Novel ultramicro neuroendoscopic surgery system with treatment function

Abstract

The invention relates to the technical field of medical equipment, and discloses a novel ultramicro neuroendoscopic surgery system with a treatment function, which comprises a coaxial mirror module integrated with a sensing function, a central processing and control imaging module and a function support module, an optical imaging unit, a bioelectrical impedance sensor and a pressure sensor are integrated at the working end of the coaxial mirror module, so that co-location and multi-mode information synchronous acquisition of target tissues is realized; a form sensing element based on a fiber bragg grating array is arranged in the lens body and is used for sensing the three-dimensional form of the lens body in real time. The central processing and control imaging module executes a multi-source information fusion and tissue classification algorithm, a lens body three-dimensional form reconstruction algorithm and a perfusion pressure closed-loop feedback control algorithm, and displays the solved tissue type and lens body three-dimensional form in real time in an augmented reality mode such as highlight color and virtual curve. The accuracy and safety of the operation are improved by quantifying the focus and accurately identifying and navigating in real time.

Description

Technical Field

[0001] This invention relates to the field of medical equipment technology, specifically to a novel ultra-micro neuroendoscopic surgical system with therapeutic functions. Background Technology

[0002] Neurosurgery is constantly developing towards minimally invasive and precise techniques. Minimally invasive procedures, such as neuroendoscopic techniques and stereotactic biopsy, play an increasingly important role in the diagnosis and treatment of diseases such as brain tumor resection, intracranial hematoma removal, and cyst drainage due to their significant advantages such as less trauma and faster recovery.

[0003] Currently, the core functions of surgical instruments used in these minimally invasive surgeries, such as endoscopes or puncture needles, primarily remain providing optical imaging and establishing instrument access channels. During the procedure, surgeons rely mainly on two-dimensional video images transmitted from the endoscopic camera, which are subject to some image distortion, to determine the boundary between the lesion (e.g., a tumor) and the surrounding normal brain tissue. This is combined with the surgeon's own experience-based perception and the vague tactile feedback transmitted through the long instrument shaft. This diagnostic method is significantly subjective, especially when dealing with lesions like gliomas, which have unclear boundaries with normal brain tissue and exhibit invasive growth. Misjudgment can easily lead to residual tumor or accidental damage to important functional areas.

[0004] Meanwhile, during the movement of the instrument into deep intracranial lesions, the determination of its spatial position and the navigation of its path mainly rely on matching the intraoperative instrument position with the preoperatively planned image-guided navigation system. However, these navigation systems are generally based on the assumption that the instrument is an ideal rigid body, and cannot track and present in real time the actual bending and morphological deviations caused by the force exerted on the instrument when traversing complex and uneven brain tissue. This deviation between the planned path displayed by the navigation and the actual physical posture of the instrument constitutes a huge potential risk of damaging key blood vessels and nerve fiber bundles in the brain.

[0005] Furthermore, continuous perfusion and suction are required during surgery to maintain a clear surgical field and perform necessary tissue irrigation. Existing perfusion systems are mostly open-loop controlled with relatively coarse flow regulation, making it difficult to accurately quantify and stably control the pressure of the microenvironment within the surgical field. Drastic pressure fluctuations or excessively high pressures can cause secondary damage to fragile brain tissue.

[0006] In summary, existing minimally invasive neurosurgical instruments are essentially passive tools with relatively limited functions, lacking integrated, multi-dimensional real-time sensing capabilities and intelligent information processing and feedback mechanisms. Therefore, there is an urgent clinical need for a new type of intelligent surgical system that deeply integrates the surgeon's subjective experience with objective sensor data, fundamentally improving surgical precision, safety, and treatment outcomes. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides a novel ultra-micro neuroendoscopic surgical system with therapeutic functions, which solves the problems in existing neurosurgical procedures caused by the lack of integrated multi-dimensional real-time sensing and feedback capabilities of surgical instruments, resulting in subjective tissue boundary definition, inaccurate instrument navigation paths, and coarse control of key surgical environment parameters.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a novel ultra-micro neuroendoscopic surgical system with therapeutic function, comprising: A coaxial endoscope module with integrated sensing function has two isolated independent channels inside the endoscope body, namely an instrument channel for accommodating treatment instruments and an infusion channel for delivering fluids. The distal end of the endoscope body is provided with a multimodal biosensing needle tip for acquiring optical image signals and at least one non-optical biosensing signal during surgery. The functional support module is used to provide energy and fluid to the coaxial mirror module with integrated sensing function; The central processing and control imaging module includes a monitor and a central processor. The central processor is used to receive and fuse the optical image signal and the non-optical biosensor signal in real time to generate diagnostic information for decision support, and to overlay the diagnostic information and the optical image signal on the monitor.

[0009] Preferably, the instrument channel is a central channel, and the infusion channel is one or more peripheral channels arranged around the central channel.

[0010] Preferably, the multimodal biosensing tip integrates: An optical imaging unit used to acquire the optical image signals; A bioelectrical impedance sensor used to acquire electrical characteristic signals as non-optical biosensing signals; A microelectromechanical system pressure sensor used to acquire force signals, which are non-optical biosensing signals.

