Open Access
Add to BibSonomy
Abstract
This work presents a wearable optical fiber sensing scheme based on an eight-figure macro-bend fiber configuration. The wearable sensor scheme utilizes a single-mode fiber deformed on an eight-figure configuration. The fabricated optical fiber sensor is mounted onto a wearable woven fabric and then garment on the elbow and knee joints of different healthy volunteers’ bodies. The proposed wearable biomechanical sensor shows an excellent sensitivity correlated with the human knee and elbow joints’ range of motion 0°-90° which is about −0.963 nm/°, with good regression coefficients (R2) exceeding 99.6%, for elbow joint flexion and sensitivity of 0.874 nm/° with a high R2 exceeding 99.4% for knee flexion. Besides, this sensor also displays high repeatability and stability and a fast response time of 1.66 ms, combined with a small standard deviation of about 2.321%. So, the planned wearable optical fiber sensor system is a practicable option for monitoring joint motion, human movement analysis, and soft robots.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
There is a growing demand in the market for the measurement of rapid changes in human joint movements due to the urgent and critical need for rehabilitation and physical therapy. Thus, observing human biomechanical motion is essential to analyzing and progressing the kinetic exercise technologies in specialized sports, physical rehabilitation, and the performance arts. The performance metrics evaluation relies on instruments for the detection and capture of motion information [1]. The principal drawbacks of currently available movement monitoring techniques are their bulk and rigidity. Particularly, traditional sensors generally need the use of complicated and inconvenient mechanical plug-ins to place sensors on clothes [2]. Also, this apparent gap has encouraged studies into portable and wearable sensing techniques that can monitor human motion, including position and speed. Commercial instruments such as goniometers are mechanical or electromechanical, which are normally implemented with a strain gauge and resistive potentiometers. The major disadvantage of using strain gauge sensors is their inaccuracy, while the potentiometer generates discomfort amongst some users, which can then limit the natural movement of the patient. With the growth of mechatronics, flexible and soft strain sensors have gained increasing attention due to their potential for application in emerging areas of health monitoring, human motion detection, and humanoid robots. To meet the demands for the abovementioned appliances, a reliable strain sensor with excellent sensibility, stability, and that is suitable for large-scale production is necessary [3–8].
Recently, optical fibers have presented a promising solution for movement monitoring sensing in terms of robustness, compactness, efficiency, and signal precision [1]. Additionally, these sensors exhibit certain advantages that have made them popular among researchers, namely their low cost, durability, that they are lightweight, flexible, chemically inert, non-toxic, and biocompatible, as well as their ability to be made portable and wearable [9–13]. In addition to showing high stability, resistance to impact, and high strain limits, which enable the fiber to bend at large angles [2,12]. Moreover, the immunity to electromagnetic interference and lack of electrical contact leads to the optical fibers being highly suited to non-invasive uses for the adjusting of the interrogation optoelectronic item in a harmless site, holding only the sensing section near the patient’s bedside, as well as the sensor, can be straightforwardly placed on the limbs [9,14]. Accordingly, optical fiber can neither harm the physiological environment nor be harmed by it. There are a plethora of interferometric sensors that are designed to measure elbow joint movement and employ various sensing configurations, such as the use of fiber Bragg gratings [15], tapered fiber [16], the Mach–Zehnder interferometer (MZI) [17], or bending optical fiber [18]. Many of these types of configurations tend to have adequate sensitivity and wide detection ranges; however, they also have certain disadvantages such as being difficult to manufacture, the requirements for additional preparation, their high cost of production, and also the need to go through several stages of fabrication. For the particular case of capturing joint angle information, optical fiber sensors based on macro-bending, such as U-shape, mono-loop, sinusoidal, and eight-figure configurations, have been the object of considerable attention, as they can be deemed one of the simplest in terms of fabrication, removing the need for preprocessing, being cost-effective, and usability. They also exhibit high sensitivity, good accuracy, stability, good measurement repeatability, as well as providing precise measurements [919–21].
