Osseosurface electronics-thin, wireless, battery-free and multimodal musculoskeletal biointerfaces

Abstract

Bioelectronic interfaces have been extensively investigated in recent years and advances in technology derived from these tools, such as soft and ultrathin sensors, now offer the opportunity to interface with parts of the body that were largely unexplored due to the lack of suitable tools. The musculoskeletal system is an understudied area where these new technologies can result in advanced capabilities. Bones as a sensor and stimulation location offer tremendous advantages for chronic biointerfaces because devices can be permanently bonded and provide stable optical, electromagnetic, and mechanical impedance over the course of years. Here we introduce a new class of wireless battery-free devices, named osseosurface electronics, which feature soft mechanics, ultra-thin form factor and miniaturized multimodal biointerfaces comprised of sensors and optoelectronics directly adhered to the surface of the bone. Potential of this fully implanted device class is demonstrated via real-time recording of bone strain, millikelvin resolution thermography and delivery of optical stimulation in freely-moving small animal models. Battery-free device architecture, direct growth to the bone via surface engineered calcium phosphate ceramic particles, demonstration of operation in deep tissue in large animal models and readout with a smartphone highlight suitable characteristics for exploratory research and utility as a diagnostic and therapeutic platform.

Fig. 1. Osseosurface electronics: concept, device architecture, and implementation strategies.

a Illustration of osseosurface electronic systems that are permanently bonded to the bone and operate wirelessly to continuously monitor biophysical signals such as bone strain, local temperature, and to deliver optical stimulation to the bone and surrounding tissues. b Photograph of an osseosurface device designed for studies in large animal models. c Layered makeup of the osseosurface system and its constituent layers. Inset features a close-up view of the multifunctional biointerface comprised of a metal-foil strain gauge, an NTC thermistor, and a µ-ILED. d Photograph of an osseosurface device conformally attached on the surface of a sheep humerus. e Functional block diagram of osseosurface electronics comprised of an external NFC reader that provides power and facilitates wireless communication, and an implanted system that contains active power management, operational control, analog front-end (AFE), and biointerface. f Photograph of a rat (2 weeks after surgery) implanted with an osseosurface device where the main electronics reside on the back while the biointerface is routed and attached on the left femur.

Fig. 2. System characteristics of osseosurface electronic systems.

a–c Device for large animal models. Device photograph (a). Harvested power and voltage as functions of electrical load (b). Spatial distribution of harvested power using a handheld primary antenna at a load of 300 Ω (c). d–f Device for small animal models. Device photograph (d). Harvested power and voltage as functions of electrical load (e). Spatial distribution of harvested power using a 45 cm × 12 cm primary antenna measured with a load of ~900 Ω (f). g Power consumption of the device operating at different modes: I. temperature sensing; II. temperature and strain sensing; and III. temperature and strain sensing, and optical stimulation. Modes II and III are represented by green and red dashed lines in (b) and (e) that indicate the value of electrical load and power consumption. h Data rate of wireless communication for a large animal device (immersed in PBS solution) read and powered by a handheld primary antenna as a function of antenna-to-device distance. Inset, 3D rendering of the experimental setup. i Demonstration of long-term data recording, for a small animal device (immersed in PBS solution), measured with the 45 cm × 12 cm primary antenna on a custom-built metal-free rat treadmill with back-and-forth motion at a speed of 25 cm s⁻¹. The wireless results are benchmarked against environmental temperature recorded by a thermometer (red data line) placed in close proximity. Inset, 3D rendering of the experimental setup.

Fig. 3. Benchtop characterization of the biointerface.

a–c Strain sensing module: comparison of strain–load curves measured using standard wired system and wireless OSE device (a); strain profile measured with the wireless device under progressively increasing load (b); and strain profiles measured with wired system and wireless device under cyclic load (c). Benchtop test setup where the strain gauges are attached on a piece of sheep bone and the load is applied with a MTS system. d–f LED module: photographs of devices inserted at the bone–tissue interface, with µ-ILED on the bottom (left), on the top (middle), and on both sides (right) (d); current profile of the µ-ILED operating at frequencies of 5, 10, and 20 Hz with pulse width of 20 ms (e); current and optical power as functions of driving voltage of the µ-ILED used (f). Scale bar in (d), 5 mm. g–i Thermal sensing module: wirelessly logged ADC values as a function of temperature (g); wirelessly measured change in temperature using the NTC thermistor with the micro-heater driven at various levels of power, demonstrating a sensing resolution of <10 mK (h); thermal impact of an µ-ILED operating at various optical powers and duty cycles in PBS solution measured wirelessly with a co-located NTC thermistor (i). Insets of (h) and (i), schematics showing the layout of thermistor-micro-heater (h) and thermistor-LED front-end (i).