[0011] Preferably, the step of the central processing and control imaging module generating the diagnostic information includes: Based on the electrical characteristic signal and the force signal, the tissue type in contact with the multimodal biosensor tip is classified in real time, and the classification result is used as part of the diagnostic and treatment information.

[0012] Preferably, the coaxial mirror module with integrated sensing function is further provided with a shape sensing element on the mirror body. The shape sensing element is used to collect signals of strain or curvature parameters that characterize the three-dimensional spatial shape of the mirror body. The strain or curvature parameter signals are part of the non-optical biosensing signals.

[0013] Preferably, the step of the central processing and control imaging module generating the diagnostic information further includes: Based on the strain or curvature parameter signal, a three-dimensional spatial morphology model of the endoscope is reconstructed, and the three-dimensional spatial morphology model is used as part of the diagnostic and treatment information.

[0014] Preferably, the shape sensing element is a fiber Bragg grating array implanted along the length of the mirror body.

[0015] Preferably, the central processing and control imaging module is further used for: A closed-loop feedback control loop for treatment parameters is established, using the force signal as feedback to automatically adjust the output pressure of the perfusion system in the functional support module.

[0016] Preferably, the superimposed display of the diagnostic information and the optical image signal on the monitor is an augmented reality display.

[0017] Preferably, the bioelectrical impedance sensor employs a four-electrode structure, which includes a pair of driving electrodes for injecting current and a pair of sensing electrodes for measuring voltage.

[0018] This invention provides a novel ultra-micro neuroendoscopic surgical system with therapeutic functions. It possesses the following beneficial effects: 1. This invention integrates a bioelectrical impedance sensor and a microelectromechanical system (MEMS) pressure sensor at the distal end of a coaxial mirror module. This enables real-time acquisition of non-optical biosensor signals, such as the electrical and mechanical properties of the tissue in contact with the surgical instrument tip. The central processing and control imaging module then fuses and intelligently identifies these multi-source sensor signals, achieving objective quantitative classification of tissue types. This provides operators with identification criteria that surpass traditional visual observation, improving the ability to distinguish between functional and diseased tissues during surgery, thereby enhancing the accuracy of surgical target localization.

[0019] 2. This invention, by incorporating morphological sensing elements on the scope of the coaxial endoscope module, enables real-time monitoring of the three-dimensional morphological changes and stress states of the scope during its puncture and movement within human tissue. The central processing and control imaging module reconstructs a spatial path model of the scope based on these parameters and provides early warnings when the scope undergoes excessive bending or the stress exceeds a preset safety threshold. This allows the operator to intuitively grasp the actual posture of the surgical instruments within the body, effectively avoiding accidental damage to vital tissues due to path deviation or instrument deformation, and enhancing the controllability and safety of the surgical procedure.

[0020] 3. This invention utilizes a microelectromechanical system (MEMS) pressure sensor integrated into the needle tip to measure the actual pressure within the surgical field in real time. This measurement value is used as a feedback signal, and the central processing and control imaging module establishes a closed-loop feedback control of the perfusion system in the functional support module. This design can automatically and accurately adjust the output pressure of the perfusion fluid, stabilizing key surgical parameters such as intra-tissue pressure within a preset safe range. This avoids complications such as tissue edema that may be caused by inaccurate or delayed manual adjustment, and realizes the automation and standardization of key surgical steps. Attached Figure Description

[0021] Figure 1 This is a perspective view of the device of the present invention; Figure 2 This is a schematic diagram of the coaxial endoscope of the present invention; Figure 3 This is a cross-sectional view of the coaxial endoscope body of the present invention; Figure 4 This is a cross-sectional view of the multimodal biosensor tip of the present invention; Figure 5 This is a schematic diagram of the bioelectrical impedance sensor of the present invention; Figure 6 This is a block diagram of the internal hardware functions of the central processing and control imaging module of the present invention.

[0022] The components include: 1. Instrument channel; 2. Infusion channel; 3. Multimodal biosensing tip; 4. Monitor; 5. Central processing unit; 6. Bioelectrical impedance sensor; 7. Microelectromechanical system pressure sensor; 8. Morphological sensing element; 9. Fiber Bragg grating array; 10. Optical imaging unit; and 11. Endoscope body. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] See attached document Figure 1 The present invention provides a novel ultra-micro neuroendoscopic surgical system with therapeutic function, which may include: a coaxial endoscope module with integrated sensing function, a function support module, and a central processing and control imaging module.

[0025] The coaxial endoscope module with integrated sensing functions is the execution end of this surgical system, and its structure is designed for minimally invasive procedures inside the human body. The distal end of this module integrates sensors for acquiring multimodal information about the surgical area and has internal channels for accommodating therapeutic instruments and delivering fluids.

[0026] The functional support module provides auxiliary energy and fluid support for the entire surgical system. Based on instructions from the central processing and control imaging module, it provides energy to therapeutic instruments passing through the instrument channels within the coaxial endoscope module, or precisely delivers fluid to the surgical area.