Generally, elbow injuries are common and account for as much as 15% of all fractures among children, second only to fractures of the lower end of the radius. On the other hand, amid other joint motions corresponding to lower limb movement, the knee joint is deemed to be amongst the most crucial and critical in terms of health evaluation due fact that this area is highly susceptible to injury. Knee injuries and problems mostly occur while practicing sports or recreational hobbies, performing work tasks, or doing housework. In the process of observing a person's movement and posture, information on the movements of the elbow and knee joints is valuable in diagnosis and rehabilitation evaluation. Measuring the range of motion (ROM) of the elbow joint after an injury or surgery in an outpatient clinic is crucial to monitoring a patient's healing progress and determining the possible need for additional treatment. Accurate assessment of ROM is also essential to determine the effectiveness of treatment in clinical research [22,23]. Herein, this work offers the design and characterization of an optical fiber sensor based on an eight-figure macro-bending optical fiber configuration for ROM monitoring. This eight-figure configuration was knit into a fabric designed to capture elbow and knee joint flexion. The designed and fabricated eight-figure optical fiber sensor probe was examined in terms of its function regarding joint angles that change from 0° to 90°, as related to the interest in the motion region range. In testing, the fabricated optical fiber sensor showed a high sensitivity and high linear regression coefficient, as well as a wide range of joint angles, can measure, which reflects its potential in terms of monitoring patients’ healing progress too. Applying this optical fiber sensor model, continuous monitoring of joint movements during daily activities can be achieved, with this sensor being capable of being taken off and put back on at any time without the demand for manual re-calibration. This permits the sensor to self-calibrate, with only simple movements of the patient.
2. Methodology
2.1 Sensor requirements, structure design concept, and working principle of the planned optical fiber Mach-Zehnder interferometer (MZI)
Firstly, sensors for joint angle monitoring of the human must be not disturbed by the motion of the joint. Wearability, low weight, compact size, comfort, elasticity, excellent efficiency, safety, and simple incorporation into rehabilitation systems are vital features that would be taken into consideration while planning sensors for the joint angle determination of the human.
The structure of the planned MZI based on an eight-figure fiber sensor and the light modes coupling within the fiber structure are illustrated in Fig. 1. The optical fiber sensor with eight-figure configuration was built using a proper length of SMF section with about 90 cm, which the middle zone of this SMF section was deformed in the shape of dual-rings via spinning one end of the SMF on the other from both sides to shape an eight-figure structure. The lead-in of the eight-figure configuration was connected to a stable broadband light source (BBS, Thorlabs SLD1550S-A1) working within the 1400 to 1600 nm spectral range, while the lead-out was united to an optical spectrum analyzer (OSA, Yokogawa AQ6370) that employed to record the transmission spectrum of the yield.
Fig. 1. The planned wearable optical fiber sensor system for determining the flexion angle (the inset figure represents the eight-figure optical fiber configuration).
Herein, as the light signal launched to the eight-figure configuration sensing element, the fundamental mode propagates through the linear part of the lead-in side of the planned structure (left pointing blue arrows of the inset figure in Fig. 1). After that, as the light signal arrives at the fiber bend zone, it partially escapes into the SMF cladding owing to the reduction of the radius of the bend curvature, and then immediately stimulates several modes in the cladding through the fiber bend zone. But as the light signal reaches to second bend region of the first loop of the planned fiber structure, some of these cladding modes will gradually return to the core and interfere with the remaining core mode. This process will be repeated in the second ring of the present structure. Due to the core mode and cladding modes own various propagation constants; an interference pattern is resulted from the propagated signal at the lead-out portion of the design sensor structure. The resulting phase difference that produced the mode interference among core mode and cladding modes can be estimated by the relation [13]:
While, the interference expression that can explain the relationship of the interference among the core and cladding modes is given as [20]:
The planned optical fiber sensor consists of simple spectral-modulated sensing in which the propagated light signal is related to effective distance variation that occurs owing to bending influence, that is, by reducing the bending radius. As the flexion/extension joint angle enlarges, the fiber bend radius reduces accordingly. Figure 2 represents the principle of joint angle sensing, in which the elbow angle variation can be evaluated also by monitoring the displacement $\Delta {d_1}$, while, the knee angle variation can be evaluated by monitoring the displacement $\Delta {d_2}$. The points ${A_1}$ and ${B_1}$ symbolize two stationary places on the skin surface close to the elbow joint, and ${d_1}$ is the displacement between the two places. As the elbow joint shifts from the location symbolized by the dashed line to the location symbolized by the solid line, so, does the point ${B_1}$ shifts to the places signified by point ${c_1}$. This activity raises displacement among ${A_1}$ and ${B_1}$ on the skin surface by the magnitude of $\Delta {d_1}$. Accordingly, the elbow joint angle can be scored indirectly by monitoring the distance $\Delta {d_1}$, via [25]:
Fig. 2. sensing principle of joint angle: (a) the relation of displacement $\Delta {d_1}$ with elbow angle variation, and (b) the relation of displacement $\Delta {d_2}$ with knee angle change.
