Multi-wavelength AI control device, method and system for minimally invasive spine surgery

Abstract

The invention is suitable for the technical field of spinal endoscopes, and provides a multi-wavelength AI control device, method and system for minimally invasive spine surgery, and the device comprises a light source module which is used for generating a plurality of irradiation light rays; the interior of the working pipeline is hollow, and the working pipeline is used for arranging an optical fiber imaging module in a penetrating mode and providing illumination for an imaging lens of the optical fiber imaging module; the AI analysis module is used for performing AI real-time analysis on the imaging image acquired by the optical fiber imaging module to obtain a tissue type in the imaging image; a plurality of light source control schemes are preset in the logic control module. Compared with the prior art, the method has the advantages that by means of the three-in-one method of multi-wavelength optical imaging, AI real-time analysis and self-adaptive light source control, the tissue identification precision, the operation view imaging stability and the operation safety of the device are remarkably improved, and the method is particularly suitable for complex scenes needing high-precision dissection and separation.

Description

Technical Field

[0001] The present invention belongs to the technical field of spinal endoscopes, and particularly relates to a multi-wavelength AI control device, method, and system for minimally invasive spinal surgery. Background Art

[0002] A spinal endoscope is a key tool for minimally invasive spinal surgery, and its structural design is aimed at entering the spinal canal or intervertebral foramen area through a small incision to achieve precise observation and operation. A spinal endoscope generally includes an optical imaging system, an illumination system, an instrument channel, a flushing / aspiration channel, an outer sheath and dilation system, and related auxiliary support systems. Through the mutual cooperation of the above structures, the spinal endoscope can complete complex operations such as discectomy, nerve decompression, and vertebral body fusion under minimally invasive conditions, significantly reducing the postoperative recovery time.

[0003] The imaging effect is a key factor in measuring the performance of a spinal endoscope system. In current relatively advanced systems, there are many auxiliary imaging means. For example, in combination with a visual navigation system, an autonomous navigation technology based on computer vision guidance is combined with a manipulator propulsion mechanism, and the intervention resistance is monitored through a force sensor to reduce the dependence on the experience of the operator. Also, for example, in combination with a deep learning algorithm to enhance the imaging image, the color restoration of the original image is optimized by AI to eliminate defects in the image such as artifacts and dark parts.

[0004] However, existing spinal endoscope imaging devices or methods often perform post-processing and data processing on the original imaging image, and cannot effectively improve the quality of the original imaging image. Therefore, it is necessary to further improve the existing spinal endoscope. Summary of the Invention

[0005] The purpose of the embodiments of the present application is to provide a multi-wavelength AI control method for minimally invasive spinal surgery, aiming to solve the problem that existing spinal endoscope imaging devices or methods often perform post-processing on the original imaging image and cannot effectively improve the quality of the original imaging image.

[0006] The embodiments of the present application are implemented as follows. A multi-wavelength AI control device for minimally invasive spinal surgery is provided, and the device includes:

[0007] A light source module for generating a plurality of irradiation lights;

[0008] A working pipeline, the inside of the working pipeline is hollow for passing through a fiber optic imaging module, and an optical fiber array is integrated on the inner wall of the working pipeline. The optical fiber array is used to guide the irradiation lights to the end outlet of the working pipeline and provide illumination for the imaging lens of the fiber optic imaging module;

[0009] An AI analysis module for performing real-time AI analysis on the imaging images obtained by the fiber optic imaging module to obtain the tissue types in the imaging images; and

[0010] A logic control module pre-set with a number of light source control schemes, each of the light source control schemes being used to control the generation of an illumination light of one wavelength or a combined illumination light of multiple wavelengths in the light source module;

[0011] The logic control module is further configured to obtain the analysis result of the tissue type and select the corresponding light source control scheme based on different analysis results.

[0012] Another object of the embodiments of the present application is to provide a multi-wavelength AI control method for spinal minimally invasive surgery, the method comprising:

[0013] Obtain a preset optical configuration, control the light source module to output a preset light based on the preset optical configuration, the preset light being an illumination light of one wavelength or a combined illumination light of multiple wavelengths, and adjust the illumination light to a preset light intensity;

[0014] Obtain the imaging images collected in real time during the operation by the imaging module, and perform real-time AI analysis on the tissue types in the imaging images;

[0015] Obtain the light source control scheme corresponding to the tissue type, a number of light source control schemes being pre-set, each being used to control the light source module to generate an illumination light of one wavelength or a combined illumination light of multiple wavelengths;

[0016] Control the light source module to output the illumination light based on the light source control scheme.