[0027] The central processing and control imaging module is the core of this surgical system for data processing and control. This module includes a central processor and a monitor. The central processor receives and processes all sensor data from the coaxial endoscope module, executes information fusion and decision-making algorithms, and generates control commands to send to the functional support module. The monitor displays real-time optical images of the surgical area and auxiliary diagnostic information generated by the central processor.

[0028] In the system's architecture, the modules are connected and communicate through specific physical interfaces. The coaxial mirror module with integrated sensing functions is connected to the central processing and control imaging module via a composite cable. This composite cable integrates a coaxial cable or fiber optic bundle for transmitting optical image signals, as well as a shielded electrical signal cable for transmitting multiple non-optical biosensor signals such as bioelectrical impedance, pressure, and morphological strain.

[0029] The coaxial endoscope module with integrated sensing capabilities is connected to the function support module via fluid lines and power cables. The fluid lines deliver perfusion fluid from the function support module to the perfusion channel of the coaxial endoscope module. The power cables transmit therapeutic energy (e.g., radiofrequency energy) generated by the function support module to the therapeutic instruments located within the instrument channel of the coaxial endoscope module.

[0030] The central processing and control imaging module is connected to the function support module via a control signal line. The central processing unit sends precise digital or analog control commands to the function support module through this signal line to adjust the power and frequency of the energy output, or to regulate the flow rate and pressure of the perfusion system.

[0031] The working principle of this invention lies in the fact that the central processing unit (CPU) performs real-time synchronization and fusion processing on multi-source heterogeneous data collected from the coaxial mirror module with integrated sensing functions. At any given time, the CPU receives a system state vector containing all sensing information. It can be represented as: ; in, This represents the two-dimensional optical image frame acquired at that moment; This represents the frequency measured at that moment. The changing bioelectrical impedance spectral vector; This represents the scalar value of the force acting on the tip of the needle measured at that moment; This represents the vector of strain parameters measured by the shape sensing elements distributed along the axial position of the mirror body at that moment.

[0032] The central processing unit (CPU) runs a pre-defined transformation function. For the system state vector Processing is performed to generate an output vector containing specific diagnostic and treatment information. and Output vector It includes information such as the classification results of the currently contacted tissue and the three-dimensional morphological model of the endoscope itself.

[0033] Ultimately, the central processing unit processes the original optical image frames. The processed diagnostic information output vector Registration and rendering are performed, and the fused image is output to a monitor in real time for overlay display. This process constitutes a complete technical loop from multimodal data acquisition to information fusion processing and visualization presentation.

[0034] See attached document Figure 2 and appendix Figure 3 The coaxial endoscope module integrating sensing functions comprises a slender, hollow tubular endoscope body with pre-defined rigidity. A handle for the operator to grip is connected to the proximal end of the endoscope body, while the distal end constitutes the working end of the invention. The entire endoscope body is made of biocompatible materials, such as medical-grade 316L stainless steel, to ensure its safety in contact with human tissue and to provide sufficient mechanical strength to resist deformation during puncture. The end face or outer surface of the working end may be coated or partially made of a biocompatible polymer with electrical insulating properties, such as polyetheretherketone (PEEK), to provide an insulating substrate for subsequent integration of electrical sensors. In one specific embodiment, the outer diameter of the working end of the endoscope body is 4.0 mm, and the total length is 150 mm.

[0035] A core structural feature of this invention lies in the specific layout of its internal channels. For example... Figure 3 As shown, the endoscope has a central channel located on its geometric axis, which serves as the instrument channel. This instrument channel has a continuous and unobstructed circular cross-section with an inner diameter designed to accommodate and guide various minimally invasive therapeutic instruments, such as biopsy forceps, radiofrequency ablation electrodes, or laser fibers, precisely reaching the surgical target. Its inner wall is polished or coated with polytetrafluoroethylene (PTFE) to reduce frictional resistance when the therapeutic instruments pass through it.

[0036] The infusion channel is designed as one or more peripheral channels arranged around the central instrument channel. In a preferred embodiment, the infusion channel system includes at least two independent peripheral channels: an inlet channel and an outlet channel. Both peripheral channels are formed within the wall structure of the endoscope body, and their cross-sections can be circular or other irregular shapes, with an equivalent diameter of 0.4 mm. The inlet channel and outlet channel 3 are symmetrically distributed in the cross-section of the endoscope body, for example, arranged at 180-degree angles.

[0037] This dual-channel irrigation system enables continuous circulation irrigation of the surgical area. The inlet channel delivers irrigation fluids such as saline from the corresponding port on the handle to the working end, where it flows out to rinse the surgical field, maintain a clear surgical view, and regulate local pressure. Simultaneously, the outlet channel removes waste fluid mixed with blood or tissue debris from the working end and discharges it through another port on the handle.

[0038] The wall structure of the endoscope, namely the annular space between the outer wall of the central instrument channel and the inner wall of the endoscope housing, is fully utilized to accommodate and house all necessary components except for the irrigation channel. These components include an illumination fiber bundle for transmitting illumination light from an external light source to the working end, a miniature coaxial cable or image transmission fiber bundle for transmitting optical image signals, and multi-core signal cables for connecting various sensors integrated on the working end. This integrated and compact layout ensures that all functions of surgical manipulation, irrigation, illumination, imaging, and multimodal sensing are achieved within a limited outer diameter of 4.0 mm.

