An intramedullary nail system for real-time monitoring of bone lengthening healing status

Abstract

This invention discloses an intramedullary nail system for real-time monitoring of bone lengthening and healing status, comprising an extendable intramedullary nail body, a signal processing module, a sensing module, and a mobile APP module. The intramedullary nail body includes a hollow proximal main nail and a distal traction nail, which are coaxially sleeved with a clearance fit. The intramedullary nail body contains a power module and a transmission module for driving the distal traction nail to extend and retract axially. The sensing module includes an axial pressure sensor and a callus multimodal sensor. The axial pressure sensor is arranged on the load transmission path inside the proximal main nail to monitor axial tensile load and weight-bearing. The callus multimodal sensor is attached to the outer surface of the proximal main nail at the fracture site to acquire external normal pressure and shear force. The signal processing module is connected to the sensing module to collect and process the signals transmitted by the sensing module and transmit them wirelessly to the APP module. The APP module is used to decouple and visualize the received signals.

Description

Technical Field

[0001] This invention relates to the field of mechanical engineering and biomedical orthopedic implant technology, specifically to an intramedullary nail system for real-time monitoring of bone lengthening and healing status based on flexible multimodal sensing and machine learning. Background Technology

[0002] Long bone defects, limb length discrepancies, and dwarfism are significant challenges in orthopedic surgery, severely impacting patients' normal physiological functions and quality of life. To address these challenges, limb lengthening surgery has emerged and become central to clinical treatment. In this procedure, intramedullary nailing is a key technique for achieving limb lengthening. This technique originates from traditional internal fixation methods for fractures, involving the implantation of a metal nail within the medullary cavity and locking it at both ends of the fracture to provide continuous and controllable traction and lengthening. This internal traction provides a stable axial and rotational environment for bone growth and healing, fundamentally overcoming the inherent drawbacks of traditional external fixation devices, such as frequent complications and poor patient compliance. Early traditional intramedullary nails only provided static support and fixation. To meet the clinical need for active traction, novel internally driven, lengthening intramedullary nails have gradually become mainstream. These devices, based on traditional nails, employ a sleeve-like structure of a "mother nail and daughter nail," integrating a sophisticated mechanical transmission or magnetic drive mechanism within. This structure can smoothly elongate the sub-screw at a preset speed, precisely pulling the fractured bone ends to induce new callus formation. However, during the bone lengthening treatment cycle, the stretching rate needs to be highly matched with the biomechanical state of callus growth. Currently, the clinical assessment of changes in internal stress and bone healing status still mainly relies on the physician's subjective experience and regular X-ray imaging examinations, lacking continuous and accurate objective monitoring of the intramedullary mechanical environment, which in turn leads to a series of complications.

[0003] Currently, flexible sensing technology can be effectively applied to biomedical monitoring, mainly including flexible electronic devices based on piezoresistive, capacitive, piezoelectric, and triboelectric principles. Among them, flexible sensors based on resistive arrays are the most commonly used technology for implantable monitoring. The core principle of this technology is to use flexible polymers such as polyimide (PI) or polydimethylsiloxane (PDMS), which have both high ductility and excellent biocompatibility, as a substrate, and integrate sensitive materials with micro- and nano-structures on it. Relying on the flexible deformation characteristics of the substrate, this sensing structure can conform to the morphological changes of irregular curved surfaces, achieving tight conformal adhesion. When subjected to external loads, the geometry or contact state of the sensitive layer changes, thereby outputting an electrical signal. It covers a wide range of monitoring indicators, including stress, strain, temperature, and biochemical indicators. Compared with traditional rigid sensors, flexible sensors have advantages such as small size, good conformal adhesion, and less stimulation to surrounding soft tissues, and therefore have received widespread attention in the intelligent upgrading of implantable medical devices.

[0004] Bone lengthening and the subsequent healing process involve complex changes in the biomechanical environment. During this process, the growth state of the callus, i.e., the evolution of its mechanical stiffness, is crucial in determining whether the lengthening rate needs adjustment, when the patient can bear weight, and when to remove the implant. Simultaneously, the nail experiences complex stresses during growth, primarily due to the normal pressure from the callus and the tangential force exerted to overcome callus friction during lengthening. Currently, clinical judgment and decision-making mainly rely on regular follow-up using X-ray or CT imaging. However, because callus mineralization often lags behind the recovery of its biomechanical properties, imaging examinations exhibit significant lag, and frequent examinations expose patients to cumulative ionizing radiation. Furthermore, imaging can only provide two-dimensional or static morphological information, making it difficult to reflect the complex dynamic mechanical environment at the fracture site in real time and quantitatively. Therefore, developing a low-cost, highly integrated intramedullary nail system capable of simultaneously decoupling and measuring normal and tangential stresses for real-time monitoring of bone lengthening healing is essential and has significant practical value in postoperative rehabilitation guidance for limb lengthening. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and solve the problem of objectively assessing the status of bone lengthening healing in clinical practice. This invention provides an intramedullary nail system for real-time monitoring of bone lengthening healing status. This system innovatively uses flexible multimodal sensors to collect the mechanical signals of callus during the bone lengthening healing process, and combines machine learning algorithms to perform in-depth processing of multidimensional data, ultimately achieving real-time, accurate and quantitative assessment of the bone lengthening healing status.