In a similar principle, the knee flexion/extension angle (${\emptyset _2}$) can be monitored, also as illustrated in Fig. 2(b), in which the relation between the displacement and the flexion/extension angle can be expressed as:
Generally, in everyday life, most movements necessitate bending angles of the elbow joint in the range of 30°−130°, and the knee joint possesses a range of 0° and 140° as a human is seated and acting in normal close-range activities [26,27]. In the current work, the biggest detectable angle of both the elbow and knee sensors was fixed to 90°.
2.2 Sensing heads characterization
Firstly, to realize a proper interference pattern for the present design sensor, an appropriate set of radii of curvatures for the bending sensor must be tested experimentally. The straight optical fiber is deformed into the eight-figure shape by crossing one side of it diagonally to the right and then passing under the loop. Follow the process of passing the optical fiber again over the ring towards the left, then pass the thread under the previous intersection. The bending radius of both rings in the planned sensor can be adjusted by pulling both ends of the optical fiber until we reach the appropriate radius for both rings. The transmission spectra of the planned sensor can be studied for various radii of bending curvature. The bending curvature varied from 1.75-0.25 cm in the radius variation step of 0.25 cm.
Figure 3 shows the examined resulting transmission spectra in the range of 1450–1600 nm of an eight-figure configuration with various radii of bending curvature. Through a suitable adjusting of the bending parameter magnitudes, the MZI has been realized, and its double optical arms are both discrete guided paths of both the core and cladding modes that result in the detection of an interference pattern. It can be noticed that as the radius of bending curvature is equal to or more than 1 cm, no clear interference fringes have been detected as a small amount of light signal guiding in the SMF core is merged into the cladding.
Fig. 3. The transmission spectrum of the eight-figure SMF configuration with various radii of bending curvature
As the radius of curvature lessens, a further signal is merged into the cladding, producing an increase of transmission loss and the interference patterns will create, but the perceiving interference fringes do not have an appropriate extinction ratio. Then, as the bending radii were lessened to 0.75 cm, interference fringes were observed at 1551.1 nm and 1592.6 nm with an extinction ratio of more than 22 dB. With more decrease in the radius of bending curvature to 0.5 cm, additional interference fringes have been detected at 1524 nm, 1564.6 nm, and 1594.6 nm with an extinction ratio of about 20 dB. So, the stimulated interference fringes reveal a smaller extinction ratio in contrast with the 0.75 cm bending radius state. If the radius of bending curvature continues to be lessened to 0.25 cm, more interference fringes will be apparent. However, these fringes exhibit a lesser extinction ratio in comparison with the two aforementioned cases. If it continues to reduce the radius of curvature more, the SMF is susceptive to crack, and the optical signal in this state is powerless due to the remaining signal is not only merged with the cladding but escapes from the cladding, initiating a big signal waste. Thus, based on the initial results obtained from this experiment, an appropriate bending radius for eight-figure configuration to produce a good interference pattern was around 0.75 cm.
Finally, to fix the structure form, both free ends of the SMF were settled by an adhesive, which via drawing each end of the SMF, the radius of bending curvature. Both the radius of bending curvature and the length of both bend rings were measured. The first bend ring owns a D1 of 0.75 cm and L1 of 1.75 cm, while the second one has a D2 of 0.7 cm and L2 of 1.8 cm.
Then the performance of the proposed sensor has been tested utilizing a flexible joint that is attuned to a range of angles between 0°−90° and the test has been executed under room temperature. Firstly, the proposed sensor was examined in the test workbench. Afterward, the planned optical fiber sensor was fixed in the horizontal position of the limbs to monitor the flexion/extension angle during mobility of the elbow or knee. To validate the measuring sensor’s stability through the flexion/extension angle, a knit textile was used to fix the textile joint mobility sensor.