[0039] On the handle, independent, standardized physical interfaces are provided for each of the aforementioned channels and cables. For example, Luer lock connectors are provided for the instrument channel, and Luer lock connectors with different color markings are provided for the water inlet and outlet channels to prevent misconnection. A composite electrical / optical connector is also provided for signal and power connection to the central processing and control imaging module.

[0040] See attached document Figure 4A core technical solution of this invention lies in the integrated multimodal biosensing tip on the working end. This integration scheme, through micromachining and precision packaging technology, highly integrates the three functional units of optical imaging, electrical property sensing, and mechanical property sensing into a very small physical space on the annular end face of the distal end of the microscope body, so as to achieve synchronous and co-located information acquisition of the same target tissue region.

[0041] The optical imaging unit is located in the central area of ​​the working end, and its optical window faces the axial direction of the instrument channel, providing a direct field of view for surgical operations.

[0042] In one specific embodiment, the optical imaging unit consists of a miniaturized CMOS image sensor chip and a matching microlens assembly. The CMOS image sensor has an effective pixel count of 400,400, or 160,000 pixels, which meets the clarity requirements of neurosurgical procedures. The sensor chip is fixed to a support structure inside the endoscope using chip-level packaging technology, and its signal is routed along the endoscope wall via a micro-coaxial cable and connected to an electrical connector on the handle. The end of an illumination fiber optic bundle is also coaxially positioned around the lens of the optical imaging unit to uniformly project light into the forward field of view.

[0043] See attached document Figure 5 A bioelectrical impedance sensor is designed and fabricated on an insulated end face of the working end. The sensor employs a four-electrode layout, including a pair of driving electrodes 2a for injecting excitation current and a pair of sensing electrodes 2b for measuring the voltage response. These two electrode pairs are fabricated directly on a PEEK insulating surface serving as a substrate using thin-film deposition processes such as magnetron sputtering or physical vapor deposition (PVD). The electrode materials are biocompatible noble metals, such as platinum or platinum-iridium alloys, to ensure their electrochemical stability in biological environments.

[0044] In one specific implementation, the driving electrode 2a is designed as the outer ring of two concentric rings, while the sensing electrode 2b is the inner ring; together, they surround the window of the optical imaging unit. The electrodes have a linewidth of 50 micrometers and a thickness of 200 nanometers. This four-electrode concentric ring configuration generates a uniformly distributed electric field in both depth and radial direction, effectively reducing the impact of contact impedance between the electrodes and tissue on measurement accuracy. Each electrode ring is connected to its independent signal transmission channel via a fine wire, which extends from inside the scope body.

[0045] A microelectromechanical system (MEMS) pressure sensor is a standalone, miniaturized sensing chip used to measure the contact force between a needle tip and tissue in real time. The sensor chip measures 1mm x 1mm x 0.5mm and is a piezoresistive pressure sensor with a range of 0-500kPa, covering the typical tissue contact pressure range in neurosurgery. For integration, a micro-groove matching the size of the sensor chip is pre-fabricated on the PEEK insulated end face of the working end using laser or precision machining.

[0046] Subsequently, the MEMS pressure sensor chip was precisely fixed into the groove using medical-grade epoxy resin, ensuring that its sensing membrane surface was flush with the end face of the working end. This embedded flush mounting method avoids additional scratching or damage to the tissue during puncture. The sensor's electrode pads are connected to micro-signal wires pre-embedded in the endoscope body using a gold wire bonding process. The sensor surface is coated with a 5-micrometer-thick, biocompatible Parylene coating to electrically isolate it from body fluids and provide long-term operational reliability.

[0047] Through the above integration scheme, this invention achieves a high degree of synergy among three sensors based on different physical principles on a single working surface. This design ensures that all acquired optical images, bioelectrical impedance spectra, and contact pressures originate from the exact same tissue microenvironment, providing high-quality, highly correlated raw data input for subsequent high-precision multimodal information fusion and analysis by the central processing unit.

[0048] See attached document Figure 4 To achieve real-time perception of the three-dimensional spatial morphology of the coaxial mirror module with integrated sensing functions within the human body, this invention incorporates morphology sensing elements within the mirror's structure. In a preferred embodiment, the morphology sensing element is a fiber Bragg grating (FBG) array.

[0049] The deployment scheme specifically includes three independent optical fibers, a, b, and c, each with multiple FBG sensing points etched along its length at a predetermined fixed interval (e.g., 10 mm). These three optical fibers are precisely embedded into the wall structure of the mirror body.

[0050] Specifically, within the wall of the endoscope, three miniature longitudinal grooves are symmetrically arranged at 120-degree angles around the axis of the central instrument channel. Three optical fibers, a, b, and c, each carrying an FBG array, are placed within these three grooves and encapsulated and fixed using biocompatible medical-grade epoxy resin. This fixation method ensures that when the endoscope bends, the resulting strain is transmitted to the optical fibers without damage and with precision.