[0017] Another object of the embodiments of the present application is to provide a multi-wavelength AI control system for spinal minimally invasive surgery, the system comprising a multi-wavelength AI control device for spinal minimally invasive surgery as described above, further comprising:

[0018] A fiber optic imaging module for collecting imaging images and simultaneously outputting the imaging images to a display and the AI analysis module.

[0019] A multi - wavelength AI control device for minimally invasive spinal surgery provided by an embodiment of the present application has the following prominent advantages: The light source module can provide combined irradiation modes of near - infrared, visible light, ultraviolet, etc. with different wavelengths, making full use of the absorption / reflection characteristic differences of different tissues for different wavelengths, and significantly enhancing the image contrast from the source. By dynamically switching the light source wavelength, layered imaging and automatic adjustment are achieved, significantly improving the imaging quality and reducing the misjudgment risk of medical staff. The AI analysis module can identify the tissue type in real - time and provide adaptive real - time light source switching control based on the actual scenario and requirements of the surgery. The trinity of multi - wavelength optical imaging, AI real - time analysis, and adaptive light source control significantly improves the tissue identification accuracy, the imaging stability of the surgical field, and the operation safety, especially suitable for complex scenarios that require high - precision anatomical separation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of a multi - wavelength AI control device for minimally invasive spinal surgery provided by an embodiment of the present application;

[0021] Figure 2 It is a schematic diagram of an imaging image of a vascular tissue provided by an embodiment of the present application;

[0022] Figure 3 It is a flowchart of a multi - wavelength AI control method for minimally invasive spinal surgery provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0024] It can be understood that the terms "first", "second", etc. used in the present application can be used in this document to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish the first unit or module from another unit or module. For example, without departing from the scope of the present application, the first light source can be called the second light source, and similarly, the second light source can be called the first light source.

[0025] In an embodiment of the present application, a multi - wavelength AI control device for minimally invasive spinal surgery is provided, and at least the following several modules are included in the device:

[0026] A light source module for generating a plurality of irradiation lights; a working pipeline, the interior of the working pipeline is hollow for threading a fiber optic imaging module, and a fiber optic array is integrated on the inner wall of the working pipeline. The fiber optic array is used to guide the irradiation lights to the end outlet of the working pipeline and provide illumination for the imaging lens of the fiber optic imaging module; an AI analysis module for performing real-time AI analysis on the imaging images obtained by the fiber optic imaging module to obtain the tissue types in the imaging images; a logic control module, which presets a plurality of light source control schemes, and each light source control scheme is used to control the generation of an irradiation light of one wavelength or a combined irradiation light of multiple wavelengths in the light source module; the logic control module is further used to obtain the analysis results of the tissue types and select the corresponding light source control scheme based on different analysis results.

[0027] As known to those skilled in the art, the current mainstream endoscopic systems often adopt a single white light or a single-wavelength near-infrared light source. However, the researchers of this application considered that since different tissues in the body, such as the ligamentum flavum, nerves, blood vessels, etc., have different absorption characteristics for the light source, the traditional endoscopic systems using a single light source are still difficult to effectively improve the imaging quality of the imaging images even in combination with various advanced auxiliary image processing methods later.

[0028] In this embodiment, the light source module uses a combined irradiation mode of near-infrared, visible light, ultraviolet, etc. with different wavelengths, makes full use of the absorption / reflection characteristic differences of different tissues for different wavelengths, significantly enhances the contrast of the image from the source, dynamically switches the light source wavelength, realizes layered imaging and automatic adjustment, significantly improves the imaging quality, reduces the misjudgment risk of medical staff, and avoids the problem of missing surgical field information caused by a single light source in traditional equipment.

[0029] Moreover, the traditional method relies on the naked-eye judgment of the operator and is easily limited by experience. The AI analysis module provided by the device in this application can identify the tissue type in real time and can provide adaptive real-time light source switching control based on the actual scenario and requirements of the operation. Compared with the traditional minimally invasive spine technology, this application significantly improves the tissue identification accuracy, the imaging stability of the surgical field, and the operation safety through the three-in-one design of multi-wavelength optical imaging, AI real-time analysis, and adaptive light source control, and is especially suitable for complex scenarios that require high-precision anatomical separation.

