Priyabrata Sethy, Santanu Kumar Behera
©SHUTTERSTOCK.COM/MICROGEN
RFID technology has boomed in the area of biomedical applications, particularly in wearable sensors and bioimplants. The placement of the tag externally or internally generates numerous challenges, such as the design procedure, fabrication method, and real-time testing, along with unique challenges specific to diagnosis and other on-body applications. This article presents a review of bioimplants for several on-body applications associated with RFID technologies and their impact on human tissue. Several fabrication methods are also discussed, with a focus on understanding flexible, conformal, and strong RFID devices. In light of the future of upcoming applications, this article also reviews transformative RFID-based solutions for wireless body area network and implantable uses. In addition, the review highlights the design and implant of prototype devices. In sum, the main focus of this article is all aspects of chipless RFID utilized in multiple places in the human body, beginning with limb sensing and tracking, which basically involves a prosthetic structural tag (PST) responsible for providing information about a prosthetic limb in space. Similarly, the ultrahigh-frequency (UHF) passive RFID implant gives information about the tag, which is placed in the limb, and the communication between the readers takes place without any physical contact. An alternative touch probe method that resonates at about 13.56 MHz improves the accuracy of the implant. Stent tags are discussed for cardiovascular systems. Other specific topics covered are body implants based on an organically modified ceramic technology (ORMOCER) substrate, myoelectric prostheses, fractal-based RFID implants for crack sensing, and the thermal effects of implants.
Wearable sensors and biological implants hold the potential to increase the accuracy of clinical diagnosis, which in turn, is increasing the demand for accurate on-body sensor applications along with the radio telemetry necessary for data recording. At a minimum, an in-body sensor must be paired with a battery source and an antenna. In this regard, chipless RFID is the most dynamic modality for sensor and in-body implant applications. At present, RFID plays a significant role in the surveillance and sensing of multiple parameters, with minimal power. The details of RFID consist of an interrogator and transponder supported by necessary middleware that plays an important role because it operates on the reader, which stores and encodes the data. The interrogator is the vital part of the carrier, where electromagnetic (EM) energy feeds to the passive tag. It is a kind of duplexer unit that transmits and receives the radio signal from the tag [1]. The reader is responsible for analyzing both incoming and outgoing signals. The salient feature of RFID is its operation without an internal battery, using a so-called passive tag. The reader antenna captures a physical parameter from the environment through various tags and stores the data for surveillance purposes. RFID is designed to operate using a frequency of more than a 1-GHz carrier signal. The availability of sensors and interfacing devices is a large part of today’s “smart” lifestyle. Figure 1 suggests how RFID sensors are utilized in multiple scenarios that require data to be acquired and kept for processing. RFID reader and tag sensors from various sources are connected. The several sources that are shown in Figure 1 are some of the primary objectives of measurements in which environmental data, causes of deforestation, wildlife biology, personal cloud data and industrial automation data, and the packaging of goods are observed. The implants installed on the patient in the figure are also sources of monitoring. The white and yellow boxes in Figure 1 represent chipless RFID sensors and implants, respectively, which are used for monitoring and observation in the desired application.
Figure 1. The sensing parameters of various tags and implants by an RFID reader.
Real-time data from multiple sources are continuously accessed by an RFID reader, and an active interaction is established [2]. In this way, the interface takes place among several devices with the help of RFID sensors, due to the advances in wearable sensors and biological implants [2]. A relatively new application of RFID technology is harvesting additional energy [4]. The overall motivation of this analysis [3] is to introduce the working principle of an automatic sensing device and its transformation to an implantable device. An important focus of this article is the requirements for the tag’s input impedance and radar cross section (RCS), which depend on the geometry of the tag as well as the nature of the substrate. RFID is also utilized for an ultrasonic mode of measurement popularly known as millimeter-scale implants, an application that is booming for interfacing with several human nerves in near proximity. RFID implants can also use piezoelectric material [6]. The purpose of using piezoelectric material is to produce an acoustically enabled antenna that conserves energy from the generated ultrasonic waves and uses that energy to send the data to a telemetry unit. Similarly, many types of implants are under study based on the characteristics and area [6] of implementation. Current RFID research and design objectives are driven by specific implants and desired results, with the need to make the implant interface smoother and more accurate.
Another major growth area for RFID-based implants is to generate an e-skin module that resembles organic skin. E-skin is a tensile ultrathin stretchable electronic technology. It is capable of self-healing properties normally integrated over the human skin. This e-skin does a good job of mimicking the sensing ability of skin tissue for all environmental factors. The methodology behind artificial skin is an array of sensors that are spread all along the module surface, which is embedded on the exterior of a prosthetic [4].
This article is organized beginning with chipless RFID systems, followed by chipless RFID wireless sensor classification. This section gives a basic idea of chipless RFID. Current design challenges are introduced next, first for RFID implants and followed by chipless RFID sensor-based implants. Then, various implants are introduced, starting with limb sensing implants, followed by UHF passive RFID implants, orthopedic knee implants, stent implants, vessel implants, body-implantable RFID tags based on ORMOCER printed circuit board (PCB) technology, a fractal RFID-based sensing tattoo for the early detection of cracks in implanted metal prostheses, the feasibility of RFID-based transcutaneous wireless communication for the control of upper-limb myoelectric prostheses, and the thermal effects of bioimplants.
An RFID system consists of a reader and a tag, as illustrated in Figure 2. The reader consists of the middleware and a hardware unit with an antenna. The reader sends an interrogation signal that is backscattered by the tag and again received by the reader by means of a frequency signature. In this case, the tag receives power from the RF. These are called passive tags. Active tags are those in which internal batteries are present, which increases the performance of the RFID system [1], [2]. Active tags include an integrated circuit (IC) chip, which is necessary for signal processing. An active tag has a battery unit along with a memory bank. In the setup for the active transponder, the signal is sent by the reader, and the signal is backscattered by the transponder with the useful information. These tags have the ability to amplify and retransmit the low-power backscattered original received signal. These tags are commonly used in secure communication and payment gateways.
Figure 2. The working principle of an RFID system [1], [2].
Initially, bar codes were used for all commercial applications. Later, chipless RFID came into the picture, with all the mobility enabled by its nonline-of-sight interpretation, beyond-visual-range tracking capability, massive data volume, and operational compatibility.
Passive wireless sensors are sensors in which the connection takes place through a wireless radio transceiver rather than any physical connection. These sensors operate on the reader’s interrogated RF signal [53]. Wireless sensors are normally categorized into two broad classes—chipped and chipless—based on their working principle and power handling abilities. A chipped RFID passive sensor is based on the principle of the modulation of the reflected RCS and works by changing impedance parameters to identify the variation in a physical situation [12]. Chipped sensors are categorized into electronic and EM. Electronic is further divided into battery, active, and passive sensors. Similarly, EM sensors are divided into passive, harmonic, and self-tuned sensors. Chipped sensors demonstrate optimum independent sustainability.
