Continuous monitoring of human physiological parameters throughout a 24-h activity cycle is becoming more common, and new application areas are continually being developed. Devices for recording vital functions have long been used outside of the doctor’s office, e.g., athletes use personal recorders to control their training intensity.1 Moreover, the development of wireless data transmission technologies and teletransmission has resulted in the creation of many home monitoring systems, which are capable of recording and independently transmitting medical parameters to a patient service center.2 If clinically significant changes are observed in the analyzed data, the supervising doctor is notified by a short message service, e-mail, or fax so that an appropriate treatment can be specified.
Similar systems have been implemented for monitoring drivers, machine operators, and technological systems operators. For example, if symptoms of fatigue, falling asleep, or fainting have been detected, an alarm signal would be triggered to stop the vehicle or device.3 In addition, both military and civil aviation utilize systems for monitoring pilots to monitor their psychophysiological state,4 ensuring pilot and passenger safety as well as the mission outcome, aircraft condition, and load being transported.
When monitoring seated individuals, the sensors and associated measuring equipment can be placed directly in the back or base of the seat, enabling monitoring without attaching electrodes or the recorder to the body.5 On the other hand, special requirements are imposed on systems that must be able to operate in harsh environments without failing, such as high-temperature or high-humidity environments, which prevail in nonair-conditioned cockpits, and environments with high electromagnetic (EM) interference caused by radio or on-board radar systems.
Magnetic resonance imaging (MRI) examination presents another situation in which devices used for monitoring physiological parameters must be capable of operating in an environment with a strong EM field.6,7 For example, monitoring changes in basic physiological parameters can enable the early detection of a claustrophobic episode, allowing the test to be immediately interrupted to remove the patient from unpleasant sensations.8
In this paper, the author justifies the need for monitoring basic physiological parameters in patients undergoing MRI scans, systematizes the knowledge of fiber-optic techniques purposed for recording life functions, which are potentially suitable for an MRI environment, and indicates the current directions of development in this area.
Importance of Monitoring the Patient During MRI
Description of the MRI Environment
The MRI environment is distinguished by high levels of magnetic induction density: 1.5 to 3.0 Tesla (T) in typical appliances. However, systems achieving densities of 9.4 T or even 21.1 T have also been developed.9,10 Strong penetration by the magnetic field is accompanied by high EM field strength, inducing radio frequency impulses. Conventional measuring equipment was not designed to work in such an environment.1112.–13 Placing metal elements, which are common electronic components in typical recorders, in a strong magnetic field can induce eddy currents in these elements and cause them to heat up. This, in turn, can cause skin burns in patients.1415.–16 Moreover, metal elements inside an MRI scanner affect the distribution of the EM field and could, therefore, impair the image quality. Additionally, the electrical signal transmitted by typical copper wires in conventional measurement equipment is sensitive to a strong EM field. As a result, the measured parameter information may be altered before reaching the MRI system operator, rendering the measurements unusable.
Optical sensors can provide an alternative solution for measurement equipment.17,18 Because optical sensors are free from metal elements, they are fully penetrable by the EM radiation used in MRI and cause no interference.19 Fiber-optic sensors2021.–22 transmit signals using optical fibers, providing them with immunity to EM fields. Moreover, there are fiber-optic sensors23 that are able to acquire a patient’s physiological signals and subsequently extract the chosen parameters2425.26.27.–28 if they are configured in an appropriate manner, i.e., connected with mechanical elements and placed near the patient.
Episodes of Claustrophobic Reactions During MRI Examinations
Along with computed tomography (CT), MRI has become one of the most common methods for detecting lesions in tissues.29 Despite its high diagnostic efficiency and prospects for further advancement, e.g., functional magnetic resonance imaging, MRI has a major disadvantage because the patient must be placed in a narrow cylindrical tube for a few dozen minutes, which can induce claustrophobia or panic attacks. Furthermore, during brain scans, the patient’s head is immobilized with a special collar, exacerbating the patient’s discomfort. Attempts to minimize these reactions using open-type MRI systems have produced poor results. The patient’s knowledge that the examination requires assuming a certain position for a length of time invades the patient’s personal and mental space. As a result, an estimated 5 to 10% of patients undergoing an MRI experience high stress related to their restriction in the MRI chamber.30,31 Moreover, 10% or more of all examinations using MRI cannot be commenced or completed due to claustrophobia.32 The acute manifestation of anxiety symptoms immediately before or during MRI examinations can also occur in persons who have been taught anxiety and stress management techniques.33 In the case of a 26-year-old man who was referred for an MRI due to back pain,34 the MRI examination very likely contributed to the patient’s development of posttraumatic stress disorder, which is, according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), an anxiety disorder caused by exposure to high stress.35
The reasons reported by patients for experiencing anxiety and (often) fear of MRI examinations include not only the space limitation and movement restriction, but also the high noise levels generated by the equipment, long duration of the examination, total isolation from the staff, and equipment that moves slowly toward the patient.36,37 As many as 30% of patients undergoing MRI examinations experience mild symptoms of anxiety and stress. Such episodes usually result in the examination being stopped or extended. Moreover, these episodes may result in the administration of anxiolytics and sedatives to the patient, which is not always advised, possible, or safe.32 Such actions may ultimately increase the examination cost as a result of the extended equipment use and involvement of additional medical and technical staff. Therefore, the early detection of anxiety or panic attacks is a significant issue in medicine and a considerable challenge for both equipment designers and all persons involved in examinations requiring any type of chamber, such as those employed in modern neuroimaging techniques.
Basic Physiological Parameters and Their Link to Claustrophobia
While a multipoint acquisition of physiological parameters, e.g., using a Holter-type system, is applied for diagnosis in hospitalized patients, the simple monitoring of basic vital functions is sufficient for persons monitored during regular activities. The respiration rate (RR), which specifies the number of ventilations per minute (vpm),38 as well as the heart rate (HR), i.e., the number of heart beats per minute (bpm),39 are basic physiological parameters. In general, RR is estimated using a respiration curve, while HR, coincident with pulse and heart rhythm, can be estimated based on an electrocardiogram, ballistocardiogram, or plethysmogram. Prophylactic examinations allow for the early detection of increased and/or irregular RR or HR, which can indicate emerging health disorders. Alarming trends in basic physiological parameters can justify a precautionary outpatient examination; likewise, similar trends observed during an MRI examination should lead to the completion or suspension of the examination and its adjustment in line with the current situation of the patient.
Respiration monitoring can be used to detect hyperventilation. Hyperventilation, i.e., an increased rate of pulmonary ventilation and considerably decreased depth of breathing, often leading to syncope,40,41 is closely related to anxiety and panic disorders.42,43 Breathing patterns in persons experiencing panic attacks are distinguished by high variability and irregularity.44 Therefore, panic attacks can be predicted or monitored in terms of the course, intensity, and the speed at which the symptoms resolve.
One aspect in the study of hyperventilation concerns the mechanisms as well as physiological and psychological factors responsible for anxiety symptoms. In addition, conscious and controlled breath management has been analyzed for the treatment of various anxiety disorders. Because attempts to control and normalize breathing helped to reduce symptoms of anxiety and panic,45 therapeutic programs have been developed to teach patients how to cope efficiently with hyperventilation during MRI examinations.46
The patient’s heart function should also be measured during examinations47 because HR indicators are considerably higher in persons suffering from general anxiety, panic, and hyperventilation in comparison with those who do not experience such discomfort. Therefore, that continuous, online monitoring of patients’ circulatory parameters during MRI examinations will allow for the prediction and assessment of anxiety, panic, and claustrophobia closely related to the circumstances of examination.
Although patients likely to experience claustrophobic anxiety attacks are the most common group classified to be monitored inside an MRI magnet, there are many other patient categories that require monitoring and support during MRI procedures. These are patients who are unable or may not be able to communicate or to use the alarm button:7
• neonatal and pediatric patients,
• sedated or anesthetized patients,
• disabled and insensible patients,
• patients in coma,
• patients with mental disorders,
• patients with impaired physiological functions,
• critically ill or high-risk patients,
• patients developing reactions to contrast media,
• patients with implanted pacemakers.
Fiber-Optic Sensors in the Measurement of Physiological Signals and Parameters
Categorization of Monitoring Principles
Most of the fiber-optic sensors23,4849.–50 configured for delivering information on vital functions have been developed over the last 20 years and the results have been published in almost 100 papers.41,5188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.22.214.171.124.126.96.36.199.100.101.102.103.188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.144.–145 Though only some of these sensors have been tested in an MRI environment, fiber-optic sensors are inherently immune to EM radiation and can be MRI-compatible. For these reasons, we consider all of the sensors discussed in the next section as potentially suitable for an MRI environment.
Forty-two fiber-optic sensors constructed by other groups41,522.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.100.101.102.103.22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.22.214.171.124.132.–133 and five in-house-designed sensors134126.96.36.199.188.8.131.52.143.144.–145 for monitoring vital signs can be easily classified into three major categories with several subcategories each, according to the location of the sensor application. In addition, we can distinguish seven sensor classes regarding the type of measurand as follows:
1. Sensors placed in (1a) a nostril or (1b) oxygen mask, measuring (I) humidity, (II) temperature, or (III) force of inhaled/exhaled air (flow);
2. Sensors embedded into (2a) expandable belts or (2b) special garments as well as affixed directly to (2c) the body, measuring (IV) the elongation/contraction, (V) strain (including curvature), or (VI) lateral pressure caused by respiration and/or heart work;
3. Sensors attached to (3a) the back of a chair, placed on/in/under (3b) a bed mattress, or embedded into (3c) a pneumatic cushion, measuring (V) the strain, (VI) pressure, or (VII) force caused by respiration and/or heart function.
Figure 1 illustrates this classification by showing the number of sensors in relation to the placement of the sensing element and the type of measurand. The largest groups are sensors embedded into a textile belt or bed mattress, which measure lung- and/or heart-induced elongation/contraction and strain, respectively.
To illustrate how trends in the area of fiber-optic sensors for recording physiological functions have changed over the last 20 years,41,5184.108.40.206.220.127.116.11.18.104.22.168.22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.22.214.171.124.126.96.36.199.188.8.131.52.100.101.102.103.184.108.40.206.220.127.116.11.18.104.22.168.22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.22.214.171.124.144.–145 Fig. 2 shows the number of articles published in any individual year in each of the three categories. In the 1990s and at the beginning of the new millennium, the first devices placed in a nostril or oxygen mask were proposed. A significant development has occurred since the mid-2000s, when it was noted that the measuring elements could be embedded into wearable textiles, ensuring the immediate proximity of the sensors to the lungs and heart. However, due to the wide range of clothing sizes required to accommodate different patients, the interest in textile-based sensors has decreased in recent years. Wearable-textile sensors are now being replaced by sensors embedded into a bed or seat that do not require preparing the patient for monitoring. In addition to the literature review, the author describes his group’s work on embedded sensors.134126.96.36.199.139.–140,142143.144.–145
Results of Literature Review
Intensity-modulated fiber-optic sensors
In recent years, different research groups have presented dozens of fiber-optic sensor designs suitable for use in an MRI environment that allow the RR and/or HR to be obtained after processing the recorded signal. An RR-monitoring device that detected the evaporated humidity from the mouth and/or nose was one of the first sensors to be studied.5152.53.54.–55 The condensed humidity substantially alters the interaction between the light from the optical fiber and the surrounding medium, which can be monitored by a photodetector. As shown in Fig. 3(a), during expiration, a water film (many hemispherical water drops) is formed at the optical fiber end with a refractive index , and the amount of light reflected back into the fiber decreases. During inspiration [Fig. 3(b)], the dry and cool air next to the fiber does not produce condensation at the tip. As the refractive index at the boundary approaches 1, the amount of light reflected back into the fiber increases. These changes in light reflection can be measured by a photodetector positioned at the opposite end of the optical fiber, and the RR can be monitored by counting the cyclic variations in the light reflection.
Table 1 lists the major properties of intensity-modulated fiber-optic sensors designed for monitoring respiration and/or heart function, which were studied in an MRI environment or are suitable for use in the presence of strong EM fields. In the first column, the category and subcategory of the sensor, according to above classification, are provided together with the reference paper(s) describing each of the mentioned devices. The second column presents the form of the sensor and the physiological parameter(s) that can be monitored. The third column summarizes the experimental studies and their results, including the number of participants, testing procedure, and quantitative data on the sensor efficiency. The last column shows the MRI compatibility of the individual sensors verified at a given magnetic induction and/or for given imaging sequences.