3. Result and analysis
The proposed sensor based on the eight-figure optical fiber sensor configuration was first examined on the test bench experiment setup by aligning it in a prototype, as shown in Fig. 4. The trial setup comprises a joint that can adjust manually to modify the angle and simulate human joints’ dynamic flexion and extension.
The response of the sensor has been assessed by investigating it's transmitting spectrum using an OSA. The calculated spectral response provided valuable insights into the sensor performance. Figure 5 illustrates the response curve of the transmission spectrum and the linear fitting graphs of the eight-figure structure. Figure 5(a) illustrates the variations in the fringe location as a function of the flexible joint angle for the sensor under research. The sensor reveals significant sensitivity, with about −0.744 nm/°, escorted by a superior regression coefficient (R2) of 99.1%.
Fig. 5. (a) The transmission spectrum response, and (b) Linear fitting of the shifted dips and flexible joint angle as a function of spectral shift for the case of test workbench.
During the experiment, the movable arm of the test bench has been moved from 0° to 90° and $\Delta \emptyset $ has been evaluated at the span of 10°. This examination has been reiterated 3-times. So, the error bar has been evaluated, as shown in Fig. 6. From this analysis, it can be found that the mean difference, standard deviation, and mean correlation coefficients for these tests are 1.45 nm, 4.902%, and 99.77%, respectively.
The calibration (work with bench) was done under controlled conditions. However, the individual volunteers’ test represents the real-world condition.
The human motion may be more complex and diverse because of joint movement variation, muscle shape and volume, properties of tissue, and individual biomechanical resulting in a large error bar.
For a comprehensive assessment of the functioning of the fabricated eight-figure optical fiber sensor, the dynamic response of the sensor has been measured. This part of the experiment can be performed via recording the transmitted light utilizing a photodiode detector (Gentec TPM300CE) coupled with a 1 GHz oscilloscope (Tektronix MDO3102). The response time, recovery time, rise time, and fall time of the sensor were measured between two rapid change flexion/extension angles. The rapid response time was 1.66 ms, the recovery time: was 4.74 ms, the rise time: was 1.64 ms, and the fall time was 3.58 ms. These measurements were carried out by changing of motion angle from 10° to 90°. The time response curve for the proposed eight-figure flexion/extension sensor is shown in Fig. 7.
Fig. 8. (a) The planned wearable optical fiber sensor system for determining the flexion angle during elbow motion, and (b) the zoom-in of the effective area of the eight-figure optical fiber sensor.
After that, the fabricated optical fiber sensor in the shape of an eight-figure structure was tried for the elbow flexion/extension motions, which was done by three volunteers (aged 20-40, before doing the tests that were section of this work, volunteers signed a written consent form) at a suitable speed and the output signal was recorded in sync with the motion. The circular protractor was employed to evaluate the elbow joint angle, as shown in Fig. 8. The fiber was settled in a fixation design and secured by a tailor applied at precise points. Mechanical stability was achieved by additional sewing layers. Comfort and skin interaction were considered throughout the process. The wearable sensor was tested and validated, with final adjustments to optimize the fixation method. The applied approach may vary based on fabric type, application, and sensor design. Extensive experimentation and testing were conducted to validate the effectiveness and durability of the fixation method. The sensor was separately positioned in joint angles to attach it exactly to the sensitive region within the joint zone. The posterior center of the wrist, the lateral epicondyle of the humerus, together with the acromion were employed to evaluate the flexion/extension angle of the elbow joint. Through the experiment, volunteers moved their elbows from 0° to 90° and the $\Delta {\emptyset _1}$ was evaluated at periods of 10°. This experiment was repeated many times for each volunteer.
The transmission spectrum is shown in Fig. 9(a), which displayed a blue shift of the interference fringes, which varied from 1545.2 nm to 1454.2 nm as the elbow angle varies from 0° to 90°. The linear relationship between the measured shifting fringes and the change in elbow flexion angle is shown in Fig. 9(b), which illustrates a high sensitivity of approximately −0.963 nm/°. This sensor shows a good linear relationship between the measured shifting fringes and the change in elbow flexion angle with R2 exceeding 99.6%, in which the linear fitting is presented for the elbow angle and the spectral wavelength shift ($\Delta {\lambda _1}$) as follows:
On the other side, as the elbow was bent 90°, the $\Delta {d_1}$ was 1.6 cm.