[0006] The objective of this invention is achieved through the following technical solution: An intramedullary nail system for real-time monitoring of bone lengthening healing status includes an extendable intramedullary nail body, a signal processing module, a sensing module, and a mobile APP module. The intramedullary nail body includes a hollow proximal main nail and a distal traction nail, which are coaxially sleeved and fitted together; the intramedullary nail body is equipped with a power module and a transmission module for driving the distal traction nail to extend and retract axially. The sensing module includes an axial pressure sensor and a callus multimodal sensor; the axial pressure sensor is arranged on the load transmission path inside the proximal main nail to monitor axial tensile load and load; the callus multimodal sensor is attached to the outer surface of the proximal main nail at the fracture site to acquire external normal pressure and shear force. The signal processing module is connected to the sensing module and is used to collect and process the signals transmitted by the sensing module, and transmit them wirelessly to the mobile APP module. The mobile APP module has a built-in decoupling algorithm for decoupling and visualizing the received signals.

[0007] Furthermore, the transmission module includes a lead screw and a lead screw nut. The lead screw is configured with a support method in which one end is fixed and the other end is movable. The fixed end adopts a combination structure of a thrust ball bearing and a bearing cup, and the movable end is provided with a deep groove ball bearing and has a clearance fit with the distal traction pin.

[0008] Furthermore, the axial pressure sensor is arranged between the proximal master pin and the bearing sleeve to monitor the axial tensile load and axial load applied by the traction mechanism.

[0009] Furthermore, the axial pressure sensor has a flexible pressure-sensitive layer with microstructure, an electrode layer, and an external encapsulation layer arranged sequentially from bottom to top, forming a sandwich encapsulation structure.

[0010] Furthermore, the callus multimodal sensor is arranged in a groove on the surface of the proximal main nail at the fracture site; the callus multimodal sensor is composed of a sandwich structure consisting of an interlocking flexible lower sensing layer with microstructure, an intermediate piezoresistive array thin film layer, and an interlocking flexible upper sensing layer with microstructure.

[0011] Furthermore, the components of the signal processing module are encapsulated in a 3D-printed hardware compartment, which is located at the front of the proximal master nail and is fixed to the inner wall of the proximal master nail by an interference fit.

[0012] Furthermore, the decoupling algorithm employs a fully connected neural network model (MLP), using the relative resistance changes of the four resistance channels of the callus multimodal sensor as input and the external forces in the x, y, and z directions after decoupling as output.

[0013] Furthermore, the mobile APP module adopts a dual-mode architecture with hierarchical access control for patients and medical staff, including a patient mode and a medical staff mode; in the medical staff mode, it can send control commands to the power module to specify the target extension length and extension rate.

[0014] Furthermore, the sensing module also includes a temperature sensor, which includes an NTC thermistor for monitoring the local temperature during bone fracture healing.

[0015] Furthermore, the power module consists of a brushless DC motor, and the transmission module includes a micro reducer and a lead screw. The output end of the brushless DC motor is connected to the micro reducer, and the output shaft of the micro reducer is connected to the lead screw through a coupling. When the lead screw rotates, it drives the distal traction pin to move axially to the distal end relative to the proximal main pin, thereby extending the limb.

[0016] Compared with the prior art, the beneficial effects of the technical solution of the present invention are: 1. A multimodal callus sensor with interlocking microstructures and an intermediate piezoresistive film layer is integrated on the surface of the main nail, and data is processed using a fully connected neural network (MLP) model. The interlocking microstructures can conform to curved surfaces and sensitively respond to three-dimensional mechanical changes. The highly aliased raw electrical signals are precisely separated into independent callus positive pressure and tangential shear force through nonlinear mapping by the MLP model, overcoming the limitations of single-dimensional monitoring and providing complete data for assessing callus stiffness, achieving in-situ, non-invasive, multidimensional complex mechanical decoupling monitoring.