[0030] In this embodiment, as Figure 1 shown, a schematic diagram of the whole and part of a multi-wavelength AI control device for spine minimally invasive surgery is given. Figure 1Part a in it is the structural model of the whole device, and parts b, c, and d are the effect diagrams of the light source in white light, blue light, and yellow light modes respectively. The working pipeline is hollow inside and is inserted into the patient's body during surgery. Its end is close to the part that needs surgery. Its interior can accommodate operating instruments, an optical imaging module, etc. The optical imaging module can adopt a fiber optic imaging module for imaging. The end of the optical fiber is an imaging lens. The AI analysis module can obtain the images collected by the imaging lens.

[0031] To further reduce the time delay of AI analysis, the AI analysis model in the AI analysis module can be run in a locally deployed manner on a general-purpose processor or a dedicated AI processing chip. The AI analysis model can adopt a pre-trained lightweight Transformer model, etc., to identify tissue types such as nerves, blood vessels, and fat in the surgical field in real time. For example, the model can automatically identify and label blood vessel tissues as shown in Figure 2 It is shown. The AI analysis model can analyze the endoscopic images in real time and detect the key anatomical structures in the field of view.

[0032] Several light source control schemes preset by the logic control module can be a logic control table, which pre-stores the optimal wavelength combinations matching each tissue type, and switches the wavelength in real time according to the signals output by the AI analysis module. At the same time, it can also adjust the light intensity of the irradiated light. That is, after obtaining the tissue type output by the AI analysis module, select the light source control scheme corresponding to this tissue type. For example, when the AI identifies that the tissue in the imaging image is nerve tissue, it automatically turns off 405nm and enables the combination of 540nm and 660nm to avoid tissue damage and improve the resolution of the image. At the same time, the light intensity is adjusted to a safe threshold. For example, the 540nm light source is adjusted to 500mW, and the 660nm light source is adjusted to 300mW. Also, for example, when it is identified that the tissue in the imaging image is the ligamentum flavum, it is adjusted to the combination of 540nm (the absorption peak of ligamentum flavum collagen) and 660nm (the low absorption area of adipose tissue), and the edge contrast is enhanced through differential absorption. Based on experimental data, the display contrast can be increased by about 40%. Also, for example, when it is identified that the tissue in the imaging image is a blood vessel, it is adjusted to the dual-wavelength combination of 760nm (the absorption peak of deoxyhemoglobin) and 850nm (the absorption peak of oxyhemoglobin), and the difference in hemoglobin absorption is used to distinguish arteries and veins, which is especially suitable for special situations such as intraoperative bleeding. The core use of the light source control scheme for the light source is to select the wavelength combination that can maximize the contrast between the target tissue and the background based on the optical characteristics of the tissue type.

[0033] Preferably, the device in the present application may further include other optical structures. For example, an optical fiber coupler: a multi-channel optical fiber bundle is adopted, preferably with a core diameter of 200 μm and NA = 0.22, which is used to couple the output light of different light sources in the light source module, such as LED light sources and laser light sources, to the same transmission optical fiber, reducing optical path crosstalk. Also for example, a collimating lens group: an aspherical lens is arranged at the front end of each light source to collimate the divergent light into a parallel light beam, improving the light energy utilization rate and having a higher coupling efficiency.

[0034] In a preferred embodiment, the light source module includes: a tunable laser and several fixed-wavelength light sources;

[0035] The wavelength range that the tunable laser can emit is 280 nm - 1600 nm;

[0036] The fixed-wavelength light sources include at least one of the 405 nm, 540 nm, 660 nm, 760 nm, 850 nm, and 940 nm wavelength light sources.

[0037] In the embodiment of the present application, the fixed-wavelength light sources can adopt an LED light source group. The LED light source group and the laser are arranged in a circular array and symmetrically distributed around the endoscopic optical channel to ensure uniform light projection onto the surgical area. The LED light source group can simultaneously include 6 groups of high-power LEDs, namely 405 nm, 540 nm, 660 nm, 760 nm, 850 nm, and 940 nm, arranged in ascending order of wavelength. Each group is equipped with an independent heat sink and is connected to a thermoelectric cooler through a copper substrate. The tunable laser can be located at the center of the array. A distributed feedback laser is adopted, with a wavelength coverage of 280 - 1600 nm, and precise wavelength tuning is achieved through a fiber Bragg grating, and the precision can be controlled within ±1 nm.