[0051] Each FBG sensing point on each optical fiber has a specific, unique original Bragg center wavelength. When a section of the mirror body bends, the three optical fibers located in that section will experience different axial strains. For example, the fiber on the outside of the bend is stretched (positive strain), while the fiber on the inside of the bend is compressed (negative strain). This strain causes a change in the grating period of the FBG, which in turn causes a shift in the center wavelength of its reflection. The drift is directly proportional to the strain, and their physical relationship can be described by the following formula: ; in, is the effective elastic-optic coefficient of the optical fiber material, which is a known physical constant.

[0052] These three optical fibers extend along the entire length of the mirror body from the working end to the handle, and are connected to an FBG demodulator in the central processing and control imaging module via a dedicated fiber optic connector. This demodulator emits a broadband light source into the three fibers and detects the center wavelength reflected from each FBG sensing point in real time with high precision. The wavelength drift at each sensing point is measured... The central processing unit can then accurately calculate the real-time strain values ​​of multiple discrete points distributed along the length of the mirror body in three different orientations based on the above formula.

[0053] This three-fiber symmetrical layout design can not only measure the amplitude of bending but also determine the direction of bending. At any cross-section of the mirror, three strain values ​​are measured. , , The differential combination uniquely determines the local curvature vector (including magnitude and direction) of the cross-section. Simultaneously, the average of these three strain values ​​corresponds to the common axial stress on the cross-section and the wavelength shift caused by temperature changes. This differential measurement and common-mode suppression calculation method effectively compensates for temperature variations when calculating curvature, thereby improving the accuracy of shape perception. These precisely calculated local curvature parameters, distributed along the entire length of the mirror, constitute the input data for subsequent algorithms for complete 3D shape reconstruction.

[0054] See attached document Figure 6 The central processing and control imaging module of this invention is physically implemented as an integrated computing and display console. This console deploys a high-performance hardware platform, specifically designed to meet the specific technical requirements of high-speed acquisition, real-time processing, complex algorithm computation, and fusion display of high-definition images and augmented reality information from multiple heterogeneous sensor data sources.

[0055] The central processing unit (CPU) is not a single processor chip, but a complete embedded industrial computer system. This system includes a CPU, a GPU, a multi-channel data acquisition card (DAQ), a fiber Bragg grating demodulator, control signal output interfaces, and a solid-state storage drive (SSD). All these components are interconnected via a high-speed PCIe bus to ensure high bandwidth and low latency for data transmission between units.

[0056] The central processing unit (CPU), such as a multi-core Intel Core i7 processor, is responsible for running the operating system, managing the task scheduling of the entire system, executing the logic of the user interface, and executing the less computationally intensive parts of algorithms.

[0057] Graphics processing units (GPUs), such as an NVIDIA RTX series computing card, are configured with a large number of parallel processing cores. They are specifically designed to perform computationally intensive parallel tasks, including: real-time image enhancement and feature extraction of raw video streams acquired from optical imaging units; executing tissue classification algorithms based on support vector machines or neural networks; and performing mirror 3D morphology reconstruction algorithms based on multi-point strain data. Offloading these computational tasks to the GPU significantly reduces the CPU load and ensures the real-time responsiveness of the entire system.

[0058] A multi-channel data acquisition card (DAQ) is a high-speed data acquisition device with multiple synchronous analog input channels. It connects to the signal output terminals of the bioelectrical impedance sensor and MEMS pressure sensor in the coaxial mirror module via shielded cables. The DAQ card incorporates a high-precision analog-to-digital converter (ADC) that converts the weak, continuous analog voltage signals output by the sensors into digital data streams at a sampling rate of at least 1MHz for subsequent processing by the CPU and GPU.

[0059] The fiber optic grating demodulator is a dedicated optoelectronic instrument that connects to the three FBG fibers (a, b, and c) in the coaxial mirror module via fiber optic connectors. Internally, the demodulator contains a broadband light source and a high-speed spectral analysis module, capable of calculating the center wavelength shift value of each FBG sensing point in real time at a frequency of at least 1 kHz, and transmitting this wavelength data to the central processing unit via USB or Ethernet interface.

[0060] A control signal output interface, such as an interface card containing a digital-to-analog converter (DAC), is used to convert digital control instructions generated by the CPU (e.g., instructions for adjusting the pressure of the infusion pump) into analog voltage or current signals and output them to the function support module.

[0061] The monitor is a 24-inch medical-grade LCD display with a physical resolution of 1920x1080 pixels or higher. This display meets medical device safety standards and features high brightness, high contrast, and a wide color gamut to ensure accurate color reproduction and detail in surgical images. In one embodiment, the monitor also integrates a capacitive touchscreen, allowing the operator to directly set parameters and select functions on the screen. Furthermore, the system is equipped with sterilizable medical-grade keyboard, mouse, and foot switch as external input devices to suit the operating room environment.

[0062] The central processing unit internally stores and executes a set of core algorithms, which form the basis for achieving the technical effects of this invention. These algorithms are designed to transform raw, discrete data streams collected from multimodal biosensing tips and morphological sensing elements into intuitive, structured diagnostic and treatment information that can be used to assist surgical decision-making.