[0017] 2. The transmission module employs a lead screw and nut mechanism with one end fixed and the other end movable. The axial pressure sensor is positioned between the proximal main screw and the bearing cup at the fixed end. Axial load can be precisely and uniquely transmitted from the traction screw through the lead screw, thrust ball bearing, and bearing cup to the axial pressure sensor. This mechanical load transmission path design not only suppresses axial movement of the lead screw to ensure extension accuracy but also allows for direct and macroscopic measurement of traction force and patient load. High fidelity of mechanical transmission is guaranteed without increasing system size.

[0018] 3. The system integrates low-power Bluetooth communication with a mobile app module offering both doctor and patient modes. Doctors can access underlying biomechanical data and real-time biomechanical curves through the doctor-patient mode, dynamically and objectively assessing bone healing status and remotely and safely issuing traction commands. Patients can receive intuitive guidance through the patient mode, including automatically calculated daily allowable weight-bearing steps and maximum weight-bearing capacity, effectively avoiding complications from excessive traction or premature closure, and significantly reducing cumulative ionizing radiation exposure from frequent X-ray examinations. This enhances the safety and personalization of clinical treatment and reduces the risk of complications.

[0019] 4. This invention uses flexible materials to prepare multimodal sensors, enabling them to conformally attach to the complex curved surfaces of the elongated intramedullary nail. Without increasing the implant volume or interfering with the mechanical movement of the device, it achieves comprehensive perception of the local mechanical microenvironment under the implantation environment, thus constructing a more comprehensive and three-dimensional multi-mechanical evaluation system for bone healing status.

[0020] In summary, this invention enables real-time, intelligent, and quantitative monitoring of the entire bone lengthening and healing process. This system transforms abstract mechanical data into intuitive bone healing indicators, significantly reducing radiation exposure for patients undergoing frequent imaging examinations during treatment. It also provides clinicians with objective numerical data to dynamically adjust traction rates and forces, effectively avoiding the risks of nonunion due to excessive traction or premature bone closure due to insufficient traction. Attached Figure Description

[0021] Figure 1 This is a system block diagram of the present invention.

[0022] Figure 2 This is a schematic diagram of the system functional components of the present invention.

[0023] Figure 3a and Figure 3b These are, respectively, a sectional view and a top view of the present invention.

[0024] Figure 4a This is an exploded view of an axial force pressure sensor. Figure 4b This is an assembly view of the axial pressure sensor. Figure 4c This is a cross-sectional view of the axial pressure sensor along the diameter direction. Figure 4d This is a circuit model for an axial force sensor.

[0025] Figure 5a Exploded view of a callus multimodal sensor. Figure 5b An assembly view of the callus multimodal sensor. Figure 5c This is a sectional view along a direction perpendicular to the interlocking structure. Figure 5d This is a circuit model for a multimodal sensor for callus.

[0026] Figure reference numerals: 1-proximal tibia, 2-proximal main nail, 3-hardware chamber, 4-callus multimodal sensor, 5-thrust ball bearing, 6-lead screw, 7-lead screw nut, 8-deep groove ball bearing, 9-coupling, 10-brushless DC motor, 11-distal traction nail, 12-distal tibia, 13-fixation screw, 14-miniature reducer, 15-bearing locking nut, 16-temperature sensor, 17-bearing locking nut, 18-cup, 19-axial pressure sensor, 20-fixation screw; 21-external encapsulation layer, 22-positive electrode layer, 23-negative electrode layer, 24-flexible pressure-sensitive layer, 25-interlocked flexible upper sensing layer, 26-piezoresistive array thin film layer, 27-interlocked flexible lower sensing layer. Detailed Implementation

[0027] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the present invention and are not intended to limit the present invention.

[0028] Example 1 This embodiment provides an intramedullary nail system for real-time monitoring of bone lengthening healing status, such as... Figure 1 As shown, it includes an extendable intramedullary nail body, a sensing module, a signal processing module, and a mobile APP module; The functions, design diagrams, information flows, and inter-system interactions of each system component are as follows: Figure 2 As shown, the intramedullary nail body primarily serves as the mechanical carrier of the sensing module, encapsulated and implanted within the long medullary cavity of the lower limb. The sensing module reads various signals, including the axial force of the intramedullary nail body, the multidimensional force exerted by the callus on the nail surface, and the temperature at the fracture site. The signal processing module processes the information transmitted from the sensing module and transmits it to the mobile APP module via Bluetooth. The APP's built-in algorithm decouples and visualizes the signals for medical personnel to interpret. Using this multidimensional data, medical personnel can input the target elongation length and elongation rate into the microcontroller of the external magnetic stator. The microcontroller then adjusts the voltage in the stator windings using a PWM wave based on the target value, thereby controlling and adjusting the nail's elongation rate.