The range of chipless RFID can be estimated on the basis of transducer performance, and the performance can be affected by many environmental factors and the maximum interfacing range. In the category of surface acoustic wave (SAW)-based sensors, the mechanism tracks the movements of an acoustic wave over a piezoelectric material. The velocity of the acoustic wave can be modified by changing the propagation delay. The SAW-based sensor is a commercial sensor based on chipless RFID technology [3]. The main disadvantages of this sensor are its nonplanar nature, fabrication complexity, and short interfacing distance of less than 10 m [6]. This is because of conversion loss when moving from an acoustic to an EM wave. On the other hand, EM transduction-based chipless RFID is able to interrogate over a longer distance (30 m), due to its lack of EM conversion loss. This specific sensor is convenient for high-frequency applications [6].
Chipless sensors present various advantages, such as interference resistance, better security, high sensitivity, and a lower price. Another important advantage of this class of sensor is its light weight. It can also offer a stretchable and long-lasting sensor model with optimum accuracy. A chipless RFID passive sensor classification is displayed in Figure 3. The EM transduction sensor group includes sensors for temperature, pressure, humidity, gas detection, strain cracking, and biosensing [7]. The sensors’ scalability enhances their usefulness in detecting structural deformation. In precision agriculture, for example, having accurate information regarding the spatial and temporal changes of soil and crop characteristics within a field is crucial [7]. EM transduction enables a chipless sensor to provide a high read range of up to 30 m, combined with the advantages of chipless sensors listed in the preceding. It provides a high read range distance of up to 30 m. Figure 4 presents various sensing materials and some of their desired applications. These materials are responsible for smart sensing and are used for many Internet of Things devices.
Figure 3. The classification of wireless passive sensors [12].
Figure 4. The classification of smart sensing materials for various sensors [7]. PEDOT: poly(3,4-ethylene dioxythiophene).
The differences between chipped and chipless RFID are listed in Table 1. The major characteristics are physical properties, frequency ranges, operational bandwidth, equivalent isotropic radiated power, average reading distance, noise and interference, handshaking protocol, working principle, and applications. From the table, the advantages of chipless sensors over chipped sensors can be analyzed based on these characteristics [1].
Table 1. The differences between chipped and chipless RFID sensors [1].
The design of RFID implants encounters numerous problems, the resolutions to which need to be discussed. Static bioimplants do not work well due to the need to surpass the elastic limit of the structure, which results in anomalous readings when the implants are utilized for diagnostics. One research intention is to determine the implant deformation at the time of rehabilitation through several patient observations. Further performance improvement may come from matching activities with the implant’s capability. The research in [6] shows that the graphical prototype limit has a bigger impact on sensor characteristics; also, the mechanical property is likely to be unstable. This phenomenon occurs when there is positive strain at the rear side of the cavity, which moves the sensor close to the neutral axis. In addition, even a small change in the neutral axis, due to manufacturing defects, changes the sensor response significantly.
The chipless RFID sensor has advantages that include passive maintenance-free operation and fully printable manufacturability. These important advantages allow the chipless sensor to achieve applications that may not be possible using conventional RFID sensors. However, power utilization is still a major concern, especially for implantable prostheses. Table 2 is a power comparison of several wireless implants.
Table 2. The wireless power technology of some implants.
The research work highlighted in this section focuses on a passive transponder, often termed a tag. It is linked with a UHF RFID protocol. The reader interacts with a small radar antenna placed on the body. The reader transmits power to the tag and receives the signal from the tag, utilizing a unique frequency signature. The design of this type of transponder for prosthetics is widely used in naval and aerospace communication. Prosthetics are artificial devices that act as replacements for lost human organs, such as digits, limbs, eyes, and many more. They can be electronically monitored by RFID, which can also be used to record various biological parameters. The transponder in [8], for use in a limb prosthesis, is similar to a funnel-shaped structure used for both the transmitter and receiver, which is normally placed on big ships and aircraft. In this work, the RFID tag structure for the prosthesis of the bone of a limb is called an intramedullary nail design. It is a titanium alloy rod specially designed for the repair of fractures of the femur and tibia, and the prosthetic is fitted to the bone by utilizing an external hole [8]. The prosthetic tag uses an antenna structure that is used in military and defense high-frequency (2–30-MHz) applications. In these applications, low-frequency radiators are used, closely adapting the radiation machinery into existing metal superstructures [12], such as the fuselage of an airplane, a funnel, and a big mast of a ship. The radiators are active parts of the radiating system. However, the design of a combination of radio sensing abilities inside a fixed medical device entails far more critical constraints. Any change to the original prosthesis cannot compromise its biocompatibility, functionality, and mechanical stability. Moreover, surface discontinuities have to be avoided since they could obstruct the correct insertion of the prosthesis into the bone. The inclusion of radiating elements is therefore required to be low profile.
The cylindrical rod prosthesis is further modified for excitation in the form of a rectangular strip having size ${W}\times{L}{.}$ It also possesses an inside notch with the dimension ${L}\times{W}\times{P}$ at a distance h from the external surface. The notch is composed of an insulating material coating to provide isolation from biological tissue, as demonstrated in Figure 5. In [14], the potentiality of a square-loop RFID tag is described for insertion in the close surrounding of a bone. In particular, the loop layout, with its intrinsic inductive reactance, was revealed to be easily compatible with the capacitive impedance of the RFID microchip even when the tag was placed in proximity to metallic objects inside a limb phantom. The square loop is now made structural with the nail by following the rationale in Figure 6. Half the loop can be replaced by the nail itself, as in towel bar antennas over aircraft [21]. To avoid a bulging structure, it is a curved structure close to the body of the cylinder. This keeps the physical endurance and steadiness of the prosthesis from being compromised. The subsequent device is hereafter denoted as a PST. It resembles a T-match formation for EM radiation [17].
Figure 5. Details of a PST: the (a) interior of the notch, which hosts the two strips where the microchip will be connected, (b) side section, and (c) external view.
Figure 6. A PST fitted into a generic biological module.
A prototype of the basic PST was fabricated using a cylindrical steel rod of total length ${l} = {290}\,{\text{mm}}$ and diameter ${D} = {12}\,{\text{mm}},$ as shown in Figure 7. The notch was centrally carved in the rod and had dimensions ${W} = {8}\,{\text{mm}},$ ${L} = {65}\,{\text{mm}},$ and ${p} = {4}\,{\text{mm}},$ which accorded to the previously simulated model. The interconnection of the microchip to the rod was achieved by means of two metal copper strips (width: 4 mm). The inner and outer dielectric coatings consist of two 2-mm-thick sheets of Teflon having a 2-mm thickness. This prototype can be considered representative of both a tibial and a femoral nail. The concept of radiation modality comes into the picture when the input parameters and radiation efficiency with respect to the frequency of the PST are analyzed. The marginal variation of the PST’s electrical parameter matched to a frequency band is provided in Figure 8. The artificial steel rod (PST) appears in Figure 7. The steel prototype of the PST consists of copper strips, Teflon, and an IC strap [8].
Figure 7. The steel prototype of a PST [8].
Figure 8. The (a) input impedance of the PST versus the frequency and (b) power transfer coefficient [8].
The rod has dimensions ${W} = {8}\,{\text{mm}},$ ${L} = {65}\,{\text{mm}},$ and ${p} = {4}\,{\text{mm}},$ which accords to the previously simulated model. The interconnection of the microchip to the rod was achieved by means of two metal copper strips (width: 4 mm). The cross section of the phantom limb is depicted in Figure 9.