[0063] The central processing unit first executes a multi-source information fusion and classification algorithm. The input to this algorithm is image frames from the optical imaging unit. Complex impedance spectroscopy data from bioelectrical impedance sensors and pressure values ​​from MEMS pressure sensors. The first step of the algorithm is feature extraction. For an image frame... The algorithm calculates its gray-level co-occurrence matrix. Then, a set of texture features, such as contrast, correlation, energy, and homogeneity, are extracted from it to form an image feature vector. For complex impedance spectroscopy data The algorithm uses a nonlinear least squares method to fit it to... Model: ; in, The imaginary unit, For frequency. Characteristic parameters obtained through fitting, including zero-frequency resistance. Infinite frequency resistor Relaxation time and coefficients characterizing the degree of dispersion Together they constitute the electrical eigenvectors Pressure value Then it is directly used as a pressure characteristic. The algorithm then fuses these features extracted from different sensor sources into a high-dimensional feature vector: The feature vector is input into a pre-trained classifier model, such as a Support Vector Machine (SVM). This SVM model uses a radial basis function (RBF) as the kernel function to construct a hyperplane in the feature space, separating feature vectors representing different tissue types (such as normal brain tissue, glioma tissue, necrotic tissue, etc.). The algorithm's final output is a classification label representing the type of tissue currently contacted by the needle tip. .

[0064] Simultaneously, the central processing unit executes a three-dimensional morphology reconstruction algorithm for the mirror body. The input to this algorithm is the wavelength drift value of each FBG sensing point along the length of the mirror body, obtained from the fiber optic grating demodulator. The first step of the algorithm is to reconstruct the morphology of each sensing point based on its wavelength drift. Calculate the corresponding axial strain The second step is to examine any cross-sectional position on the mirror body. The algorithm utilizes the strain values ​​of the three optical fibers at that location. To solve for the local curvature vector of the cross section Assume the radial distance of the three optical fibers from the neutral axis of the mirror is... It is relative to a local coordinate system The angular positions are respectively, then the curvature components can be solved by the following system of linear equations: ; The third step of the algorithm involves analyzing the discrete curvature vectors distributed along the entire length of the mirror body. Numerical integration is performed to reconstruct the three-dimensional spatial coordinates of the mirror's central axis. This process begins at the base of the mirror ( ), with increments of _____. Iterative calculation. In the first... The position of each node and tangent vector Can be derived from the first The state of the node and the first The curvature of the segment is approximately calculated. This iterative process ultimately generates a series of three-dimensional coordinate points, which, when connected, constitute a real-time three-dimensional morphological model of the endoscope in the surgical space.

[0065] In addition, the central processing unit executes a closed-loop feedback control algorithm for the infusion pressure. The input to this algorithm is the real-time pressure measurement value from the MEMS pressure sensor. and a target pressure value preset by the operator. The algorithm first calculates the pressure error at the current moment. Subsequently, a digital PID (proportional-integral-derivative) controller calculates a control output signal based on this error. : ; These are the pre-tuned proportional, integral, and derivative gain coefficients, respectively. This output signal... The signal is converted into a specific voltage or current signal via the control signal output interface, which is used to directly drive the infusion pump in the function support module, thereby precisely regulating its output flow rate and adjusting the surgical field pressure. Dynamically approaching and stabilizing at the target value nearby.

[0066] The central processing and control imaging module ultimately fuses all processed diagnostic and treatment information with the original optical image signals and visualizes them on a monitor in an augmented reality manner.

[0067] The augmented reality display is implemented by the graphics processing unit (GPU) in the central processing unit. The GPU receives a real-time video stream from the optical imaging unit as a background layer. Simultaneously, it receives structured diagnostic information generated by the core algorithm, including tissue classification results and a set of coordinate points for the endoscopic 3D morphological model. Based on preset rendering rules, the GPU generates virtual graphical elements from this structured information and precisely overlays them onto the corresponding positions in the real-time video stream.

[0068] To achieve precise spatial registration between virtual information and real images, the system pre-calibrates the internal parameters (such as focal length and principal point coordinates) and external parameters (its orientation relative to the working end coordinate system) of the optical imaging unit. For the three-dimensional morphological model of the mirror body, the coordinate point set generated by the algorithm is directly projected onto the two-dimensional image plane, forming a virtual curve corresponding to the actual bending shape of the mirror body. This curve is superimposed on the video image in a semi-transparent manner.

[0069] For the tissue classification results, the system maps them to a predefined color lookup table. For example, normal tissue corresponds to green, and tumor tissue corresponds to red. The GPU renders a circular highlighted area corresponding to the classification result color in the center of the image displayed on the monitor (i.e., the area where the needle tip touches). The transparency and size of this highlighted area can be adjusted by the operator.

[0070] The entire human-computer interface displayed on the monitor is divided into multiple functional areas. The main display area, occupying most of the screen, is used to present the aforementioned augmented reality fusion images. The status and data display area, usually located at the edge of the screen (such as the top or sidebar), displays key surgical parameters in real time in the form of digital and graphical gauges, such as real-time pressure values ​​measured by MEMS pressure sensors. The target pressure value set by the operator And the maximum curvature of the mirror body, etc.