[0029] The extendable intramedullary nail body mainly consists of a hollow proximal master nail 2, a sliding distal traction nail 11, and a built-in power module and transmission module. For example... Figure 3aAs shown, the proximal main nail 2 and the distal traction nail 11 adopt a coaxial sleeve structure design, forming a clearance fit that allows for free expansion and contraction along the axial direction. The power module and transmission module are sealed and fixed in the internal cavity of the distal traction nail 11. Its core includes a brushless DC motor (including stator windings and a permanent magnet rotor), a miniature reduction gearbox, and a precision lead screw and nut shaft system. The lead screw and nut mechanism adopts a support configuration with one end fixed and the other end movable. Its movable end is equipped with a deep groove ball bearing with a clearance fit to the traction nail, allowing relative sliding between the bearing and the inner wall of the traction nail. The fixed end adopts a combination structure of double thrust ball bearing and bearing cup to ensure that the lead screw achieves high-precision centering rotation under load conditions. The intramedullary nail body serves as the mechanical carrier and actuator of the entire system, providing initial stability and controllable bone traction extension power for both ends of the fracture, while providing safe conformal attachment and wiring space for flexible sensor components and internal wiring harnesses.

[0030] The sensing module includes an axial pressure sensor 19, a callus multimodal sensor 4, and a temperature sensor 16; the axial pressure sensor 19 is integrated in the proximal bearing area, and the temperature sensor 16 and the callus multimodal sensor 4 are integrated in the callus growth area of ​​the side wall.

[0031] Specifically, such as Figure 4a As shown, the axial pressure sensor 19 mainly consists of an outer encapsulation layer 21, a flexible pressure-sensitive layer 24 with microstructures, an electrode layer, and wires. This part of the sensor, consisting of the microstructured flexible pressure-sensitive layer, the electrode layer, and the outer encapsulation layer arranged sequentially from bottom to top, forms a "sandwich" encapsulation structure, positioned between the proximal master nail and the bearing cup. This axial pressure sensor is primarily responsible for macroscopic mechanical safety monitoring and overall load measurement. During bone lengthening, it is used to monitor the axial tensile load applied by the traction mechanism in real time; during rehabilitation, it is used to accurately read the macroscopic axial load applied to the affected limb when the patient walks.

[0032] like Figure 5a As shown, the callus multimodal sensor 4 mainly consists of an interlocking flexible upper sensing layer 25 with microstructures, an interlocking flexible lower sensing layer 27 with microstructures, an intermediate piezoresistive array thin film layer 26, and wires. This part of the sensor has a "sandwich" structure consisting of upper and lower double encapsulation layers covering the double electrode layer and the central interlocking layer, arranged in a groove on the surface of the proximal main nail at the fracture site. The intermediate piezoresistive film is attached between the interlocking microstructures to obtain external normal pressure and shear force. The interlocking flexible sensing layer with microstructures is prepared in the same way as the flexible layer and encapsulation layer of the axial pressure sensor. This callus multimodal sensor mainly performs multidimensional stress decoupling of the microscopic healing environment.

[0033] Temperature sensor 16 mainly consists of an NTC thermistor and enameled wire. This temperature sensor primarily monitors the local temperature during bone fracture healing to prevent local inflammation. To capture minute temperature changes in the fracture area, parameters are selected... R 0 = 100 ± 1% kΩ B An NTC thermistor with K = 3950 ± 1% K has a temperature excitation-resistance response curve that satisfies the formula: (1) in, R T Resistance value at temperature T (unit: Kelvin K). R 0: At the reference temperature T The nominal resistance value at 0 (usually 25 ℃, i.e. 298.15 K).

[0034] The signal processing module comprises hardware components primarily used for in-situ acquisition of multimodal sensor data, signal conditioning, analog-to-digital conversion, and low-power wireless transmission. These hardware components mainly include a microcontroller unit integrating a microcontroller (MCU) core and a low-power Bluetooth RF transceiver, along with a miniature power supply module. To ensure effective sensor signal access, the flexible ribbon cable output from the sensor is inserted through pre-drilled micro-holes at the bottom of the hardware compartment and connected to the internal acquisition interface. To ensure the overall airtightness of the structure, waterproof adhesive is applied to the mating surfaces of the hardware compartment and the intramedullary nail. Furthermore, to accommodate the extremely limited cylindrical space within the hardware compartment during implantation, the circuit board of the signal processing module preferably employs a multilayer PCB structure. This circuit board is meticulously divided into an RF communication area, a digital core processing area, and an analog acquisition area to reduce signal crosstalk. All hardware components are highly integrated and sealed within a 3D-printed hardware compartment at the front of the proximal main nail 2; the outer wall of the hardware compartment and the inner wall of the proximal main nail are press-fitted, ensuring stable fixation within the intramedullary nail body.