[0038] In a preferred embodiment, the working pipeline is transparent or semi-transparent and is made of a composite material of polypropylene and polyamide; an anti-reflection film is plated on the inner wall of the working pipeline.

[0039] In the embodiment of the present application, the working pipeline is a blend of polypropylene (PP) and polyamide (PA6) in a ratio of 7:3, and 1 wt% of nano-SiO2 is also added to improve wear resistance. The whole is injection molded at a temperature of 220 °C and a pressure of 80 MPa, and an anti-reflection film of MgF2 with a thickness of 100 nm is plated on the inner wall.

[0040] In a preferred embodiment, as Figure 3 shown, a multi-wavelength AI control method for spinal minimally invasive surgery is proposed. It can be understood that this method can be applied and implemented in the above-mentioned multi-wavelength AI imaging device for spinal minimally invasive surgery. This method can run within a general-purpose processor or a dedicated AI processing chip or module in the device, and specifically may include the following steps:

[0041] Step S10, obtain a preset optical configuration, control the light source module to output preset light rays based on the preset optical configuration, where the preset light rays are illumination light rays of one wavelength or a combined illumination light ray of multiple wavelengths, and adjust the illumination light rays to a preset light intensity.

[0042] In this embodiment, the method for obtaining the preset optical configuration is as follows: pre-load tissue optical parameters according to the patient's CT / MRI data. For example, according to the preoperative CT / MRI data, quantify the absorption ability of the tissue for a specific wavelength, evaluate the tissue scattering characteristics through diffuse optical tomography (DOT), optimize the light source penetration depth, and generate an initial wavelength scheme.

[0043] In the embodiment of the present application, the calculation method of the initial light intensity can be: estimate the blood loss (i.e., depth) according to the preoperative CT / MRI data, and then calculate the penetration depth that different incident light intensities can have according to the Beer-Lambert law to obtain the initial light intensity.

[0044] Step S20, obtain the imaging images collected in real time during the operation by the imaging module, and analyze the tissue types in the imaging images in real time based on AI.

[0045] In this embodiment, the AI real-time analysis model significantly improves the accuracy and safety of spinal endoscopic surgery through anatomical structure recognition and wavelength dynamic matching.

[0046] Step S30, obtain the light source control scheme corresponding to the tissue type, where there are several preset light source control schemes, and each is used to control the light source module to generate an illumination light ray of one wavelength or a combined illumination light ray of multiple wavelengths.

[0047] Step S40, control the light source module to output illumination light rays based on the light source control scheme.

[0048] In the embodiment of the present application, the AI model identifies the tissue type through real-time image analysis, dynamically matches the optimal light source scheme, and combines multi-wavelength combined illumination, which can simultaneously enhance the imaging effects of different tissues. The light source module can provide combined illumination modes of different wavelengths such as near-infrared, visible light, and ultraviolet, making full use of the absorption / reflection characteristic differences of different tissues for different wavelengths, and significantly enhancing the image contrast from the source. By dynamically switching the light source wavelength, layered imaging and automatic adjustment are realized, significantly improving the imaging quality and reducing the misjudgment risk of medical staff. The AI analysis module can identify the tissue type in real time and provide adaptive real-time light source switching control based on the actual scenario and requirements of the operation. The trinity of multi-wavelength optical imaging, AI real-time analysis, and adaptive light source control significantly improves the tissue identification accuracy, the imaging stability of the surgical field, and the operation safety, especially suitable for complex scenarios that require high-precision anatomical separation.

