Wireless sensor implant for nerve monitoring

Abstract

An implantable assembly for detecting electrical activity in nerves or tissue for real time monitoring of neural regeneration interfaces with a nerve conduit surgically located at treatment site. A flexible cuff with an electrode assembly detects electrical signals generated within the nerve conduit and communicates the signals via a flexible interconnect to a control / sensor device with a power source, an analog front end, and a wireless communications module that transmits output signals to an external device. The flexible cuff, interconnect, and a control / sensor device are encapsulated in a flexible, biocompatible material that can be implanted at or near the treatment site.

Description

[0001] WIRELESS SENSOR IMPLANT FOR NERVE MONITORING RELATED APPLICATIONS

[0002] This application claims the benefit of the priority of U.S. Provisional Application No.

[0003] 63 / 764,548, filed February 27, 2025, which is incorporated herein by reference in its entirety.

[0004] GOVERNMENT RIGHTS

[0005] This invention was made with government support under HD 112026 awarded by the National Institutes of Health. The government has certain rights in the invention.

[0006] FIELD OF THE INVENTION

[0007] The invention relates to a wireless implantable sensor designed to interface with a peripheral nerve to monitor recovery after injury or therapy. More specifically, the device detects electrical impulses across the recovering nerve and wirelessly transmits those data to user device.

[0008] BACKGROUND

[0009] Infants and children are particularly susceptible to peripheral nerve injuries (PNIs) through the birthing process, trauma, tumors, and underlying medical conditions such as diabetes and autoimmune disorders. PNIs can affect a developing child’s quality of life, as sensorimotor interactions are a fundamental part of learning and development. Current repair strategies for PNIs after a complete nerve transection involve either suturing the distal and proximal nerve ends without introducing tension or placing an autologous nerve graft (autograft) harvested from some other part of the body to treat larger defects. Although the autograft is the current "gold standard,” it requires additional procedures to harvest the graft and often leads to neuroma formation and loss of function at the donor site. This approach is often a challenge in infants / children due to the growth, particularly when there is an injury of the brachial plexus or sciatic nerve, which can result in a defect up to a meter long. In such cases, there are limited donor sites for an autograft in infants and children, restricting the reconstructive options.

[0010] Because the fundamental goal of tissue engineering is to create materials that can replace or repair injured tissues, having tissue-engineered constructs that mimic the architecture of native tissues is desirable. However, current technology cannot recreate complex architectures that are three-dimensional (3D), span multiple length scales, haveinterconnected pores and features, are constructed from natural biomaterials, and provide feedback on functional recovery to limit the time between subsequent surgical interventions.

[0011] The ability to fabricate designer scaffolds with complex architectures (such as branches) that can be utilized for regeneration of more complicated nerve gaps, for example, nerve gaps joining the proximal stump of the common digital nerve with two distal stumps of the proper distal nerve, or to reconstruct the complex branching pattern of the extratemporal facial nerve. The goal of surgical nerve repair and implantation of NGCs is to regain function. Because the distance of the repair is often far from the end target organ muscle, the time to functional recovery can be up to 6 to 12 months. As such, especially when the initial therapeutic intervention was unsuccessful, a patient’s surgical options may be limited because of atrophy of the end organ muscle (resulting from denervation), and an additional surgical nerve repair attempt is typically not pursued. The pediatric patient is therefore forced to live with the debilitating paralysis or will undergo a sub-optimal surgical option, such as a static sling.

[0012] Projection-style 3D printing, using a digital micro-mirror array device (DMD) for massively parallel light projection has been previously disclosed. See, for example, the descriptions provided in International Publications No. WO2024 / 118942 and WO2015 / 179572, which are incorporated herein by reference. Advantages of the projection printing system lie in its capability to fabricate 3D scaffolds with complex microarchitectures that mimic the intricate tissue structures having high resolution (i.e., ~ 1 pm). In addition, it prints features in each layer simultaneously without point-by-point scanning and builds a 3D structure continuously from the first layer to the last, to provide much faster printing speeds. This projection printing approach can improve the mechanical integrity of the 3D-printed parts because the process does not create artificial interfaces within the part. This is particularly important for a nerve conduit, as the implant may experience external forces in the body, particularly among infants and children who will be growing over the period of functional recovery.

[0013] Beyond the ability to generate synthetic nerve structures, there remains a need for approaches to improve the time between surgical repair and the assessment of functional recovery, i.e., monitoring of nerve and muscle electrical activity. Such an approach should be designed to be implanted either independently or in tandem with a therapeutic autograft, decellularized allograft / xenograft or synthetic graft (e.g., nerve guidance conduit). Nerve regeneration post injury in human patients can take a year or longer to achieve maximumfunctional recovery. Therefore, the implantation of a wireless, battery-free sensor with a therapeutic intervention would enable measuring the progress and robustness of neural regeneration in real-time, compared to waiting for six or more months to determine whether functional regeneration is occurring as the physician expects for a given therapeutic approach. If sufficient nerve function can be confirmed at an early stage across the surgical repair site, the likelihood of surgical success is high, and no further intervention would be necessary until the nerve regeneration to target muscle is complete. On the other hand, if neural regeneration does not occur within the predicted timeframe, then knowing this earlier than currently possible would allow for an alternate surgical strategy (e.g., a second nerve graft) within the window of time before muscle atrophy occurs.

[0014] SUMMARY

[0015] An implantable assembly is provided for detecting electrical activity in damaged nerves that have been repaired by direct reconstruction via suturing or by a therapeutic autograft, decellularized allograft / xenograft or synthetic graft, collectively a “neural conduit”, for monitoring neural regeneration and functional recovery. A flexible cuff with an electrode assembly detects electrical signals generated within the nerve bundle and communicates the signals via a flexible interconnect to a sensor device with a power source, an analog front end and a wireless communications module that transmits output signals to an external device. The flexible cuff, interconnect, and a sensor device are encapsulated in a flexible, biocompatible material that can be implanted in close proximity to the treatment site.

[0016] In one aspect, the inventive monitoring scheme may further employ artificial 3D guidance conduits using a composite material of a synthetic poly(ethylene glycol) derivative (PEGDA) and naturally derived gelatin (GelMA) with intraluminal channels that match the specific size and shape of a patient’s nerve defect. These conduits are formed from native extracellular matrix (ECM) components and have linear micro-channels along the length for axon growth and side micro-holes for vascularization. The inventive printing approach also offers flexibility to create designer scaffolds, such as branched tubes (e.g., to mimic branched nerves at a plexus), with varying mechanical properties and gradients of various biomolecules. Conduits also will be embedded with a wireless sensor to allow for enhanced monitoring and early detection of functional recovery.Three-dimensional (3D) bioprinted engineered nerve guidance conduits have been successfully used for some peripheral nerve repairs and are a potential solution for some of the hurdles encountered in infants and children. An ideal nerve conduit is marked by several intrinsic properties including biocompatibility, vascularizable, patient specific, fit to match the size and shape of the damaged site of a specific patient, especially in a growing child, and intelligent, with an embedded wireless sensor for on-demand monitoring of recovery.