[0071] The interaction and control area allows operators to set and adjust system parameters via the monitor's touchscreen or external input devices. Operators can input or modify target pressure values ​​in this area. This value will serve as the input to the closed-loop feedback control algorithm for perfusion pressure. Operators can also selectively enable or disable specific augmented reality display elements in this area; for example, hiding the 3D morphological model curve of the endoscope during a specific operational phase to obtain an unobstructed view. Furthermore, this area provides functions such as starting / stopping surgical data recording, image freezing, and taking snapshots. When the system detects an abnormal state, such as pressure exceeding a safety threshold or excessive endoscope curvature, the status and data display area will issue a visual alarm with a highlighted or flashing indicator, accompanied by an audible prompt, to alert the operator.

[0072] The functional support module of this invention, serving as the power and fluid supply unit for the entire surgical system, is integrated into a separate chassis distinct from the central processing and control imaging module. This module is designed to receive precise control commands from the central processor and accordingly provide adjustable energy and fluid to the coaxial endoscope module with integrated sensing capabilities. The module primarily comprises an energy platform and a perfusion system.

[0073] In one specific embodiment, the energy platform is a medical-grade high-frequency radio frequency (RF) generator. This generator is capable of producing a continuously adjustable radiofrequency current with a frequency range of 400kHz to 500kHz and an output power between 0 and 50 watts. The platform is connected to a radiofrequency ablation electrode located within the instrument channel of the coaxial endoscope module via a dedicated energy cable. The control circuitry within the energy platform is connected to the control signal output interface of the central processing and imaging module via control signal lines, receiving digital instructions from the central processor. These instructions specifically define the required power level, duration of action, and operating mode (such as cutting or coagulation), thereby achieving precise control of the treatment process by the system software.

[0074] The perfusion system is designed for precise fluid management in the surgical area. It includes a high-precision peristaltic pump driven by a stepper motor, which serves as the inlet unit. The pump's inlet is connected to a saline bag via sterile tubing, and its outlet is connected to an interface on the coaxial endoscope module handle corresponding to the inlet channel. The stepper motor driver directly receives control signals from the central processing unit via a control signal output interface.

[0075] The control signal is the physical manifestation of the output signal calculated by the aforementioned closed-loop feedback control algorithm for perfusion pressure, such as a pulse width modulation (PWM) signal or a specific analog voltage. The duty cycle or voltage amplitude of this signal has a precise linear relationship with the rotational speed of the peristaltic pump, thereby directly controlling the flow rate of the perfusion fluid, with an adjustment range of 0 to 100 ml / min. In this way, the real-time feedback from the MEMS pressure sensor is integrated into the control loop, achieving automated and precise control of the surgical field pressure.

[0076] The irrigation system also includes a negative pressure suction unit for connection to the interface on the handle of the coaxial endoscope module corresponding to the water outlet channel. This suction unit can be connected to the central negative pressure suction system of the operating room and the suction flow rate can be controlled manually or by the system via an electronically controlled regulating valve to coordinate with the irrigation process and maintain the cleanliness and pressure stability of the surgical field.

[0077] To further illustrate the present invention, the overall workflow of the novel needle-type neuroendoscopic direct visualization surgical system of the present invention will be described below in conjunction with a specific surgical application scenario.

[0078] Before the surgery begins, the operator first prepares and calibrates the system. The coaxial mirror module with integrated sensing functions, the functional support module, and the central processing and control imaging module are connected via their respective cables and conduits. After the system starts, the central processor automatically executes the initialization program. At this time, the coaxial mirror module's mirror body is in a naturally straight state, and the fiber optic demodulator records the initial center wavelength of all FBG sensing points as the zero-point reference for strain calculation. Simultaneously, the MEMS pressure sensor performs zero-point calibration while in contact with the atmosphere. The operator sets the target pressure value for the surgery and various safety warning thresholds in the interaction and control area through the human-machine interface.

[0079] The surgery begins with the operator holding the handle of the coaxial endoscope module and, with the assistance of the image navigation system, proceeding to puncture the target lesion area. During this process, as the endoscope moves within the tissue and may bend, the morphology sensing element (i.e., the FBG array) detects the strain distribution of each segment of the endoscope in real time and transmits the data to the central processing unit. A three-dimensional morphology reconstruction algorithm for the endoscope is executed in real time, and its calculation result is superimposed on the real-time endoscopic video stream in the main display area of ​​the monitor as a semi-transparent virtual curve. This curve visually reproduces the actual three-dimensional path and shape of the endoscope within the invisible tissue, providing the operator with real-time navigation feedback to avoid critical functional areas.

[0080] When the working tip approaches or contacts the target tissue, the multimodal biosensing tip begins to synchronously acquire data. The optical imaging unit provides a high-resolution image of the tissue surface; the bioelectrical impedance sensor, under the control of the central processing unit, injects a set of weak sinusoidal currents at a preset frequency (e.g., 1 kHz to 1 MHz) into the tissue and measures the corresponding voltage response to obtain a complex impedance spectrum; the MEMS pressure sensor measures the contact pressure between the tip and the tissue.