Fig. 9. (a) The transmission spectrum response, and (b) Linear fitting of the shifted dips and flexible joint angle as a function of spectral shift for the case of elbow flexion/extension motion.
The error bar has been also evaluated, as shown in Fig. 10. From this analysis, it can be found that the mean difference, standard deviation, and mean correlation coefficients for these tests are 3.62 nm, 2.095%, and 99.66%, respectively. The slope of the fitting graph for all volunteers ranged between −0.96048 and −0.96867.
Fig. 10. Error bar for the proposed sensor in the case of elbow flexion/extension motion for three different volunteers.
After that, a similar experiment was conducted for the knee joint. The planned wearable optical fiber sensor system for determining the flexion angle during knee motion is shown in Fig. 11. The regarded transmission spectrum of this sensor is shown in Fig. 12, which displayed a redshift of the interference fringes, starting from 1494 nm to 1572.4 nm as the knee angle varies from 0° to 90°. The linear relationship between the measured fringe shifts and the change in knee flexion angle established a high sensitivity of approximately 0.874 nm/°. A linear fitting curve has been performed for the knee flexion angle as a function of the shifted fringes, this sensor shows a good linear relationship between the measured shifting fringes and the change in elbow flexion angle with R2 exceeding 99.4%, which the linear fitting is presented for the elbow angle and the spectral wavelength shift ($\Delta {\lambda _2}$) as follows:
As the knee was bent 90°, the $\Delta {d_2}$ was 27 cm. After increasing the span by 10%.
Fig. 11. (a) The planned wearable optical fiber sensor system for determining the flexion angle during knee motion, and (b) the zoom-in of the effective area of the eight-figure optical fiber sensor.
Fig. 12. (a) The transmission spectrum response, and (b) Linear fitting of the shifted dips and flexible joint angle as a function of spectral shift for the case of knee flexion/extension motion.
The error bar for all volunteers is shown in Fig. 13. From this graph, it can be found that the mean difference, standard deviation, and mean correlation coefficients for these tests are 4.44594 nm, 2.321%, and 99.66%, respectively. The slope of the fitting graph for all volunteers ranged between 0.84339 and 0.87442.
Fig. 13. Error bar for the proposed sensor in the case of knee flexion/extension motion for three different volunteers.
The present technique for evaluating motion angles employing the transmission spectrum disparity is a very practicable alternative as it displays acceptable errors and is more sensitive when compared to other sensors, such as the light intensity variation [12], relative resistance [28], and Fiber Bragg grating types [15,29,30], with the advantage of being easy to fabricate in comparison to other optical fiber sensors. The fiber sensitivity obtained in this work ranged between −0.744 nm/° to −0.963 nm/° which is increased more than 35 times, in correspondence with previously published work [15,30–33]. Moreover, in contrast to other presented traditional sensors, like the inertial measurement unit sensor [34]. The currently planned sensor displays only small errors, confirming the prospect of its being employed for similar goals, besides, it exhibits immunity to electromagnetic interference. On the other hand, in the recently planned optical fiber sensor, a referral technique is not utilized, i.e., the one employed in Ref. [35] was included a fiber and a receptor, causing the cost-effectiveness and compactness of the planned sensor. Further, the calibration of the sensor would not be needed for every use. The sensor demands only a single calibration on the workbench and can subsequently be used without additional calibration, displaying its extensive applicability even beyond controlled circumstances. To the best of our knowledge, this is the first study of human joint movement monitoring by employing an optical sensor based on an eight-figure configuration.
4. Discussion
The demonstration of a flexible wearable optical fiber sensor as a novel platform for precise, comfortable, simple for the user, eco-friendly, and longitudinal optical fiber sensing may offer new prospects for personal aids apparatuses to monitor human movement and physiological signals.
The study relies mainly on measuring the spectral response of the planned eight-figure fiber sensor configuration, using tunning bend angles beginning at 0°. As the fiber is curved, the light waves move at various distances and undergo various phase shifts. This leads some of the light waves to cancel each other out, causing fringes to appear in the spectrum. The transmission spectra of the sensing head under varying degrees of limb flexion/extension angle and for various cases are illustrated in Figs. 9–12.
The main reason for using an eight-figure macro bend fiber in our sensor is to create a suitable interference pattern (dips) and display wavelength shifts clearly with bending variations This structure contributes significantly to enhancing the sensor’s sensitivity.