[0017] Nerve regeneration post injury in human patients can take a year or longer to achieve maximum functional recovery. A key goal of the inventive approach is to reduce the time between surgical implantation and the assessment of functional recovery. The inventive implantable wireless, battery -free device for monitoring nerve and muscle electrical activity can be implanted either independently or in tandem with a therapeutic autograft, decellularized allograft / xenograft or synthetic graft (e.g., nerve guidance conduit). The sensor facilitates measurement of the progress and robustness of neural regeneration in realtime, compared to waiting for six or more months to know if functional regeneration is occurring as the physician expects for a given therapeutic approach. If sufficient nerve function can be confirmed at an early stage across the surgical repair site, the likelihood of surgical success is high, and no further intervention would be necessary until the nerve regeneration to target muscle is complete. However, if neural regeneration does not occur within the predicted timeframe, then knowing this earlier than currently possible would allow for an alternate surgical strategy (e.g., a second nerve graft) within the window of time before muscle atrophy occurs.

[0018] In some embodiments, through integration of tissue engineering, 3D bioprinting, advanced biomaterials, and sensing, the inventive scheme provides functional nerve conduits for patient-specific nerve repair and regeneration. A key component of this approach includes an artificial nerve conduit 3D-printed as a hydrogel scaffolds that can be tuned by varying the bio-ink composition and bioprinting parameters. These conduits incorporate “vascularization passages” with micropores along the outer wall of the implants to act as passages for the penetration of vessel networks. Integration of a wireless sensor for on-demand, non-invasive electrophysiological data output has not previously been disclosed. In some embodiments, the sensor may be battery powered, either wirelessly rechargeable or non-rechargeable, or it may be wirelessly-powered and battery-free, powered through inductive coupling. Currently, no clinical solution exists for monitoring the success of a neural graft therapy in the early stages after surgery. Such an implantablesensor should be bio-inert and functional throughout the duration of recovery (i.e., months in mice and over a year for humans).

[0019] By integrating a wireless sensor with a nerve conduit, whether in the form of a therapeutic autograft, a decellularized allograft / xenograft, or a 3D-printed scaffold, the timing and robustness of neural regeneration can be measured within a defined time that is shorter than waiting for the nerve to reach the target muscle. If neural regeneration is confirmed across the surgical repair site, then the prospect of surgical success is high, and no further intervention should be necessary until the nerve regeneration to target muscle is complete. On the other hand, if neural regeneration is determined to incomplete within the predicted time, this knowledge allows for an alternate surgical strategy to be taken (e.g., a second nerve graft) within the critical window of time before muscle atrophy occurs.

[0020] The implantable control / sensor device comprises three primary components: electronics, serpentine (flexible interconnect), and electrode(s), each fabricated and encapsulated according to its functional role. A key feature of the control / sensor device lies in its effectiveness in enabling chronic experiments under in vivo conditions through the selection of layer-by-layer materials. Using a combination of Parylene, epoxy, and silicone, the electronic components are protected from bodily fluids, while the silicone coating provides a modulus similar to that of muscle and biological tissues.

[0021] The flexible cuff portion of the probe addresses the mechanical stresses applied to the device due to tensile forces arising from patient movement and internal anatomical bending. In addition, several processes are performed to allow the silicone layer to be properly encapsulated.

[0022] In one aspect, an implantable assembly for detecting electrical activity in nerves for monitoring neural regeneration includes: a flexible cuff comprising an electrode assembly configured for detecting electrical signals generated within the nerve conduit having a shape and dimensions configured to match a lesion to be treated for neural regeneration; a control device comprising a power source, an analog front end configured for receiving and processing the electrical signals to generate processed output signals, and a wireless communications module configured for transmitting processed output signals to an external device; and a flexible interconnect disposed between the electrode assembly and the control device, wherein the flexible interconnect and a control device are each encapsulated in a flexible, biocompatible material. In some embodiments, the nerve conduit can be a therapeutic autograft, a decellularized allograft / xenograft, a synthetic graft, or acombination thereof.. The power source may be an integrated battery, which may be configured for wireless recharging, or it may be an inductive resonant wireless power transfer device. In some embodiments, the nerve conduit comprises a 3D-bioprinted nerve conduit bioprinted from a polymer solution comprising PEGDA-GelMA and may comprise a plurality of microchannels wherein micropores are formed through sidewalls of the microchannels.

[0023] In some embodiments, the digital signal processing is configured to execute one or more algorithms for signal conditioning and signal filtering. Signal conditioning includes one or more of timestamp interpolation, detection of missing data, data filling or concatenating, and baseline removal, while signal filtering includes one or more of bandpass filtering, harmonic removal, and separation of frequency components based on spectral characteristics. Signal filtering may further comprise applying a Fourier transform to a filtered signal to generate an output power spectrum.

[0024] In another aspect, a method for in vivo monitoring neural regeneration within a nerve defect at a treatment site includes: implanting an assembly for detecting electrical activity in nerves in associated with the treatment site, the assembly comprising: a nerve conduit having a shape and dimensions configured to match a lesion corresponding to the treatment site; a flexible cuff comprising an electrode assembly configured for detecting electrical signals generated within the nerve conduit; a control device comprising a power source, an analog front end configured for receiving and processing the electrical signals to generate processed output signals, and a wireless communications module configured for transmitting processed output signals to an external device; and a flexible interconnect disposed between the electrode assembly and the control device, wherein the flexible cuff and a control device are encapsulated in a flexible, biocompatible material; and transmitting the processed output signals to an application configured for execution on an external device, wherein the processed output signals are stored for further processing and analysis. The nerve conduit may be a therapeutic autograft, a decellularized allograft / xenograft, a synthetic graft, or a combination thereof..

[0025] In some embodiments, the nerve conduit is a 3D-bioprinted scaffold, wherein, prior to implanting, the method includes generating an magnetic resonance image (MRI) of the lesion to determine the defect volume, and wherein a 3D bioprinter is programmed to generate the 3D bioprinted scaffold to match the defect volume. Additionally, prior toimplanting, the method may include quantifying 3D spatial overlap of the defect and the 3D bioprinted scaffold.

[0026] The power source may be an integrated battery, which may be configured for wireless recharging, or it may be an inductive resonant wireless power transfer device. In some embodiments, the 3D bioprinted scaffold is bioprinted from a polymer solution comprising PEGDA-GelMA and may comprise a plurality of microchannels wherein micropores are formed through sidewalls of the microchannels.

[0027] In some embodiments, the digital signal processing is configured to execute one or more algorithms for signal conditioning and signal filtering. Signal conditioning includes one or more of timestamp interpolation, detection of missing data, data filling or concatenating, and baseline removal, while signal filtering includes one or more of bandpass filtering, harmonic removal, and separation of frequency components based on spectral characteristics. Signal filtering may further comprise applying a Fourier transform to a filtered signal to generate an output power spectrum.

[0028] BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic illustration of a 3D bioprinting set-up for use in embodiments incorporating an artificial nerve conduit.

[0029] FIGs. 2A-2E show sample masks for an exemplary 3D-bioprinted conduit design and results of printing using such masks, where FIG.2A shows a CAD photomask of inner microchannel cross section for a conduit, FIG. 2B is a brightfield image of printed conduit cross section (scale bar: 100 pm); FIG. 2C is a brightfield image of conduit side view with hollow sleeves on each end and regularly spaced 50-pm pores along the central microchannel region (scale bar: 500 pm); FIG.2D is a CAD schematic of conduit; and FIG.

[0030] 2E is a zoomed image of 50-pm pores (scale bar: 200 pm).

[0031] FIG. 3A is a diagrammatic view of an embodiment of the inventive implantable assembly with exemplary 3D bioprinted nerve conduit(s) and control / sensor module; FIG.