[0049] In a preferred embodiment, the light source control scheme includes:

[0050] When the tissue type is identified as nerve, the wavelength of the irradiation light output by the light source module is 660 nm, and the irradiation light with a wavelength in the ultraviolet band is disabled;

[0051] When the tissue type is identified as ligamentum flavum, the wavelengths of the irradiation light output by the light source module are 540 nm and 660 nm;

[0052] When the tissue type is identified as blood vessel, the wavelengths of the irradiation light output by the light source module are 760 nm and 860 nm;

[0053] When the tissue type is identified as intervertebral disc, the wavelength of the irradiation light output by the light source module is 980 nm;

[0054] When the tissue type is identified as collagen denaturation region, the wavelengths of the irradiation light output by the light source module are 280 and 980 nm;

[0055] When the tissue type is identified as adipose tissue, the wavelength of the irradiation light output by the light source module is 660 nm;

[0056] When the tissue type is identified as bony structure, the wavelength of the irradiation light output by the light source module is 850 nm.

[0057] In the embodiment of the present application, corresponding control schemes for the irradiation light are respectively set for different tissue types to further improve the degree of intelligence and the safety of lighting. For example, if nerve tissue is detected, the ultraviolet band is automatically blocked, and the safe wavelength of 660 nm can be selected to be enabled.

[0058] As a preferred embodiment of the present application, when there are multiple types of tissues in the field of view, after the tissue type is judged, a contrast judgment is also required. Only when the contrast is lower than the threshold value, the switching of the irradiation light is performed to avoid frequent switching.

[0059] In this embodiment, an intelligent decision is constructed. First, based on real-time AI analysis, the most important or key tissue type in the imaging image is determined, and then the contrast is evaluated. The contrast index of the current wavelength combination, such as Weber contrast, is calculated in real time. If the contrast meets the requirements, even if the tissue type in the field of view changes, the switching can be temporarily not performed. When the contrast is lower than the preset contrast threshold value, a switching instruction is triggered to switch to other light source irradiation modes to make the contrast greater.

[0060] In a preferred embodiment, the system also includes overheat protection. For example, if the proportion of saturated pixels in any wavelength image > 5%, the light intensity is automatically reduced to 80% of the original value; when the temperature sensor detects high temperature, a power reduction mode is triggered.

[0061] In a preferred embodiment, the different light source control schemes are preset with the same or different control priorities; the light source control scheme corresponding to the tissue type of arteriovenous bleeding has the highest priority;

[0062] When there are multiple different tissue types in the imaging image simultaneously, the light source control scheme with the highest control priority is preferentially selected.

[0063] In the embodiment of the present application, in order to cope with sudden emergencies such as bleeding or preferentially display key tissue types, different priorities are set. For example, if it is necessary to resect the intervertebral disc, 980nm (water absorption peak) penetration imaging is enabled to locate the lesion area in combination with 280nm (collagen characteristic absorption); if vascular bleeding is recognized at this time, it is automatically switched to the dual-wavelength mode of 760nm (deoxyhemoglobin absorption peak) + 850nm (oxyhemoglobin absorption peak) to enhance vascular imaging.

[0064] In a preferred embodiment, the method for analyzing the tissue type in the imaging image is as follows:

[0065] Analyze the tissue type based on a pre-trained lightweight Transformer model;

[0066] The encoder of the lightweight Transformer model is used to extract local features of the multi-wavelength image, the attention mechanism is used to weighted focus on the key areas in the imaging image, and the classification head is used to output the tissue type and the judgment probability.

[0067] In the embodiment of the present application, by using a lightweight Transformer model, it is possible to identify tissue types such as nerves, blood vessels, and fat in the surgical field in real time with extremely low latency. The model can determine the tissue type through the following feature analyses: texture analysis, extracting tissue texture through the gray-level co-occurrence matrix, such as the fibrous structure of the ligamentum flavum and the uniformity of fat. Edge gradient, calculating the image gradient amplitude to identify the sharpness of the boundary between the ligamentum flavum and the surrounding tissues. Color distribution, in RGB or multi-spectral images, statistically analyzing the pixel intensity distribution at different wavelengths.

[0068] In the embodiment of the present application, based on the above model, the system can perform dynamic feature capture: for example, real-time bleeding detection, by real-time monitoring the pixel mutation in the red channel of the image to judge the bleeding area. Evaluating the dynamic change of tissue water content through the light attenuation rate of the 940nm near-infrared band to obtain the edema change of the key part.

[0069] In a preferred embodiment, the method further includes:

[0070] Obtain the tissue type in the imaging image obtained by real-time acquisition, obtain the thermal damage threshold corresponding to the tissue type, and set the safety range of the irradiation light intensity based on the thermal damage threshold.