[0035] The mobile app module is primarily used for decoupling and visualizing the received signals, displaying the final results after signal processing for medical personnel to assess treatment measures. The app employs a dual-mode architecture with tiered patient and medical staff permissions: In "Patient Mode," to ensure the absolute safety of the implant and avoid causing patients comprehension burdens or anxiety due to professional data, the system physically shields the underlying multimodal raw waveform data view and completely locks the drive extension permissions of the intramedullary nail's internal power module. The UI design of this interface emphasizes intuitive graphical representation, only directly pushing rehabilitation guidance indicators processed by the system to the patient. Specific displayed content includes: a visualized progress bar for bone lengthening and bone healing, dynamically calculated daily allowable weight-bearing steps and weight-bearing limits for the affected limb based on the system's assessment of the current callus stiffness, and reminders for follow-up visits and imaging examinations automatically triggered by the system calendar; "Medical Staff Mode" is mainly for attending physicians and clinical researchers, who, after the system identifies their identities, will be granted the highest level of system operation and data access permissions. In this mode, all underlying biomechanical data and intermediate algorithm variables are viewable and editable. Medical staff and researchers can freely access the raw mixed electrical signals of the sensor array, the decoupled real-time tangential mechanical curves, and long-term callus stiffness evolution scatter plots, providing comprehensive data support for in-depth patient diagnosis or research retrospection. Furthermore, this mode allows medical staff to perform intramedullary nail extension control operations, issuing personalized distraction displacement or velocity adjustment commands to the internal hardware.

[0036] Example 2 Based on Example 1, this embodiment provides a detailed description of the specific structural features and working mechanism of the extendable intramedullary nail body.

[0037] See Figure 3a and Figure 3b The extendable intramedullary nail body provided in this embodiment is implanted into the medullary cavity of the human body, such as the tibia. The proximal main nail 2 is placed within the medullary cavity of the proximal tibia 1 and is laterally locked by two fixing screws 20. In this embodiment, the top of the proximal main nail 2 has a connecting hole for the fixing screws 20 to pass through, and the upper part has a slot for fixing the hardware compartment 3. The middle part of the proximal main nail 2 at the end of the slot has a connecting hole for a cable to pass through, and the tail of the proximal main nail 2 has a cylindrical groove for fitting the distal traction nail and accommodating the lead screw 6.

[0038] The distal traction nail 11 is placed within the medullary cavity of the distal tibia 12 and laterally locked by two fixing screws 13. A coaxial sleeve fit is formed between the proximal main nail 2 and the distal traction nail 11, with the tip of the distal traction nail 11 inserted into the lower opening of the proximal main nail 2, allowing for axial relative extension and retraction. A brushless DC motor 10 serves as the power source, its output connected to a miniature reducer 14 to convert high-speed, low-torque rotational motion into low-speed, high-torque driving force. The output shaft of the miniature reducer 14 is fixedly connected to the end of the lead screw 6 via a coupling 9, thereby driving the lead screw 6 to rotate synchronously. A lead screw nut 7 is threaded onto the threaded section of the lead screw 6, and the outer circumference of the lead screw nut 7 is in contact with the inner walls of the distal traction nail 11 and the proximal main nail 2. Because the entire intelligent intramedullary nail is subjected to axial pressure, the lead screw nut 7 remains in contact with the distal traction nail 11 and will not disengage. When the lead screw 6 rotates, the lead screw nut 7 will drive the distal traction pin 11 to move axially towards the distal end relative to the proximal main pin 2, thereby achieving limb extension.

[0039] Meanwhile, to ensure the rotational stability of the lead screw 6 under heavy axial loads, this embodiment employs a specific bearing support scheme. At the top of the lead screw 6, a thrust ball sleeve 18 and a thrust ball bearing 5 are provided. The thrust ball bearing 5 is configured to withstand the axial load generated by the tension of the limb's soft tissue and is preloaded and positioned by a bearing locking nut 17. When the intramedullary nail is subjected to an axial load, this load is transmitted through the shoulder of the lead screw 6 to the thrust ball bearing 5, then sequentially through the bearing locking nut 17, the sleeve 18, and the axial pressure sensor 19, and finally to the proximal master nail 2. Therefore, the entire transmission module is always secured by axial pressure, thereby suppressing axial movement of the lead screw 6. At the end of the lead screw 6, a deep groove ball bearing 8 and a bearing locking nut 15 are provided. The deep groove ball bearing 8 is installed between the end of the lead screw 6 and the inner wall of the distal traction nail 11, mainly serving a radial support and guiding function, and allowing relative sliding between the distal traction nail 11 and the deep groove ball bearing 8. Therefore, the transmission path of the axial load is: distal traction pin 11—screw nut 7—screw 6—thrust ball bearing 5—cup 18—axial pressure sensor 19—proximal main pin 2.