[0032] 3B is a block diagram of an embodiment of the wireless implantable control / sensor for the recording of ENG signals; FIG. 3C diagrammatically illustrates an embodiment of tripolar Pt-black electrodes embedded in a soft and flexible nerve cuff; FIG.3D is a block diagram of another embodiment of the wireless implantable control / sensor; and FIGs. 3E - 3F are circuit schematics of an analog front end and main control unit, respectively, of the control device according to an embodiment.FIG. 4 is a block diagram showing an exemplary sequence for assembly of an embodiment of the control device.

[0033] FIG. 5 is a diagram of the process flow of an electronics module encapsulation according to an exemplary embodiment.

[0034] FIGs. 6A-6B are diagrams of an exemplary fabrication sequence for the serpentine interconnect as well as preparation of the electrodes before and after connection, where FIG.

[0035] 6A provides a top-down view of the process and FIG. 6B diagrammatically depicts stacked layering.

[0036] FIG. 7 provides photographic images of an embodiment of the control / sensor assembly with details enlarged to show attachment of the serpentine interconnect to the main electronics unit.

[0037] FIG. 8 is a block diagram depicting software macro-stages for analyzing electroneurogram (ENG) data collected by the inventive control device.

[0038] DETAILED DESCRIPTION OF EMBODIMENTS

[0039] The inventive scheme and method employ multiple technological innovations. A key element of the inventive approach involves employing an implantable wireless control / sensor for on-demand, non-invasive electrophysiological data output for monitoring the success of neural graft therapy, where an autograft, decellularized allograft / xenograft, or synthetic graft has been surgically implanted at an injury site for repair of a nerve defect. This implantable control / sensor is placed in electrical communication with a neural conduit via a flexible cuff with electrodes configured for detecting electrical signals within the neural conduit. In some embodiments, the nerve conduit is a synthetic graft in the form of 3D printed scaffold generated using a digital micro-mirror array device (DMD), such as described in International Publications No. WO2024 / 118942 and WO2015 / 179572, which are incorporated herein by reference.

[0040] A key innovation disclosed herein is an implantable wireless control / sensor device, which is disclosed in detail below. This device is designed to be implanted independently or in tandem near the site of implantation of a nerve conduit. For purposes of the present disclosure, the phrase “nerve conduit” refers to a therapeutic autograft, a decellularized allograft / xenograft, a synthetic graft, which may include a 3D-bioprinted scaffold as described further herein, or a combination thereof. While specific examples described and / or illustrated herein specifically relate to a synthetic nerve conduit, it will be recognizedby those in the art that the inventive control / sensor device is applicable for interfacing with all types of nerve conduits, natural or synthetic, for detection of electrical activity within the conduit. In some cases, it may even be used for monitoring of electrical activity in a nerve that has not necessarily been subject to injury, but may be suspected of having some other defect that is interfering with normal operation. In this case, it may be possible to simply expose an intact nerve conduit through an incision in the patient’s skin, apply the flexible cuff to the exposed conduit, and implant the device for extended monitoring of the electrical nerve transmission to, for example, generate data to assist in identifying the source of abnormal operation. In the examples discussed and illustrated herein, flexible cuff is depicted as a soft sheet-like structure that is bendable to conform to the physical features of the tissue in which electrical signals are to be detected. For example, as shown in FIG.3A, the cuff is shown as a ribbon-like structure with electrodes formed on the inner surface that contacts the nerve conduits by wrapping or looping the “ribbon” to at least partially encircle the nerve conduit. In such an application, the approach to affixed the cuff in place should avoid potential injury to the contact surface. However, it should be noted that the depicted physical configuration of the cuff in these examples as a wrap-around surrounding the conduit is not intended to be limiting. Rather, the shape and attachment approach for securing the cuff in place at or near the treatment site may vary depending on the application. For example, the flexible cuff can be in the form of a patch that is affixed to muscle tissue via a tissue-adhesive material or mechanism, e.g., sutures or staples. Accordingly, as used herein, the phrase “flexible cuff’ is not limited to the illustrated examples, but includes any electrode-supporting flexible assembly that can be secured in place in the desired position for in-vivo detection of electrical signals that are (or are expected to be) transmitted through the area of interest.

[0041] Certain aspects of the inventive approach relate to the design of an overall engineered nerve regeneration scheme for repair of nerve injury or defect employing a nerve conduits. FIG.3A, discussed below, provides a diagrammatic illustration of such a scheme. Before launching into the details of the implantable control / sensor device itself, the following section provides a description of an exemplary approach to nerve generation via an artificial nerve conduit.3D-bioprinted conduits

[0042] FIG. 1 provides a schematic diagram of proposed bioprinting system for implementation of the inventive scheme with a 3D-bioprinted nerve conduit. A visible LED light source projects onto a DMD chip which reflects a digitally controlled pattern through an optical rail. The patterned light projects into the prepolymer vat, curing the solution into a patterned hydrogel solid. The motorized build platform lifts synchronously with the projected light images to produce a smooth true 3D construct.

[0043] The 3D printer system is based on digital light projection (DLP), for rapid, parallel photopolymerization of bioinks, to fabricate complex 3D scaffolds. An example of an appropriate bioprinter is commercially available under the name BIONOVA X™ from Cellink (a BICO Company, Gothenburg, Sweden). The projection based DMD printing process produces a structure continuously in a layerless fashion without artificial interfaces for better structural integrity. Briefly, the 3D printer system 100 includes a short wavelength (e.g., blue light (405 nm)) source 102 for photopolymerization, a computer for sliced imageflow generation 104 and system synchronization, a Digital Micromirror Device (DMD) 106 for optical pattern generation, a set of projection optics 108, and a high-precision x-y-z stage 110. The DMD chip 106, which comprises approximately 4 million micro-mirrors, modulates the uniform incident light and projects an optical pattern dictated by user-defined images onto the photocrosslinkable solution supported in a vat or similar container on stage 110. Areas illuminated by light will crosslink and form the hydrogel network (cell-laden construct) within seconds for the entire layer, while leaving the unexposed regions uncrosslinked (as liquid). By moving the motorized build platform 112, a 3D construct can be fabricated continuously. This printing platform provides a printing speed that is 1,000 times faster than a traditional extrusion-based bioprinter and offers a printing resolution as small as 1 micron. Using 3D bioprinting system, PEGDA-GelMA microchannel conduits were generated for a sciatic nerve injury model. The system has been optimized to produce longitudinal microchannels with a 10-pm wall thickness (FIGs. 2A-2B) and 50-pm wide micropores along the side wall (FIGs.2C-2E). The printing solution contains 25% PEGDA (700 Da), 7.5% GelMA, 1% lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP), and 0.075% tartrazine, a photoabsorber (all percentages, unless otherwise stated, are weight by volume percent). In a pilot study in four mice (N = 2 per group) to evaluate the effect of a microporous versus non-porous conduit side wall on vascular infiltration and integration with the regenerating proximal axons, the microporous conduit had appreciably greatervascularization along its entire length as compared to the non-porous conduit. The morphology and extent of vascularization after one-month post-intervention in the microporous group appeared to at least match if not exceed that of the 3 -month postintervention autograft group.