These results display a significant blue shift of the interference dips for the test workbench and elbow joint angle monitoring and an obvious red shift in the case of knee joint angle monitoring.
The red and blue shifts perceived in the fringe location of the transmission spectrum are due to the variation in the effective refractive index of the fiber, which bending the fiber results in compression on the inside of the core and stretching on the outside of the bend, causing in a change in the refractive index on the inner and outer sides. As the effective refractive index rises, a negative wavelength deviation (blue shift) will produce while in the case of a reduction tendency in the refractive index, it has a positive wavelength deviation (red shift). Besides, when the bending angle enlarges, the compression and stretching of the fiber rise, leading to a larger variation in the effective length of the fiber and a greater wavelength deviation [36]. This is the reason why the detected fringes observed in the OSA exhibit a shift in response to tuning bending angle. Also, this influence is useful for forming sensors capable of monitoring tunes in bending or strain.
On the other side, when the bending angle of the optical fiber varies from 0° to 90°, a wave intensity loss is dominant owing to the raised attenuation. This is because as the fiber is curved, light signals can leak out the fiber’s core and be transmitted to the cladding, causing a reduction in overall light wave intensity. Moreover, excessive bending can also cause macro-bends which can increase attenuation. For all these measurement cases, it is clear that the deviation in the spectral fringes exhibits a good linear correlation with the tuning in bend angles.
From the resulting data, it can be perceived how the transmission signal from the sensor varies as the arm or leg main in a bent or straightened state. An overshoot can be observed as an abrupt action change. When the strain is suddenly swapped with each movement of the stretching phase, the planned optical fiber sensor quickly liberates the stress these experiences through mechanical distortions [28,37]. In the flexion/extension activity, some minuscule muscles shake and contract. Accordingly, some minor signal fluctuations can be detected, but these can be straightforwardly eliminated via a low-pass filter.
The difference in the transmission spectra was obvious and corresponded to both elbow and knee joint movements. Owing to the various ranges of mobility of the different human joints, the transmission spectral signal of the optical fiber sensor also varied and accordingly exhibited a sensitivity alteration. Additionally, the differences in muscle movements of the human joint, besides, the flexion/extension activity of each part rotation are different [28].
The macro bend structure provides real benefits in compactness as well the use of thin optical fiber is necessary for reducing the modal noise which is highly present in fiber bending and consequently increases the sensitivity.
5. Conclusion
In this work, the design of a wearable human joint movement optical fiber sensor has been displayed. A single-mode fiber is deformed into an eight-figure configuration and then incorporated into a woven fabric to house the textile joint motion sensor. This method fits the human joint, as well as confirms the stability of the sensor measurement during a change in flexion angle. The fabricated optical fiber sensor showed an excellent sensitivity in the range between −0.744 nm/° to −0.963 nm/°, which means the fabricated sensor significantly improves the optical sensing signal to more than 35 times in correspondence with many previous published works. After preliminary examination including a woven fabric sensing clothing for lower and upper human body monitoring, it has been observed that this practice is applicable for joint motion monitoring of both the elbow and knee. The accuracy and stability of the planned optical fiber sensor in monitoring joint angle were determined several times by different volunteers, the resultant data of which prove that the planned sensor is reliable and precise. Moreover, the sensor showed high stability and reliability across numerous trials, confirming excellent repeatability.
Funding
Ministry of Higher Education and Scientific Research; University of Baghdad.
Acknowledgments
This work was supported by the Ministry of Higher Education and Scientific Research (MOHESR), University of Baghdad (UoB). This work was approved by the institutional review board of the Institute of Laser for Postgraduate Studies, University of Baghdad.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. Y. D’Mello, J. Skoric, L. Moukarzel, S. Hakim, and D. V Plant, “Wearable fiber optic sensors for biomechanical sensing via joint angle detection,” in 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2019, pp. 32221–33225.