Figure 9. The detailed geometry of the phantoms of the limb [8].
The limb phantom was assembled from a glass cylindrical pipe, with a diameter of 140 mm, filled with minced meat (moisture of muscle with 35% fat) up to a height of 350 mm, and it is displayed in Figure 10. The PST was implanted inside the medullary canal of a cow bone (with an average diameter of 52 mm and bone thickness of 20 mm) placed at two different distances from the lateral surface of the cylinder [8].
Figure 10. (a) A PST placed inside the medullary canal of a cow bone. (b) The measurement setup, including the limb phantom and the bone [8]. PIFA: planar inverted-F antenna.
A first configuration, with ${h}_{a} = {70}\,{\text{mm}},$ emulates a deep (femoral) implant, as shown in Figure 9(a). In a second configuration, the bone was placed at a closer distance from the pipe surface ${(}{h}_{b} = {25}\,{\text{mm}}{)}$ to simulate a more superficial (tibial) implant, as in Figure 10. The turn-on power is the minimum power emitted by the reader that is needed to energize the tag in the specific configuration. The upper bound of ${P}_{{t}{o}}$ is the maximum power that the reader may provide to the reader antenna. The direct link margin (in decibels) can be expressed as \[{M}_{\text{DL}} = {PR}\rightarrow\left|{{db}{-}{p}_{\text{chip}}}\right|{db} \tag{1} \] \[{P}_{R}\rightarrow{T} = {G}_{T}{P}_{\text{av}}{.} \tag{2} \]
The power is transmitted by the reader to the IC when the reader emits the maximum available power ${P}_{{av}}$ (here, 30 dB with the available hardware), and ${G}_{{T}}$ is the transducer power gain [19]. If ${M}_{DL}{>}{0},$ it ensures that energy power transmitted by the interrogator is enough to energize the chip [8]. The previously obtained descend link can also be obtained in the case of an implanted tag. Therefore, the preceding boundary may be checked to see whether it is enough to judge the possibility of a link. The higher the direct margin, the more excess power the chip of the tag will harvest, excess power that could be useful to energize additional onboard sensors. During the measurement, the power emitted by the reader was increased until the chip was activated, thus obtaining the turn-on power. The transducer gain may be written as \[{G}_{T} = \frac{{p}_{\text{chip}}}{{p}_{\text{to}}}{.} \tag{3} \]
The direct link margin can be expressed from (1), (2), and (3): \[{M}_{\text{DL}} = {P}_{{\text{av}}{|}{\text{db}}}{-}{P}_{{\text{to}}{|}{\text{db}}}{.} \tag{4} \]
The major observed factor here is the power calculation and gain for the two proposed implants, which are the femoral nail in Figure 9(a) and the tibial nail in Figure 9(b). It is also responsible for establishing a numeric simulation that accounts for the entire RFID link. The purpose is to correctly predict the EM coupling among the various devices without using any far-field assumptions; simulations included the limb phantom, the cow bone, and both the PST and the antenna of the reader, as well. The values of the simulated and measured turn-on powers show that the RFID communication was correctly established. The most challenging case was the deeper implant in the center of the phantom, with the reader placed at the longer distance of ${d} = {10}{c}{m}{.}$ The direct link margin was ${M}_{DL} = {4}{dB}$ in the European RFID band and up to 6 dB in the U.S. band (902–928 MHz).
In both cases, the measurements are in reasonable agreement with the simulations even if a mutual frequency shift is apparent and is probably due to the approximate knowledge of the real parameters of the used biologic materials and the imperfection in the manual fabrication of the prototype. The values of the transducer gain are on the same order (−40 to −30 dB) as those found in [15] for a square-loop tag that was simply placed on the lateral surface of a cow bone enveloped by an aluminum sheet. In a second experimental session, the angular dependency of the RFID link was evaluated by moving the antenna of the reader around the phantom at a fixed 5-cm distance from its surface when the PST was still implanted, as depicted in Figure 10(b). The resulting angular read region was centered around the notch ${(}{-}{45}\leq\varphi\leq{45}{),}$ while the prosthesis was fully unreachable when interrogated from the lateral and rear sides, as in Figure 11.
Figure 11. An RFID implant and reader communication with a distance of 5 cm [8].
The work covered in this section concentrates on studying and analyzing the viability of EM radiation through the body. In this case, a UHF RFID link is established externally, while the reader and tag are implanted inside the limb, without any physical contact in the area where the prosthesis is normally positioned. Along with this, an RFID microchip is placed over the square-loop antenna used in modeling and simulations with various tests on phantoms [9]. The actual RFID interfaces are implemented for calculating the power needed for the utmost achievable read range for various kinds of implants. In this case, the active link response of the body is observed with respect to the position between the reader and tag. The human body is sensitive to power, and, because of internal resistance, there is a certain amount of attenuation in power such that the estimated read range distance can be uncertain, and the far-field calculation may be affected [17]. Hence, the EM interface of the prosthetic places the implanted transponder over the tissue and the outer interrogator. This entire setup is analyzed using the conventional two-port network method, enabling the configuration of the parameters of the power to be easily determined. Modeling and simulations are executed over an anthropomorphic structure [21].
The method of measurement is an interesting one in which minced meat, bone, and metallic tape fill a cylindrical vessel. The performance of an RFID implant depends upon the particular configuration, ${\psi}{(}\mbox{\textquoteright}\Omega,{d},{h}{),}$ where $\mbox{\textquoteright}\Omega$ represents the particular region of the body where the tag is implanted. Similarly, d is the distance between the surface of the antenna and the host where the antenna is fitted, and h is the depth of the implant measured from the body, surface as described in Figure 12.
Figure 12. A radiative RFID link involving an implanted passive RFID tag inside a limb and a contacting reader antenna [24].
The aspect of the RFID radio channel linking implanted tags subjected to the basic arrangement in which the implanted tag is utilized, along with the range between the surface of a human body and the reader antenna, is available in Figure 13. Another factor that must be considered is the penetration of the placed implant and the point from which it is measured. Here, the coupling factor of the EM field between the tag and reader is obtained by a two-port network system, which is identified by means of an impedance matrix between the reader and the tag terminal as input (port 1) and output (port 2). This equivalent model works more generally with respect to the Friis equation since it considers all near-field communication.
Figure 13. A two-port network representation of a through-the-body passive RFID link [24].
A typical linearly polarized stacked planar inverted-F antenna with a Teflon substrate of dimension ${13}\times{20}{cm}$ and gain of 4 dB along the broadside is chosen along with its reflection coefficient, as presented in Figure 14. The proposed tag for a square-loop planar circuit with dimensions of ${12}\times{18}\,{\text{mm}}^{2}$ was designed on a polyvinyl chloride (PVC) substrate of 3-mm thickness and permittivity ${\varepsilon}_{r} = {2}{.}{3}$ and loss tangent $\tan{\delta} = {2}\times{10}^{{-}{4}}$ in the UHF band. In this case, the material was selected for the purpose of optimizing manufacturing. To achieve biocompatibility in real-time applications, it is important to integrate the tag into the prosthetic device. Initially, copper was chosen for the material, but, later, it was replaced by a titanium alloy for various parts of the prosthesis. Here, the electrospinning technology is adapted to achieve better results, which leads to a thickness on the order of nanometers.