[0071] In the embodiments of the present application, according to the ISO15004-2 standard, the light power density on the tissue surface ≤ 1 W / cm². For example, when the spot diameter of 760 nm is 3 mm, 800 mW corresponds to a power density ≈ 1.13 W / cm², approaching the safety upper limit.

[0072] In a preferred embodiment, the method further includes:

[0073] Obtain the contrast data and signal-to-noise ratio data of the imaging image obtained by real-time acquisition, and control the irradiation light intensity of the irradiation light;

[0074] The control modes for controlling the irradiation light intensity include a contrast priority mode and a signal-to-noise ratio priority mode;

[0075] In the signal-to-noise ratio priority mode, obtain the signal-to-noise ratio histogram of the imaging image based on the signal-to-noise ratio data, and adjust the irradiation light intensity based on the signal-to-noise ratio histogram to reduce the image signal-to-noise ratio;

[0076] In the contrast priority mode, based on the contrast data, use the gradient descent algorithm to adjust the irradiation light intensity to improve the contrast of the image.

[0077] In the embodiments of the present application, the Weber contrast of the dual-wavelength image can be calculated. If the contrast is less than a certain value, the light intensity is adjusted according to the gradient descent algorithm. For example, for 760 nm, it increases by 50 mW per gradient, and for 850 nm, it decreases by 30 mW per gradient. The signal-to-noise ratio can be obtained by analyzing the image histogram. The ultimate goal of the adjustment is to make the image clearer.

[0078] In the embodiments of the present application, the corresponding curve of the relationship between the light intensity and the contrast can be obtained based on experiments. For example, in vitro blood vessel experiments show that when the light intensity of 760 nm increases from 500 mW to 800 mW, the contrast increases from about 0.4 to about 0.75.

[0079] In a preferred embodiment, the system includes a multi-wavelength AI control device for minimally invasive spinal surgery as described above, and further includes:

[0080] An optical fiber imaging module, which is used to acquire the imaging image and output the imaging image to the display and the AI analysis module simultaneously.

[0081] In this embodiment, the system is integrated with an imaging system and an imaging AI lighting system. The advantage of this system is that the light source module can provide combined irradiation modes of near-infrared, visible light, ultraviolet, etc. with different wavelengths, making full use of the absorption / reflection characteristic differences of different tissues for different wavelengths, and significantly enhancing the image contrast from the source. By dynamically switching the light source wavelength, layer imaging and automatic adjustment are achieved, significantly improving the imaging quality and reducing the misjudgment risk of medical staff. The AI analysis module can identify the tissue type in real time and provide adaptive real-time light source switching control based on the actual scenario and requirements of the operation. The trinity of multi-wavelength optical imaging, AI real-time analysis, and adaptive light source control significantly improves the tissue identification accuracy, the imaging stability of the surgical field, and the operation safety, and is particularly suitable for complex scenarios that require high-precision anatomical dissection.

[0082] It should be understood that although the steps in the flowcharts of the embodiments of the present application are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limitation, and these steps can be executed in other orders. Moreover, at least a part of the steps in each embodiment may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

[0083] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0084] The technical features of the above-described embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0085] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A multi-wavelength AI control device for minimally invasive spinal surgery, characterized in that, The device includes: A light source module for generating a plurality of irradiation lights; A working pipeline, which is hollow inside for passing through a fiber optic imaging module. An optical fiber array is integrated on the inner wall of the working pipeline. The optical fiber array is used to guide the irradiation lights to the end outlet of the working pipeline and provide illumination for the imaging lens of the fiber optic imaging module; An AI analysis module for performing real-time AI analysis on the imaging images obtained by the fiber optic imaging module to obtain the tissue types in the imaging images; and A logic control module, which presets a plurality of light source control schemes. Each of the light source control schemes is used to control the generation of an irradiation light of one wavelength or a combined irradiation light of multiple wavelengths in the light source module; The logic control module is further used to obtain the analysis result of the tissue type and select the corresponding light source control scheme based on different analysis results.

2. The multi-wavelength AI control device for spinal minimally invasive surgery according to claim 1, characterized in that, The light source module includes: a tunable laser and a plurality of fixed-wavelength light sources; The wavelength range that the tunable laser can emit is 280nm - 1600nm; The fixed-wavelength light sources include at least one of the light sources with wavelengths of 405nm, 540nm, 660nm, 760nm, 850nm, and 940nm.