Intensity-modulated fiber-optic sensors for respiration monitoring.
|Category; reference||Sensing element; parameter||Characteristics of subjects; testing procedure; quantitative data on sensor efficiency||MRI compatibility|
|1a.I; 51–53||Plastic tube; RR||10 healthy volunteers (M), age: 24 to 42 years; supine position, recording time: 100 breaths (each subject), normal ventilation (9 to 17 vpm); Wilcoxon test: no statistically significant differences between reference and measured RR values (p>0.05), apnea and tachypnea: events detected immediately||Not tested|
|1a.I; 54||As above||289 patients (151 M, 138 F), age: 8 to 90 years, mean 50.1 years; eight technicians estimated clinical utility of sensor: very valuable in 17 patients (5.9%), valuable in 77 patients (26.6%), not necessary in 195 patients (67.5%)||Yes (0.2 T, 1.0 T, and 1.5 T)|
|1a.I; 55||As above||18 healthy volunteers, age: 9 to 83 years; recording time: 4×3 min periods (each subject), normal ventilation; t test: sensor significantly overestimated RR mean value (p<0.001), 95% confidence interval: sensor did not overestimate number of breaths for individual registration (0.5 to 1.5 vpm, mean 14 vpm)||As above|
Note: MRI, magnetic resonance imaging; RR, respiration rate; vpm, ventilations per minute.
Interferometric fiber-optic sensors
At the same time, solutions also emerged that were based on phase interferometry, where the optical signal that was directed to the measurement area of the optical fiber (placed near the monitored person, e.g., in the back or base of the seat) was compared with the output signal, which was modified by disrupting factors, such as respiration and heartbeats.56188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.73.74.–75 Body movements caused by the function of the lungs and heart cause phase changes between signals (sensing and reference) and could be identified as a series of time-variable interference patterns. Thus far, in research on the acquisition of physiological signals, fiber-optic interferometers have been tested in their most popular configurations, i.e., Michelson, Mach-Zehnder, and Sagnac interferometers.5859.–60 The popular Michelson interferometer has been the most widely studied.622.214.171.124.126.96.36.199.70.71.–72 As shown in Fig. 4, the Michelson interferometer consists of two optical paths (arms), i.e., the sensing and reference fibers. Both fibers have mirrored ends, and the sensing region is formed by the differential path length between the two parallel fibers. Minute lung- and/or heart-induced changes in the sensing fiber length cause significant changes in the phases of reflected light, which subsequently produces changes in the optical power monitored by the photodetector and allows for the determination of RR and/or HR.
Human vital signs have also been monitored using fiber-optic speckle interferometry.7374.–75 Speckle patterns can be observed in the laser light output of coherently illumined multimode fibers (MMFs), as shown in Fig. 5. The fiber output is projected onto a complementary metal-oxide semiconductor imaging sensor. Consecutive speckle images arise from the interference between propagating fiber modes and are highly sensitive to external fiber perturbations, such as those caused by the mechanical impact of human activity.
Recently, interferometric sensors have been developed in which a segment of photonic crystal fiber (PCF) was used as the measuring element.76,77 As shown in Fig. 6, the collapsed region within the PCF allows the excitation and recombination of modes, thus creating an interferometer. The reflection spectrum of this device displays a sinusoidal interference pattern that instantly shifts when water molecules, which are present in exhaled air, are adsorbed on or desorbed from the PCF surface. Therefore, these sensors detect the difference between the humidity of inhaled and exhaled air, which allows a respiration curve and RR to be obtained.
A cornerstone for the development of the photonic crystal based interferometer was the Fabry-Pérot (FP) etalon interferometer shown in Fig. 7, which was formed by placing alternating multilayer thin films of polymers and inorganic nanoclusters on the cleaved end of a single-mode fiber (SMF).7879.–80 However, the patient is required to wear a mask with built-in sensors, which presents a notable obstacle of the PCF- and FP-based fiber-optic interferometers for respiration monitoring. Table 3 lists the major results obtained using these devices.
The key advantage of interferometric methods is their high sensitivity; however, this can imply instabilities in the signal amplitude if the monitored person moves, changes his/her position, or simply coughs.59,60 Then, the signal amplitude can reach relatively high values, which bring difficulties in the demodulation of the phase signal. Therefore, systems for the elimination or compensation of body movements are required, and advanced signal analysis techniques have to be able to detect miniature signal changes carrying information on respiration activity and heart function. Major outcomes obtained using these interferometric fiber-optic sensors for respiration and/or heart function monitoring are listed in Table 2, using a column layout identical to that presented in Table 1.
Interferometric fiber-optic sensors for respiration and/or heart function monitoring.
|Category; reference||Sensing element; parameter||Characteristics of subjects; testing procedure; quantitative data on sensor efficiency||MRI compatibility|
|2b.V; 56,57||Mach-Zehnder, Michelson, or Sagnac interferometer embedded into elastic pad; RR or HR||No data (patent description)||Yes (no data)|
|3b.V; 56,57||As above||As above||As above|
|3a.V; 58–60||As above||Insufficient data||Not tested|
|3b.V; 61–72||Michelson interferometer embedded into elastic mat; RR and HR||Several research sessions with varying numbers of participants and signal analysis techniques. Most recent experimental study: (1) rest: 20 healthy volunteers (19 M, 1 F), age: 31.0±8.15 years; supine position, recording time: 60 s; sensitivity: 97.64±7.28%, precision: 99.38±2.80%, maximum relative error (MRE): 7.35±7.20% for RR detection, sensitivity: 99.46±1.11%, precision: 99.60±1.05%, MRE: 3.16±2.32% for HR detection; (2) cycling ergometer: 10 volunteers (9 M, 1 F), age: 31.30±10.45 years, supine position, recording time: 60 s, sensitivity: 92.05±6.10%, precision: 93.45±3.08%, MRE: 16.28±6.25% for RR detection, sensitivity: 96.22±2.96%, precision: 95.35±3.03%, MRE: 9.56±3.67% for HR detection. In general, MRE: >30% for RR detection, <17% for HR detection||As above|
|3b.V; 73||Speckle interferometer embedded into elastic mat; RR and HR||No data (patent description)||As above|
|3b.V; 74,75||As above||Most recent experimental study: 10 healthy volunteers, age: 31.8±8.1 years; rest, supine position, recording time: 120 s; sensitivity: 95.3±3.0%, precision: 97.9±2.3% for RR detection, sensitivity: 99.4±0.6%, precision: 98.8±1.5% for HR detection||As above|
|1b.I; 76||Photonic crystal fiber interferometer inside capillary tube; RR||Two volunteers; sensor was tested for different conditions: normal breathing, accelerated breathing after physical exercise, stopped breathing for few seconds; basic evaluations only||As above|
|1b.I; 77||As above||No information on number of participants and their age; sensor was tested for normal and irregular breathing; evaluation of sensor construction; no data on sensor efficiency||As above|
|1b.I; 78–80||Fabry-Pérot interferometer inside plastic tube; RR||Tests were carried out using a simulated breathing flow system, with relation to thin-film refractive indices and thickness; no data on sensor efficiency||As above|
Note: HR, heart rate.
Sensors in modalmetric configuration
While the vital activity information is contained within the phase changes in fiber-optic interferometers, the so-called modalmetric configuration of a sensor detects the vibrations related to vital functions as light intensity modulations proportional to those functions (sensor category: 3a.V).60,81 Such sensors consist of (1) an SMF transmitting segment, (2) an MMF segment sensitive to mechanical stress, and (3) an SMF receiving segment, as shown in Fig. 8. The single-mode transmitting fiber delivers optical radiation to the MMF section, in which the layout of the mode field is dependent on mechanical deflections caused, for example, by respiration and heart function. In other words, the signal loss is primarily due to the bending loss when the MMF section is deformed. The amount of light reaching the single-mode receiving fiber reflects the layout of the mode field. Thus, the receiving section of the sensor serves as a detector for the radiation propagated within the MMF section, i.e., the measurement head. As described in the solutions involving interferometry, the measurement of vital functions with a modalmetric sensor is accomplished using ballistography.60,81 The main disadvantage in the use of this type of sensor is that it is difficult to simultaneously record respiration and heart function. So far, no comprehensive evaluation of the modalmetric sensor has been performed in the MRI environment.
Micro- and macrobending fiber-optic sensors
In sensors measuring transmission loss caused by slight (miniscule) deflections of an optical fiber or in fiber bends with a radius of curvature well above the fiber diameter (), i.e., in so-called micro-82188.8.131.52.87.–88 and macrobend89184.108.40.206.220.127.116.11.18.104.22.168.22.214.171.124.106.–107 sensors, respectively, the intensity of the light reaching the receiver is also measured. As in the cases described above, body movements (caused by respiration and heart function, among other things) are a complicating factor since these movements produce micro- and macrobends of fibers with a variable radius of curvature. Thus, along the axis of a bent fiber, the layout of the mode fields continuously changes as the energy radiates; this is seen in the form of light intensity changes at the receiver. The measuring system for micro- and macrobend sensors is distinguished by a simple design for the sensors alone as well as for the associated transceiver modules (light source and photodetector). Nevertheless, due to their relatively low sensitivity, the sensors have to be embedded in a special mat82,83 or cushion,8586.87.–88 on which the patient lies during monitoring, or in textiles, such as a vest, T-shirt, or belt, which are worn by the monitored person,89126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.106.–107 to ensure immediate proximity to the lungs and heart. An example of a microbend sensor system is shown in Fig. 9(a). The sensor mat is constructed by means of two microbenders with MMF passed between them, as shown in Fig. 9(b). The whole setup is covered with a textile material to protect the sensor. A macrobend sensor embedded into an expandable belt is illustrated in Fig. 10 in its most common arrangements, i.e., half-loop (half-oval-form), figure-eight loop, and U-shape.
Another method for improving the sensitivity is the use of a standard MMF41 or SMF108 into which a short section of SMF with a narrower diameter core is inserted. In this hetero-core sensor segment, the transmitted beam is partially leaked into the cladding layer. Additionally, as shown in Fig. 11, the leaked light changes with the bending action of the hetero-core segment, and the bending property can be measured from the changes in transmission.
Table 3 characterizes the major properties of these micro- and macrobend fiber-optic sensors obtained during experimental studies.
Micro- and macrobending fiber-optic sensors for respiration and/or heart function monitoring.
|Category; reference||Sensing element; parameter||Characteristics of subjects; testing procedure; quantitative data on sensor efficiency||MRI compatibility|
|3b.VII; 82 and 83||Sandwich microbender structure; RR||Most recent experimental study: 20 healthy volunteers (10 M, 10 F), age: 24 to 61 years; supine position, recording time: 30-min MRI session (each subject), normal ventilation (7 to 21 vpm); Pearson test: statistically significant correlation in reference and measured RR values (p<0.05), accuracy: ±2 vpm||Yes (1.5 T, spin-echo, T2-weighted)|
|3b.VII; 84||Two microbenders embedded into elastic mat; RR and HR||Clinical trials: 11 healthy subjects (6 M, 5 F), age: 26 to 62 years; supine position, recording time: 30-min MRI session (each subject); Pearson test: good agreement in simultaneous RR and HR measurements between sensor and reference system, statistically significant correlations (r=0.963, p<0.001 for RR; r=0.997, p<0.001 for HR), maximum errors: ±1.7 vpm for RR and ±1.2 bpm for RR and HR||Yes (3.0 T)|
|3c.VII; 85–88||Microbender array; HR||Several research sessions with different number of participants and various algorithms for signal analysis. Most recent experimental study: three volunteers, no age data; headrest position, recording time: 3×2 min periods; HR range: 69 to 72 bpm, error: <0.4 bpm||Not tested|
|2a.IV; 89 and 90||Optical fiber; RR||Four piglets; tests aimed at monitoring chest wall motion during high-frequency oscillatory ventilation||As above|
|2a.IV; 91||As above||Eight healthy volunteers (M), age: 25 to 56 years; sitting position, after 10 min rest, recording time: 3 to 4×1 min periods (each subject)||As above|
|2a.IV; 92–94||Optical fiber in half-loop or figure-eight loop configurations; RR||Insufficient information on number of participants and research methodology; correlation coefficient: 0.7558||As above|
|2a.IV; 95–106||Optical fiber in half-loop or U-shaped configurations; RR or HR (at apnea)||Four volunteers, waistline range: 80 to 110 cm; sensor was tested for different conditions: normal breathing, slow and deep breathing, stepping on ledge, filling dishwasher; elongations of the abdominal circumference: <1%||Yes (no data)|
|2b.VI; 107||Optical fiber in half-oval-form geometry; RR||One volunteer (M), age: 28 years; Four custom-made optical fibers were tested in five placements and with three types of breathing: diaphragmatic, upper costal, and mixed breathing; respiration-induced signal amplitude: max 60 mV0−pk||Not tested|
|1b.I; 41||Hetero-core fiber-optic arrangement (core: 50-3-50 μm); RR||One volunteer; recording time: 30 s, sensor was tested with five types of respiration: breathing by mouth, by nose, heavily, deeply, not breathing||As above|
|3b.VI; 108||Hetero-core fiber-optic arrangement (core: 9-5-9 μm) embedded inside plastic sheet with eight sensing elements; RR||Seven volunteers; during sleep, recording time: >6 h (each subject), sensor was tested for different breathing conditions, e.g., apnea, and various configurations of single sensing elements||As above|
Note: bpm, beats per minute.