Figure 14. (a) A stacked planar inverted-F antenna over a Teflon substrate, used for interrogation. (b) The simulated reflection coefficient in free space [24].
The RFID performance parameter of various implanted regions of a numeric anthropomorphic model is based on the voxel dataset of the Visible Human (VH) project [20]. The tag is suitably positioned inside the body of the VH phantom with respect to the bone. Here, four distinctive loci for the implant are tested: 1) the shoulder (superior humerus), 2) the elbow (inferior humerus), 3) the hip (superior femur), and 4) the knee (inferior femur). The integration of the implants varies based on the location, ranging from ${h} = {27}\,{\text{mm}}$ for the elbow to 107 mm for the hip, where the layer of the muscle tissue is thicker. The standard reader’s spiral planar inverted-F antenna (SPIFA) is engaged at distance of ${d} = {90}\,{\text{mm}}$ from a plane touching the front end of the phantom.
The simulated results are obtained by investigating the variation of the link output of RFID implants based on the change of the human structure. The entire process is performed over a standardized stratified cylinder-like model of a limb that resembles the knee (Figure 15). The variable dimension of muscle and fat derives physical (normal/muscular build) as well as pathological (obesity) conditions. The equivalent EM constraints are obtained and equate the gain of the transducer calculated for the series of three configurations at distance [28]. The deviation of the worst case (muscular/obese) and the common build is about 1 dB, due to the different absolute depth of the implant and the different percentage of surrounding muscle fat.
Figure 15. A (a) simulated transducer and (b) round-trip gain for implants in knee-like cylindrical stratified phantoms corresponding to three cases of normal, muscular, and obese builds [24]. GT: transducer gain; GRT: round trip gain.
The practical setup process (Figure 16) includes prototypes of the SPIFA and the square-loop tag in Figure 17. The link antenna is connected to a ThingMagic M5e RFID reader. A cylindrical-shaped phantom, with a height of 150 mm, thickness of 2 mm, and diameter of 120 mm, and a PVC container were chosen.
Figure 16. (a) A limb phantom simulating muscle and internal bone. (b) The tag is placed over the lateral surface of the bone so that the implant depth is h = 40 mm [24].
Figure 17. The measured transducer gain around the phantom [24].
A test is conducted in which the reader is displaced from the phantom, and the transducer gain is observed by the turn-on method, and the corresponding gain distance is found to be a 0.3-db/cm attenuation, as illustrated in Figure 18. Another test is conducted in which the SPIFA of the reader moves around the phantom and is placed 22 cm from the implanted tag. The obtained angular movement of the transducer gain is automatic when the reader is opposite the implant, as indicated in Figure 17.
Figure 18. The measured transducer gain for the cylindrical laboratory phantom versus the SPIFA cylinder [24].
Most implants with an implantable RFID tag inserted on bone and muscle are inefficient due to high attenuation from metallic interference. The researchers in [9] recommend a technique of working an implantable passive RFID tag by utilizing a touch probe at 13.56 MHz to increase the response accuracy. The method is based on the electric field interaction among the active electrodes (Figure 19); one of the electrodes is on the touch probe integrated over the surface of the tissue, and another portion of the tag is mounted inside the tissue [34]. By using a traditional RFID antenna, preferably a loop antenna, this methodology has an improved output in the near-field communication range and decreases attenuation with the orthopedic prosthesis. To achieve better efficiency and avoid signal attenuation, there should be proper matching between the tag and tissue.
Figure 19. Two pairs of electrodes coupling with each other for energy transfer [9].
The movement of ions inside tissue fluid conducts electrical current, which can transfer data and energy [25]. If two pairs of electrodes are placed on a piece of tissue, with one electrode placed on the inside and the other on the outside, then a transfer of energy will take place due to capacitive coupling between the electrodes. Capacitive coupling plays an important role for the conduction of current and transfer of signals inside biological tissue [46].
The change in RF frequency inside biological tissue causes attenuation of the signal. To analyze the role of metal in the reception of the signal, two experiments have been conducted.
In [9], a vessel was filled with saline solution, and two holes were drilled at the front side through the center of the vessel (Figure 20). This allowed an exterior electrode to be attached to the vessel wall by using a screw. Two electrodes were placed 40 and 50 mm from the center of the container so that the obtained space would cause the proper distribution of an RF field in the saline solution [45]. A flexible panel was used to manage the distance between the electrodes [9]. The interior and exterior electrodes were placed in such a way that their flat surface maintained a constant distance from the bottom of the saline container. A 13.56-MHz sinusoidal signal was applied to the exterior electrode, and the terminal voltage between the two electrodes (interior electrode and exterior electrode) was measured. When the signal passed through the container, it became attenuated, and the attenuation can be expressed by the following equation: \[{A} = {20}\log{10}\frac{{V}_{\text{EX}}}{{V}_{M}} \tag{5} \]
Figure 20. A saline container with electrodes [9].
where ${V}_{EX}$ and ${V}_{M}$ are the terminal voltage measured across the interior and exterior terminals.
The attenuation of the signal without a metal plate between the inner and outer electrodes is available in Figure 21(a) and (b). Here, L denotes the space between the interior electrodes, and D represents the diameter of the interior electrode. The rise of the signal attenuation level averaged 2.38 dB when a 5-mm-thick ${50}\times{80}{-}{\text{mm}}^{2}$ steel plate was inserted between the electrodes to provide interference, which equaled a 24% loss of voltage efficiency caused by interference, and a 13.56-MHz sinusoidal signal was applied to the exterior electrode and the terminal voltage between the two electrodes (interior and exterior) [Figure 21(c) and (d)]. A ${50}\times{80}{-}{\text{mm}}^{2}$ steel plate, having a thickness of 5 mm, is attached between the movable panels.
Figure 21. Signal attenuation with and without a metal plate [9]. (a) Signal attenuation without the metal plate: D is fixed to be 10 mm. (b) Signal attenuation without the metal plate: L is fixed to be 30 mm. (c) Signal attenuation with the metal plate: D is fixed to be 10 mm. (d) Signal attenuation with the metal plate: L is fixed to be 30 mm. Letter “D” represents the diameter of the interior electrodes and “L” represents the distance between the interior electrodes.
The researchers in [9] also developed an analytical model for tissue to test a touch probe method, as described in Figure 22. In this case, pigskin serves as the tissue, and two electrodes are placed on the external part of the tissue, and the other two are injected inside the tissue. These two pairs of electrodes serve as the active carrier for communication. Impedance matching is established to avoid attenuation [28]. The first of each pair of matching electrodes is placed externally, which is exclusively for the reader, and the other one is an interior electrode just for the chipless tag. The use of chipless RFID implants creates the actual operative condition for the touch probe method and validation of its results when projected through tissue.
Figure 22. A knee implant RFID system [9].