[0040] Preferably, the core of this embodiment lies in the integration of multimodal sensing capabilities. First, a hardware compartment 3 is provided on the upper part of the proximal main nail 2. Second, an axial pressure sensor 19 is provided on the force transmission path of the lead screw 6, and the axial pressure sensor 19 is arranged between the proximal main nail 2 and the cup 18. The axial pressure sensor 19 is used to directly measure the axial traction force transmitted through the lead screw and the axial load when the patient bears weight. Finally, on the outer surface of the proximal main nail 2 within the gap to be lengthened in the bone, in this embodiment, the outer surface of the proximal main nail 2 has a long strip-shaped groove that runs through a certain length, for attaching the callus multimodal sensor 4 side by side. This sensor adopts a flexible array structure and is connected to the signal processing module of the hardware compartment 3 through leads. When callus forms at the fracture site, the callus multimodal sensor 4 can sense the positive pressure and tangential shear force at the contact interface and transmit the multidimensional signals to the signal processing module for processing and wireless transmission.

[0041] Example 3 This embodiment, based on embodiment 1, provides a detailed description of the specific sensors involved in the sensing module.

[0042] Combination Figures 4a to 4d It can be seen that the axial pressure sensor 19 is composed of three parts arranged from top to bottom: an outer encapsulation layer 21, an electrode layer, and a flexible pressure-sensitive layer 24 with microstructures, forming a sandwich structure. The electrode layer includes a positive electrode 22 and a negative electrode 23; the flexible pressure-sensitive layer is composed of a flexible body and a conductive film of a certain thickness. The current of the measuring circuit flows from the positive electrode to the conductive film, and then to the negative electrode. Therefore, the positive electrode 22, the conductive film of the flexible pressure-sensitive layer 24, and the negative electrode form a circuit. The circuit resistance is mainly composed of the electrode resistance, the film resistance, and the contact resistance between the film and the electrode. The electrode resistance is composed of the resistance of the positive electrode and the resistance of the negative electrode. The contact resistance between the film and the electrode includes the contact resistance between the film and the positive electrode and the contact resistance between the film and the negative electrode. Therefore, the sensor circuit can be regarded as... Figure 4d The sensor takes the form of an axial force. When subjected to axial force, the microstructure undergoes significant deformation, resulting in a marked change in the contact area between the sensor and the electrodes. This area gradually increases with the increase of the external excitation, leading to a substantial change in the contact resistance. However, the resistances of the other two parts remain essentially unchanged. The magnitude of the axial force can be measured by detecting the total resistance of the sensor. The axial pressure sensor 19 has a total thickness of approximately 2 mm, a flexible layer thickness of approximately 1 mm, and a micro-pyramidal structure integrated on the flexible layer with a bottom width of approximately 0.5 mm, a top width of approximately 0.2 mm, and a height of approximately 0.5 mm.

[0043] Preferably, the axial pressure sensor 19 is prepared by the following method: (1) The outer encapsulation layer 21 is made of polydimethylsiloxane (PDMS) material and is manufactured using a template method. Specifically, a certain amount of PDMS base solution and crosslinking agent are accurately measured, with a mass ratio of 10:1 between the base solution and the crosslinking agent. The mixture is stirred until it is fully homogeneous. To ensure the stability of the encapsulation layer, the mixture is placed in a vacuum chamber and left to stand for half an hour to completely remove internal air bubbles. After vacuum degassing, the solution is poured into a reverse template with microstructure morphology imprinted on it. After a certain period of time, the liquid is allowed to flow fully in the template. Then, an adjustable preparation device is used to scrape off the excess liquid. Finally, the template is placed on a heater and heated under constant temperature conditions until the mixed solution is completely solidified. After solidification, the template is demolded to obtain the final encapsulation layer product.

[0044] (2) Compared to the encapsulation layer, the flexible pressure-sensitive layer 24 needs to be sprayed with a conductive solution. Specifically, a conductive solution of carbon nanotube composite deionized water with a ratio of 20:1 is prepared; the solution is sprayed onto the surface of the flexible pressure-sensitive layer using a spray gun; after standing on a heating platform at 45°C for about 30 minutes, a conductive film of a certain thickness can be formed. After this step, a flexible microstructure covered with a conductive film can be prepared.