Plastic optical fiber-based sensors
A variation of the microbend sensor is a system in which the intensity of light falling onto a plastic (polymer) optical fiber (POF) head is measured; this light was previously reflected from a mirror surface whose distance from the fiber front changes according to the respiration rhythm.109 Figure 12(a) shows the structure of the abdomen-attached respiration sensor. The polymethyl-methacrylate (PMMA) tube fastens the aluminum mirror and one end of the spring, while the other end of the PMMA tube is attached to the POF tip.
Other solutions utilizing POF allow RR monitoring based on the temperature difference between inhaled and exhaled air109,110 or the force with which the air is exhaled immediately in front of the nose or mouth.111 In the first case, shown in Fig. 12(b), the intensity of the reflected light is changed upon the color variation of the temperature-sensing film, and the modulated light is guided to a photodetector via the POF link. In the second device, depicted in Fig. 12(c), the force of the exhaled air pushes a vertical flap, and the coupling of light from the fixed fiber becomes disturbed. The change in coupled power can be easily measured by a photodetector.
Breathing condition sensors also utilized POF as a substrate for the cladding layer, which swells when it comes into contact with water molecules within the exhaled air, as shown schematically in Fig. 13.112,113 Under the dry-state conditions in Fig. 13(a), the refractive index of cladding layer, , is slightly larger than that of the PMMA fiber core (), and the sensor head is a leaky-type head. For the humid state in Fig. 13(b), the sensor head becomes a guided-type head as decreases, and the output light power, , through the sensor head will increase remarkably, i.e., .
Table 4 characterizes the major outcomes obtained using the above-mentioned POF-based sensors.
The plastic (polymer) optical fiber based sensors for respiration monitoring.
|Category; reference||Sensing element; parameter||Characteristics of subjects; testing procedure; quantitative data on sensor efficiency||MRI compatibility|
|2a.IV; 109||PMMA tube; RR||Insufficient data||Yes (0.32 T, spin-echo T1-weighted, gradient-echo T2-weighted)|
|1a.II; 109 and 110||Thermochromic-pigment applied on PMMA tube; RR||As above||As above|
|1a.II; 110||Silver halide applied on PMMA tube; RR||As above||As above|
|1b.III; 111||Sensor head; RR||No information on number of participants and their age; sensor was tested for normal breathing, evaluations of several sensor designs; maximum sensitivity: 1.2 vpm at resolution >1 Hz||Not tested|
|1b.I; 112 and 113||Hydroxyethylcellulose and polyvinylidenefluoride-based sensor head; RR||No information on number of participants and their age; sensor was tested for normal breathing disturbed by coughs, basic evaluations of the sensor construction, response time: <1 s||As above|
Note: PMMA, polymethyl-methacrylate.
Optical time domain reflectometry based sensors
Another known solution is based on optical time domain reflectometry (OTDR), where the propagation of a light impulse (introduced into the POF and dispersed with partial reflection at the location of the mechanical disturbance of the fiber) is measured as a function of time (sensor category: 2a.IV).9618.104.22.168.–101,114 Figure 14 illustrates a basic system employing a POF-OTDR-type sensor for respiration monitoring. The preliminary results show sufficient sensitivity for detecting respiratory movements, i.e., cyclic strain between 0 and 3%. Moreover, steps of 0.25% strain could be measured, demonstrating a quadratic correlation between the sensor signal and the applied strain. However, the sensor has not been evaluated in the MRI environment. The disadvantages of this type of sensor are a relatively low sampling rate for the signal and the need for complex and expensive transceivers.
The list of sensors allowing the acquisition of the respiration curve with the ballistocardiographic (BCG) signal is completed by solutions based on fiber Bragg gratings (FBGs) whose properties are described in the next section. The author also devotes Sec. 5.1 and 5.2 to explore the literature and present his work on FBG-based vital signs sensors, respectively.
Fiber Bragg Gratings
Principle of Operation of FBG Sensors
In recent years, FBGs have been the subject of much research and are some of the most widely utilized optical elements, serving a wide range of applications in telecommunications as well as in laser and measurement technology. The applications in which FBGs serve solely as sensors cover many fields,146 e.g., the monitoring of structural health,147148.149.–150 operating conditions of electrical plants,151152.–153 or life processes in the human body.2627.–28,154155.156.157.–158 Furthermore, FBGs are popular for their resistance to modulations in the optical signal intensity, which is possible because of the so-called spectral encoding, i.e., the information is contained not in the light intensity but in a spectral parameter (wavelength).159 This property, along with its so-called self-reference capability, i.e., a return to the referential values of spectral parameters after achieving the initial conditions, allow multiple Bragg gratings to be contained within one optical fiber.160,161 Exhibiting the standard advantages of fiber technologies, FBG-based sensors are immune to EM effects, chemically inert, small-scale, lightweight, and provide electrically safe modes of operation.162,163
A typical FBG sensor operates as a stop-band filter, where part of the light falling onto the grating is not retained but reflected. The Bragg grating is formed by a portion of the SMF core where periodic modulation is introduced in the refractive index value.164 Beams of light reflected from individual grating layers, with the same phases, are subject to interference. This phenomenon occurs at only one wavelength, called the Bragg wavelength or the central wavelength, :165 and Λ is the modulation period of this index.
The spectrum of light reflected from the grating forms a pointed intensity peak, the maximum value of which is located at the Bragg wavelength, whereas the spectrum of light transmitted through the grating is a narrow intensity trough with its minimum value at the Bragg wavelength. An example of the signal reflected from a Bragg grating simulated using the T-matrix method166167.–168 in Mathcad software is shown in Fig. 15. The simulated FBGs have the Bragg wavelength of 1550 nm with 4000 [Fig. 15(a)] or 12,000 [Fig. 15(b)] index modulation periods, which is equivalent to a grating length of 2 or 6 mm, respectively. The peak sharpness, full-width at half-maximum (FWHM), and the corresponding reflectance can be adjusted by the number of refractive index perturbations. In most sensor applications, the central wavelength is determined from the peak position because the shape of the reflected signal is easier to demodulate. Therefore, FBG-based sensors have one crucial advantage: they can be connected to a transceiver with a single fiber cable. This means that the sensor can be connected using only one of its terminals.
Strain and Temperature Sensitivity of FBGs
The FBG susceptibility to strain, which is manifested as a shift of the Bragg wavelength, , motivated our first FBG sensor model for monitoring respiration and heart function. The Bragg shift148,169 has the following relationship:169 The strain change, expressed in the units of microstrain (), is given by:168,170
Due to the phenomenon of thermal expansion, the Bragg wavelength is also affected by temperature changes, , as follows:150,153,171172.–173 Therefore, for a wavelength of 1550 nm, the temperature change is given by: 150,174,175 Due to the considerable difference between the fiber diameter ( for common optical fibers) and the FBG length (5 to 10 mm for a standard grating inscribed in an SMF core), the effects of the fiber diameter are assumed to be negligible. Finally, the expected strain- and temperature-induced wavelength shift of an FBG can be calculated as follows:
As the heart and lungs work, deformations occur on the surface of the torso, which can be measured by strain gauges, such as FBG sensors. Preliminary testing with an FBG applied to the chest near the heart showed that heartbeats cause peak-to-peak strains with values not less than . At the same time, breathing-induced strains were detected at least four times greater than the heart beats, i.e., . These values were taken as the amplitudes of the BCG signal and the respiration curve, respectively, to demonstrate how the working heart and lungs affects the Bragg wavelength when an FBG makes physical contact with the subject’s torso. Figures 16Fig. 17–18 show the results of simulations performed using Mathcad software. Figure 16 demonstrates changes in the Bragg wavelength caused only by cardiac function, e.g., during apnea. The upper graph shows a 1-s window (2-s window shown in the still image) representing the animated curve of the heart-induced strain during the 30-s measurement duration. The signal was obtained by the superposition of several sine functions whose amplitudes and periods were chosen so that the resultant signal shape was similar to a ballistocardiogram.176,177 In other words, the BCG signal expressed in microstrains with a amplitude and 1-s period (60 bpm) is presented in the upper graph. The middle graph shows the strain-induced Bragg wavelength shifts of the FBG peak synchronized with the strain changes. To facilitate the calculation, the reflected spectrum is represented by a Gaussian with a reflectance close to 1 and with an FWHM of . The lower graph demonstrates how changes in the heart-induced strain modulate the Bragg wavelength in time. The axes are intentionally inverted to emphasize synchronization with the middle graph. Moreover, this result reveals that the maximum value of the FBG reflected spectrum could be used for drawing the ballistocardiogram shape in wavelengths. According to Eq. (3), the strain-induced wavelength shift changes linearly with the heart-induced strain, hence the heart-induced Bragg wavelength shift.
Figures 17 and 18 show the same BCG signal additionally supplemented with the effects of breathing and temperature changes, respectively. The videos are presented in a similar convention as Fig. 16, except that the upper graph in Fig. 18 shows a 5-s window representing the animated curve of the heart- and lung-induced strain along with the temperature trace (rate increase of ). The characteristic shape for the breathing signal was obtained using a superposition of the sine function and its second harmonic. The resultant signal has a amplitude and 5-s period (12 vpm). Finally, the ballistocardiogram and respiration curve signals are superposed to form the time-varying trace of the strain exerted on the FBG sensor. Fig. 18 clearly shows that gradual heating of the grating does not disturb the measurement of heart- and lung-induced strain because the temperature growth influences the mean value of the measured wavelength (signal), whereas the heart- and lung-induced vibration causes changes to the instantaneous value. The temperature-induced drift can be removed by appropriate signal processing, and hence, further considerations could treat the effects associated with temperature fluctuations without affecting the heart- and lung-induced Bragg wavelength shift measurements.
FBG-Based Sensors for Monitoring Vital Signs
The literature provides a number of descriptions in which a single FBG115116.117.–118 or multiple FBGs9722.214.171.124.126.96.36.199.–106,114,119120.121.122.–123 act as the main sensing elements for acquiring body movements caused by respiratory and/or heart function. In both cases, FBGs can be embedded in the clothing worn by the patient during monitoring 97188.8.131.52.184.108.40.206.–106,115220.127.116.11.120.121.–122 or mounted in the bed on which the person lies, as shown in Fig. 19.123
Furthermore, FBGs are used as sensors for detecting sounds produced by a working heart.124,125 Preliminary studies of long-period gratings (LPGs), a variation of Bragg gratings, for recording ventilatory movements and/or volumes have been performed because these gratings are particularly sensitive to any deformations in the LPG section.126127.128.129.130.131.–132 Change in thorax curvature is sensed as bending of an LPG attached to the thorax. Bending induces strain and stress across the fiber and, in consequence, a change in its refractive index via the strain-optic effect, which manifests as a change in position, shape and amplitude of resonant bands in the transmission spectrum. The cladding modes generated by an LPG are more affected by bend-induced refractive index changes across the fiber than the backpropagating core modes generated by an FBG. Until recently, complex and expensive transceiving systems were used to handle LPG-based sensors, although the latest studies present a monochromatic measurement scheme that requires only a photodiode as a detector.132 Moreover, a fiber LPG curvature sensing scheme has been presented lately as a system capable for detecting the mechanical motions and/or sounds of the human heart from surface of the torso.133 Table 5 lists the major properties of FBG-based sensors for respiration and/or heart function monitoring. In general, fiber gratings provide many benefits with relatively few disadvantages, therefore, these elements have been of interest to the author for application to physiological measurements.