[0044] Based on preliminary data, the materials selected to fabricate conduits were from a PEGDA-GelMA composite hydrogel network. PEGDA (700 Da, Sigma Aldrich) provides the mechanical integrity and GelMA provides integrin binding domains (i.e., RGD) for cell adhesion and motility. GelMA is synthesized from bioreagent grade gelatin Type A, 300g Bloom (Sigma Aldrich). Briefly, 25-27, 10% (w / v) gelatin is dissolved in 0.25 M 3:7 carbonate-bicarbonate buffer solution (pH 9) at 50°C. Methacrylic anhydride is added dropwise at a volume of 0.085 mL / (gram gelatin), and the reaction is allowed to run for 1 hour. The reaction then is quenched with HC1, the solution is dialyzed, frozen overnight at -80°C, lyophilized for 3 days, and stored at -80°C until further use. Ideally, a commercial grade version of GelMA will be used. Established commercial sources that are currently available include BICO and Sigma Aldrich. The prepolymer printing solution consists of a sterile 0.22-pm filtered solution of 25% PEGDA, 7.5% GelMA, 1% LAP (photoinitiator, Sigma Aldrich), and 0.075% tartrazine (biocompatible photoabsorber, Neta Scientific) prepared in PBS. Tartrazine is a ubiquitous nontoxic food additive commonly known as FD&C Yellow No. 529, which limiting the penetration of polymerization-inducing light into the bio-ink, leading to higher light absorption relative to scattering and enhanced the spatial resolution of the printing. Shape fidelity and mechanical properties can be tuned by modulating not only PEGDA, GelMA, and LAP concentrations, but also printing parameters, such as the light intensity and the light exposure time. The stiffness of the nerve conduits can be measured by a nanoindenter (Optics 11 Life Piuma) and / or micromechanical tester (CellScale MicroTester).

[0045] Using the micron scale printing resolution of the bioprinter, micropores are incorporated on the outer walls of the scaffolds to act as vascularization passages and facilitate the infiltration of the host vasculature, as shown in FIGs. 2C-2E. The rapid printing speed and digitalized process allows adjustment and optimization to various pore sizes (e.g., 5 pm to 50 pm) and densities (i.e., total number of pores by modifying pore-to-pore distance) for the optimal vascularization, while maintaining the structural integrity. In contrast to conventional circular microchannels, the conduit was designed with 300-pm-wide hexagonal microchannels to maximize the amount of longitudinal microchannels, which, in turn, provides more surface area topological guidance for regenerating axons.

[0046] The microporous conduit encourages increased vascularization of the regenerating nerve as is observed in autograft treatment and more so than the non-porous conduit group. Moreover, this increased vascularization leads to more robust neural regeneration, which in turn results in greater restoration of motor function. The rate of degradation of the scaffolds can be customized by using PEGDA of different molecular weights (i.e., 700Da, 3.4KDa, lOKDa, 20KDa) in combination with adjustment of printing parameters such as light exposure time and intensity to tune the crosslinking density and, therefore, optimize the degradation rate. If the conduit degradation rate needs to be increased beyond the PEGDA-GelMA composite material limits, it may be desirable to incorporate a hydrolytically degradable biopolymer poly(glycerol sebacate) acrylate (PGSA) into the prepolymer solution. Procedures for synthesizing and 3D printing with PGSA and PGSA-PEGDA composite are described by P. Wang, et al., in “3D Printing of a Biocompatible Double Network Elastomer with Digital Control of Mechanical Properties,” Adv. Funct Mater.

[0047] 2020 Apr;30(14), which is incorporated herein by reference.

[0048] While the transection nerve injury model provides a simplified model to assess neural regeneration capability, real world injuries rarely are as clean. These injuries comprise various types, both complete and partial and of various sizes, due to complex bone fractures, knee dislocations, lacerations, and ballistic trauma. Pediatric PNI adds additional complications for standardized treatment due to large variations in anatomical dimensions across both biological sex and age. For an off-the-shelf conduit therapeutic, the manufacturing method must be adaptable to and dependent on the size and scope of each individual patient’s volumetric defect. A robust and rapid 3D printing platform is well-suited to meet these dynamic manufacturing constraints. Due to the flexibility of the bioprinting system, any MR-imaged nerve defect region can be converted into a customsized hydrogel nerve conduit with micro-topological features.

[0049] J. Koffler J, et al. in “Biomimetic 3D-printed scaffolds for spinal cord injury repair,” Nature Medicine, 2019 Feb;25(2):263-269, incorporated herein by reference, describe the use of retrospective MRI scans of adult human spinal cord injury patients to 3D print a polymeric scaffold that tightly matches the injury’s volumetric defect. First, the dimensions of the volumetric defect are segmented and measured. Then, a CAD object is created within the bounds of the defect dimensions, where microchannels are specified along thelongitudinal axis. Using this procedure, 3D-printed polymeric scaffolds can be designed and executed using the 3D bioprinting system with not only varying overall dimensions, but also complex, irregular geometry that closely matches the defect volume, as determined by the MRI scan. Additionally, the bioprinting system can be used to produce human-scale branched peripheral nerve channeled conduit architecture, as described by W. Zhu W, et al. in “Rapid continuous 3D printing of customizable peripheral nerve guidance conduits,” Materials Today, 2018 Nov 1 ;21(9):951-959, incorporated herein by reference.

[0050] MRI data from clinical peripheral nerve injuries has demonstrated the generalizability of conduit design and fabrication for most peripheral nerve injury defects. Each nerve injury image is converted into a personalized digital print file that can be used to produce a geometrically matching microchannel nerve conduit for the injury. The printed construct dimensions can then be compared against the MRI volumetric defect via quantifying 3D spatial overlap. Glass and knife lesions are the most common cause of nerve injury among children. Other common nerve injuries tend to be caused by bone fracture or other blunt force trauma. Notably, complete nerve transections exhibit worse outcomes than crush injuries. In newborns, brachial plexus injuries can occur during childbirth and may require surgery to prevent permanent loss of function. Pediatric peripheral nerve injuries can occur in the head and neck, arms, and legs. Each region requires a unique conduit geometry. The combination of patient-specific data from MRI studies with the highly customizable printing capabilities of the inventive bioprinting system in terms of size and mechanical scaffold characteristics, provides for personalized design and fabrication of patient-specific nerve conduits for treatment and repair of peripheral nerve injuries.

[0051] FIG. 3A diagrammatically illustrates the basic elements of the inventive implantable scheme employing with 3D bioprinted nerve conduits 150, flexible nerve cuff 222 (with electrodes 212) positioned to detect electrical signals transmitted through the nerve conduits. Electrodes 212 are connected via an expandable, flexible serpentine interconnect 224 to implantable control / sensor device 202. For illustration purposes, the conduits and cuff are shown within lesion 130, with the conduit closely positioned relative to the lesion boundaries. The conduits are shown as elongated cylinders for ease of illustration, however, as described above, the versatility of the 3D printing system enables the conduits to be custom printed to closely fit the dimensions (depth, width, length, shape) of the defect. Control / sensor device 202 may be implanted near to, but generally not within the lesion itself so as not to impact the regeneration in the lesion. As previously noted, the 3Dbioprinted nerve conduit depicted in the figure provides an example of a nerve conduit and is not intended to be limited. As will be recognized by those of skill in the art, the nerve cuff, electrodes, and sensor device are designed for use with any type of nerve conduit, synthetic or natural.

[0052] Integrated Implantable Control / Sensor

[0053] A key advancement of the inventive scheme is the integration of the nerve conduits with a wireless sensor to enable assessment, in real-time or near real-time, of the nerve fiber growth across the surgical repair site. By integrating a wireless sensor with the nerve conduit through the use of a flexible nerve cuff, the timing and robustness of neural regeneration can be monitored within a defined time that is significantly less than waiting for the nerve to reach the target muscle. If neural regeneration is confirmed across the surgical repair site, the prospect of surgical success is high, and no further intervention should be necessary until the nerve regeneration to target muscle is complete. On the other hand, if neural regeneration is confirmed to not be occurring within the predicted time, this knowledge allows for an alternate surgical strategy to be taken (e.g., a second nerve graft) within the window of time before muscle atrophy occurs.