3. The multi-wavelength AI control device for spinal minimally invasive surgery according to claim 1, wherein The working pipeline is transparent or semi-transparent and is made of a composite material of polypropylene and polyamide; an anti-reflection film is coated on the inner wall of the working pipeline.

4. A multi-wavelength AI control method for minimally invasive spinal surgery, characterized in that, The method includes: Obtaining a preset optical configuration, controlling the light source module to output a preset light based on the preset optical configuration. The preset light is an irradiation light of one wavelength or a combined irradiation light of multiple wavelengths, and adjusting the irradiation light to a preset light intensity; Obtaining the imaging images collected in real time during the operation by the imaging module, and performing real-time AI analysis on the tissue types in the imaging images; Obtaining the light source control scheme corresponding to the tissue type. There are a plurality of preset light source control schemes, and each is used to control the generation of an irradiation light of one wavelength or a combined irradiation light of multiple wavelengths by the light source module; Controlling the light source module to output irradiation light based on the light source control scheme.

5. A multi-wavelength AI control method for spinal minimally invasive surgery according to claim 4, characterized in that, The light source control scheme includes: When the tissue type is recognized as nerve, the wavelength of the irradiation light output by the light source module is 660nm, and the irradiation light with a wavelength in the ultraviolet band is disabled; When the tissue type is recognized as ligamentum flavum, the wavelengths of the irradiation light output by the light source module are 540nm and 660nm; When the tissue type is recognized as blood vessel, the wavelengths of the irradiation light output by the light source module are 760nm and 860nm; When the tissue type is recognized as intervertebral disc, the wavelength of the irradiation light output by the light source module is 980nm; When the tissue type is recognized as a collagen denaturation region, the wavelengths of the irradiation light output by the light source module are 280 and 980nm; When the tissue type is recognized as adipose tissue, the wavelength of the irradiation light output by the light source module is 660nm; When the tissue type is recognized as bony structure, the wavelength of the irradiation light output by the light source module is 850nm.

6. A multi-wavelength AI control method for spinal minimally invasive surgery according to claim 4, characterized in that Different light source control schemes preset the same or different control priorities; the light source control scheme corresponding to arteriovenous hemorrhage has the highest priority; When there are multiple different tissue types in the imaging image, the light source control scheme with the highest control priority is preferentially selected.

7. The multi-wavelength AI control method for minimally invasive spinal surgery according to claim 4, characterized in that The method for analyzing the tissue types in the imaging image is as follows: Analyze the tissue types based on a pre-trained lightweight Transformer model; The encoder of the lightweight Transformer model is used to extract local features of the multi-wavelength image, the attention mechanism is used to weight and focus on the key areas in the imaging image, and the classification head is used to output the tissue types and judgment probabilities.

8. A multi-wavelength AI control method for spinal minimally invasive surgery according to claim 4, characterized in that, The method further includes: Obtain the tissue types in the imaging image collected in real time, obtain the thermal damage threshold corresponding to the tissue types, and set the safety range of the irradiation light intensity based on the thermal damage threshold.

9. The multi-wavelength AI control method for spinal minimally invasive surgery according to claim 4, wherein, The method further includes: Obtain the contrast data and signal-to-noise ratio data of the imaging image collected in real time, and control the irradiation light intensity of the irradiation light; The control modes for controlling the irradiation light intensity include a contrast priority mode and a signal-to-noise ratio priority mode; In the signal-to-noise ratio priority mode, obtain the signal-to-noise ratio histogram of the imaging image based on the signal-to-noise ratio data, and adjust the irradiation light intensity based on the signal-to-noise ratio histogram to reduce the image signal-to-noise ratio; In the contrast priority mode, adjust the irradiation light intensity using the gradient descent algorithm based on the contrast data to improve the contrast of the image.

10. A multi-wavelength AI control system for spinal minimally invasive surgery, characterized in that, The system includes a multi-wavelength AI control device for minimally invasive spinal surgery according to any one of claims 1-3, and further includes: An optical fiber imaging module, which is used to collect imaging images and output the imaging images to the display and the AI analysis module at the same time.

Images