The fiber Bragg grating (FBG) based sensors for respiration and/or heart function monitoring.
|Category; reference||Sensing element; parameter||Characteristics of subjects; testing procedure; quantitative data on sensor efficiency||MRI compatibility|
|2a.V; 115||Single FBG; RR||One volunteer (M), 26-year old; sitting position, during rest, sensor was tested for different conditions: slow and deep breathing, fast and shallow breathing||Not tested|
|2b.IV; 116 and 117||Single FBG embedded into polymeric foil; RR and HR||12 healthy volunteers, age: 20 to 30 years; standing position, during rest, sensor was tested when the subjects breathed naturally and when they stopped breathing for few seconds, recording time: 60 s, respiration-induced Bragg wavelength shift amplitude: max 153 pmpk-pk, heartbeat-induced Bragg wavelength shift amplitude: max 12 pmpk-pk||As above|
|2a.IV; 118||Single FBG; RR||No information on number of participants and research methodology; respiration-induced Bragg wavelength shift amplitude: max 60 pmpk-pk (according to graphical data), designers concentrated mainly on signal processing||Yes (no data)|
|2a.VI; 119||Two FBGs written in Er-doped fiber, laser cavity fixed to plastic plate; RR||Two healthy volunteers (M), age: 26 and 24 years; sensor was tested in four conditions, including normal breathing (mean: 12.60 and 19.32 vpm for first and second volunteer, respectively), breath apnea, breath in different gestures (sitting and standing), and breath recovery procedure after strenuous activity, recording time: 60 and 210 s; elongation ratio for the flexible textile: 15 and 9% (first subject with waistline: 78 cm)||Not tested|
|2a.IV; 97–106, 114, 120, and 121||Several FBGs; RR and HR||Healthy adult volunteers, sensor was tested for normal breathing and when breath was stopped for a few seconds; elongations of the abdominal circumference: 1 to 3%, respiration-induced Bragg wavelength shift amplitude: max 152 pmpk-pk, heartbeat-induced Bragg wavelength shift amplitude: max 25 pmpk-pk||Yes (no data)|
|2b.IV; 122||Array of 40 FBGs; three-dimensional volumetric changes of the human torso||Five volunteers, age: 27 to 56 years, waistline range: 73.7 to 96.5 cm; no information on recording time; mean Pearson coefficient for correlation in the reference and measured volume values: 0.86±0.03 (p<0.01), maximum volumetric error: 15%||Not tested|
|3b.V; 123||12 FBGs (3×4 array), arc-shaped metal bridges; RR||10 subjects, no age data; supine position, no information on recording time, the sensor was tested for normal breathing (10 to 25 vpm); maximum error: ±1 vpm when compared with manual counting||As above|
|2c.IV; 124||Single FBG, stethoscope structure; HR||Sensor was attached on speaker membrane emulating heart sounds; heartbeat sound-induced Bragg wavelength shift amplitude: max 10 pmpk-pk, proof-of-concept demonstration only||As above|
|2c.V; 125||Single FBG; blood flow and HR||FBG was fixed to speaker emulating the effect of blood flow-induced wrist motions; amplitude of heartbeat spectrum: 5 dB higher than the floor (poor selectivity), preliminary demonstration only||As above|
|2c.V; 126 and 127||Array of sensing platforms consisted of two LPGs (each), flexible pad; RR||Sensor was tested at five different localizations on resuscitation training manikin; respiration-induced Bragg wavelength shift amplitude: max 3.18 pmpk-pk, depending on the change in the torso circumference||As above|
|2a.V; 128–131||LPG-based curvature sensor array, flexible band; RR||Six volunteers (M), age: 29 to 65 years, different body dimensions with marked variation in body shape; standing position, recording time: 5×1 min periods that consisted of typically 10 breaths (each subject); maximum volumetric error: 6% when compared with spirometer||As above|
|2c.V; 132||Single LPG; volumetric changes of the human torso||18 healthy volunteers (9 M, 9 F), age: (33±8) years, BMI: (24±5) kg/m2; supine position, recording time: not less than 30 s (each subject), natural and shallow breathing (typical tidal volumes: 400 to 800 ml); minute volume error: (8.7±4.4) % for natural breathing, (10.1±5.8) % for shallow breathing, when compared with a spirometer, paired t test: no statistically significant differences between the minute volumes measured by the sensor and spirometer, mean tidal volume error: (10.5±3.8) % for natural breathing, (15.0±4.8) % for shallow breathing, good agreement between the mean tidal volumes measured by the two methods||As above|
|2c.V; 133||3 or 4 LPGs attached to the human torso; motion of the heart wall next to the apex region of the heart||2 subjects, age: 34 and 43 years., chest diameter: 87 and 113 cm, respectively; neutral spine position, no information on recording time; sensor were tested at 6 different locations on the human torso, seismocardiographic signals up to 54 Hz were obtained, the sensing scheme can resolve individual components of the mechanical wave generated by the human heart revealing common frequencies of 1.3, 7.6, 10.8, and 22.1 Hz present in tests on both subjects||As above|
Note: LPG, long-period grating.
Development of In-House-Designed Sensors
The preprototype sensor base was formed by a pneumatic cushion, which served as a medium for transmitting strains caused by the movements of the person leaning against or lying on the cushion.134135.136.–137 As shown in Fig. 20, an FBG attached inside the cushion records the strains in the form of instantaneous Bragg wavelength changes. The strains can be calibrated to represent the respiration curve and ballistocardiogram. Adequate signal filtration allows reading of RR and HR.
The commercially available sm130-700 four-channel device by Micron Optics178 was used as a transceiver system, which is also called the interrogation system or interrogator, and was equipped with a broadband light source and a tunable FP filter allowing the 1510 to 1590 nm band to be scanned at a frequency of 1 kHz. This interrogation system type is one of the most popular methods for analyzing the reflected FBG spectrum. The measurement is performed by placing a peak from the FBG (or several peaks from an array of FBGs) within the free spectral range, i.e., within the interval between successive interference maxima of the FP filter, and monitoring the changes in the Bragg wavelength by means of spectral tuning, as shown in Fig. 21. The peak-to-peak resolution of the sm130-700 interrogator is 1 picometer (pm), which, upon conversion according to Eq. (3), ensures dynamic measurement of the strain changes with a resolution of .
The initial sensor tests were conducted to assess whether the magnitude of the respiration- and heart-function-induced strains is sufficient for detection in the signal from the FBG-based sensor. The evaluation was performed at a test facility, where the monitored person sat in an office chair with a firm back. The sensor (cushion) was placed between the back of the monitored person and the back of the chair at heart height. In all recorded cases, the signal consisted of two fundamental components according to the simulations presented earlier: a slowly changing wave reflecting lung function, i.e., the respiration curve, and cyclic disturbances to this curve reflecting heart activity, i.e., the ballistocardiogram. As shown in Fig. 22(a), the signal wave amplitude was within 20 to , which corresponds to strain values of 16.5 to . The heart function disturbed the signal envelope with an amplitude of , corresponding to a strain of . A clear upward trend in the signal indicates the temperature increase of the FBG element due to the proximity of the subject’s body. In order to demonstrate that the sensor can simultaneously obtain RR and HR, the signal was subjected to parabolic detrending and Fourier transform. Figure 22(b) shows the Fourier spectrum in which two peaks at frequencies of 0.35 and 1.45 Hz are easily observable. They correspond to the RR (21 vpm) and HR (87 bpm), respectively. An additional peak at 2.90 Hz reflects the sinusoidal component of the BCG signal with a frequency twice higher than the HR. The magnitude increase toward zero frequency (constant component) indicates deficiencies in signal detrending.
Given that the preprototype sensor was an ad hoc design for initial laboratory evaluation, we subsequently designed improved versions. The optimization work was based on modeling in the SAP2000 environment. Using the finite element method (FEM),179 we analyzed how individual sensor component parameters influenced the quality of the signal obtained. Numerical models were developed to analyze the effects of different dimensions, thicknesses, and cushion wall rigidity as well as the effects of several varieties of adhesives (considering the adhesive layer thickness, Young’s modulus, and thermal expansion coefficient when set) on the sensor properties. In this manner, another cushion-based prototype sensor137138.–139 and its simplified version using an elastic plate as the medium for transferring the strains to the measuring element137,140 were constructed. Additionally, a special stethoscope with a built-in FBG, enabling the monitoring of vibrations caused by the function of the cardiovascular system, e.g., in the cervical artery region, was designed as the source of the reference signal.135136.137.138.–139,141
The experiences discussed above allowed us to design an FBG sensor suitable for MRI environments. The development of the sensor was preceded by FEM simulations to determine the deflection lines that react to heart and lung movements. Theoretical results obtained from FEM simulations and confirmed with laboratory tests have shown a trade-off between highly accurate parameter measurements and the comfort of the monitored person. An elastic plate with larger dimensions provides a better transfer of body vibrations and, thus, produces greater amplitudes due to the extraction/contraction of the FBG measuring element. Nevertheless, the large sensor dimensions require a thicker plate to prevent it from being cracked under the pressure exerted by a larger body surface, which impairs the comfort of the monitored person. On the other hand, a smaller, thinner plate allows the sensor to be placed on a thin, adequately profiled mattress on the MRI scanner track table, providing comfort for the monitored person.
Thus, in the MRI-compatible sensor, a Perspex plate with dimensions of and a thickness of 1.5 mm was used.140,142143.144.–145 A single FBG was attached to the central part of the plate (along the longitudinal axis) by a thin layer of epoxy adhesive that exhibited considerable rigidity when set and had a thermal expansivity coefficient comparable to that of the glass used in the fiber. The sensor structure is shown in Fig. 23. One of the FBG terminals was spliced with a 10-m optical fiber terminated with an ferrule connector/angled physical contact connector. The fiber splice protector was fitted with a ceramic strengthening core instead of the standard (steel) core. The redundant fiber between the FBG and the splice protector was formed into a multiturn loop and attached to the plate.
Before performing experiments on the FBG sensor in its natural working conditions, the frequency response of the measuring system was determined, and the signals within the BCG band, i.e., 1 to 20 Hz,65,180 were recorded without a loss in quality.144 The BCG band was observed well below the sensor’s resonance frequency, which was confirmed when the numerical modeling results were obtained.
The research on the sensor was performed in an Achieva 1.5-T MRI scanner chamber by Philips,181 which was available at the Military Institute of Aviation Medicine, Warsaw, Poland. As depicted in Fig. 24(a), the sensor was placed in the mattress of the MRI scanner track table in a position under the patient’s back near the heart. The sensor output signal was transmitted from the scanner to the interrogator located in the MRI system operation room via a 10-m fiber-optic cable running through a wall opening. The signals were visualized online and archived in a database using a portable computer that was also located in the operation room [Fig. 24(b)] and connected to the interrogation system.
To acquire the reference respiration curve and electrocardiographic (ECG) signal, we used modules of different patient monitoring systems, such as Precess 3160 by Invivo,182 which were designed to operate in the MRI environment. The respiration and ECG signals were obtained using pneumatic bellows and carbon electrodes, respectively, attached to patient. The data were transmitted wirelessly from the modules to the computer using a 2.4-GHz radio link. Figure 25 demonstrates an example of the recorded respiration and heart activity using the FBG-based sensor and the patient monitor, obtained during an MRI scan. The BCG signal was obtained by subjecting the source signal to band filtering in the range of 2 to 60 Hz, which did not eliminate the physiological information carried in the signal. As shown in Fig. 25, the reference signals are clearly better than the signals from the fiber-optic sensor. This results from resolution limitations of the interrogation system as well as the susceptibility to loss of contact between the monitored person and the sensor based on a single FBG. Regarding the first of the mentioned shortcomings, only a few solutions may offer slightly better resolution than the interrogator used, but this usually results in a reduced scanning speed.183,184 High-speed systems that are not commercially available have also been proposed.185186.–187 On the other hand, the sensor reliability can be easily improved through the multiplexing of FBGs, acquiring signals from several different places of the body, addressing all the FBGs with one optical fiber, and handling such an arrangement with one interrogation unit.