[0045] (3) The electrode layer is made of copper foil of a specific shape with a thickness of 0.01 mm, which is prepared by a triaxial cutting platform.

[0046] Combination Figures 5a to 5d It is known that the callus multimodal sensor 4 consists of three parts arranged from top to bottom: an interlocking flexible upper sensing layer 25 with microstructures, an intermediate piezoresistive array thin film layer 26, and an interlocking flexible lower sensing layer 27 with microstructures. The piezoresistive array thin film layer 26 is located on the surface where the interlocking flexible upper sensing layer 25 and the interlocking flexible lower sensing layer 27 contact each other, forming a sandwich structure. The piezoresistive array thin film layer 26 is composed of twelve pressure-sensitive thin film units (a1). The sensor surface has multiple sets of interlocking structures in the form of micro-prisms. Flexible pressure-sensitive films are attached to the two sides of each set of interlocking micro-prisms. There are three sets of micro-prisms arranged in an L-shape, with the inclined surfaces facing four different directions: +x, -x, +y, and -y. The three pressure-sensitive films on the inclined surfaces of each direction are connected in series, such as: piezoresistive resistors AC connected in series, DF connected in series, and so on. Every three piezoresistive thin films connected in series with the measurement circuit form a loop. Therefore, the twelve piezoresistive thin film units constitute four groups of four loops, each loop forming a piezoresistive array group. There are four simplified circuit models in total, resulting in four output resistors. When the sensor surface is subjected to a three-dimensional force, the a1 piezoresistive thin film units in the piezoresistive array thin film layer 26 will respond differently depending on the form of the force. By decoupling these signals using machine learning algorithms, the positive pressure and shear pressure of the callus on the sensor can be decomposed.

[0047] Preferably, the temperature sensor 16 is located above the callus multimodal sensor 4. By reading the resistance value of the NTC thermistor in real time, the temperature at the fracture site can be calculated using formula (1), thereby enabling real-time monitoring.

[0048] Example 4 This embodiment provides a detailed description of the mobile APP module based on Embodiment 1.

[0049] See Figure 1 and Figure 2 In this embodiment, the mobile APP module includes a data receiving unit, a data processing unit, a control unit, and a visualization unit; The data receiving unit is equipped with a Bluetooth interaction module and corresponding communication protocol to receive signals output from the signal processing module.

[0050] The control unit allows doctors or patients to input control commands through the interface of a mobile app module. The power module then responds by rotating the rotor and driving the lead screw to move the distal traction pin to open the osteotomy gap. At the same time, the sensing module reads the axial load and callus stress in real time and feeds the multimodal sensing data back to the mobile app module via Bluetooth. The data processing unit then decouples and processes the data, and finally the visualization unit outputs visualized data to assist in the diagnosis of healing status.

[0051] The data processing unit's internal processing algorithms are primarily used for nonlinear decoupling of multimodal sensors. The decoupling model employs a small fully connected neural network (MLP) for joint training on experimental data. The sensor is loaded using a universal mechanical testing platform, and by changing the loading method, different forces are applied to the sensor, resulting in approximately 100 sets of data. This method first obtains the initial reference resistance values ​​of the four resistance input channels under no external force conditions. R 10 , R 20 , R 30 , R 40 Subsequently, approximately 100 sets of calibration data were collected under different known three-dimensional force loading conditions. Each set of data included four resistance signals R1, R2, R3, and R4 and their corresponding three-dimensional force outputs. F x , F y , F z To reduce the impact of individual differences and environmental drift, the original resistance signals of each channel are converted into relative resistance changes, as shown in the following formula: Use it as a four-dimensional input vector; and the corresponding three-dimensional forceF x , F y , F z The output vector is constructed. Based on the above input-output samples, a small fully connected neural network model (MLP) is established. Its input layer contains four neurons, corresponding to the relative changes in the four resistance channels, and the output layer contains three neurons, corresponding to the decoupled external forces in the x, y, and z directions. The hidden layer adopts a two-layer lightweight structure; for example, the first hidden layer has 12 neurons, and the second hidden layer has 8 neurons, using the tanh activation function. The output layer uses a linear activation function to achieve continuous force value regression. During training, mean squared error is used as the loss function, and the input and output data are standardized. Through this calibration model, a nonlinear mapping relationship between the four-channel resistance response and the three-dimensional external force can be established, achieving accurate decoupling and prediction of multi-channel resistance signals to three-dimensional forces, thereby improving the calibration accuracy and force sensing reliability of the sensor under multi-directional force conditions.