2. T.-W. Shyr, J.-W. Shie, C.-H. Jiang, and J.-J. Li, “A textile-based wearable sensing device designed for monitoring the flexion angle of elbow and knee movements,” Sensors 14(3), 4050–4059 (2014). [CrossRef]
3. T. Li, “Sodium alginate reinforced polyacrylamide/xanthan gum double network ionic hydrogels for stress sensing and self-powered wearable device applications,” Carbohydr. Polym. 309, 120678 (2023). [CrossRef]
4. D. Kong, “Highly sensitive strain sensors with wide operation range from strong MXene-composited polyvinyl alcohol/sodium carboxymethylcellulose double network hydrogel,” Adv. Compos. Hybrid Mater. 5(3), 1976–1987 (2022). [CrossRef]
5. Y. Wu, “Melamine sponge skeleton loaded organic conductors for mechanical sensors with high sensitivity and high resolution,” Adv. Compos. Hybrid Mater. 6(1), 4 (2023). [CrossRef]
6. X. Chang, L. Chen, J. Chen, Y. Zhu, and Z. Guo, “Advances in transparent and stretchable strain sensors,” Adv. Compos. Hybrid Mater. 4(3), 435–450 (2021). [CrossRef]
7. M. Hu, “High-performance strain sensors based on bilayer carbon black/PDMS hybrids,” Adv. Compos. Hybrid Mater. 4(3), 514–520 (2021). [CrossRef]
8. N. Jiang, “Ionic liquid enabled flexible transparent polydimethylsiloxane sensors for both strain and temperature sensing,” Adv. Compos. Hybrid Mater. 4(3), 574–583 (2021). [CrossRef]
9. D. I. Al-Janabi, A. M. Salman, and A. Al-Janabi, “All fiber, highly sensitive sensor based on gold nanoparticle-coated macrobent single mode fiber for human temperature monitoring,” J. Nanophotonics 14(04), 1–14 (2020). [CrossRef]
10. Huda Alsweefe, Sarah Kadhim Al-Hayali, and Abdulhadi Al-Janabi, “Enhanced relative humidity sensor via diameter of no-core fiber structure,” Iraqi J. Laser 18(1), 19–23 (2019).
11. S.A. Mohammed and A.H. Al-Janabi, “All Fiber Chemical Liquids Refractive Index Sensor Based on Multimode Interference,” Iraqi J. Laser 17(A), 33–40 (2019). [CrossRef]
12. A. Rezende, C. Alves, I. Marques, M. A. Silva, and E. Naves, “Polymer optical fiber goniometer: A new portable, low cost and reliable sensor for joint analysis,” Sensors 18(12), 4293 (2018). [CrossRef]
13. V. H. R. Cardoso, P. Caldas, M. T. R. Giraldi, O. Frazão, J. C. W. A. Costa, and J. L. Santos, “Optical Strain Gauge Prototype Based on a High Sensitivity Balloon-like Interferometer and Additive Manufacturing,” Sensors 22(19), 7652 (2022). [CrossRef]
14. T. Li, “A Skin-Like and Highly Stretchable Optical Fiber Sensor with the Hybrid Coding of Wavelength – Light Intensity,” 2022. [CrossRef]
15. Z. A. Abro, C. Hong, N. Chen, Y. Zhang, R. A. Lakho, and S. Yasin, “A fiber Bragg grating-based smart wearable belt for monitoring knee joint postures,” Text. Res. J. 90(3-4), 386–394 (2020). [CrossRef]
16. Q.-Q. Ge, M.-Y. Chen, and T.-Y. Gong, “Investigation of a tapered polymer optical fiber sensor for bending angle measurement,” Proceedings of the SPIE 12169, 444 (2022). [CrossRef]
17. L. Ding, L. Yu, G. Hu, Q. Xie, and L. Zhu, “Knee joint curvature detection system based on fiber optic Mach-Zendler interferometric curvature sensor,” IEEE Sens. J. 21(24), 28017–28024 (2021). [CrossRef]
18. D. Z. Stupar, J. S. Bajic, L. M. Manojlovic, M. P. Slankamenac, A. V. Joza, and M. B. Zivanov, “Wearable low-cost system for human joint movements monitoring based on fiber-optic curvature sensor,” IEEE Sens. J. 12(12), 3424–3431 (2012). [CrossRef]
19. A. S. Silva, A. Catarino, M. V. Correia, and O. Frazão, “Design and characterization of a wearable macrobending fiber optic sensor for human joint angle determination,” Opt. Eng. 52(12), 126106 (2013). [CrossRef]
20. A. M. Salman, S. K. Al-Hayali, and A. H. Al-Janabi, “Wide-range and highly sensitive pH sensor based on a figure-eight fiber structure coated with copper/polyvinyl alcohol hydrogel,” Opt. Mater. Express 12(9), 3763–3775 (2022). [CrossRef]
21. D. I. A.-J. and A. S. and A. H. Al-Janabi, “High sensitivity balloon-like thermometric sensor based on bent single mode fiber,” Meas. Sci. Technol., 2020, [Online]. Available: http://iopscience.iop.org/10.1088/1361-6501/ab9458
22. D. P. H. Koong, J. Lee, T. L. Cheng, and D. G. Little, “Validity and reliability of smartphone inclinometer applications for measurement of elbow range of motion in paediatric patients,” J. Child. Orthop. 14(5), 488–494 (2020). [CrossRef]
23. G. M. Salim and M. A. Zawawi, “Knee joint movement monitoring device based on optical fiber bending sensor,” J. Telecommun. Electron. Comput. Eng. 10(1–3), 25–29 (2018).
