I could see hooking this up with BCI caps and thus controlling peripheral nerves that no longer work(e.g. fingers). But with NO STROKE LEADERSHIP OR STRATEGY, there is nothing that can be done until we create stroke survivors in charge.
Wireless endovascular nerve stimulation with a millimeter-sized magnetoelectric implant
Abstract
Implanted bioelectronic devices have the potential to treat disorders that are resistant to traditional pharmacological therapies; however, reaching many therapeutic nerve targets requires invasive surgeries and implantation of centimeter-sized devices. Here we show that it is possible to stimulate peripheral nerves from within blood vessels using a millimeter-sized wireless implant. By directing the stimulating leads through the blood vessels we can target specific nerves that are difficult to reach with traditional surgeries. Furthermore, we demonstrate this endovascular nerve stimulation (EVNS) with a millimeter sized wireless stimulator that can be delivered minimally invasively through a percutaneous catheter which would significantly lower the barrier to entry for neuromodulatory treatment approaches because of the reduced risk. This miniaturization is achieved by using magnetoelectric materials to efficiently deliver data and power through tissue to a digitally-programmable 0.8 mm2 CMOS system-on-a-chip. As a proof-of-principle we show wireless stimulation of peripheral nerve targets both directly and from within the blood vessels in rodent and porcine models. The wireless EVNS concept described here provides a path toward minimally invasive bioelectronics where mm-sized implants combined with endovascular stimulation enable access to a number of nerve targets without open surgery or implantation of battery-powered pulse generators.
Competing Interest Statement
The authors have declared no competing interest.
Bioelectronic modulation of neural activity is a powerful tool for treating many disorders, especially when these disorders cannot be effectively managed with conventional therapies. For example, electronic devices that stimulate neural activity are effective for treating disorders like Parkinson’s Disease, epilepsy, chronic pain, hearing loss and paralysis [1–7]. These devices are most effective when implanted in the body where they can selectively stimulate the desired nerve targets; however, the invasiveness of the implantation can introduce additional risk for the patient. Invasive implants can also lead to complications such as chronic inflammation which can further degrade device functionality and lead to failure [8–10].
The vascular system that accompanies nerves as a part of the neurovascular bundle, provides a less invasive route for approaching nerve targets [11]. Existing neural implants for nerve targets such as the dorsal root ganglion (DRG) can suffer from site infection that results in device explantation and follow-up surgeries [12]. Endovascular neural stimulators (EVNS) delivered via an intravascular catheter to deep tissue targets with a minimally invasive procedure through the blood vessels within the body would leave the tissue target undisturbed. As a result, endovascular deployment of devices is often associated with significantly lower risk compared to open surgical approaches: recovery times are drastically reduced and site infections are extremely uncommon [11]. Given these advantages, an endovascular approach to neural stimulation would be attractive for the multitude of central and peripheral nerve targets that are adjacent to vascular structures, such as targets in deep brain, peripheral nerves, and the heart. [13–15]. Recently, several new endovascular bioelectronic devices have been developed that exemplify the benefits of stimulating neural tissue through the vasculature [16–19]. However, these devices have stimulation leads that are wired to pulse generators or centimeter-sized inductive coils. The long lead wires and implantation of centimeter-sized devices create additional failure points and require an open surgery that reduces some of the benefits of an endovascular surgical approach [20].
By miniaturizing the bioelectronic implants to a diameter of a few millimeters it would be possible to deliver endovascular neuromodulation therapies entirely with minimally invasive procedures that rely on percutaneous catheters. In order to sufficiently miniaturize the device to the size constraints of the catheter (< 3 mm in diameter), some form of wireless power is necessary to replace the bulkier batteries if we expect long-term operation. While several innovative wireless power transfer modalities have been demonstrated including far-field RF radiation, near-field inductive coupling, mid-field electromagnetics with hybrid inductive and radiative modes, ultrasound, and light; there has yet to be a demonstration of a millimeter-sized wireless neural stimulator that operates at a depth of several cm in a large animal model [21–34].
Here we turn to magnetoelectrics (ME) as a wireless data and power transfer technology due to its large power densities, high tolerance for misalignment, and ability to operate in deep tissue when compared to alternative wireless power technologies for bioelectronic implants [35].
Our results show for the first time that it is possible to safely stimulate peripheral nerves using electrodes placed inside the blood vessels, and that we can deliver the stimulation using a millimeter sized bioelectronic implant. By combining ME data and power delivery with a custom application specific integrated circuit (ASIC) we achieve a miniature device that is only 3 × 2.15 × 14.8 mm when fully encapsulated. Compared to miniature ultrasound powered devices, our MagnetoElectric-powered Bio ImplanT (ME-BIT) maintains functional power levels over a larger range of translational and angular misalignment, and does not need ultrasound gels or foams to couple energy from the transmitter. We also demonstrate a robust communication protocol that allows us to adjust the ME-BIT stimulation parameters. As a proof-of-concept, we show that these ME-BITs can be powered several centimeters below the tissue surface and can electrically stimulate peripheral nerve targets through the vasculature in a large animal model. These proof of principle studies open the door to minimally invasive bioelectronic therapies based on EVNS.
