another type of imaging bundle based on internal coating of a polygon structure of 900 channels produced by collimated holes (CA).
The coating method used is the same as applied in the single-tube coating of fused silica or Teflon, but in this case, each element was 45 or 65 \(\mu m\) in diameter. The coated bundle is shown in Fig. 17. An image of a heated hot wire is shown in Fig. 18. The bundle created in this method is still not very flexible, and only about 50-cm long pieces are fabricated at this time. One of the great advantages of this bundle is the zero crosstalk between the elements.
Infrared Imaging and Tomography in Minimally Invasive Procedures
The investigation of \(\text{IR}\) imaging and/or thermocouple-based thermography using minimally invasive techniques has just begun. Several studies from various medical disciplines that have used experimental apparatuses are briefly summarized here.
Ogan et al. have exploited trans-laparoscopic \(\text{IR}\) imaging to monitor surface renal temperatures during \(\text{RF}\) ablation. They claimed that \(\text{IR}\) imaging had enabled them to assess treatment adequacy and ablation margins. \(\text{In}\) \(\text{vivo}\)
coronary bypass minimally invasive surgeries in beagles has been carried out by Nakagawa et al. using \(\text{IR}\) imaging. In addition, a physiological saline of different temperatures was injected through the endoscope channel to see the changes in cortical images.
They concluded that transendoscopic \(\text{IR}\) imaging (7–14 \(\text{mm}\)) may provide a noninvasive functional angiography. Stefanadis et al. have developed a thermocouple-based thermography catheter and demonstrated in a human \(\text{in}\) \(\text{vivo}\) trial that thermal gradients between healthy and neoplastic tissue zones could be a useful criterion in the diagnosis of malignancy in tumors of the bladder.
They found significant temperature differences (e.g., \(1\;^\circ\text C\)) between patients with benign and malignant tumors. Endodontists McCullagh et al. compared thermocouple-based thermography versus IR imaging of temperature rise on the in vitro root surface during the continuous wave of condensation technique. They found \(\text{IR}\) imaging a useful tool for mapping temperature change over a large area.
In 2001, Cadeddu et al. reported nine laparoscopic urologic procedures in patients, using \(\text{IR}\) imaging (3–5 \(\mu m\)) laparoscopically (i.e., via a rigid endoscope). They stated that \(\text{IR}\) imaging proved useful in differentiating between blood vessels and other anatomic structures and that, in contrast to conventional endoscopy, vessel identification, assessment of organ perfusion, and transperitoneal localization of the ureter were successful in all instances. They concluded that \(\text{IR}\) imaging is a potentially powerful adjunct to laparoscopic surgery.
Dayan et al. have developed a method where thermal imaging can be used as a feedback mechanism to optimize laser ablation within body cavities. All these examples prove the necessity of a flexible imaging bundle in minimally invasive medical procedures for diagnostics and treatments.