[0081] Upon receiving this synchronized multi-source data, the central processing unit immediately initiates a multi-source information fusion and tissue classification algorithm. The algorithm extracts feature vectors from the image, electrical impedance spectrum, and pressure data, and fuses them into a high-dimensional feature vector, which is then input into a pre-trained support vector machine (SVM) classifier. The classifier outputs the classification result of the currently contacted tissue in real time, such as normal tissue or tumor tissue. This result is then rendered as a highlighted area of ​​a specific color, superimposed on the center of the image in the main display area of ​​the monitor, providing the operator with objective and quantitative evidence for tissue identification.

[0082] After confirming that the lesion area has been reached, the operator inserts the treatment device, such as a radiofrequency ablation electrode, through the instrument channel. Simultaneously, the perfusion system is activated, infusing physiological saline into the surgical field through the inlet channel. At this point, the perfusion pressure closed-loop feedback control algorithm is activated. MEMS pressure sensors continuously monitor the actual pressure within the surgical field. The PID controller within the central processing unit compares this real-time value with a preset target value and dynamically calculates a control signal based on the error. This control signal output interface precisely adjusts the rotational speed of the peristaltic pump in the function support module. This closed-loop control loop automatically maintains the surgical field pressure within the set safe range. The operator initiates radiofrequency treatment through the interactive interface, and the central processing unit sends instructions to the energy platform to precisely control the treatment power and duration.

[0083] After the treatment was completed, the operator smoothly withdrew the coaxial mirror module under the continuous guidance of the 3D morphological model. All sensor data, processing results, and fusion videos throughout the entire surgical process were recorded by a solid-state storage drive for subsequent analysis and archiving.

Claims

1. A novel ultramicro neuroendoscope surgery system with therapeutic function, characterized in that, include: The coaxial mirror module with integrated sensing function has two independent channels inside the mirror body (11), which are isolated from each other. These are an instrument channel (1) for accommodating therapeutic instruments and an infusion channel (2) for delivering fluid. The distal end of the mirror body (11) is provided with a multimodal biosensing tip (3) for acquiring optical image signals and at least one non-optical biosensing signal during surgery. The functional support module is used to provide energy and fluid to the coaxial mirror module with integrated sensing function; The central processing and control imaging module includes a monitor (4) and a central processor (5). The central processor (5) is used to receive and fuse the optical image signal and the non-optical biosensor signal in real time, generate diagnostic information for decision support, and display the diagnostic information and the optical image signal superimposed on the monitor (4).

2. The novel ultramicro neuroendoscope system with therapeutic function according to claim 1, characterized in that, The instrument channel (1) is a central channel, and the infusion channel (2) is one or more peripheral channels arranged around the central channel.

3. The novel ultramicro neuroendoscope system with therapeutic function according to claim 1, characterized in that, The multimodal biosensing tip (3) integrates the following internally: An optical imaging unit used to acquire the optical image signals; Bioelectrical impedance sensor (6) for acquiring electrical characteristic signals as non-optical biosensing signals; Microelectromechanical system pressure sensor (7) for acquiring force signals as non-optical biosensing signals of the aforementioned biosensing signals.

4. A novel ultra-micro neuroendoscopic surgical system with therapeutic function according to claim 3, characterized in that, The steps by which the central processing and control imaging module generates the diagnostic information include: Based on the electrical characteristic signal and the force signal, the tissue type contacted by the multimodal biosensor tip (3) is classified in real time, and the classification result is used as part of the diagnostic and treatment information.

5. A novel ultra-micro neuroendoscopic surgical system with therapeutic function according to claim 1, characterized in that, The coaxial mirror module with integrated sensing function is also provided with a shape sensing element (8) on the mirror body (11). The shape sensing element (8) is used to collect signals of strain or curvature parameters that characterize the three-dimensional spatial shape of the mirror body (11). The strain or curvature parameter signals are part of the non-optical biosensing signals.

6. A novel ultra-micro neuroendoscopic surgical system with therapeutic function according to claim 5, characterized in that, The step of generating the diagnostic information by the central processing and control imaging module further includes: Based on the strain or curvature parameter signal, a three-dimensional spatial morphology model of the mirror body (11) is reconstructed, and the three-dimensional spatial morphology model is used as part of the diagnostic and treatment information.

7. A novel ultra-micro neuroendoscopic surgical system with therapeutic function according to claim 5, characterized in that, The morphological sensing element (8) is a fiber Bragg grating array (9) implanted along the length direction of the mirror body (11).

8. A novel ultra-micro neuroendoscopic surgical system with therapeutic function according to claim 4, characterized in that, The central processing and control imaging module is also used for: A closed-loop feedback control loop for treatment parameters is established, using the force signal as feedback to automatically adjust the output pressure of the perfusion system in the functional support module.

9. A novel ultra-micro neuroendoscopic surgical system with therapeutic function according to claim 1, characterized in that, The superposition display of the diagnostic information and the optical image signal on the monitor (4) is an augmented reality display.

10. A novel ultra-micro neuroendoscopic surgical system with therapeutic function according to claim 3, characterized in that, The bioelectrical impedance sensor (6) adopts a four-electrode structure, which includes a pair of driving electrodes for injecting current and a pair of sensing electrodes for measuring voltage.

Images