To enhance the coupling for better signal reception on the interior side, the exterior and interior electrodes are organized so that they are in parallel and their symmetric axes overlap. The coupling between electrodes inside the tissue is detailed in Figure 23. To validate the capability of the bioimplant tag, the touch probe technique is followed. The system under test goes through a high-frequency reader and an application console. This prototype uses the touch probe method for the implantable RFID tag. For this purpose, a 1-KB RFID chip with a 96-KB/s data rate is chosen. The tag is integrated into the knee implant along with the touch probe connected at the center of the knee above the pigskin.
Figure 23. The equivalent touch probe method [9].
The measurement of the S-parameters (Figure 24) is obtained with 10 mm of pigskin between the touch probe and a metal plate. Here, ${S}_{11}$ is considered the return loss of the touch probe, and ${S}_{21}$ is ${-}{11}{.}{8}{dB}$ at the center frequency; ${S}_{22}$ is normalized to ${S}_{21},$ as shown in Figure 25.
Figure 24. An experiment for reading a tag through pigskin [9].
Figure 25. The measured S-parameters of the system [9].
The research discussed in this section aims to improve integrated antennas for implants and observe the effect of changes in dielectric properties inside the body. To realize simultaneous sensing and communication, additional analysis follows the implant process to diagnose any endovascular issues. Passive stent tags are able to detect issues in the vessel present at the site where the implant is integrated [10]. The real-time installation and monitoring of the proposed model (which the authors termed STENTag) required the development of a simulation model that resembles the human neck and is designed to monitor a complication of implanted stents, called in-stent restenosis (ISR). Digitized body phantoms (one for all, the VH [7]) are readily accessible and provide validated databases of various parameters of the EM radiation of human organs. The finite-difference time domain method is proposed here to estimate the tag behavior with respect to the evolution of restenosis. A cylindrical model is used to set up an interface between the human body and the implants. It provides a sensitive measurement of the restenosis grade, which can be controlled in a simple manner and tested with high accuracy.
The evolution of ISR is given in Figure 26. Stenosis is a narrowing of a blood vessel or other tubular bodily structure, caused by hyperproliferation of neointimal cells (similar to muscular tissue) [40]. An expandable mesh called a stent can reopen the vessel, but sometimes tissue reaccumulates in and around the stent, a recrudescence named ISR. Continuous sensing of the stent with an RFID tag can give early warning of ISR. From a medical diagnosis and prognostic prospective, ISR can be classified on the basis of its progression pattern and type. Generally, a 40% decrement of the vessel is problematic for the health of the patient, while a decrement in excess of 80% must be considered a serious issue that requires further surgical therapy. The severe form of ISR occurs when the restenosis is large (>10 mm long) and spread beyond the limits of the stents. This developed version is referred as diffuse proliferative ISR.
Figure 26. The (a) stenosis, (b) stenting procedure, and (c) ISR of a biological duct [10].
Stents are made of nickel–titanium alloy and can be fabricated physically for supporting the biocompatibility between tissue and implant. The material possesses conductive features as well as a radiating element that favors a wireless telemetry system. The wireless technology of the stent in [10] is designed to detect the state of a blood vessel. The concept of a stent that is used as a sensor is elaborated in [50] by correlating cell multiplication and tissue growth in a low-frequency range (0.1 Hz to 10 MHz). Later, the stent was further enhanced to have both mechanical and sensing abilities by inserting a tag into the stent, renamed STENTag (Figure 27).
Figure 27. The STENTag prototype, with details of the integration of the RFID IC and the inductor [10].
Thus far, we have reviewed several implants and their range of frequency and application purposes. Table 3 lists a number of implantable prosthetics and their frequency of operation.
Table 3. RFID implants, with their frequencies and applications.
An ORMOCER substrate is preferred for the design of compatible RFIDs that can be interfaced with biological systems (Figure 28). It is basically compatible in terms of chemical analysis, is a good insulator, and is convenient for body area applications. This section discusses RFID implemented as a biocompatible tag printed over a substrate. A number of implants have been developed to serve several physiological, biochemical, and physical needs for medical diagnosis and surveillance. Due to the versatile applications and different requirements of these implant devices, several material designs have been electrically verified. For today’s use, system integration PCBs are required for their superior durability compared to conventional PCB technology.
Figure 28. The steps of an OrmoComp fabrication process [48]. (a) Liquid Ormocomp is sandwiched between two glass plates with 250 ${\mu}$m spacers to control the thickness of the substrate. (b) The liquid Ormocomp is cured for 5 min. using an LED UV lamp at a wavelength of about 400 nm. (c) The cured substrate is removed from the glass plates. (d) A small amount of Ormocomp is applied to a 25-${\mu}$m thick copper foil or a 15-${\mu}$m thick aluminum foil. (e) The Cu or Al foil is pressed against the Ormocomp substrate prepared in (c) between glass plates to bond the foil with the substrate by UV-curing the Ormocomp applied in (d) for 1 min. (f) The metal-cladding substrate is detached from the glass plates. (g) Conductor traces are patterned by traditional photolithography, and electronic components (e.g., an RFID control IC, Alien Higgs 3 [17], in this study) are attached and wire-bonded. h) To encapsulate the Ormocomp PCB, liquid Ormocomp is poured on the component side of the PCB and UV-cured between glass plates for 3 min. (i) Finally, the encapsulated Ormocomp PCB is diced for individual body-implantable devices. It is noted that the encapsulation can also be conducted at the device level by dicing the PCB in (g) first.
Microcracks can arise in implants, due to manufacturing defects and aging (Figure 29). A surface fault can be detected by means of a crack sensor, and the issue can be solved with the replacement of the medical device. In [47], a fractal-based tattoo-like sensing method is integrated on a medical device and coupled to a zero-power RFID transponder. This enables the detection of the early formation of cracks, which is communicated to an external part of the body by means of backscattering communication (Figure 30).
Figure 29. Cracked orthopedic prostheses: the (a) fracture of a femoral stem, (b) failure in a stainless femoral plate, and (c) breakage of a nail.
Figure 30. An RFID-powered prosthesis with self-detection of surface damage consisting of a space filling curve [47].
To achieve lost biomechanics functionality after the loss of an arm or a hand, several dynamic hand prostheses that are capable of myoelectric control with an electronic actuator can be used (Figure 31) [49]. To mimic the electromyography signal of muscle tissue, a transcutaneous telemetry link based on UHF RF is obtained. The active communication takes place through the link between the reader and the implant antenna.
Figure 31. The transcutaneous communication link for the control of an upper-limb myoelectric prosthesis [49].
Bioimplants can have thermal effects on the body by increasing the temperature of surrounding tissues. Some probable causes for the rise in temperature inside the tissue are given in Figure 32. The impact of an increase in temperature inside the tissue that is due to implantable devices is a major concern. The power dissipated by the implant is due to the presence of a microchip, a telemetry coil, and electrodes. In addition, the impact of EM fields on the human body should be investigated. Similarly, [50] details the various sources of temperature increases and possible methods to compute and measure them [50]. Table 4 summarizes of the works covered in this article, with their references.
Figure 32. The causes of temperature increases in the human body associated with the operation of implantable devices [50].
Table 4. The works in the literature covered in this article.