[0054] Prior work has demonstrated advanced wireless epidermal technologies that record physiological signals, such as electromyography (EMG) in humans. See, e.g., Y. Liu, et al., “Intraoperative monitoring of neuromuscular function with soft, skinmounted wireless devices,”, Digital Med. 2018 May 23; 1(1): 1—10.3 (incorporated herein by reference). The devices consist of (1) a flexible printed circuit board where all electrical components are mounted and (2) a set of Ag / AgCl electrodes that interface with the skin. Devices then are fully encapsulated in a soft, moisture resistant elastomeric layer. This device contains a system on a chip with BLUETOOTH® low energy (BLE) wireless communication capability. The BLE chip includes a 64-bit processor and 12-bit analog-to-digital converter (ADC) used to operate the device and collect data from the sensors. The epidermal sensors, coupled to the targeted tissue via a pair of low impedance, Ag / AgCl electrodes (single lead configuration), collect high quality EMG signals with amplitudes of a small fraction of a millivolt. These data are further processed on-board with analog front-end (AFE) modules, which include (1) a Notch 60 Hz rejection filter to remove electrical noise from power lines and (2) a 0.5-125 Hz bandpass filter and an amplifier. The resulting signal is digitalized with the ADC at 250 samples per second and 12-bit resolution. This technology platform as anassistive tool in intraoperative neurological procedures in humans. In one case, the devices record EMG signals from the tibialis anterior muscle group during stimulation of the common peroneal nerve in real-time, while stimulating the L5 spinal nerve during spinal cord surgery. The results exhibit signal quality and signal-to-noise ratio (SNR) that are comparable to those of both clinical gold standard needle electrodes with wired interfaces and expensive data acquisition systems. Additional research resulted in an implantable wireless Bluetooth device that records continuous electroencephalogram (EEG) signals from a rat in vivo model. Such a device samples EEG signals at 256 samples per second from subdermal stainless-steel screw electrodes attached to the skull and digitalizes with 12-bit resolution. Similar wireless, bio-integrated devices, capable of recording diverse types of physiological signals (ECG, EEG, SpO2, RR, and HR) provide the technological foundations for the inventive system.

[0055] Adopting these basic principles and building further on them, embodiments of the inventive control / sensor device can implemented in two main design architectures: 1) a battery-powered device, which employs a lithium battery as the primary power source and includes two sub-variants: wireless rechargeable battery and non-rechargeable; and 2) wirelessly powered (battery-free) device. This latter embodiment operates without a battery and relies on near-field inductive coupling for power delivery. It integrates a Near-Field Communication (NFC)-based wireless power transfer (WPT) module, eliminating the need for onboard energy storage.

[0056] A key innovation of this approach is the implementation of a fully wireless neural recording and stimulation system capable of operating either battery-free or with wireless recharging, while maintaining stable analog performance in a compact form factor. Achieving high-fidelity neural acquisition under power constraints represents a significant engineering challenge addressed by the inventive design.

[0057] The inventive control / sensor device includes three primary components: electronics, serpentine (flexible interconnect), and electrode(s), each fabricated and encapsulated according to its functional role.

[0058] An important feature of this invention lies in enabling chronic experiments under in-vivo conditions through the selection of layer-by-layer materials. By combining Parylene, epoxy, and silicone, the electronic components are protected from body fluids, while the silicone coating provides a modulus similar to that of muscle and biological tissues.Referring to FIG.3B, the cuff portion of the probe addresses the mechanical stresses applied to the device due to tensile forces arising from subject movement and internal anatomical bending. In addition, several processes are performed to allow the silicone layer to be properly encapsulated.

[0059] FIG. 3C provides a block diagram of the basic elements of the wireless implantable control / sensor 202. In some embodiments, the sensor device architecture employs a low-power BLUETOOTH® low energy (BLE) system on a chip 210 (e.g., CC2640R2F, Texas Instruments) that supports bidirectional wireless data transmission with one or more external data logging device 220, such as a smart phone, tablet, or a computer.

[0060] The BLE connectivity enables transmission of collected data to a mobile application (e.g., iOS), which allows users to control device settings and record the wirelessly transmitted data using device 220. The application can define measurement parameters, receive data transmitted by the device via BLE, visualize in real time the transmitted signal, and store the data in a spreadsheet format for further analysis. In some implementations, a predictive algorithm may be included in an application on the external device to anticipate at an early time point the quality of end-point functional healing to determine if a follow-up interventional surgery would be needed for ensuring expected functional outcome.

[0061] The implantable device form factor includes a centralized unit, on the order of 10 mm x 16 mm in size, formed from a thin polyimide substrate that contains all the electronic components and a peripheral probe that connects a pair of platinum black electrodes. The device is encapsulated with materials that are FDA-approved for implantable medical device applications. First, a layer of 15 pm parylene C is conformally deposited on the device, and then a polydimethylsiloxane (PDMS) is used to cover the device with a soft encapsulation. The exposed epineural black platinum electrodes 212, in tripolar configuration, embedded in a mechanically compliant soft cuff, interfaces with the sciatic nerve to record electrical impulses, either unstimulated or externally stimulated. FIG. 3C provides a diagram of an assembly for the flexible cuff. The neural electrical signals are conditioned on board with an analog front-end (AFE) module 214, which may include a rejection Notch filter that removes the 60 Hz electrical noise, a band pass filter (100Hz - 2KHz), and an amplifier to extract the neural electrical signal envelope.

[0062] In some embodiments, AFE module 214 may include an optional electrostimulation unit that generates a stimulating electrical pulse output to electrodes for chronic therapeutic treatment during monitored recovery. In embodiments with the stimulation option, one ormore second (independent) flexible cuff may be connected to same control device for multiregion interfacing of the injured tissue. The first cuff may provide for delivering stimulating electrical pulse output for chronic therapeutic treatment during monitored recovery and for capturing stimulated compound muscle action potential (CMAP) data. The second cuff may be used to monitor electrical activity for recording both stimulated CMAP and unstimulated electromyogram (EMG) data.

[0063] In some embodiments, device is designed to operate in a battery-free fashion through resonant inductive wireless power transfer, to enable a thin form-factor (~2 mm) to facilitate implantation conditions. Referring to FIG. 3D, wireless power transfer is implemented with: (1) a transmission antenna (within wireless power transfer module 230), in the form of a coil driven with a radio frequency (RF) current, and (2) a receiving antenna 218, embedded in the implanted device 202, i.e., control / sensor device 202, in the form of a planar coil, in resonance with the transmission antenna, to harvest electrical power. This power is rectified and continuously stored in a capacitor bank (i.e., two supercapacitors of 11 mF each and a 220 pF ceramic capacitor) to power the device during operation.

[0064] In both the wireless rechargeable and wirelessly powered versions, power management is based on an NFC wireless power transfer module operating at 13.56 MHz. Energy is transferred via resonant magnetic induction between a planar receiving antenna 218 integrated on the device, and an external transmitting antenna driven by an RF power source 230. The received AC (alternating current) signal is converted to DC (direct current) using a half-wave rectifier composed of two Schottky diodes and a smoothing capacitor. The rectified voltage feeds an ultra-low-noise, high power-supply rejection ratio (PSRR), and low-dropout regulator (TPS7A20), which generates stable DC supply voltages for the analog and digital circuitry. The selection of a low-noise, high-PSRR regulator was critical to prevent power supply noise from degrading the analog front-end performance.