The experiments consisted of two phases: (1) a pilot study, in which 5-min intervals of MRI examinations of the spine were recorded where the sensor’s center of symmetry was placed in the region of the vertebrae Th2 and Th3144 and (2) clinical research, in which full imaging of the thoracic spine (vertebrae Th1 to Th12) was recorded for several tens of minutes.145 The pilot study was used to analyze the images obtained with magnetic resonance and demonstrate that the MRI system EM radiation fully penetrated the sensor. Another goal was to confirm that the sensor did not affect the monitored person and to determine whether the sensor worked normally in a magnetostatic field at an induction level of 1.5 T as well as in gradient magnetic fields and pulse-modulated EM fields. Twelve healthy volunteers participated in the pilot research. Tests in a real MRI environment showed that the sensor was safe for the monitored person and did not affect the quality of imaging. In particular, the sensor did not introduce artifacts into the spin echo imaging sequence in T1- and T2-weighted times or into the gradient echo sequences in T2-weighted times. Moreover, the results showed that the optical signal delivered by the sensor was not adversely impacted by the high EM field strength.
The second phase of the experiments determined if it would be possible to utilize the FBG-based sensor for lengthy measurements of the mechano-physiological signals for a patient lying on the sensor and to examine the possible impact of the sensor on the patient’s comfort during a full MRI examination. The clinical research was performed with three patient-volunteers who underwent standard MRI examinations of the thoracic spine. No problems were reported in relation to the operation of the sensor during MRI. None of the examined patients complained of discomfort related to the presence of the sensor under his/her back during a 40-min examination.
The results of these experiments on the sensor showed that the maximum errors calculated with respect to the reference values of the respiration curve and ECG signal did not exceed 8% for RR and 7% for HR. The Bland-Altman method188,189 was applied to statistically assess the accuracy of specifying characteristic points used in determining RR and HR, i.e., the local maxima of the respiration curve (source signal) and the local maxima of the ballistocardiogram (filtered signal), respectively. In the first case, the times of occurrence of the successive characteristic points were compared with the times of occurrence of the successive local maxima of the referential respiration curve. In the second case, the times of occurrence of the successive characteristic points were compared with the times of occurrence of the successive R waves of the ECG signal. Interpretation of the Bland-Altman analysis, shown in Fig. 26, shows that this method exhibits no systematic error. For the entire data set (clinical research, three subjects), the sensor demonstrates good reproducibility. The agreement range limits are as narrow as 0.188 and 0.039 s, and 95.91 and 96.24% of the sample values lie within the limits when specifying characteristic points needed for the RR and HR determination, respectively. Moreover, the error is not proportional and does not depend on how frequently the characteristic points occur.
The RR and HR measurements obtained with this FBG sensor demonstrated that its accuracy was comparable for all examined persons. The detailed results of these experiments in the MRI environment are presented elsewhere.142143.144.–145 In addition to the BCG measurements obtained in the supine position, measurements were also recorded for sitting and standing persons.140 The major properties of the in-house-designed sensors are listed in Table 6.
In-house-designed fiber-optic sensors for respiration and heart function monitoring.
|Category; reference||Sensing element; parameter||Characteristics of subjects; testing procedure; quantitative data on sensor efficiency||MRI compatibility|
|3c.V; 134–137||Single FBG; RR and HR||Seven healthy volunteers, age: 26 to 39 years; during rest, sitting position, recording time: 60 s (each subject); respiration-induced Bragg wavelength shift amplitude: max 30 pmpk-pk (24.8 μstrainpk-pk), heartbeat-induced Bragg wavelength shift amplitude: max 10 pmpk-pk (8.3 μstrainpk-pk), relative error: <14% for HR detection||No (metallic parts)|
|3c.V; 137–139||Two FBGs; RR and HR||Seven healthy volunteers, age: 25 to 37 years; during rest after physical exercises, sitting position, recording time: 10 min (each subject); respiration-induced Bragg wavelength shift amplitude: max 150 pmpk-pk (124 μstrainpk-pk), heartbeat-induced Bragg wavelength shift amplitude: max 10 pmpk-pk (8.3 μstrainpk-pk), relative error: <12% for HR detection||As above|
|3b.V; 137 and 140||Two FBGs; RR and HR||The most recent experimental study includes two phases during rest: (1) standing position, one healthy volunteer (M), age: 36 years; recording time: 120 s; relative error: max 6.55% for HR detection; (2) sitting position, one healthy volunteer (F), age: 26 years; recording time: 120 s; relative error: max 6.24% for HR detection||Yes (not tested)|
|2c.IV; 135–139 and 141||Single FBG, stethoscope structure; HR||Sensor was coupled with speaker membrane emulating heart sounds; heartbeat sound-induced Bragg wavelength shift amplitude: 5 pmpk-pk, frequency range: 25 to 135 Hz, amplification: 9.2 dB at 100 Hz||No (metallic parts)|
|3b.V; 140 and 142–145||Single FBG; RR and HR||Two phases during MRI examination: (1) pilot studies, 12 healthy volunteers (8 M, 4 F), age: 33.25±6.06 years; recording time: 5 min (each subject); relative error: <7% (MRE: 5.24±1.02%) for RR detection, <6.5% (MRE: 5.71±0.61%) for HR detection; (2) clinical studies, 3 healthy volunteers (2 M, 1 F), age: 31.33±5.03 years; total recording time: 95 min 25 s; relative error: max 7.67% for RR detection, max 6.61% for HR detection||Yes (1.5 T, spin-echo T1- and T2-weighted, gradient-echo T2-weighted)|
Summary and Conclusions
This paper provides an overview of fiber-optic sensors for recording physiological signals caused by mechanical movements of the lungs and heart for sensors discussed in the literature and those studied by the author. Moreover, we focus on FBGs, a special group of optical elements, whose unique properties make them ideal for applications in harsh environments, including the EM radiation-saturated MRI environment. In the literature, fiber grating-based sensors are the most frequently reported devices for recording life functions. However, comparatively strong competitors are micro- and macrobend sensors due to their simple design and ease of signal demodulation. Numerous publications have also reported on interferometric sensors, which are unbeatable in terms of precision but require complex signal processing techniques. Other fiber-optic measurement methods have been explored to a much lesser extent, and the small number of publications has provided poor data.
The need to monitor basic vital functions and relate these measurements to the patient’s emotional state during an MRI examination has been noted by many researchers,190 who have further stressed that claustrophobia symptoms should be detected as early as possible.191 New heart rate variability techniques based on the ECG signal make it possible to assess the activity of the autonomic nerve system in an accurate and reliable manner.192 In this area, sensors employing FBGs can play an important role because, if properly configured, they can record RR and HR of the patient through body movements caused by respiratory and heart function.
In contrast to other groups working on optical fiber sensors for monitoring vital functions, 41,518.104.22.168.22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.100.101.102.103.18.104.22.168.22.214.171.124.126.96.36.199.188.8.131.52.184.108.40.206.220.127.116.11.18.104.22.168.132.–133 in our research,140,142143.144.–145 we show that noninvasive, simultaneous acquisition of the respiration curve and BCG signal for a patient undergoing an MRI examination is possible with a single FBG. This method’s originality and usefulness are illustrated by the fact that it is not necessary to use any textiles to ensure immediate proximity of the lungs and heart to the sensor. Therefore, there is no need to perform any additional actions to prepare the patient for monitoring, which may require additional medical staff. During the course of our work, we have shown that the FBG-based sensors are safe for patients in the chamber of a working MRI scanner and that the sensor has no impact on the quality of the imaging. This type of sensor can also be used during CT scans, where there are no similar demanding requirements arising from the presence of a magnetic field, but where respiration and heartbeat should be continuously monitored.193 Recently, this sensor has been applied in research on claustrophobia disorders with a specific focus on hyperventilation for persons undergoing MRI and CT examinations.
The research work presented in this article was part of Project No. UDA-POIG.01.03.01-14-136/08, supported in part by the resources of the EU’s European Regional Development Fund under the Innovative Economy Operational Programme 2007–2013, and this research has been continued within Project No. DOBR/0052/R/ID1/2012/03, financed by the resources of the Program for National Defense and Security, Poland.
N. OliverF. Flores-MangasR. de Oliveira, “Towards wearable physiological monitoring on a mobile phone,” in Mobile Health Solutions for Biomedical Applications, P. OllaJ. Tan, Eds., pp. 208–243, IGI Global, Hershey, New York, NY (2009).Google Scholar
A. Mülleret al., “Remote monitoring in patients with pacemakers and implantable cardioverter-defibrillators: new perspectives for complex therapeutic management,” in Modern Pacemakers—Present and Future, M. K. Das, Ed., pp. 147–166, InTech, Rijeka, Croatia (2011).Google Scholar
B.-G. LeeW.-Y. Chung, “Driver alertness monitoring using fusion of facial features and bio signals,” IEEE Sens. J. 12(7), 2416–2422 (2012).ISJEAZ1530-437Xhttp://dx.doi.org/10.1109/JSEN.2012.2190505Google Scholar
A. Marqueset al., “Light aviation and flight safety: monitoring system for unpressurised cabins. Flight and physiological data acquisition,” 2012, http://www-sre.wu.ac.at/ersa/ersaconfs/ersa12/e120821aFinal00856.pdf (28 December 2014).Google Scholar
Ł. DziudaF. SkibniewskiW. Torbicz, Non-Contact Methods for Monitoring Psychophysiological Activity: Selected Papers, Akademicka Oficyna Wydawnicza EXIT, Warszawa, Poland (2011).Google Scholar
F. G. ShellockS. M. MyersK. J. Kimble, “Monitoring heart rate and oxygen saturation with a fiber-optic pulse oximeter during MR imaging,” AJR Am. J. Roentgenol. 158(3), 663–664 (1992).AJROAM0092-5381http://dx.doi.org/10.2214/ajr.158.3.1739016Google Scholar
