Resistive sensors operate on the principle that the resistance in a circuit changes in response to variations in a physical quantity. Featuring a simple construction, durability, wide dynamic range, and low cost, they are suitable for applications such as temperature and strain sensing. The most common resistive sensors include resistance temperature detectors (RTDs), thermistors, and strain gauges. The key feature enabling temperature measurement in RTDs and thermistors is their high linearity between resistance and temperature. Typically, RTDs have a resistance temperature coefficient value of 3.33 × 10⁻³/°C with an accuracy of ± 0.2°C, over a measurement range of 20°C–100°C (Kim et al., 2020b). Therefore, RTDs and thermistors are used to measure implant temperatures, with a temperature increase indicating a possible implant infection (Yu et al., 2017). Strain gauges are usually attached to the surface of a metal elastomer. As the elastomer undergoes deformation from an external force, the strain is transmitted through the adhesive to the resistive strain gauge, resulting in a change in the resistance value of the strain gauge (Zhao et al., 2020). Strain gauges are the most widely used sensors for measuring a variety of complex forces in vivo across different scenarios. These applications include forces in joints, displacement during fracture healing, forces acting on dental implant prostheses, and bladder pressure (Kim et al., 2012; D'Lima et al., 2013; Cho et al., 2014; Pelham et al., 2017). Fukase et al. (2022) developed a smart bone plate to wirelessly monitor healing utilizing electrical impedance spectroscopy to provide real-time data on tissue composition within the fracture callus. The sensor is wireless and can take readings up to 3 m from the implanted bone for up to 8 weeks.

The capacitor is the key electronic component in a capacitive sensor. Typically, a capacitor comprises two parallel conductive plates at the top and bottom, with a dielectric material positioned in between. When the measured physical quantity changes, alterations occur in the distance between the electrodes, their overlapping area, or the dielectric properties of the material (Qin et al., 2021). Selecting materials with a low temperature coefficient for electrodes in capacitive sensor construction ensures they are almost temperature independent (Sander et al., 1980). Moreover, capacitive sensors offer advantages such as simple device structure, low power consumption, versatile applications, rapid dynamic response, and high durability (Sun et al., 2014; Guo et al., 2019; Liu et al., 2020). They are commonly employed for level, strain, force, pressure, and humidity monitoring. Because of their high sensitivity, capacitive sensors can detect small force magnitudes, proving valuable in applications such as the histomorphometric evaluation of dental implants, and tracking bone growth surrounding them over time (D’Lima et al., 2013). In musculoskeletal system, capacitive sensors are often used to monitor implant loosening and fracture healing. Sorriento et al. (2021) proposed a system based on capacitive sensor technology, able to measure, quantitatively, the relative pins displacements in bone fractures treated with external fixators with a resolution of 0.5 mm and 0.5°.

Inductive sensors are typically composed of a magnetic core with an air gap and a coil (Veletić et al., 2022). The magnetic resistance is determined by the length of the air gap, with longer gaps resulting in higher magnetic resistance. Even a very small change in the air gap size can cause a significant change in magnetic resistance. Because of their high sensitivity, inductive sensors are frequently used to measure small changes such as displacement and strain and are often integrated into an inductor-capacitor circuit (Chen et al., 2014b; Burton et al., 2021). Bennett et al. (2021) developed a sensor to detect the depth of a locking pin in the shuttle lock of a transtibial socket and to monitor the small motions between ratchet clicks during ambulation. The sensor demonstrated a root mean square error of 0.21% of the full-scale output and it is sufficiently accurate. Additionally, these sensors can be used for wireless communication and power transmission, making them particularly promising for the remote monitoring of implants.

Si is commonly used as a structural material or bulk substrate in implantable sensors. SiO₂ is a rigid material that is also widely used in implantable sensors, particularly for blood pressure measurements (Liang et al., 2011; Ye et al., 2011; Murphy et al., 2013). Different types of sensors are made by etching various patterns on the surface of Si. Although Si is inherently hard and stiff, it can be used as a thin membrane in implantable pressure sensors, which convert physiological pressure into a measurable capacitive change or piezoelectric signal (Yoon et al., 2004).