[0065] The wirelessly powered (battery-free) embodiment version includes a capacitor bank 206 to maintain device operation during temporary drops in transferred wireless power.

[0066] Wireless rechargeable battery-powered embodiment replaces the capacitor bank 206 with a lithium battery charger (LTC4065) for charging battery 208. A non-rechargeable battery-powered embodiment uses a lithium battery directly regulated by the TPS7A20 LDO. Regardless of whether it is rechargeable or not, the battery life should provide sufficient operational power for at least 16 weeks, and preferably up to 6 months or longer.The analog front-end (AFE) 214 is responsible for neural signal acquisition and electrical stimulation. A key design goal is to design a simple and compact circuit capable of maximizing common-mode rejection and minimizing baseline noise, despite size and power constraints. The AFE consists includes an instrumentation amplifier stage, an active band-pass filter, and an electrical stimulation module. An exemplary circuit schematic for the AFE is shown in FIG. 3E, which provides values for individual components.

[0067] Neural signals are low-amplitude and highly susceptible to common-mode interference and baseline noise. A key performance parameter in this context is the Common-Mode Rejection Ratio (CMRR). CMRR is the ratio between differential gain and common-mode gain. It quantifies the ability of the amplifier to reject signals that appear simultaneously at both inputs (e.g., environmental noise, motion artifacts, and power-line interference). A higher CMRR results in improved noise rejection and signal integrity. The ability to maintain high CMRR in a small, power-constrained, wireless system was achieved through use of high-precision resistors and capacitors to improve input balance at the instrumentation amplifier. A low-noise, high-PSRR regulator is provided to prevent supply ripple from coupling into the signal. Passive high-pass filters (10 Hz cutoff) are employed at both INA inputs to provide a defined bias current return path avoiding signal degradation, reduce electrode polarization effects, minimize baseline drift and improve effective CMRR.

[0068] Signals from the electrodes 212 first pass through passive high-pass filters (10 Hz cutoff frequency) before entering the AD8426B instrumentation amplifier. This amplifier was selected for high input impedance, high intrinsic CMRR, and low noise performance.

[0069] An active band-pass filter based on the NSC2333 is used, with cutoff frequencies of 100 Hz (low) and 2 kHz (high). This stage isolates the frequency band associated with nerve activity while attenuating low-frequency drift and high-frequency noise.

[0070] An electrical stimulation module within the AFE is in communication with a microcontroller in module 210, which controls a digitally implemented R-2R ladder digital -to-analog converter (DAC), enabling programmable stimulation waveforms. This allows a controlled amplitude stimulation, and configurable pulse width and timing. The integration of stimulation and recording within the same wireless, battery-free / rechargeable platform increases the versatility of the system for neuromodulation applications.

[0071] An exemplary circuit schematic for main control unit (MCU) 211) of the Bluetooth chipset 210 is provided as FIG. 3F, including individual component values. Within MCU 211, analog-to-digital conversion is performed using the built-in ADC of the low-powermicrocontroller (CC2640R2F). The ADC supports configurable sampling rates ranging from 1 kSps up to 10 kSps, allowing the system to capture biosignals across different temporal resolutions depending on the application. This flexibility enables trade-offs between signal fidelity, power consumption, and data throughput. The digitized signals are then processed and transmitted wirelessly. Additionally, optimization of ADC acquisition parameters (integration time and sample averaging) further improved the effective signal-to-noise ratio.

[0072] The inventive control / sensor device is designed with a fully reconfigurable data acquisition framework, enabling adaptation to a wide range of experimental scenarios and application requirements. Key acquisition parameters can be dynamically configured via firmware and remote BLE commands, including sampling rate: IKSps up to lOKSps; integration time: ADC sampling window during which the input signal is integrated to form a single digital sample. Ranging from 5.3 ps to 21.3 ps; and sample averaging: configurable number of averaged samples to improve signal -to-noise ratio (SNR).

[0073] Communication and control is based on the low power microcontroller CC2640R2F within MCU 211. It coordinates all device functions and provides Bluetooth Low Energy (BLE) connectivity, enabling the transmission of collected data to a mobile application 220 (e.g., iOS). Through the mobile application, data can be visualized in real time and exported for further analysis. Device parameters and operating modes can also be remotely configured.

[0074] Performance parameters for an embodiments of the control / sensor device are listed in Table 1 below:

[0075]

[0076]

[0077] TABLE 1

[0078] The electronics module, which includes AFE 214, Bluetooth chipset 210 and power source 216, is implemented on a two-layer polyimide (PI) flexible PCB (FPCB), referred to collectively as electronics module 240. Electronic components are assembled using soldering paste, melted at approximately 160 °C to form reliable connections. After assembly, the firmware is flashed and functional verification is performed. Once operation is confirmed, epoxy is applied on both sides as a first encapsulation step, followed by Parylene-C coating as a secondary encapsulation.

[0079] In an exemplary embodiment, the serpentine (flexible interconnect) 224 is fabricated on a three-layer Cu-Polyimide-Cu flexible printed circuit board (FPCB) and laser-cut to the desired mechanical pattern, which has a serpentine or zig-zag shape that is capable of flexing, expanding and bending to physically conform to the area at which the device is to implanted. Fabrication details are provided in the discussion referring to FIGs. 6A-6B below. The serpentine construction provides flexibility and strain relief, mechanically and electrically interfacing the electronics with the electrode. The electrode assembly 212 is the neural interface that directly contacts the nerve to provide electrical conduction between the control / sensor and the nerve. Electrodes are laser-cut from platinum (Pt) foil to the required geometry. The electrodes support a pre-stop function for accommodating variations in nerve diameter. The electrode thicknesses are on the order of 25 pm.

[0080] Electrodes 212 may be fabricated using a variety of different materials depending on application needs. In some embodiments, gold (Au) and platinum (Pt) are used. For gold electrode fabrication, gold can be deposited via e-beam evaporation under vacuum using fabrication procedures that are known in the art. Patterning is achieved using a shadow mask,eliminating the need for a separate etch step. Alternative procedures for deposition and patterning of gold are well known in the art. For platinum foil electrode fabrication, Pt foil is laser-cut to the target electrode geometry and dimensions.

[0081] Referring to FIG. 4, electrical and mechanical connection between the electrode assembly 212 and the main FPCB (via the serpentine interconnect 224) is achieved by soldering or welding in step 250, selected based on assembly constraints and reliability targets. Electrode assembly 212 supports a pre-stop feature that enables compatibility with a range of nerve diameters. The electrode thickness is selected to be approximately 25 pm to balance conformability and handling.

[0082] In step 252, encapsulation is tailored for the different device parts to (i) protect the system from body fluids, (ii) reduce foreign body response upon implantation, and (iii) improve mechanical integration with surrounding tissue over the long term. Different material stacks are applied to the electronics on the main FPCB, serpentine 224, and electrode assembly 212 elements to achieve these goals.

[0083] The electronics module encapsulation process 252 employs a three-layer sequence shown in FIG. 5. In step 260, medical epoxy (e.g., Loctite EA-M31CL) is applied and cured at approximately 75 °C to create a robust primary moisture and mechanical barrier. In step 262, a uniform Parylene-C film is deposited following epoxy cure to enhance dielectric isolation and moisture resistance. The target thickness for this film is on the order of 10 pm. To enhance adhesion of parylene, the following can be performed: application of A-174 Silane (methacryloxyproplytrimethoxysilane), dip coating (i.e., surface treatment), vapor deposition through parylene coating, etc. This process can generate covalent bonds between the coating target and parlyene, as parylene does not typically bond well inherently with other materials. In step 264, the electronics module is connected to the (previously coated) serpentine interconnect 224 after which, in step 266, a layer of biocompatible silicone, such as Ecoflex™ 00-30 silicone (from Smooth-On, Inc.) or similar, is mold-cured to a target thickness of 1.5 mm in a mold conforming to the device geometry. This layer provides improving softness and tissue compatibility.