Shellock R & D Services, Inc. and F. G. Shellock, “Monitoring patients in the MRI environment,” 2012, http://www.mrisafety.com/SafetyInfov.asp?SafetyInfoID=186 (28 December 2014).Google Scholar
I. Eshedet al., “Claustrophobia and premature termination of magnetic resonance imaging examinations,” J. Magn. Reson. Imaging 26(2), 401–404 (2007).1053-1807http://dx.doi.org/10.1002/(ISSN)1522-2586Google Scholar
C. Qianet al., “A volume birdcage coil with an adjustable sliding tuner ring for neuroimaging in high field vertical magnets: ex and in vivo applications at 21.1 T,” J. Magn. Reson. 221, 110–116 (2012).JOMRA40022-2364http://dx.doi.org/10.1016/j.jmr.2012.05.016Google Scholar
M. Folkeet al., “Critical review of non-invasive respiratory monitoring in medical care,” Med. Biol. Eng. Comput. 41(4), 377–383 (2003).MBECDY0140-0118http://dx.doi.org/10.1007/BF02348078Google Scholar
T. NiendorfL. WinterT. Frauenrath, “Electrocardiogram in an MRI environment: clinical needs, practical considerations, safety implications, technical solutions and future directions,” in Advances in Electrocardiograms—Methods and Analysis, R. M. Millis, Ed., pp. 309–324, InTech, Rijeka, Croatia (2012).Google Scholar
S. JonesW. JaffeR. Alvi, “Burns associated with electrocardiographic monitoring during magnetic resonance imaging,” Burns 22(5), 420–421 (1996).BURND80305-4179http://dx.doi.org/10.1016/0305-4179(95)00160-3Google Scholar
Y. TakemuraJ. SatoM. Nakajima, “A respiratory movement monitoring system using fiber-grating vision sensor for diagnosing sleep apnea syndrome,” Opt. Rev. 12(1), 46–53 (2005).1340-6000http://dx.doi.org/10.1007/s10043-005-0046-6Google Scholar
S. J. Calaet al., “Chest wall and lung volume estimation by optical reflectance motion analysis,” J. Appl. Physiol. 81(6), 2680–2689 (1996).JAPYAA0021-8987Google Scholar
ICT Results, “Optical sensors make MRI scans safer,” Science Daily, 20 September 2008, http://www.sciencedaily.com/releases/2008/09/080918091609.htm (28 December 2014).Google Scholar
L. G. LindbergH. UgnellA. Oberg, “Monitoring of respiratory and heart rates using a fibre-optic sensor,” Med. Biol. Eng. Comput. 30(5), 533–537 (1992).MBECDY0140-0118http://dx.doi.org/10.1007/BF02457833Google Scholar
M. A. ZawawiS. O’KeeffeE. Lewis, “Intensity-modulated fiber optic sensor for health monitoring applications: a comparative review,” Sens. Rev. 33(1), 57–67 (2013).SNRVDY0260-2288http://dx.doi.org/10.1108/02602281311294351Google Scholar
F. Taffoniet al., “Optical fiber-based MR-compatible sensors for medical applications: an overview,” Sensors 13(10), 14105–14120 (2013).SNSRES0746-9462http://dx.doi.org/10.3390/s131014105Google Scholar
P. Rorizet al., “From conventional sensors to fibre optic sensors for strain and force measurements in biomechanics applications: a review,” J. Biomech. 47(6), 1251–1261 (2014).JBMCB50021-9290http://dx.doi.org/10.1016/j.jbiomech.2014.01.054Google Scholar
Magnetic resonance imaging (MRI) equipment—a global strategic business report, Global Industry Analysts, Inc., San Jose, CA (2013).Google Scholar
J. C. MeléndezE. McCrank, “Anxiety-related reactions associated with magnetic resonance imaging examinations,” J. Am. Med. Assoc. 270(6), 745–747 (1993).JAMAAP0098-7484http://dx.doi.org/10.1001/jama.1993.03510060091039Google Scholar
L. M. HarrisJ. RobinsonR. G. Menzies, “Predictors of panic symptoms during magnetic resonance imaging scans,” Int. J. Behav. Med. 8(1), 80–87 (2001).1070-5503http://dx.doi.org/10.1207/S15327558IJBM0801_06Google Scholar
F. T. Tschirchet al., “Low-dose intranasal versus oral midazolam for routine body MRI of claustrophobic patients,” Eur. Radiol. 17(6), 1403–1410 (2007).EURAE31432-1084http://dx.doi.org/10.1007/s00330-006-0457-1Google Scholar
R. LukinsI. G. P. DavanP. D. Drummond, “A cognitive behavioural approach to preventing anxiety during magnetic resonance imaging,” J. Behav. Ther. Exp. Psychiatry 28(2), 97–104 (1997).JBTEABhttp://dx.doi.org/10.1016/S0005-7916(97)00006-2Google Scholar
American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition: DSM-IV-TR, American Psychiatric Publishing, Washington, DC (2000).Google Scholar
S. A. Sarjiet al., “Failed magnetic resonance imaging examinations due to claustrophobia,” Australas. Radiol. 42(4), 293–295 (1998).AURDAWhttp://dx.doi.org/10.1111/j.1440-1673.1998.tb00525.xGoogle Scholar
S. Wright, “All embracing,” Nurs. Stand. 19(50), 32–33 (2005).NSTAEU0029-6570Google Scholar
S. R. Braun, “Respiratory rate and pattern,” in Clinical Methods: The History, Physical, and Laboratory Examinations, H. K. WalkerW. D. HallJ. W. Hurst, Eds., pp. 226–230, Butterworth Publishers, Stoneham, MA (1990).Google Scholar
A. H. Roscoe, “Assessing pilot workload. Why measure heart rate, HRV and respiration?,” Biol. Psychol. 34(2–3), 259–287 (1992).BLPYAX0301-0511http://dx.doi.org/10.1016/0301-0511(92)90018-PGoogle Scholar
P. Váradyet al., “A novel method for the detection of apnea and hypopnea events in respiration signals,” IEEE Trans. Biomed. Eng. 49(9), 936–942 (2002).IEBEAX0018-9294http://dx.doi.org/10.1109/TBME.2002.802009Google Scholar
A. E. Meuretet al., “Voluntary hyperventilation in the treatment of panic disorder—functions of hyperventilation, their implications for breathing training, and recommendations for standardization,” Clin. Psychol. Rev. 25(3), 285–306 (2005).CPSRDZhttp://dx.doi.org/10.1016/j.cpr.2005.01.002Google Scholar
C. L. Badouret al., “Specificity of peritraumatic fear in predicting anxious reactivity to a biological challenge among traumatic event-exposed adolescents,” Cognit. Ther. Res. 36(4), 397–406 (2012).CTHRD80147-5916http://dx.doi.org/10.1007/s10608-011-9380-0Google Scholar
D. Rosenfieldet al., “New breathing therapy reduces panic and anxiety by reversing hyperventilation” (2010) http://www.sciencedaily.com/releases/2010/12/101220200010.htm.Google Scholar
E. WollburgW. T. RothS. Kim, “Effects of breathing training on voluntary hypo- and hyperventilation in patients with panic disorder and episodic anxiety,” Appl. Psychophysiol. Biofeedback 36(2), 81–91 (2011).1090-0586http://dx.doi.org/10.1007/s10484-011-9150-5Google Scholar
A. Pittiget al., “Heart rate and heart rate variability in panic, social anxiety, obsessive-compulsive, and generalized anxiety disorders at baseline and in response to relaxation and hyperventilation,” Int. J. Psychophysiol. 87(1), 19–27 (2013).IJPSEE0167-8760http://dx.doi.org/10.1016/j.ijpsycho.2012.10.012Google Scholar
S. YinP. B. RuffinF. T. S. Yu, Fiber Optic Sensors, 2nd ed., CRC Press, Boca Raton, FL (2008).Google Scholar
M. Vegforset al., “Presentation and evaluation of a new optical sensor for respiratory rate monitoring,” Int. J. Clin. Monit. Comput. 11(3), 151–156 (1994).IJMCEJ0167-9945http://dx.doi.org/10.1007/BF01132363Google Scholar
H. Petterssonet al., “Optical aspects of a fibre-optic sensor for respiratory rate monitoring,” Med. Biol. Eng. Comput. 34(6), 448–452 (1996).MBECDY0140-0118http://dx.doi.org/10.1007/BF02523849Google Scholar
C. Larssonet al., “Respiratory monitoring during MR imaging. The clinical value of a new fibre-optical monitor,” Acta Radiol. 40(1), 33–36 (1999).ACRAAX0284-1851http://dx.doi.org/10.1080/02841859909174399Google Scholar
C. LarssonP. Staun, “Evaluation of a new fibre-optical monitor for respiratory rate monitoring,” J. Clin. Monit. Comput. 15(5), 295–298 (1999).1387-1307http://dx.doi.org/10.1023/A:1009967122272Google Scholar
D. VarshneyaJ. L. Maida, “Fiber-optic monitor using interferometry for detecting vital signs of a patient,” U.S. Patent No. 6498652 (2002).Google Scholar
D. VarshneyaJ. L. MaidaL. A. Jeffers, “Fiber-optic interferometric vital sign monitor for use in magnetic resonance imaging, confined care facilities and in-hospital,” U.S. Patent No. 6816266 (2004).Google Scholar
M. ŻyczkowskiB. Uziębło-Życzkowska, “Human psychophysiology activity monitoring methods using fiber optic sensors,” Proc. SPIE 7838, 78380W (2010).PSISDG0277-786Xhttp://dx.doi.org/10.1117/12.864667Google Scholar
M. Życzkowskiet al., “Interferometric fiber optics based sensor for monitoring of the heart activity,” Acta Phys. Pol. A 120(4), 782–784 (2011).ATPLB60587-4246Google Scholar
M. Życzkowski, “The measurement of respiration and heart work using fiber-optic sensors,” in Non-Contact Methods for Monitoring Psychophysiological Activity: Selected Papers, Ł. DziudaF. SkibniewskiW. Torbicz, Eds., Akademicka Oficyna Wydawnicza EXIT, Warszawa, Poland (2011).Google Scholar
S. ŠpragerD. ĐonlagićD. Zazula, “Monitoring of basic human vital functions using optical interferometer,” in Proc. of 10th IEEE Int. Conf. on Signal Processing, Vol. 2, pp. 1738–1741, Institute of Electrical and Electronics Engineers, New York, NY (2010).Google Scholar
S. ŠpragerD. ĐonlagićD. Zazula, “Estimation of heart rate, respiratory rate and motion by using optical interferometer as body sensor,” in Proc. of the 13th Int. Conf. on Signal and Image Processing, pp. 280–287, ACTA Press , Anaheim, California (2011).Google Scholar
S. ŠpragerD. ĐonlagićD. Zazula, “Heart beat monitoring using optical interferometric signal and pseudo Wigner-Ville distribution,” in Proc. of the 7th Int. Conf. on Information Technology and Applications, pp. 271–275, IEEE New South Wales (NSW) Section, Sydney, Australia (2011).Google Scholar
D. ZazulaD. ĐonlagićS. Šprager, “Unobtrusive monitoring of biomedical signals in home environment,” in Proc. of 19th Telecommunications Forum, pp. 31–34, Institute of Electrical and Electronics Engineers, New York (2011).Google Scholar
D. ZazulaD. ĐonlagićS. Šprager, “Application of fiber-optic interferometry to detection of human vital signs,” J. Laser Health Acad. 2012(1), 27–32 (2012).Google Scholar
S. ŠpragerD. ĐonlagićD. Zazula, “Heartbeat detection applying activity index on optical interferometric signal,” in Proc. of the 11th WSEAS Int. Conf. on Instrumentation, pp. 77–82, World Scientific and Engineering Academy and Society Press, Rovaniemi, Finland (2012).Google Scholar
D. ZazulaD. ĐonlagićS. Šprager, “Fibre-optic interferometry as a means for the first heart sound detection,” in Advances in Applied Information Science, M. DemiralpA. N. Pisarchik, Eds., pp. 31–35, World Scientific and Engineering Academy and Society Press, Greece (2012).Google Scholar
S. ŠpragerD. Zazula, “Optimization of heartbeat detection based on clustering and multimethod approach,” in Proc. of Annual Int. Conf. of the IEEE Engineering in Medicine and Biology Society, pp. 5–8, Institute of Electrical and Electronics Engineers, New York, NY (2012).Google Scholar
S. ŠpragerD. ĐonlagićD. Zazula, “Overnight heartbeat monitoring by using unobtrusive optical interferometric measurements,” in Proc. of IEEE EMBS Conf. on Biomedical Engineering and Sciences, pp. 847–852, Institute of Electrical and Electronics Engineers, New York, NY (2012).Google Scholar
S. ŠpragerD. Zazula, “Heartbeat and respiration detection from optical interferometric signals by using a multimethod approach,” IEEE Trans. Biomed. Eng. 59(10), 2922–2929 (2012).IEBEAX0018-9294http://dx.doi.org/10.1109/TBME.2012.2213302Google Scholar