Si₃N₄ is a thermally stable, non-toxic, and biocompatible ceramic material, making it more suitable than other ceramic materials for use in orthopedic implants. In addition, Si₃N₄ can be deposited on both sides of a silicon wafer to serve as a hard mask during etching processes, which is particularly useful in the design of implantable sensors for cerebrospinal fluid (CSF) flow monitoring (Raj et al., 2015; Apigo et al., 2016; Apigo et al., 2017). This pressure sensor is capable of detecting CSF pressure, hydrostatic pressure, and flow by measuring the deflection of a flexible membrane. However, it exhibits relatively low sensitivity and is primarily suitable for measuring slow-moving fluids in clinical contexts, such as for CSF.

Metallic substrates are also used in the fabrication of implantable sensors, with stainless steel being a common choice. Chen et al. (2017) developed a capacitive pressure sensor on a stainless steel chip, designed for seamless integration with inductive stents using micro welding. The use of stainless steel as the substrate improved the sensor’s reliability. In addition, various metals, including Zn, Cu, Al, Au, and Ag are used as conductive materials in the construction of implantable sensors (Heller, 2006; Soebadi et al., 2020; Yang et al., 2020; Lee et al., 2022).

The application of rigid materials in implantable sensors poses many challenges. Firstly, the majority of rigid materials are not biocompatible, necessitating additional encapsulation materials to reduce the risk of harming the body. The bonding of the encapsulation material to the sensor increases the complexity of the sensor manufacturing process. Secondly, the sharp edges of hard materials may pose a risk of injuring soft tissues during implantation into the body.

PDMS is a type of silicone elastomer with a low Young’s modulus, fluid impermeability, high dielectric strength, low chemical reactivity, and excellent biocompatibility (Lee and Choi, 2008). PDMS is used in sensitive capacitive sensors designed for measuring pressure and oxygen levels in the heart and blood vessels, as well as for monitoring nerve tissue health (Chiang et al., 2007; Koley et al., 2009). The low Young’s modulus of PDMS makes it suitable for use as a flexible substrate for sensors. Even under bending or stretching, PDMS maintains its structural integrity without undergoing permanent deformation. Chen et al. (2022) proposed a fabrication method for a double-layer piezoresistive pressure sensor, aiming to achieve a wide sensing range and high sensitivity. A conductive pathway was constructed on PDMS films and the resulting sensor exhibited a high sensitivity of 2.6 kPa⁻¹ within a wide linear range of 0–30 kPa, along with fast response and recovery times of 40/40 ms. It demonstrated excellent reproducibility and was successfully applied to the real-time detection of radial artery heart rate, limb movement, handwriting, and vocal cord vocalization.

Medical-grade silicone, a type of silicone elastomer approved for use in biomedical implants by the Food and Drug Administration, has excellent biocompatibility. In comparison to PDMS, it has high tear strength and superior elasticity across a broad temperature range. Medical-grade silicone is an ideal candidate for substrate materials in implantable sensors. This is demonstrated in the fabrication of soft contact lens sensors, where resonance circuits are embedded within layers of medical-grade silicone for continuous intraocular pressure monitoring (Chen et al., 2014a; Farandos et al., 2015). In addition, medical-grade silicone can be used to make strain gauge housing strips for measuring blood pressure through the deformation of blood vessels (Bingger et al., 2012). It has a similar Young’s modulus to that of blood vessels which minimizes interference when wrapped around the vessel.

Parylene-C is widely used as a substrate or encapsulation material for implantable sensors because of its biocompatibility, chemical and biological inertness, effective barrier properties with low water permeability and absorption, and its functionality as an electrical insulator. The electrochemical impedance of microbubbles can be measured through the electrolysis of a platinum electrode in contact with the Parylene-C surface. This unique property enables the use of Parylene-C in the preparation of standard sensors for monitoring hydrocephalus treatment and shunt performance (Yu et al., 2015; Kim et al., 2016). Using Parylene-C as a foundation, Yao et al. (2020) designed an ultrathin, flexible, waterproof, and robust micro-nano composite coating for encapsulating an implantable pressure sensor. Experimental results demonstrate its functional performance in both simulated body fluid environments and animal experiments.

PI is a type of polymer composed of imide monomers and is known for its high heat resistance and glass transition temperatures, remaining stable up to 440°C (Liaw et al., 2012). Its biocompatibility was tested using standard methods according to International Organization for Standardization 10,993, and the results indicated that the electrodes were non-cytotoxic, non-mutagenic, non-clastogenic, and non-hemolytic (Kullmann et al., 2022). PIs are widely used in sensors as passivation or insulation materials and substrate layers because of their excellent thermal and chemical stabilities, low dissipation factors, and low dielectric constants. PIs are commonly used as thin films, serving as structural components in implantable sensors. In a wireless intraocular pressure sensor, the copper inductor pattern is deposited on top of the PI membrane (Kang et al., 2013). A thin PI film is also used as the diaphragm in capacitive pressure transducers for implantable cardiovascular applications (Hasenkamp et al., 2012). Zhu et al. (2022) introduced a PI-based film bulk acoustic resonator humidity sensor with high sensitivity and stability. This sensor was applied for the first time in the real-time monitoring of human respiration.