In day-to-day life, implants are an important part of many medical applications. Adding sensor capability to medical implants could achieve many advantages. These sensor implants encounter several challenges when integrated on the body, such as
Tissue variability is defined as the radiation variation through tissue when several implantable antennas are integrated and inserted inside the tissue. Tissue variations are often observed by using square-loop meander dipoles for an external implanted antenna [14]. Tissue variability is essential for robust implantable diagnosable medical applications engaging transcutaneous power. The physiological properties of the tissue tend to vary in the patient with the position of the body and over time [7]. The operating power of the implant is a major concern, so the power gain is dependent upon the antenna topology. Impedance matching between the tissue and antenna can help avoid attenuation, which also depends upon the size of the antenna, characteristics of the tissue, and depth of the implantation [12].
RFID sensors are chosen for their convenience in the manufacturing of implants. The main challenge is to manufacture low-cost chipless RFID sensors even though the fabrication technique is currently more expensive. Substrate selection for interfacing is also a vital point, and the most common method is a photolithographic process. This method is preferably accomplished with dry etching and chemical etching. Dry etching eliminates the extra copper by degrading the tag response, and chemical etching is not suitable for commercial applications [13]. Manufacturing the implants can also add substantial benefits, such as being cost-effective, stable, and user-friendly. The manufacturing technology of implants is still dealing with several challenges, including tissue compatibility and radiation efficiency.
The most useful characteristic of a smart sensing material is its carrier mobility ${\mu}{.}$ It is well defined as the proportionality constant between the applied electric field, E, and the corresponding average carrier drift velocity [30]. Usually, the carrier mobility of these materials is quite low, and the materials are not suitable for RF applications. However, they can be introduced as the sensing materials that change the RF responses of microwave devices under the influence of changing physical parameters. The chipless sensor plays a primary role in environmental interfacing and connecting devices. The most preferable sensing materials, such as polyvinyl alcohol (PVA), phenanthrene, poly(3,4-ethylene dioxythiophene), metallic oxide, and single-wall carbon nanotubes, are very sensitive materials. A material’s sensing properties can be checked by actual scrutiny of the material and by metal characterization using X-ray diffraction, atomic force microscopy, scanning electron microscopy, and transmission electron microscopy as well as by verifying microwave properties, such as the loss tangent, Q factor, dielectric constant, and permeability, using a vector network analyzer.
Currently used implant devices are well known for their accuracy and body-centric applications. Robust flexible and user-friendly implants are required for maintaining secure communication between the body and the reader. It is a tough task to design a sensing device for sensing multiple parameters, such as temperature, acidity, strain, and humidity, on a single tag. This type of challenge can be avoided with the addition of multiple sensors that have multiple parameter sensing abilities. The sensitivity of the sensor is an important parameter to be optimized. Graphene is the best alternative for temperature sensing. It is a suitable material for electrical conductivity and high mechanical stability. It is also sensitive to temperature in biosensors. PVA is mostly preferred for sensing humidity and can be easily fabricated on a single chip, which lends itself to compactness and is less expensive.
The preceding characteristic is the major concern for designing and measuring bioimplants. In modern biomedical applications, implant integration is a complex process. Therefore, adequate knowledge is important for the commercial approach. Along with that, tag and reader placement, material impact on living cells, wireless system architecture, and data mining features are important considerations in bioimplants.
The main aim of this article is to connect emerging RFID technology and prosthetic applications in medical science. The review details several scenarios where chipless RFID sensors are employed in various medical applications with desired frequency ranges for smooth operation. So, this article will add a new spotlight on the area of chipless RFID sensors and their application in bioimplants.
This review presented multiple chipless RFID sensors involved in bioimplants, which supply accurate and secure interaction between the reader and the implant attached to the patient. Loop tags can be used in UHF band applications over a distance of 35 cm, with the sensitivity of microchips. The distance obtained between the body and the reader is 10 cm. Similarly, in a limb prosthesis, active transcutaneous communication is activated with external UHF RFID. This analysis holds good for experimentation with the new era of sensor-based RFID chips for the absolute measurement of strain and temperature for the detection of inflammation. Another tag has been designed using a touch probe method for implants inserted into the human body. Its operation was validated in pigskin. An active observation showed a decreasing level of interference of the RF signal over a metallic implant. This analysis demonstrates the advantages of the touch probe method. STENTag is another enhanced bioimplant setup for the detection of vascular lesions. The insertion of an RFID sensor offers autonomous measurements, with backscattered power and turn-on power obtained over varying frequencies and flexible interfaces. Implantable RFID sensors using ORMOCER technology are introduced for a PCB that is better for chemical analysis and insulation, which is friendly to the human body. Similarly, fractal-based sensing tattoos are introduced for crack detection in bone prostheses. These are normally called crack sensors. Myoelectric prosthesis is also discussed, in which myoelectric signals from the tissue can be gathered. Another important factor is the thermal impacts of implants, which is a consequences of bioimplants on the human body that must be addressed. This article focused on the advanced techniques of various bioimplants. A comparison table containing different applications of bioimplants reported previously is incorporated at the end of the article. Table 4 gives an overall view of this article. The challenge of real-time implant installation is a future challenge, as well. This article will definitely be beneficial for young researchers.
The authors would like to thank Durga Prasad Mishra and Subhasish Pandav for their support during the preparation of this article.
[1] N. C. Karmaker, “Tag, you’re it radar cross section of chipless RFID tags,” IEEE Microw. Mag., vol. 17, no. 7, pp. 64–74, Jul. 2016, doi: 10.1109/MMM.2016.2549160.
[2] S. K. Behera and N. C. Karmakar, “Wearable chipless radio-frequency identification tags for biomedical applications: A review [Antenna Applications Corner] ,” IEEE Antennas Propag. Mag., vol. 62, no. 3, pp. 94–104, Jun. 2020, doi: 10.1109/MAP.2020.2983978.
[3] A. Yakovlelv, S. Kim, and A. Poon, “Implantable biomedical devices: Wireless powering and communication,” IEEE Commun. Mag., vol. 50, no. 4, pp. 152–159, Apr. 2012, doi: 10.1109/MCOM.2012.6178849.
[4] I. Balbin and N. C. Karmakar, “Multi-antenna backscattered chipless RFID design,” in Handbook of Smart Antennas for RFID Systems, N. C. Karmakar, Ed. Hoboken, NJ, USA: Wiley, 2010, pp. 413–443.
[5] G. Khadka, M. A. Bibile, L. M. Arjomandi, and N. C. Karmakar, “Analysis of artifacts on chipless RFID backscatter tag signals for real world implementation,” IEEE Access, vol. 7, pp. 66,821–66,831, May 2019, doi: 10.1109/ACCESS.2019.2917757.
[6] E. M. Amin, N. Karmakar, and S. Preradovic, “Towards an intelligent EM barcode,” in Proc. 7th Int. Conf. Electr. Comput. Eng., 2012, pp. 826–829, doi: 10.1109/ICECE.2012.6471678.
[7] E. M. Amin, J. K. Saha, and N. C. Karmakar, “Smart sensing materials for low-cost chipless RFID sensor,” IEEE Sensors J., vol. 14, no. 7, pp. 2198–2207, Jul. 2014, doi: 10.1109/JSEN.2014.2318056.