S. ŠpragerD. Zazula, “Detection of heartbeat and respiration from optical interferometric signal by using wavelet transform,” Comput. Methods Programs Biomed. 111(1), 41–51 (2013).CMPBEK0169-2607http://dx.doi.org/10.1016/j.cmpb.2013.03.003Google Scholar
S. ŠpragerD. Zazula, “Optimization of heartbeat detection in fiber-optic unobtrusive measurements by using maximum a posteriori probability estimation,” IEEE J. Biomed. Health Inform. 18(4), 1161–1168 (2014).2168-2194http://dx.doi.org/10.1109/JBHI.2013.2282403Google Scholar
A. B. NafarrateE. G. Rawson, “Fiber optical monitor for detecting normal breathing and heartbeat motion based on changes in speckle patterns,” U.S. Patent No. 5291013 (1994).Google Scholar
J. Škrabecet al., “Preliminary detection of periodic perturbations using speckle imaging and interframe gradient,” in Recent Advances in Applied & Biomedical Informatics and Computational Engineering in Systems Applications, M. Lazardet, Ed., pp. 251–256, World Scientific and Engineering Academy and Society Press, Florence, Italy (2011).Google Scholar
P. Podbrezniket al., “Cost-efficient speckle interferometry with plastic optical fiber for unobtrusive monitoring of human vital signs,” J. Biomed. Opt. 18(10), 107001 (2013).JBOPFO1083-3668http://dx.doi.org/10.1117/1.JBO.18.10.107001Google Scholar
F. C. FaveroJ. VillatoroV. Pruneri, “Microstructured optical fiber interferometric breathing sensor,” J. Biomed. Opt. 17(3), 037006 (2012).JBOPFO1083-3668http://dx.doi.org/10.1117/1.JBO.17.3.037006Google Scholar
Y. Kanget al., “Nanostructured optical fibre sensors for breathing airflow monitoring,” Meas. Sci. Technol. 17(5), 1207–1210 (2006).MSTCEP0957-0233http://dx.doi.org/10.1088/0957-0233/17/5/S44Google Scholar
Z. ChenJ. T. TeoX. Yang, “In-bed fibre optic breathing and movement sensor for non-intrusive monitoring,” Proc. SPIE 7173, 71730P (2009).PSISDG0277-786Xhttp://dx.doi.org/10.1117/12.807924Google Scholar
D. Lauet al., “Intensity-modulated microbend fiber optic sensor for respiratory monitoring and gating during MRI,” IEEE Trans. Biomed. Eng. 60(9), 2655–2662 (2013).IEBEAX0018-9294http://dx.doi.org/10.1109/TBME.2013.2262150Google Scholar
Z. Chenet al., “Simultaneous measurement of breathing rate and heart rate using a microbend multimode fiber optic sensor,” J. Biomed. Opt. 19(5), 057001 (2014).JBOPFO1083-3668http://dx.doi.org/10.1117/1.JBO.19.5.057001Google Scholar
C. J. Deepuet al., “A smart cushion for real-time heart rate monitoring,” in Proc. of IEEE Conf. on Biomedical Circuits and Systems, pp. 53–56, Institute of Electrical and Electronics Engineers, New York, NY (2012).Google Scholar
Y. Zhuet al., “Ballistocardiography with fiber optic sensor in headrest position: a feasibility study and a new processing algorithm,” in Proc. of the 35th Annual Int. Conf. of the IEEE Engineering in Medicine and Biology Society, pp. 5203–5206, Institute of Electrical and Electronics Engineers, New York, NY (2013).Google Scholar
C. DavisA. MazzoliniD. Murphy, “A new fibre optic sensor for respiratory monitoring,” Australas. Phys. Eng. Sci. Med. 20(4), 214–219 (1997).AUPMDI0158-9938Google Scholar
C. Daviset al., “A new sensor for monitoring chest wall motion during high-frequency oscillatory ventilation,” Med. Eng. Phys. 21(9), 619–623 (1999).MEPHEO1350-4533http://dx.doi.org/10.1016/S1350-4533(99)00094-6Google Scholar
A. Babchenkoet al., “Fiber optic sensor for the measurement of respiratory chest circumference changes,” J. Biomed. Opt. 4(2), 224–229 (1999).JBOPFO1083-3668http://dx.doi.org/10.1117/1.429909Google Scholar
A. T. AugoustiA. RazaM. Graves, “Design and characterization of a fiber-optic respiratory plethysmograph (FORP),” Proc. SPIE 2676, 250–257 (1996).PSISDG0277-786Xhttp://dx.doi.org/10.1117/12.238805Google Scholar
A. T. AugoustiF.-X. MaletrasJ. Mason, “The use of a figure-of-eight coil for fibre optic respiratory plethysmography: geometrical analysis and experimental characterisation,” Opt. Fiber Technol. 11(4), 346–360 (2005).1068-5200http://dx.doi.org/10.1016/j.yofte.2005.02.001Google Scholar
A. T. AugoustiF.-X. MaletrasJ. Mason, “Improved fiber-optic respiratory monitoring using a figure-of-eight coil,” Physiol. Meas. 26(5), 585–590 (2005).PMEAE30967-3334http://dx.doi.org/10.1088/0967-3334/26/5/001Google Scholar
L. T. D’Angeloet al., “A system for respiratory motion detection using optical fibers embedded into textiles,” in Proc. of the 30th Annual Int. Conf. of the IEEE Engineering in Medicine and Biology Society, pp. 3694–3697, Institute of Electrical and Electronics Engineers, New York, NY (2008).Google Scholar
A. Grilletet al., “Optical fibre sensors embedded into medical textiles for monitoring of respiratory movements in MRI environment,” Proc. SPIE 6619, 66191R (2007).PSISDG0277-786Xhttp://dx.doi.org/10.1117/12.738631Google Scholar
A. Grilletet al., “Optical fiber sensors embedded into medical textiles for healthcare monitoring,” IEEE Sens. J. 8(7), 1215–1222 (2008).ISJEAZ1530-437Xhttp://dx.doi.org/10.1109/JSEN.2008.926518Google Scholar
J. Wittet al., “Smart medical textiles with embedded optical fibre sensors for continuous monitoring of respiratory movements during MRI,” Proc. SPIE 7653, 76533B (2010).PSISDG0277-786Xhttp://dx.doi.org/10.1117/12.866434Google Scholar
J. Wittet al., “Medical textiles with embedded fiber-optic sensors for monitoring of respiratory movement,” IEEE Sens. J. 12(1), 246–254 (2012).ISJEAZ1530-437Xhttp://dx.doi.org/10.1109/JSEN.2011.2158416Google Scholar
K. Krebber, “Smart technical textiles based on fiber optic sensors,” in Current Developments in Optical Fiber Technology, S. W. HarunH. Arof, Eds., pp. 319–344, InTech, Rijeka, Croatia (2013).Google Scholar
J. De Jonckheereet al., “Optical fibre sensors embedded into technical textile for a continuous monitoring of patients under magnetic resonance imaging,” in Proc. of the 30th Annual Int. Conf. of the IEEE Engineering in Medicine and Biology Society, pp. 5266–5269, Institute of Electrical and Electronics Engineers, New York, NY (2008).Google Scholar
J. De Jonckheereet al., “OFSETH: smart medical textile for continuous monitoring of respiratory motions under magnetic resonance imaging,” in Proc. of the 31st Annual Int. Conf. of the IEEE Engineering in Medicine and Biology Society, pp. 1473–1476, Institute of Electrical and Electronics Engineers, New York, NY (2009).Google Scholar
F. Narbonneauet al., “OFSETH: optical technologies embedded in smart medical textile for continuous monitoring of respiratory motions under magnetic resonance imaging,” Proc. SPIE 7715, 77151D (2010).PSISDG0277-786Xhttp://dx.doi.org/10.1117/12.854386Google Scholar
J. De Jonckheereet al., “OFSETH: a breathing motions monitoring system for patients under MRI,” in Proc. of the 32nd Annual Int. Conf. of the IEEE Engineering in Medicine and Biology Society, pp. 1016–1019, Institute of Electrical and Electronics Engineers, New York, NY (2010).Google Scholar
J. Wittet al., “Fiber optic heart rate sensor for integration into personal protective equipment,” in Proc. of IEEE Int. Workshop on BioPhotonics, pp. 1–3, Institute of Electrical and Electronics Engineers, New York, NY (2011).Google Scholar
M. NishyamaM. MiyamotoK. Watanabe, “Respiration and body movement analysis during sleep in bed using hetero-core fiber optic pressure sensors without constraint to human activity,” J. Biomed. Opt. 16(1), 017002 (2011).JBOPFO1083-3668http://dx.doi.org/10.1117/1.3528008Google Scholar
W. J. Yooet al., “Development of respiration sensors using plastic optical fiber for respiratory monitoring inside MRI system,” J. Opt. Soc. Korea 14(3), 235–239 (2010).1226-4776http://dx.doi.org/10.3807/JOSK.2010.14.3.235Google Scholar
W. J. Yooet al., “Development of optical fiber-based respiration sensor for noninvasive respiratory monitoring,” Opt. Rev. 18(1), 132–138 (2011).1340-6000http://dx.doi.org/10.1007/s10043-011-0010-6Google Scholar
S. MutoH. SatoT. Hosaka, “Optical humidity sensor using fluorescent plastic fiber and its application to breathing-condition monitor,” Jpn. J. Appl. Phys. 33(Part 1, No. 10), 6060–6064 (1994).JJAPA50021-4922http://dx.doi.org/10.1143/JJAP.33.6060Google Scholar
M. MorisawaS. Muto, “A novel breathing condition sensor using plastic optical fiber,” in Proc. of IEEE Sensors, pp. 1277–1280, Institute of Electrical and Electronics Engineers, New York, NY (2004).Google Scholar
G. Wehrleet al., “A fiber optic Bragg grating strain sensor for monitoring ventilatory movements,” Meas. Sci. Technol. 12(7), 805–809 (2001).MSTCEP0957-0233http://dx.doi.org/10.1088/0957-0233/12/7/309Google Scholar
A. F. Silvaet al., “Simultaneous cardiac and respiratory frequency measurement based on a single fiber Bragg grating sensor,” Meas. Sci. Technol. 22(7), 075801 (2011).MSTCEP0957-0233http://dx.doi.org/10.1088/0957-0233/22/7/075801Google Scholar
C. Zhanget al., “Smart textile sensing system for human respiration monitoring based on fiber Bragg grating,” Proc. SPIE 7381, 738104 (2009).PSISDG0277-786Xhttp://dx.doi.org/10.1117/12.834970Google Scholar
J. Woet al., “Noninvasive respiration movement sensor based on distributed Bragg reflector fiber laser with beat frequency interrogation,” J. Biomed. Opt. 19(1), 017003 (2014).JBOPFO1083-3668http://dx.doi.org/10.1117/1.JBO.19.1.017003Google Scholar
J. De Jonckheereet al., “OFSETH: optical fibre embedded into technical textile for healthcare, an efficient way to monitor patient under magnetic resonance imaging,” in Proc. of the 29th Annual Int. Conf. of the IEEE Engineering in Medicine and Biology Society, pp. 3950–3953, Institute of Electrical and Electronics Engineers, New York, NY (2007).Google Scholar
J. De Jonckheereet al., “FBG-based smart textiles for continuous monitoring of respiratory movements for healthcare applications,” in Proc. of the 12th IEEE Int. Conf. on e-Health Networking Applications and Services, pp. 277–282, Institute of Electrical and Electronics Engineers, New York, NY (2010).Google Scholar
T. Allsopet al., “Respiratory function monitoring using a real-time three-dimensional fiber-optic shaping sensing scheme based upon fiber Bragg gratings,” J. Biomed. Opt. 17(11), 117001 (2012).JBOPFO1083-3668http://dx.doi.org/10.1117/1.JBO.17.11.117001Google Scholar
D. GurkanD. StarodubovX. Yuan, “Monitoring of the heartbeat sounds using an optical fiber Bragg grating sensor,” in Proc. of IEEE Sensors, pp. 306–309, Institute of Electrical and Electronics Engineers, New York, NY (2005).Google Scholar
D. TosiM. OliveroG. Perrone, “Low-cost fiber Bragg grating vibroacoustic sensor for voice and heartbeat detection,” Appl. Opt. 47(28), 5123–5129 (2008).APOPAI0003-6935http://dx.doi.org/10.1364/AO.47.005123Google Scholar
T. Allsopet al., “The application of a long period grating array for respiratory function monitoring,” in Proc. of Conf. on Lasers and Electro-Optics Europe, p. 508, Institute of Electrical and Electronics Engineers, New York (2003).Google Scholar