PVDF and its copolymer, polyvinylidene fluoride-trifluoroethylene (PVDF-TrFE), have a wide range of applications because of their piezoelectric response, thermal resistance, and chemical stability. PVDF is a thermoplastic semi-crystalline polymer which, in more of being cheap and easy-to-process, exhibits attractive electroactive properties (Song et al., 2022a; Li et al., 2022; Yuan et al., 2022). PVDF scaffolds for tissue engineering were proven to have good biocompatibility (Kitsara et al., 2022). Their piezoelectric properties make them valuable components in acoustic sensors. PVDF-TrEE is a lightweight material and can sustain higher strains (Zhang et al., 2002). The PVDF-TrFE diaphragm exhibits high linearity in response to small pressure changes, high sensitivity, and insensitivity to ambient temperature changes (Li et al., 2010). This diaphragm can be integrated with a catheter for intravascular measurements and used for monitoring heartbeat and respiration by measuring deformation of the chest wall (Chiu et al., 2013; Sharma et al., 2013).

LCP belongs to the family of aromatic polymers. LCP has been widely employed as an insulating and substrate material in the implantable devices due to its biocompatibility and low moisture absorption rate (<0.04%) compared to other polymers such as PI, parylene-C, or silicon elastomer (Kim et al., 2020a; Yun et al., 2022; Ahn et al., 2023). The use of LCP encapsulation provides long-term reliability without electrical degradation. In vivo studies demonstrate minimal fibrous encapsulation and inflammation in LCP-packaged devices (Chow et al., 2010). Sattayasoonthorn et al. (2019) developed a biocompatible, miniature, and implantable wireless intracranial pressure monitoring device. The device was primarily fabricated using LCP through standard microelectromechanical system procedures. The results demonstrated the sensor’s capability to perform intracranial pressure measurements in a humid environment within the range of 0–30 mmHg.

In 1966, the first implantable sensor was introduced for application in hip implants. Rydell (1966) placed a strain gauge on the neck of the metal prosthesis to measure the forces acting on a mobile patient’s femur prosthesis. The system involved connecting one end of a wire to the prosthesis and the other end to an external data recorder through a percutaneous connection. While not a fully implantable sensor, this research marked a significant milestone. It was not until 1974 that a truly wireless implantable sensor was developed. Carlson et al. (1974b) placed 14 pressure sensors on the hemispheres of a prosthesis to measure pressure on the surface of the hip cartilage. Subsequent applications of similar sensors have extended to the measurement of hip pressure during daily activities using remote sensing or battery power (English and Kilvington, 1979; Davy et al., 1988; Bergmann et al., 2004; Damm et al., 2010).

Post-operative monitoring extended from 12 days to 36 months (Carlson et al., 1974a; Hodge et al., 1989; Cristofolini et al., 2000; Wei et al., 2022). During extended walks, hip prostheses experienced heat generation because of friction. Prolonged exposure to intra-articular heat may inhibit the growth of periarticular cells, potentially resulting in bone resorption or implant loosening (Pritchett, 2011). Bergmann et al. (2012) conducted a clinical trial involving 100 individuals, during which a temperature sensor was placed in the neck of the prothesis (Force sensor and IMU sensor). The sensor measured implant temperature within the range of 20°C–58°C, with an accuracy of 0.1°C. Some implants can incorporate dual sensors for simultaneous force and temperature measurement (Graichen et al., 1999). In addition, Lange et al. (2021) designed a piezoelectric sensor-based hip implant powered by the pressure exerted within the joint. This innovative approach could pave the way for future developments in implantable sensor design.