[0052] The visualization unit outputs schematic diagrams of the decoupling results of multidimensional mechanics and the assessment results of the healing status. After feature mapping processing by the built-in machine learning decoupling model, the four-channel highly aliased raw electrical signals acquired by the flexible sensor array are precisely separated into two independent mechanical characteristic curves: positive pressure and tangential pressure. The healing of fractures is mainly divided into three stages: the "prolongation stage," the "cartilage callus stage," and the "bone callus mineralization stage," each corresponding to different mechanical characteristics. By collecting and comparing sensor signals over several weeks, the healing status of the fracture can be determined, thus providing real-time data support for medical personnel.

[0053] This invention is not limited to the embodiments described above. The above description of specific embodiments is intended to illustrate and explain the technical solutions of this invention. The specific embodiments described above are merely illustrative and not restrictive. Without departing from the spirit and scope of the claims, those skilled in the art can make many specific modifications based on the teachings of this invention, and these modifications all fall within the scope of protection of this invention.

Claims

1. An intramedullary nail system for real-time monitoring of bone lengthening healing status, characterized in that, It includes an extendable intramedullary nail body, a signal processing module, a sensing module, and a mobile APP module; The intramedullary nail body includes a hollow proximal main nail (2) and a distal traction nail (11), which are coaxially sleeved and fitted together; the intramedullary nail body is provided with a power module and a transmission module, which are used to drive the distal traction nail (11) to extend and retract axially. The sensing module includes an axial pressure sensor (19) and a callus multimodal sensor (4); the axial pressure sensor (19) is arranged on the load transmission path inside the proximal main nail (2) to monitor axial tensile load and load; the callus multimodal sensor (4) is attached to the outer surface of the proximal main nail (2) at the fracture site to obtain external positive pressure and shear force. The signal processing module is connected to the sensing module and is used to collect and process the signals transmitted by the sensing module, and transmit them wirelessly to the mobile APP module. The mobile APP module has a built-in decoupling algorithm for decoupling and visualizing the received signals.

2. The intramedullary nail system according to claim 1, characterized in that, The transmission module includes a lead screw (6) and a lead screw nut (7). The lead screw (6) is configured with a support method in which one end is fixed and the other end is movable. The fixed end adopts a combination structure of a thrust ball bearing (5) and a sleeve cup (18). The movable end is provided with a deep groove ball bearing (8) and is clearance-fitted with the distal traction pin (11).

3. The intramedullary nail system according to claim 1, characterized in that, The axial pressure sensor (19) is arranged between the proximal master nail (2) and the sleeve (18) to monitor the axial tensile load and axial load applied by the traction mechanism.

4. The intramedullary nail system according to claim 1 or 2, characterized in that, The axial pressure sensor (19) consists of a flexible pressure-sensitive layer with microstructure, an electrode layer, and an external encapsulation layer arranged sequentially from bottom to top, forming a sandwich encapsulation structure.

5. The intramedullary nail system according to claim 1, characterized in that, The callus multimodal sensor (4) is arranged in the groove on the surface of the proximal main nail (2) at the fracture site; the callus multimodal sensor (4) is composed of a sandwich structure consisting of an interlocking flexible lower sensing layer with microstructure, an intermediate piezoresistive array thin film layer, and an interlocking flexible upper sensing layer with microstructure.

6. The intramedullary nail system according to claim 1, characterized in that, The components of the signal processing module are encapsulated in a 3D-printed hardware compartment (3). The hardware compartment (3) is located at the front of the proximal master nail (2) and is fixed to the inner wall of the proximal master nail (2) by interference fit.

7. The intramedullary nail system according to claim 1, characterized in that, The decoupling algorithm uses a fully connected neural network model MLP, with the relative resistance change of the four resistance channels of the callus multimodal sensor (4) as input and the external forces in the three directions of x, y, and z after decoupling as output.

8. The intramedullary nail system according to claim 1, characterized in that, The mobile APP module adopts a dual-mode architecture with hierarchical access control for medical staff and patients, including a patient mode and a medical staff mode. In the medical staff mode, it can send control commands to the power module to set the target extension length and extension rate.

9. The intramedullary nail system according to claim 1, characterized in that, The sensing module also includes a temperature sensor (16), which includes an NTC thermistor for monitoring the local temperature during bone fracture healing.

10. The intramedullary nail system according to claim 1, characterized in that, The power module consists of a brushless DC motor, and the transmission module includes a micro reducer and a lead screw. The output end of the brushless DC motor is connected to the micro reducer, and the output shaft of the micro reducer is connected to the lead screw through a coupling. When the lead screw rotates, it drives the distal traction nail (11) to move axially to the distal end relative to the proximal main nail (2), thereby extending the limb.

Images