[0084] FIGs. 6A-6B provide fabrication sequence for the serpentine interconnect as well as preparation of the electrodes before and after connection. FIG. 6A provides a top-down view of the process, while FIG. 6B diagrammatically depicts the stacked layering. Note that not all steps are shown in each figure. Starting at step 271, the base starting film for the serpentine is a flexible Cu-Polyimide-Cu printed circuit board (FPCB). In step 272, theconductor is deposited or applied to and patterned on the FPCB using a process appropriate for the selected conductor. In the illustrated example, the conductor is platinum, which is applied to the FPCB as a foil that is laser-cut to the required geometry. In step 273, the FPCB material is laser cut to define the dimensions of the flexible cuff by tracing the outline of the conductive lines defined in step 272. In step 274, the cuff electrode joint between the serpentine and the electrode assembly is formed by soldering or welding, which procedure is selected based on electrical, thermal, and mechanical criteria.

[0085] In step 275, index-matching metals such as titanium (Ti) or chromium (Cr) are deposited on the bottom of the electrode(s) to enhance layer adhesion, acting as a cushion between different mechanical properties of silicon and metal. Specifically, this metal layer, which may be on the order of 10 nm in thickness, is introduced to improve interfacial bonding — facilitating lattice / chemical compatibility between metallic and non-metallic layers. The index-matching metal can be deposited using a known technique, for example, as e-beam or sputter deposition.

[0086] In step 276, silicon dioxide (SiCh) is deposited on the electrode backside by e-beam, sputter, spin-coating, or other appropriate technique as is known in the art. The Si O2 layer, with a target thickness of approximately 50 nm, mirrors the serpentine strategy. The outermost silicone surrounding the electrode region is fabricated to a target thickness of 100 pm to achieve a thin, compliant interface.

[0087] In step 277, a spin-coated silicone film is UVO-bonded to the backside of both the serpentine and electrodes to establish a strong base silicone layer. The UVO treatment generates an OH' group on both the silicone and SiO2 layer so that both layers form a physical bond after the treatment. In step 278, a top silicone layer is subsequently applied either by brushing or by hot pressing at ~ 80°C - 100°C, forming a complete top-bottom silicone encapsulation around the joined subassembly. These steps help establish robust electrode-interconnect bonding and enable strong silicone-inorganic interfaces through UVO-assisted covalent bonding.

[0088] Referring to FIG. 7, the serpentine 224 is then soldered to the electronics FPCB 240 to establish electrical continuity. The joint region is reinforced with epoxy 242 to improve mechanical durability and long-term reliability under flexure and handling. The assembled device is placed into a custom silicone mold, and silicone (Ecoflex) is injected and cured to finalize the outer encapsulation. This step yields an integrated implantable device exhibiting biocompatibility, flexibility, and environmental sealing suitable for operation inphysiological conditions. In some implementations, the interconnect wire and cuff may be bioresorbable, which may include programmable degradation through the application of an external stimulus.

[0089] Data Analysis Workflow

[0090] The software algorithm for data analysis was written in MATLAB® and includes the following modules: preprocessing, processing, and analytics. The preprocessing module consists of first a Hampel filter to eliminate extraneous spikes in the raw signal, then the baseline is estimated using one of three methods: envelope averaging, global average, or low-pass filtering. The envelope averaging consists of determining the upper and lower envelopes of the signal and averaging them to obtain the baseline. The global average consists of calculating the mean value of the entire signal. The low-pass filtering employs a Butterworth low-pass filter to smooth the signal and estimate its baseline. The estimated baseline is then subtracted from the raw signal to center the data around zero. The processing module consists of multiple filter passes to reduce any noise. A notch filter is first applied for removing 60 Hz environmental noise. A band-stop filter is then applied to remove oscillatory noise found between 100 Hz and 200 Hz. Then a band-pass filter from 100 Hz to 2 kHz is applied to isolate the neural activity. A wavelet filter is next applied to extract axon potential waveforms; the signal is decomposed into its wavelet components and reconstructed using only those that passed a specified threshold.

[0091] An analytics module performs peak sorting, signal identification, and signal measurement. Signal peaks are chosen that have the characteristic frequency components of axon potential, are classified as nerve pulses, and are displayed to the user. The processed signal containing only validated nerve pulses is saved and identified as an electroneurogram (ENG) signal in a .m or .csv file. A tool within the module enables loading, visualizing, and analyzing previously stored electroneurogram signals, providing measurements of amplitude, duration, and frequency components.

[0092] Referring to the block diagram in FIG. 8, software for analyzing electroneurogram (ENG) data comprises two macro-stages: signal conditioning 900 and signal filtering / transformation 920. In the first stage 900, the data are organized and cleaned, intermediate values are computed, sampling irregularities are detected, missing-data intervals are corrected, and the signal is normalized via baseline removal. In the second stage 920, sequential digital filters are applied to suppress undesired components, includingButterworth filtering 922, a notch filter 924, and wavelet filtering 926. Thereafter, a spectral representation is computed using a Fourier transform 928. As a result, the user can visualize both the processed signal 930 and its power spectrum 932, facilitating identification of relevant ENG parameters.

[0093] In signal conditioning block 900, the raw data are stored in a 121 -column matrix for timestamp interpolation in block 902. The first column corresponds to a series of timestamps; the second column contains a code associated with the data format; the third column indicates the channel in which the measurement was performed; and columns 4 through 121 (a total of 118 columns) contain the raw samples measured by the device. In this manner, each timestamp is associated with 118 measurements.

[0094] To enable the application of subsequent signal conditioning and filtering algorithms, the system estimates the time instant associated with each of the 118 measurements. In one implementation, the algorithm divides the time interval between two consecutive timestamps into 118 sub-intervals and assigns a time value to each sample using linear interpolation. This procedure is repeated for each pair of consecutive timestamps throughout the record, thereby transforming the original matrix into a time vector and a signal amplitude (sample) vector.

[0095] During temporal interpolation 902, one or more data packets (each packet contains 118 samples) may be lost, such that the separation between two consecutive timestamps is greater than the expected separation. Based on this deviation, the algorithm automatically identifies segments with missing data in block 904. Once detected, the algorithm reconstructs the signal using one of two user-selectable strategies in block 906.

[0096] In one mode, the missing segments are filled with zeros (or an equivalent neutral value) to preserve the original length of the vector and maintain the original signal timing. This mode is useful when preserving the signal’s temporal and frequency characteristics is prioritized, for example, for event counting over fixed time intervals, comparison between sessions performed over the same time interval, or spectral analysis of the signal.

[0097] In an alternative approach, the algorithm avoids inserting artificial samples and instead rearranges the available samples, assuming continuous measurements and prioritizing that the end time of the reconstructed signal is the same as that of the raw data. This mode prioritizes rapid inspection and visibility of the overall signal morphology, facilitating qualitative trace review and visual identification of trends or segments of interest.After handling missing data, the algorithm performs baseline removal in block 908 to center the signal around the zero line, thereby reducing effects caused by motion, changes in the electrode / nerve contact, impedance variations, or system offsets. Baseline removal also improves the robustness of subsequent filtering and stabilizes the thresholds used in wavelet filtering.