Results
Magnetoelectrics combined with a custom ASIC enables a millimeter-sized neural stimulator
To overcome the challenge of wireless data and power delivery to miniature bioelectronic implants, we developed a data and power delivery system based on magnetoelectrics, which achieves high power densities within the safety limits for human exposure [36]. Magnetoelectric materials provide efficient power delivery for bioelectronic implants by directly converting magnetic fields to electric fields based on the material properties [35, 37]. In our case we use a laminated bi-layer material that consists of Metglas, a magnetostrictive layer and lead zirconium titanate (PZT), a piezoelectric layer. When we apply a magnetic field to the material, the magnetostrictive material generates a strain that is coupled to the piezoelectric layer that, in turn, generates an electric field [35]. Thus, by applying an alternating magnetic field at the acoustic resonant frequency of the film we can efficiently deliver power to our implant [35,36,38, 39]. In addition to delivering power, we can also transmit data to our implant by modulating the frequency of the applied magnetic field. The frequency shift results in a change in the amplitude of the received voltage, which can be interpreted as a digital bit sequence that specifies the stimulation parameters for the implant [38,39]. Taken together, the complete wireless EVNS system consists of an external magnetic field transmitter, a ME film that harvests power and data from the magnetic field, and a custom integrated circuit that interprets the digital data and generates the electrical stimulus delivered by the electrodes (Fig. 1a). Figure 1b demonstrates a conceptual overview of the system implemented in a large animal model where a surface coil can be used to wirelessly transmit a magnetic field to power and program the implant for endovascular stimulation.
a. The overall EVNS system includes a magnetic field driver connected to a resonant coil that is tuned to the ME film resonant frequency. The coil outputs a frequency modulated magnetic field and switches between three different frequencies to modulate data and power received by the ME-BIT implant. The magnetic field is converted to an electric field by the magnetoelectric film. Specifically, the magnetostrictive layer, Metglas, mechanically deforms under the magnetic field and transfers the resulting strain to the piezoelectric layer, PZT, and induces a voltage. The amplitude of the voltage is then modulated by shifting the frequency of the applied field. The resulting voltage modulation received by the ME-BIT programs the custom IC to output the desired stimulus waveform. b. For our proof-of-concept experiments the ME-BIT is implanted proximally to a blood vessel deep within tissue and wirelessly powered through a magnetic coil in a pig. The implant’s stimulation lead is introduced into the vessel to stimulate nearby nerve targets. c. A rendering of the implant is shown with all the external components including the SoC, external capacitor, and the ME transducer. d. Photograph of the device fully packaged inside a 3D printed capsule resting in a clear sheath. The implant is encapsulated with a non-conductive epoxy prior to being implanted into the body and has the potential to be delivered endovascularly.
The ME-BIT itself consists of a magnetoelectric film with a size of 1.75 mm × 5 mm and a thickness of 0.3 mm for wireless power and data transfer, an ASIC for modulating the ME power and stimulation, and an external capacitor for energy storage as shown in the rendering in Fig 1c, which can be packaged to fit within an 11 French catheter. For our experiments we packaged the ME-BIT within a custom 3D printed PLA capsule with on-board electrodes that can also be used to power external electrodes. (Fig 1d) With this design, the miniature capsule can not only be delivered through a minimally invasive catheter, but also serve as a complete neuromodulatory device that can receive power, undergo programming, and transmit stimulation to neural tissue.
A custom magnetic field transmitter enables data and power transfer at centimeter depths within safety limits
To deliver data and power to the implant we designed a magnetic field transmitter that drives a high-frequency biphasic current into a resonant coil [38]. By maintaining transmitter power levels below 1W, we can achieve field strengths of > 1mT sufficient to power the ME-BIT at depths of 4 cm within the safety limits.
Because the amplitude of the ME voltage peaks at the acoustic resonant frequency, we can send digital signals to our ME-BIT by detuning the applied magnetic field frequency. Figure 2 shows our communication protocol with charging, data transfer, and stimulation phases. As seen in Figure 2b we can select 3 frequencies to transmit digital data. The first frequency “Data 1” corresponds to the mechanical resonance (345 kHz). This is the frequency of maximum voltage (and maximum power transfer), which we use as a digital 1, and for the charging and stimulation phases. The second frequency “Data 0” is detuned by ~ 5 kHz. This frequency of 350 kHz produces a lower amplitude voltage, which is used as a digital 0. The third frequency is detuned by 55 kHz from the resonance peak and produces an even lower voltage than the “Data 0” signal. This “Notch Frequency” of 400 kHz is used to indicate the start of the data transfer and stimulation phases. By using the mechanical properties for the ME film to receive data based on frequency modulation we can avoid turning the transmitter coil on and off, which would require a settling time of 100 us for our resonant transmitters. Given the fast settling time of this frequency modulation scheme we find that 64 cycles of the carrier frequency can reliably transmit one bit, resulting in a 4.6-kbps data rate. We used a digital payload of 18-bits per stimulation, which accounts for a preamble and real-time calibration of the demodulation reference. This payload combined with the charging phase yields a maximum stimulation rate of 1 kHz, which is well within the range of typical neural stimulation applications [38].
No comments:
Post a Comment