T. Allsopet al., “Embedded progressive-three-layered fiber long-period gratings for respiratory monitoring,” J. Biomed. Opt. 8(3), 552–558 (2003).JBOPFO1083-3668http://dx.doi.org/10.1117/1.1578497Google Scholar
T. Allsopet al., “The application of a long period grating sensors to human respiratory plethysmography,” Proc. SPIE 6631, 66311G (2007).PSISDG0277-786Xhttp://dx.doi.org/10.1117/12.727821Google Scholar
M. D. Petrovićet al., “Non-invasive respiratory monitoring using long-period fiber grating sensors,” Biomed. Opt. Express 5(4), 1136–1144 (2014).BOEICL2156-7085http://dx.doi.org/10.1364/BOE.5.001136Google Scholar
Ł. Dziudaet al., “The device for measuring vital signs,” Patent No. 217840, Patent Office of the Republic of Poland (2010).Google Scholar
Ł. Dziudaet al., “Fiber-optic sensor for monitoring respiration and cardiac activities,” Polski Przegląd Medycyny i Psychologii Lotniczej 17(3), 267–284 (2011).Google Scholar
L. Dziudaet al., “Fiber-optic sensor for monitoring respiration and cardiac activity,” in Proc. of IEEE Sensors, pp. 413–416, Institute of Electrical and Electronics Engineers, New York, NY (2011).Google Scholar
Ł. DziudaJ. Lewandowski, “Fiber Bragg grating-based sensor for monitoring respiration and cardiac activity,” in Non-Contact Methods for Monitoring Psychophysiological Activity: Selected Papers, Ł. DziudaF. SkibniewskiW. Torbicz, Eds., pp. 7–43, Akademicka Oficyna Wydawnicza EXIT, Warszawa, Poland (2011).Google Scholar
Ł. Dziudaet al., “Fiber Bragg grating-based sensor for monitoring respiration and cardiac activity,” Elektronika: Konstrukcje, Technologie, Zastosowania 52(8), 98–106 (2011).Google Scholar
L. Dziudaet al., “Monitoring respiration and cardiac activity using fiber Bragg grating-based sensor,” IEEE Trans. Biomed. Eng. 59(7), 1934–1942 (2012).IEBEAX0018-9294http://dx.doi.org/10.1109/TBME.2012.2194145Google Scholar
Ł. DziudaF. W. Skibniewski, “A new approach to ballistocardiographic measurements using fiber Bragg grating-based sensors,” Biocybern. Biomed. Eng. 34(2), 101–116 (2014).0208-5216http://dx.doi.org/10.1016/j.bbe.2014.02.001Google Scholar
J. LewandowskiŁ. DziudaK. Celiński-Spodar, “Mechano-acoustic sensor of the cardiovascular system activity,” Elektronika: Konstrukcje, Technologie, Zastosowania 53(1), 77–80 (2012).Google Scholar
Ł. Dziudaet al., “Fiber-optic sensor for breathing and heart rate to be applied in the MRI environment,” Przegląd Elektrotechniczny 88(10b), 120–125 (2012).Google Scholar
L. Dziudaet al., “Fiber-optic sensor for respiration and heart rate monitoring in the MRI environment,” Procedia Eng. 47C, 1291–1294 (2012).PERNBE1877-7058http://dx.doi.org/10.1016/j.proeng.2012.09.391Google Scholar
Ł. Dziudaet al., “Fiber Bragg grating-based sensor for monitoring respiration and heart activity during magnetic resonance imaging examinations,” J. Biomed. Opt. 18(5), 057006 (2013).JBOPFO1083-3668http://dx.doi.org/10.1117/1.JBO.18.5.057006Google Scholar
Ł. DziudaM. KrejF. W. Skibniewski, “Fiber Bragg grating strain sensor incorporated to monitor patient vital signs during MRI,” IEEE Sens. J. 13(12), 4986–4991 (2013).ISJEAZ1530-437Xhttp://dx.doi.org/10.1109/JSEN.2013.2279160Google Scholar
M. Majumderet al., “Fibre Bragg gratings in structural health monitoring—present status and applications,” Sens. Actuators A: Phys. 147(1), 150–164 (2008).SAAPEB0924-4247http://dx.doi.org/10.1016/j.sna.2008.04.008Google Scholar
F. SurreT. SunK. T. Grattan, “Fiber optic strain monitoring for long-term evaluation of a concrete footbridge under extended test conditions,” IEEE Sens. J. 13(3), 1036–1043 (2013).ISJEAZ1530-437Xhttp://dx.doi.org/10.1109/JSEN.2012.2234736Google Scholar
V. KumarB. C. ChoudharyU. K. Tiwari, “FBG temperature sensor for avionics display system,” Adv. Aerosp. Sci. Appl. 3(1), 1–11 (2013).Google Scholar
L. Dziudaet al., “Hybrid fiber optic voltage sensor for remote monitoring of electrical submersible pump motors,” Opt. Eng. 44(6), 064401 (2005).OPEGAR0091-3286http://dx.doi.org/10.1117/1.1926947Google Scholar
L. Dziudaet al., “Laboratory evaluation of the hybrid fiber-optic current sensor,” Sens. Actuators A: Phys. 136(1), 184–190 (2007).SAAPEB0924-4247http://dx.doi.org/10.1016/j.sna.2006.11.003Google Scholar
R. C. S. B. Allilet al., “Application of fiber Bragg grating sensors in power industry,” in Current Trends in Short- and Long-Period Fiber Gratings, C. Cuadrado-Laborde, Ed., pp. 133–166, InTech, Rijeka, Croatia (2013).Google Scholar
H. Xiaet al., “Fiberoptic patient health multi-parameter monitoring devices and system,” U.S. Patent No. 8009946 (2011).Google Scholar
S. C. Hoet al., “FBG sensor for contact level monitoring and prediction of perforation in cardiac ablation,” Sensors 12(1), 1002–1013 (2012).SNSRES0746-9462http://dx.doi.org/10.3390/s120101002Google Scholar
E. Al-FakihN. A. Abu OsmanF. R. Mahamd Adikan, “The use of fiber Bragg grating sensors in biomechanics and rehabilitation applications: the state-of-the-art and ongoing research topics,” Sensors 12(10), 12890–12926 (2012).SNSRES0746-9462http://dx.doi.org/10.3390/s121012890Google Scholar
Y. Yuet al., “Fiber Bragg grating sensor for simultaneous measurement of displacement and temperature,” Opt. Lett. 25(16), 1141–1143 (2000).OPLEDP0146-9592http://dx.doi.org/10.1364/OL.25.001141Google Scholar
S. Abadet al., “Fiber Bragg grating-based self referencing technique for wavelength-multiplexed intensity sensors,” Opt. Lett. 27(4), 222–224 (2002).OPLEDP0146-9592http://dx.doi.org/10.1364/OL.27.000222Google Scholar
P. OrrP. Niewczas, “A robust, multiplexable fiber sensor for simultaneous measurement of bend and strain,” IEEE Sens. J. 11(2), 341–342 (2011).ISJEAZ1530-437Xhttp://dx.doi.org/10.1109/JSEN.2010.2072497Google Scholar
P. Niewczaset al., “Design and evaluation of a preprototype hybrid fiber-optic voltage sensor for a remotely interrogated condition monitoring system,” IEEE Trans. Instrum. Meas. 54(4), 1560–1564 (2005).IEIMAO0018-9456http://dx.doi.org/10.1109/TIM.2004.851072Google Scholar
R. A. Bettset al., “Nonlinear refractive index in erbium doped optical fiber: theory and experiment,” IEEE J. Quantum Electron. 27(4), 908–913 (1991).IEJQA70018-9197http://dx.doi.org/10.1109/3.83325Google Scholar
R. Kashyap, Fiber Bragg Gratings, Academic Press, San Diego, CA (1999).Google Scholar
A. M. Abdiet al., “Modeling, design, fabrication, and testing of a fiber Bragg grating strain sensor array” Appl. Opt. 46(14), 2563–2574 (2007).APOPAI0003-6935http://dx.doi.org/10.1364/AO.46.002563Google Scholar
Ł. Dziuda, “Novel approaches to hybrid voltage and current sensing for condition monitoring of remotely operated electrical plant,” PhD Dissertation, Dept. Elect. Electric. Engineering, Univ. of Strathclyde, Glasgow, UK (2007).Google Scholar
M. G. Xu, “The measurement of physical fields using optical fibers and fiber gratings,” PhD Dissertation, Dept. Electron, Comput. Sci., Univ. Southampton, Hampshire, UK (1995).Google Scholar
B. T. Meggittet al., “A novel digital phase tracking algorithm for a high resolution fibre Bragg grating based sensor system,” J. Phys.: Conf. Ser. 15, 74–82 (2005).1742-6588http://dx.doi.org/10.1088/1742-6596/15/1/013Google Scholar
Q. ChenP. Lu, “Fiber Bragg gratings and their applications as temperature and humidity sensors,” in Atomic, Molecular, and Optical Physics: New Research, L. T. Chen, Ed., pp. 235–260, Nova Science Publishers, Inc., Hauppauge, New York, NY (2008).Google Scholar
A. Rahmanet al., “Anomalous thermal dynamics of Bragg gratings inscribed in germanosilicate optical fiber,” J. Optoelectron. Adv. Mater. 3(1), 17–23 (2009).1454-4164Google Scholar
J. Mandalet al., “Bragg grating-based fiber-optic laser probe for temperature sensing,” IEEE Photon. Technol. Lett. 16(1), 218–220 (2004).IPTLEL1041-1135http://dx.doi.org/10.1109/LPT.2003.820099Google Scholar
K. T. Kimet al., “A temperature-insensitive cladding-etched fiber Bragg grating using a liquid mixture with a negative thermo-optic coefficient,” Sensors 12(6), 7886–7892 (2012).SNSRES0746-9462http://dx.doi.org/10.3390/s120607886Google Scholar
R. González-LandaetaO. CasasR. Pallàs-Areny, “Heart rate detection from an electronic weighing scale,” Physiol. Meas. 29(8), 979–988 (2008).PMEAE30967-3334http://dx.doi.org/10.1088/0967-3334/29/8/009Google Scholar
Y. B. Linet al., “Packaging methods of fiber-Bragg grating sensors in civil structure applications,” IEEE Sens. J. 5(3), 419–424 (2005).ISJEAZ1530-437Xhttp://dx.doi.org/10.1109/JSEN.2005.844539Google Scholar
Y. Kimet al., “Evaluation of unconstrained monitoring technology used in the smart bed for u-Health environment,” Telemed. J. E Health 17(6), 435–441 (2011).1530-5627http://dx.doi.org/10.1089/tmj.2010.0211Google Scholar
M. D. ToddG. A. JohnsonC. C. Chang, “Passive light intensity-independent interferometric method for fibre Bragg grating interrogation,” Electron. Lett. 35(22), 1970–1971 (1999).ELLEAK0013-5194http://dx.doi.org/10.1049/el:19991328Google Scholar
P. OrrP. Niewczas, “High-speed, solid-state, interferometric interrogator and multiplexer for fiber Bragg grating sensors,” J. Lightwave Technol. 29(22), 3387–3392 (2011).JLTEDG0733-8724http://dx.doi.org/10.1109/JLT.2011.2169044Google Scholar
P. Orret al., “Solid state interferometric interrogator and multiplexer for high-speed dynamic and absolute FBG wavelength measurement,” Proc. SPIE 8794, 87493X (2013).PSISDG0277-786Xhttp://dx.doi.org/10.1117/12.2026772Google Scholar
J. M. BlandD. G. Altman, “Statistical methods for assessing agreement between two methods of clinical measurement,” Lancet 327(8476), 307–310 (1986).LANCAO0140-6736http://dx.doi.org/10.1016/S0140-6736(86)90837-8Google Scholar
F. D. McGlynnet al., “Component fears of claustrophobia associated with mock magnetic resonance imaging,” J. Anxiety Disord. 21(3), 367–380 (2007).JADIE80887-6185http://dx.doi.org/10.1016/j.janxdis.2006.06.003Google Scholar
J. Enderset al., “Reduction of claustrophobia with short bore versus open magnetic resonance imaging: a randomized controlled trial,” PLoS One 6(8), e23494 (2011).1932-6203http://dx.doi.org/10.1371/journal.pone.0023494Google Scholar
J. H. Shinet al., “Heart rate variability analysis using a ballistocardiogram during Valsalva manoeuvre and post exercise,” Physiol. Meas. 32(8), 1239–1264 (2011).PMEAE30967-3334http://dx.doi.org/10.1088/0967-3334/32/8/015Google Scholar
Philips Healthcare, “Respiratory motion management for CT,” 2013, http://healthmanagement.org/uploads/Respiratory%20Motion%20Management%20for%20CT_white%20paper.pdf (28 December 2014).Google Scholar
Łukasz Dziuda is an associate professor at the Military Institute of Aviation Medicine in Warsaw, Poland. He received his PhD degree from the University of Strathclyde in Glasgow, United Kingdom in 2007 and his DSc degree from the Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences in Warsaw in 2014. He has published nearly 40 papers on the advancement of optical sensing techniques in the fields of biomedical and power engineering.