[0098] Signal filtering block 920 includes Butterworth filter 922 in which the conditioned signal is processed to suppress components outside the band of interest (e.g., 100 Hz to 10 kHz), as well as to isolate specific bands that contribute noise to the signal. The user has control over these filters and may apply multiple filters if desired. This step yields a cleaner signal containing primarily relevant information.

[0099] Notch filter 924 is applied to attenuate narrowband periodic interference associated with environmental sources (for example, power-line coupling or other dominant sources), or signal clipping events that generate harmonics. The user also has control over this stage and may apply as many filters as needed at specific frequencies and their multiple harmonics.

[0100] Wavelet filter 926 is used to separate frequency components whose amplitude, at certain times, becomes relevant and should be considered. Thresholds were estimated based on the spectral characteristics of electroneurogram signals. In principle, this final filter retains primarily information associated with the ENG signal while reducing the associated noise. The user may also adjust these thresholds if desired.

[0101] The primary output of the data analysis algorithm is filtered data 930, which generated after the signal conditioning and digital filtering stages. This signal is suitable for event detection, pulse quantification, time-window analysis, and computation of derived metrics relevant to clinical assessment and monitoring. The filtered signal also serves as a reference trace for visual inspection and for verifying the effect of user-selected filtering parameters.

[0102] In parallel with the primary output, the system may generate a power spectrum 932 from the processed signal using a Fourier transform 928, providing a quantitative spectral characterization. This representation may be used to: (i) validate signal quality and identify interference, (ii) identify dominant bands associated with noise or physiological activity, and (iii) support the selection and tuning of processing parameters and thresholds (e.g., notch frequencies or filter bands) according to the acquisition context.

[0103] Electrode integration with soft tissues is challenging because of the large mechanical moduli mismatch, especially in chronic implantation, which could lead to tissue damage.The foregoing description details an inventive approach in which nerve conduits, whether natural or artificial, are integrated with a wireless control / sensor that provides robust ENG data output across regenerating sciatic nerves. The use of soft elastomeric materials for construction of cuff electrodes provides for the a safe, biocompatible interface for chronic implantation. The inventive approach represents a significant advancement over the existing art in that it is designed to be implanted in the same surgical intervention as the primary therapy (e.g. nerve graft, surgical reconstruction). The device can remain implanted at the site of injury until the patient has fully recovered or the device has been fully degraded and even possibly resorbed (typically one year after surgical intervention). The wireless interface with a dedicated application downloaded on an external device such as a smart phone, tablet, or personal computer allows a patient to personally monitor their recovery or provides quick and easy monitoring and data tracking capability for a medical professional during routine follow-up visits.

Claims

CLAIMS:

1. An implantable assembly for detecting electrical activity in nerves for monitoring neural regeneration, the device comprising:a flexible cuff comprising an electrode assembly configured for detecting electrical signals generated within the nerve conduit having a shape and dimensions configured to match a lesion to be treated for neural regeneration;a control module comprising a power source, an analog front end configured for receiving and processing the electrical signals to generate processed output signals, and a wireless communications module configured for transmitting processed output signals to an external device; anda flexible interconnect disposed between the electrode assembly and the sensor device, wherein the flexible interconnect and a sensor device are each encapsulated in a flexible, biocompatible material.

2. The implantable assembly of claim 1, wherein the nerve conduit is a therapeutic autograft, a decellularized allograft / xenograft, a synthetic graft, or a combination thereof.

3. The implantable assembly of claim 1, wherein the power source comprises an integrated battery.

4. The implantable assembly of claim 3, wherein the integrated battery is configured for wireless recharging.5 The implantable assembly of claim 1, wherein the power source comprises an inductive resonant wireless power transfer device.

6. The implantable assembly of claim 1, wherein the nerve conduit is synthetic graft comprising a 3D scaffold bioprinted from a polymer solution comprising PEGDA-GelMA.

7. The implantable assembly of claim 1, wherein the flexible interconnect is configured as a serpentine structure for expansion and strain relief.

8. The implantable assembly of claim 1, wherein the analog front end is configured to execute one or more algorithms for signal conditioning and signal filtering.

9. The implantable assembly of claim 8, wherein signal conditioning includes one or more of timestamp interpolation, detection of missing data, data filling or concatenating, and baseline removal.

10. The implantable assembly of claim 8, wherein signal filtering includes one or more of bandpass filtering, harmonic removal, and separation of frequency components based on spectral characteristics.

11. The implantable assembly of claim 8, wherein signal filtering further comprising applying a Fourier transform to a filtered signal to generate an output power spectrum.

12. The implantable assembly of claim 1, wherein the electrode assembly further comprises one or more stimulation electrode associated with at least one second flexible cuff, and the control module further includes an electrostimulation unit configured for generating simulation pulses to the one or more stimulation electrodes.

13. A method for in vivo monitoring neural regeneration within a nerve defect at a treatment site, the method comprising:implanting an assembly for detecting electrical activity in nerves in associated with the treatment site, the assembly comprising:a nerve conduit having a shape and dimensions configured to match a lesion corresponding to the treatment site, wherein the nerve conduit is configured to promote vascularization and integration with tissue adjacent the nerve defect;a flexible cuff comprising an electrode assembly configured for detecting electrical signals generated within the nerve conduit;a sensor device comprising a power source, an analog front end configured for receiving and processing the electrical signals to generate processed output signals, and a wireless communications module configured for transmitting processed output signals to an external device; anda flexible interconnect disposed between the electrode assembly and the sensor device, wherein the flexible cuff and a sensor device are encapsulated in a flexible, biocompatible material;transmitting the processed output signals to an application configured for execution on an external device, wherein the processed output signals are stored for further processing and analysis.

14. The method of claim 13, wherein the nerve conduit is a therapeutic autograft, a decellularized allograft / xenograft, a synthetic graft, or a combination thereof.

15. The method of claim 13, further comprising prior to implanting, generating an magnetic resonance image (MRI) of the lesion to determine the defect volume, wherein the nerve conduit is a synthetic graft comprising a 3D scaffold bioprinted from a polymer solution comprising PEGDA-GelMA, and wherein a 3D bioprinter is programmed to generate the 3D scaffold to match the defect volume.

16. The method of claim 15, further comprising, prior to implanting, quantifying 3D spatial overlap of the defect and the 3D scaffold.

17. The method of claim 15, wherein the 3D scaffold comprises linear microchannels extending along a length of the lesion and having side micro-pores through channel sidewalls for vascularization.

18. The method of claim 13, wherein the integrated battery is configured for wireless recharging.

19. The method of claim 13, wherein the power source comprises an inductive resonant wireless power transfer device.

20. The method of claim 13, wherein the power source comprises an integrated battery.

21. The method of claim 13, wherein the flexible interconnect is configured as a serpentine structure for expansion and strain relief.

22. The method of claim 13, wherein the analog front end is configured to execute one or more algorithms for signal conditioning and signal filtering.

23. The method of claim 22, wherein signal conditioning includes one or more of timestamp interpolation, detection of missing data, data filling or concatenating, and baseline removal.-SO- 24. The method of claim 22, wherein signal filtering includes one or more of bandpass filtering, harmonic removal, and separation of frequency components based on spectral characteristics.

25. The method of claim 22, wherein signal filtering further comprising applying a Fourier transform to a filtered signal to generate an output power spectrum.

Images