[8] R. Lodato and G. Marrocco, “Close integration of a UHF-RFID transponder into a limb prosthesis for tracking and sensing,” IEEE Sensors J., vol. 16, no. 6, pp. 1806–1813, Mar. 2016, doi: 10.1109/JSEN.2015.2503887.
[9] X. Liu, J. L. Berger, A. Ogirala, and M. H. Mickle, “A touch probe method of operating an implantable RFID tag for orthopedic implant identification,” IEEE Trans. Biomed. Circuits Syst., vol. 7, no. 3, pp. 236–242, Jun. 2013, doi: 10.1109/TBCAS.2012.2201258.
[10] C. Occhiuzzi et al., “Design of implanted RFID tags for passive sensing of human body: The STENTag,” IEEE Trans. Antennas Propag., vol. 60, no. 7, pp. 3146–3154, Jul. 2012, doi: 10.1109/TAP.2012.2198189.
[11] S. Kurtz, K. Ong, E. Lau, F. Mowat, and M. Halpern, “Projections of primary and revision hip and knee arthroplasty in the united states from 2005 to 2030,” J. Bone Joint Surg., Amer., vol. 89, no. 4, pp. 780–785, Apr. 2007, doi: 10.2106/JBJS.F.00222.
[12] G. W. Wood II, “Intramedullary nailing of femoral and tibial shaft fractures,” J. Orthopaedic Sci., vol. 11, no. 6, pp. 657–669, Dec. 2006, doi: 10.1007/s00776-006-1061-6.
[13] W. Xu, “Instrumentation and experiment design for in-vitro interface temperature measurement during the insertion of an orthopaedic implant,” in Proc. 10th Int. Conf. Contr., Autom., Robot. Vis. (ICARCV), Dec. 2008, pp. 1773–1778, doi: 10.1109/ICARCV.2008.4795796.
[14] J. F. Drazan et al., “Archimedean spiral pairs with no electrical connections as a passive wireless implantable sensor,” J. Biomed. Technol. Res., vol. 1, no. 1, pp. 1–8, Aug. 2014.
[15] H. Chen et al., “Low-power circuits for the bidirectional wireless monitoring system of the orthopedic implants,” IEEE Trans. Biomed. Circuits Syst., vol. 3, no. 6, pp. 437–443, Dec. 2009, doi: 10.1109/TBCAS.2009.2026283.
[16] F. Burny et al., “Concept, design and fabrication of smart orthopedic implants,” Med. Eng. Phys., vol. 22, no. 7, pp. 469–479, Sep. 2000, doi: 10.1016/S1350-4533(00)00062-X.
[17] D. Dobkin, The RF in RFID. Burlington, MA, USA: Elsevier, 2007.
[18] M. A. Gibney, C. H. Arce, K. J. Byron, and L. J. Hirsch, “Skin and subcutaneous adipose layer thickness in adults with diabetes at sites used for insulin injections: Implications for needle length recommendations,” Current Med. Res. Opinion, vol. 26, no. 6, pp. 1519–1530, Jun. 2010, doi: 10.1185/03007995.2010.481203.
[19] G. Marrocco, “RFID antennas for the UHF remote monitoring of human subjects,” IEEE Trans. Antennas Propag., vol. 55, no. 6, pp. 1862–1870, Jul. 2007, doi: 10.1109/TAP.2007.898626.
[20] M. J. Ackerman, “The visible human project,” Proc. IEEE, vol. 86, no. 3, pp. 504–511, Mar. 1998, doi: 10.1109/5.662875.
[21] G. Marrocco and L. Mattioni, “Naval structural antenna systems for broadband HF communications,” IEEE Trans. Antennas Propag., vol. 54, no. 4, pp. 1065–1073, Apr. 2006, doi: 10.1109/TAP.2006.872559.
[22] G. Marrocco and P. Tognolatti, “New method for modelling and design of multiconductor airborne antennas,” Inst. Elect. Eng. Proc. Microw. Antennas Propag., vol. 151, no. 3, pp. 181–186, Mar. 2004, doi: 10.1049/ip-map:20040178.
[23] H. Lin et al., “Characteristics of electric field and radiation pattern on different locations of the human body for in-body wireless communication,” IEEE Trans. Antennas Propag., vol. 61, no. 10, pp. 5350–5354, Oct. 2013, doi: 10.1109/TAP.2013.2272672.
[24] R. Lodato, V. Lopresto, R. Pinto, and G. Marrocco, “Numerical and experimental characterization of through-the-body UHF-RFID links for passive tags implanted into human limbs,” IEEE Trans. Antennas Propag., vol. 62, no. 10, pp. 5298–5306, Oct. 2014, doi: 10.1109/TAP.2014.2345586.
[25] S. A. Hackworth, “Design, optimization, and implementation of a volume conduction energy transfer platform for implantable devices,” Ph.D. dissertation, Dept. Elect. Eng., Univ. Pittsburgh, Pittsburgh, PA, USA, 2010.
[26] T. E. Tice, “An overview of radar cross section measurement techniques,” IEEE Trans. Instrum. Meas., vol. 39, no. 1, pp. 205–207, Feb. 1990, doi: 10.1109/19.50445.
[27] J. Han, G. Wang, and J. Sidn, “Fragment type UHF RFID tag embedded in QR barcode label,” Electron. Lett., vol. 51, no. 4, pp. 313–315, Feb. 2015, doi: 10.1049/el.2014.4355.
[28] F. Costa, S. Genovesi, and A. Monorchio, “Chipless RFIDs for metallic objects by using cross polarization encoding,” IEEE Trans. Antennas Propag., vol. 62, no. 8, pp. 4402–4407, Aug. 2014, doi: 10.1109/TAP.2014.2326421.
[29] P. R. Foster and R. A. Burberry, “Antenna problems in RFID systems,” in Proc. Inst. Elect. Eng. Colloq. RFID Technol., 1999, pp. 3/1–3/5, doi: 10.1049/ic:19990676.
[30] F. Babaeian and N. C. Karmakar, “Compact multi-band chipless RFID resonators for identification and authentication applications,” Electron. Lett., vol. 56, no. 14, pp. 724–727, Jul. 2020, doi: 10.1049/el.2020.0707.
[31] G. Xiao et al., “Printed UHF RFID reader antennas for potential retail applications,” IEEE J. Radio Freq. Identif., vol. 2, no. 1, pp. 31–37, Mar. 2018, doi: 10.1109/JRFID.2018.2823640.
[32] T. Hänel, A. Bothe, R. Helmke, C. Gericke, and N. Aschenbruck, “Adjustable security for RFID-equipped IoT devices,” in Proc. IEEE Int. Conf. RFID Technol. Appl. (RFID-TA), Warsaw, Poland, 2017, pp. 208–213, doi: 10.1109/RFID-TA.2017.8098883.
[33] I. Farris, A. Iera, and S. C. Spinella, “Introducing a novel ‘Virtual Communication Channel’ into RFID ecosystems for IoT,” IEEE Commun. Lett., vol. 17, no. 8, pp. 1532–1535, Aug. 2013, doi: 10.1109/LCOMM.2013.070913.130392.
[34] A. Kiourti, “RFID antennas for body-area applications: From wearable to implants,” IEEE Antennas Propag. Mag., vol. 60, no. 5, pp. 14–25, Oct. 2018, doi: 10.1109/MAP.2018.2859167.