In addition to measuring temperature, force, and pressure, implantable sensors play an important role in monitoring the stability of implants by detecting any signs of loosening (Ruther et al., 2011). Heilemann et al. (2020) developed a pin-socket sensor system capable of monitoring micro-movements within a range of 59 ± 2 μm to 222 ± 5 μm on the implant surface. However, the drawback of this approach is that many holes need to be drilled in the implant surface to mount the sensor, which may cause structural damage to the implant. Mohammadbagherpoor et al. (2020) designed a wireless inductive proximity sensor system for detecting early implant loosening (Inductive-capacitive passive sensors). Its key advantage lies in the high sensitivity of the inductive sensor, which is capable of monitoring loosening at a distance of 20 mm with a resolution of 100 μm. de Sousa et al. (2021) designed a surface capacitive architecture sensing system to detect bone implant loosening, achieving favorable outcomes in both in vitro experiments and numerical simulations. Another study found that extracorporeal informatic systems enable continuous patient monitoring through cosurface capacitive networks, with or without hydroxyapatite-based layers (Peres et al., 2022). This could represent a significant advancement in the design of multi-functional smart implants. Additionally, the team proposed a new cosurface-based capacitive system designed to deliver controllable and personalized electric field stimuli to target tissues (Soares Dos Santos et al., 2016; Bernardo et al., 2019; Ramos and Soares Dos Santos, 2023). This system not only enhanced the predictability of sensing bone-implant fixation states but also delivered personalized stimulation to peri-implant tissues (de Sousa et al., 2021; Ramos and Soares Dos Santos, 2023).

Some sensors can enhance intra-operative precision and aid surgeons in refining procedures. Chen et al. (2015) designed a visual-aided wireless monitoring system composed of two key components: the sensors and the display. The sensors include both contact and image sensors, enabling the estimation of the relative position between the femoral head and acetabular components. Manupibul et al. (2019) developed the Force-PRO device, which assists doctors in evaluating the force on the surface of the acetabular liner and the angle of the implant during surgery. The optimal angle reduces the risk of post-operative implant dislocation. Wijayaratna et al. (2021) designed a sensor attached to the hip prosthesis for measuring the pH of the joint fluid. When the pH decreases, it signals to the doctor that the hip prosthesis may be infected and requires prompt treatment. Recent advances in implantable hip sensors are detailed in Table 2, with representative studies highlighted in Figure 1.

Severe knee osteoarthritis can result in disability, significantly affecting quality of life. Several studies have demonstrated a significant link between knee joint forces and osteoarthritis (Amin et al., 2004; Niu et al., 2009). Therefore, understanding the forces acting on the knee joint following total knee arthroplasty (TKA) is crucial. These forces can impact the longevity of the implant and the integrity of the interface between the implant and the bone (Song et al., 2020). To gain insights into the mechanisms of knee disorders and post-operative knee care, direct measurement of the force in the knee joint can be achieved through implantable sensor technology. Recent advances in implantable sensors for knee applications are summarized in Table 3, with representative studies highlighted in Figure 2.

In 1998, sensors were integrated into a distal femoral prosthesis to measure the forces acting on the femoral shaft in two patients throughout their daily activities for up to 2.5 years. Subsequently, these measurements were used to calculate the forces exerted on the knee joint (Taylor et al., 1998; Taylor and Walker, 2001). In 2004, the first direct in vivo measurement of knee forces was performed using a uniaxial force measurement device (D’Lima et al., 2005). However, this device could not measure anterior shear and axial torque, which led to the development of a new generation of knee sensors in 2005. The second-generation sensor design incorporated a redundant array of 12 strain gauges, collectively providing measurements for the three components of force and three moments (Kirking et al., 2006). In 2014, piezoelectric sensors were incorporated into knee implants, exploiting their self-powering capability (Strain gauge sensor made of PI) (Forchelet et al., 2014). In 2017, Safaei et al. (2017) introduced a knee bearing with four embedded piezoelectric transducers capable of measuring knee joint pressure (The piezoelectric transducer, diameter 8 mm, thickness 3 mm). They investigated the fatigue behavior of the bearing and the electromechanical performance of the embedded piezoelectric components (The piezoelectric transducer, diameter 8 mm, thickness 3 mm) (Safaei et al., 2017; Safaei et al., 2018; Ponder et al., 2019; Safaei et al., 2020). The findings revealed that a peak voltage of 3 V can be generated with a load resistor of 1 MΩ, and a peak average power of 5.5 μW can be obtained under a resistive load of 171 kΩ. Three preparation processes were also compared. Fused deposition modeling proved to be fast and cheap, but the results were not satisfactory. The longer the stereolithography print time, the higher the quality of the final product. Alternatively, computer numerical control machining provided superior material characteristics but was more labor intensive. Finally, they evaluated the distribution of piezoelectric sensors on the bearing using finite element analysis. The results indicated a maximum error rate of 2.6% for the device.