24. K. Tian, “Simultaneous measurement of displacement and temperature based on two cascaded balloon-like bent fibre structures,” Opt. Fiber Technol. 58, 102277 (2020). [CrossRef]
25. Y. Koyama, M. Nishiyama, and K. Watanabe, “A motion monitor using hetero-core optical fiber sensors sewed in sportswear to trace trunk motion,” IEEE Trans. Instrum. Meas. 62(4), 828–836 (2013). [CrossRef]
26. M. Li, B. He, X. Zhao, J. Xie, W. Yao, and G. Xu, “A wearable fiber-optic sensor for monitoring human elbow and wrist joint motion,” Adv. Robot. 35(7), 400–412 (2021). [CrossRef]
27. L. C. Pereira, S. Rwkabayiza, E. Lécureux, and B. M. Jolles, “The knee smartphone-application goniometer is a more reliable tool than the standard goniometer in acute orthopaedic settings,” Physiotherapy 101, e1192–e1193 (2015). [CrossRef]
28. J. Zhang, “Human motion monitoring in sports using wearable graphene-coated fiber sensors,” Sens. Actuators, A 274, 132–140 (2018). [CrossRef]
29. S. Umesh, S. Padma, T. Srinivas, and S. Asokan, “Fiber Bragg grating goniometer for joint angle measurement,” IEEE Sens. J. 18(1), 216–222 (2018). [CrossRef]
30. T. Apiwattanadej, B. J. Chun, H. Lee, K. H. H. Li, and Y.-J. Kim, “Stability test of the silicon fiber Bragg grating embroidered on textile for joint angle measurement,” in Fifth International Conference on Optical and Photonics Engineering, 2017, vol. 10449, pp. 47–52.
31. L. Li, “Embedded FBG-based sensor for joint movement monitoring,” IEEE Sens. J. 21(23), 26793–26798 (2021). [CrossRef]
32. C. K. Jha, A. L. Chakraborty, and S. Agarwal, “A fiber Bragg grating-based sensing glove with a sensitivity of 18.45 pm/degree to accurately assess finger flexure,” in International Conference on Optical Fiber Sensors, 2018, p. WB4. [CrossRef]
33. A. B. Socorro-Leranoz, “Optical system based on multiplexed fbgs to monitor hand movements,” IEEE Sens. J. 21(13), 14081–14089 (2021). [CrossRef]
34. L. S. Vargas-Valencia, A. Elias, E. Rocon, T. Bastos-Filho, and A. Frizera, “An IMU-to-body alignment method applied to human gait analysis,” Sensors 16(12), 2090 (2016). [CrossRef]
35. L. Bilro, J. G. Oliveira, J. L. Pinto, and R. N. Nogueira, “A reliable low-cost wireless and wearable gait monitoring system based on a plastic optical fibre sensor,” Meas. Sci. Technol. 22(4), 045801 (2011). [CrossRef]
36. A. M. Salman, S. K. Al-Hayali, and A. Al-Janabi, “Sensitivity-enhanced moisture sensor based on θ-shape bending fiber coated with copper-polyvinyl alcohol thin film,” IEEE Sens. J. 21(19), 21546–21554 (2021). [CrossRef]
37. A. D. Smith, “Electromechanical piezoresistive sensing in suspended graphene membranes,” Nano Lett. 13(7), 3237–3242 (2013). [CrossRef]