[35] D. B. Mitzi, “Solution-processed inorganic semiconductors,” J. Mater. Chem., vol. 14, no. 15, pp. 2355–2365, 2004, doi: 10.1039/b403482a.
[36] R. Bashirullah, “Wireless implants,” IEEE Microw. Mag., vol. 11, no. 7, pp. 14–23, Dec. 2010.
[37] M. Levine, B. Adida, K. Mandl, I. Kohane, and J. Halamka, “What are the benefits and risks of fitting patients with radiofrequency identification devices?” PLoS Med., vol. 4, no. 11, pp. 1709–1711, Jul. 2007, doi: 10.1371/journal.pmed.0040322.
[38] T. Ativanichayaphong et al., “Development of an implanted RFID impedance sensor for detecting gastroesophageal reflux,” in Proc. IEEE Int. Conf. RFID, Grapevine, TX, USA, Mar. 2007, pp. 127–133, doi: 10.1109/RFID.2007.346160.
[39] H. Ishihata et al., “A radio frequency identification implanted in a tooth can communicate with the outside world,” IEEE Trans. Inf. Technol. Biomed., vol. 11, no. 6, pp. 683–685, Nov. 2007, doi: 10.1109/TITB.2007.891926.
[40] R. Das, F. Moradi, and H. Heidari, “Biointegrated and wirelessly powered implantable brain devices: A review,” IEEE Trans. Biomed. Circuits Syst., vol. 14, no. 2, pp. 343–358, Apr. 2020, doi: 10.1109/TBCAS.2020.2966920.
[41] G. Shin et al., “Flexible near-field wireless optoelectronics as subdermal implants for broad applications in optogenetics,” Neuron, vol. 93, no. 3, pp. 509–521, Feb. 2017, doi: 10.1016/j.neuron.2016.12.031.
[42] K. L. Montgomery et al., “Wirelessly powered, fully internal optogenetics for brain, spinal and peripheral circuits in mice,” Nature Methods, vol. 12, no. 10, pp. 969–974, Oct. 2015, doi: 10.1038/nmeth.3536.
[43] V. T. Vu, D. N. Nehru, M. I. Pettersson, and T. K. Sjögren, “An experimental ground-based SAR system for studying SAR fundamentals,” in Proc. Asia-Pacific Conf. Synthetic Aperture Radar (APSAR), 2013, pp. 424–427.
[44] T. Tokuda et al., “1 mm3-sized optical neural stimulator based on CMOS integrated photovoltaic power receiver,” AIP Adv., vol. 8, no. 4, Apr. 2018, Art. no. 045018, doi: 10.1063/1.5024243.
[45] M. M. Ghanbari et al., “A Sub-mm ultrasonic free-floating implant for multi-mote neural recording,” 2019, arXiv:1905.09386.
[46] U. M. Jow and M. Ghovanloo, “Modeling and optimization of printed spiral coils in air, saline, and muscle tissue environments,” IEEE Trans. Biomed. Circuits Syst., vol. 3, no. 5, pp. 339–347, Oct. 2009, doi: 10.1109/TBCAS.2009.2025366.
[47] S. Nappi, L. Gargale, F. Naccarata, P. P. Valentini, and G. Marrocco, “A fractal-RFID based sensing tattoo for the early detection of cracks in implanted metal prostheses,” IEEE J. Electromagn., RF, Microw. Med. Biol., vol. 6, no. 1, pp. 29–40, Mar. 2022, doi: 10.1109/JERM.2021.3108945.
[48] G. Scotti, S.-Y. Fan, C.-H. Liao, and Y. Chiu, “Body-implantable RFID tags based on ormocer printed circuit board technology,” IEEE Sens. Lett., vol. 4, no. 8, pp. 1–4, Aug. 2020, doi: 10.1109/LSENS.2020.3009126.
[49] C. Miozzi, S. Guido, G. Saggio, E. Gruppioni, and G. Marrocco, “Feasibility of an RFID-based transcutaneous wireless communication for the control of upper-limb myoelectric prosthesis,” in Proc. 12th Eur. Conf. Antennas Propag. (EuCAP), 2018, pp. 1–5, doi: 10.1049/cp.2018.0483.
[50] G. Lazzi, “Thermal effects of bioimplants,” IEEE Eng. Med. Biol. Mag., vol. 24, no. 5, pp. 75–81, Sep./Oct. 2005, doi: 10.1109/MEMB.2005.1511503.
[51] Y. Muraki and M. Mori, “Effective tooth excitation position for an implanted dental-bone conduction hearing aid,” in Proc. IEEE 9th Global Conf. Consum. Electron. (GCCE), 2020, pp. 24–28, doi: 10.1109/GCCE50665.2020.9291837.
[52] N. A. Quadir, L. Albasha, M. Taghadosi, N. Qaddoumi, and B. Hatahet, “Low-power implanted sensor for orthodontic bond failure diagnosis and detection,” IEEE Sensors J., vol. 18, no. 7, pp. 3003–3009, Apr. 2018, doi: 10.1109/JSEN.2018.2791426.
[53] V. Viikari, J. Song, N. Pesonen, P. Pursula, and H. Seppä, “Review of passive wireless sensors utilizing the intermodulation communication,” in Proc. IEEE RFID Technol. Appl. Conf. (RFID-TA), 2014, pp. 56–61, doi: 10.1109/RFID-TA.2014.6934200.
[54] F. Costa, S. Genovesi, M. Borgese, A. Michel, F. A. Dicandia, and G. Manara, “A review of RFID sensors, the new frontier of Internet of Things,” Sensors, vol. 21, no. 9, Apr. 2021, Art. no. 3138, doi: 10.3390/s21093138.
[55] S. I. Park et al., “Soft, stretchable, fully implantable miniaturized optoelectronic systems for wireless optogenetics,” Nature Biotechnol., vol. 33, no. 12, pp. 1280–1286, Dec. 2015, doi: 10.1038/nbt.3415.
[56] B. Tehrani, R. Bahr, D. Revier, B. Cook, and M. Tentzeris, “The principles of ‘Smart’ encapsulation: Using additive printing technology for the realization of intelligent application-specific packages for IoT, 5G, and automotive radar applications,” in Proc. IEEE 68th Electron. Compon. Technol. Conf. (ECTC), 2018, pp. 111–117, doi: 10.1109/ECTC.2018.00025.
[57] E. Moradi, L. Sydänheimo, G. S. Bova, and L. Ukkonen, “Measurement of wireless power transfer to deep-tissue RFID-based implants using wireless repeater node,” IEEE Antennas Wireless Propag. Lett., vol. 16, pp. 2171–2174, May 2017, doi: 10.1109/LAWP.2017.2702757.
[58] A. Adeyeye, C. Lynch, A. Eid, J. Hester, and M. Tentzeris, “5.8-GHz low-power tunnel-diode-based two-way repeater for non-line-of-sight interrogation of RFIDs and wireless sensor networks,” IEEE Microw. Wireless Compon. Lett., vol. 31, no. 6, pp. 794–797, Jun. 2021, doi: 10.1109/LMWC.2021.3064734.
Digital Object Identifier 10.1109/MMM.2023.3265465