Implantable sensors are widely used to measure contact forces within joints during daily activities. Fregly et al. (2009) tested the medial contact force of the knee joint in a patient who had undergone knee arthroplasty and observed that, when the patient used a cane, this force was reduced by 27% and 16% compared with that when the patient was standing and walking without a cane, respectively. However, the benefits of using a cane were contingent on the side on which it was used. Walking with contralateral hand crutches resulted in a 43% reduction in knee contact force, while walking with ipsilateral hand crutches lead to an average increase of 9% (K-Scan sensors, Tekscan, South Boston, MA) (D’Lima et al., 2012).

In addition to monitoring everyday contact forces in the knee joint, implantable sensors can help doctors assess the condition around the prosthesis after TKA (MacDessi et al., 2020; Song et al., 2022b). Lavdas et al. (2021) implanted a sensor package into the femoral osteotomy of a cadaver (Temperature sensor, length 16.5 mm, width 23 mm). The sensor demonstrated an accuracy and precision of ± 0.24°C and 0.09°C, respectively. While additional testing is necessary for clinical application, the feasibility of this approach is evident. Sarpong et al. (2022) used sensors for soft tissue balancing during TKA, revealing that they did not provide additional benefits in clinical outcomes. Moreover, it significantly increased surgery time and incurred additional costs.

In recent years, researchers have focused on the development of multi-functional and smart implants. Yamomo et al. (2021) introduced a 3D-printed titanium interpositional device designed to integrate triboelectric generators into a commercially available TKA. The study evaluated the device’s stiffness, durability, and effectiveness as a housing for triboelectric generators.

Implantable sensors hold great promise for various commercial applications in the knee joint. Their integration into existing joint replacement components involves encapsulating the sensor, telemetry technology, and power supply system within the tibial tray, thereby increasing economic feasibility. A tibial tray integrated with sensors for the intra-operative monitoring of tibiofemoral force distribution and quantification of soft tissue balance enables the collection of extensive data on internal forces during the post-operative period. Surgeons find significant value in measuring in vivo forces and strains, while patients benefit by gaining insights into the forces within their knees, aiding in the identification of potentially risky activities.

The shoulder is one of the most complex major joints in the body. Unlike the hip joint, the shoulder joint is characterized by exceptional flexibility and is primarily stabilized by muscles, while the hip joint is constrained by bony structures. Forces exerted in and around the shoulder are directly related to many common shoulder disorders, and some inappropriate forces generated during daily and sport activities can lead to shoulder injuries. Therefore, shoulder arthroplasty provides a convenient opportunity to implant sensors into the shoulder to monitor forces during complex activities. In 2007, Bergmann et al. (2007) pioneered the in vivo measurement of shoulder forces using sensors (Strain gauges, SG, type KSP 1-350-E4, Kyowa). To measure the contact force of the joint, six sensors were placed on the humeral stem during shoulder arthroplasty. Two years later, they conducted a comparative analysis of the contact forces in the shoulder joint during everyday activities using data from four patients with the aforementioned implants. The study revealed that the highest joint forces were recorded while steering a car with one hand, placing a 1.5 kg load on a table, and lifting a 2 kg weight to a high shelf (Westerhoff et al., 2009). Their findings strongly indicate that patients with shoulder problems or those recovering from shoulder surgery should avoid certain activities, such as lifting heavy objects with the arm outstretched. Furthermore, their research revealed that the frictional force within the shoulder joint exceeded previous estimates made using models that neglected its significance in shoulder calculations.

Reverse total shoulder arthroplasty (RTSA) is mainly applicable to cases of rotator cuff tear arthropathy that are not suitable for conventional total shoulder arthroplasty (TSA). However, the sensor designed by Bergmann et al. (2007) targets the TSA and may not be optimized for RTSA. Farmer et al. (2020) integrated four strain gauge sensors with a trial glenosphere to measure intra-operative glenohumeral contact forces during RTSA. Their findings indicate the feasibility of measuring joint contact forces intra-operatively in RTSA. Building on this, Leinauer et al. (2021) replaced the resistive sensors with capacitive sensors, citing their perceived benefits of enhanced stability and high sensitivity (Capacitive sensor made of PDMS). However, it is important to note that these sensors are intended solely for intra-operative measurements and must be removed before the procedure is completed. To date, there have been no studies that demonstrate the post-operative use of pressure sensors for shoulder measurements.