Fig. 4. Mechanical design and characterization of osseosurface electronic devices designed for rodents.

a Micro-CT scan of a rat implanted with an osseosurface electronic device. The device is highlighted in yellow. b Photograph of the self-similar serpentine interconnects stretched to 250% (upper) and FEA strain profile of the serpentine interconnects stretched to 250% (lower). c Resistance of the interconnects at 0% and 250% strain during cyclic stretching to 250% for 10,000 cycles. d FEA of the biointerface when the bone is compressed by 1000 με while the serpentine interconnects are in two states: i. relaxed (0% strain); and ii. stretched (250% strain). e Wirelessly recorded ADC values when the strain gauge is cyclically loaded and unloaded while the serpentine interconnect is strained to 250%, the inset shows the test setup with 250% strain applied. f Wirelessly recorded ADC values when the strain gauge is cyclically loaded and unloaded after the serpentine interconnect has been strained for 0, 5000, and 10,000 cycles.

Fig. 5. In vivo studies of OSE in rats.

a Photograph of the rear part of a rat featuring the µ-ILED attached on the femur illuminating through muscle and skin. b Temperature profile recorded in vivo during the period of ~11.5 h following the implantation surgery. Inset, schematic of the rat residing in the home cage. c Time-synchronized strain recording and corresponding video frames of the rat gait cycle. d, e Comparison of the rat’s gaits at different stages of the study: before surgery, 1 week after surgery, and 2 weeks after surgery: photograph of rats walking on the treadmill overlaid with the trajectory of the ankle. Color of the dots represents the velocity of the ankle (d). Key parameters: stride frequency (before: mean = 2.4293, SD = 0.4057, n = 7; 1 week: mean = 2.1843, SD = 0.1692, n = 7; 2 week: mean = 2.3336, SD = 0.197, n = 7); stride length (before: mean = 9.3812, SD = 1.749, n = 7; 1 week: mean = 9.1175, SD = 2.2686, n = 7; 2 week: mean = 8.8466, SD = 1.0577, n = 7); and duty factor (before: mean = 0.6558, SD = 0.2689, n = 7; 1 week: mean = 0.6189, SD = 0.118, n = 7; 2 week: mean = 0.6182, SD = 0.0585, n = 7) that characterizes the rat gait before and after implantation (e).

Fig. 6. Promotion of osteogenesis using surface-engineered calcium phosphate ceramic coatings.

a Photograph of the bottom coated with CPC and a side view diagram of layer composition of the strain gauge sensor with CPC. b Image of the device 3 weeks post implantation. c Wireless strain gauge data collected over 2 weeks post implantation showing stable CPC adhesion. d Single electron microscope image of cross-section of CPC adhesion to bone. e Animal healing after implantation.

Fig. 7. In situ studies in sheep and diagnostic application.

a Photograph of a wirelessly powered osseosurface electronic device attached on a sheep humerus. b Temperature profile during implantation surgery on sheep humerus recorded wirelessly through tissue. c Strain profile of 3-point bending test of sheep humerus recorded wirelessly through tissue. d Tools and strategy to attach osseosurface devices on the bone: i. bone surface is cleaned by sanding; ii. device with applied glue is brought into contact with the bone surface with applicator; iii. device is pressed firmly with finger; and iv. applicator is removed. e Photos of demonstration of real-time signal recording through porcine skin (~1 cm thick) using an NFC-enabled smartphone: i. device attached on the sheep humerus; and ii. device covered by porcine skin and real-time signal readout with NFC-enabled smartphone. f Relative data rate and power availability as functions of tissue thickness. Inset shows photograph of experimental setup.