Ultrasonic knife energy control method and system based on artificial intelligence and ultrasonic knife equipment

Abstract

The invention discloses an ultrasonic scalpel energy control method and system based on artificial intelligence and ultrasonic scalpel equipment. The method comprises the following steps that image or video data of an operation area are obtained in real time; inputting the image or video data into a pre-trained artificial intelligence model, analyzing the operation image or video data, and obtaining a tissue state parameter set output by the artificial intelligence model; based on the organization state parameter set, a target energy interval is determined according to a preset energy mapping rule, and the target energy interval is limited by a minimum target energy value and a maximum target energy value; and controlling the energy output of the ultrasonic knife to enable the energy output to be restrained in the target energy interval. The tissue type is identified in an artificial intelligence mode, so that energy output is controlled, the tissue separation condition is identified in the artificial intelligence mode, energy output is automatically controlled, and a prompt is sent out.

Description

Technical Field

[0001] This invention relates to the field of ultrasonic scalpel energy control, and in particular to an ultrasonic scalpel energy control method, system, and ultrasonic scalpel device based on artificial intelligence. Background Technology

[0002] Ultrasonic scalpels can replace ordinary scalpels to cut and separate diseased tissues or organs, and to coagulate blood vessels or bleeding points to achieve surgical treatment. However, during ultrasonic scalpel cutting and coagulation, if the excitation is not stopped promptly after the tissue or blood vessel is severed, the scalpel tip will come into contact with the tissue pad, accelerating tissue pad wear and shortening the ultrasonic scalpel's lifespan. Simultaneously, the friction between the tissue pad and the scalpel tip will generate a large amount of heat, causing the tip temperature to rise sharply, affecting subsequent surgical procedures.

[0003] Current technologies set the impedance to a fixed range. As long as the impedance value falls within this range, it is considered that the tissue has been severed. If the severance value range is set to [a1, a2], then the current tissue severance must satisfy the impedance value a0∈[a1, a2] to determine tissue severance.

[0004] This method has significant limitations and requires high-quality materials.

[0005] For different organizations, there may be different organizational boundaries, requiring too many matching ranges and limiting the scope of practicality. Summary of the Invention

[0006] The purpose of this invention is to provide an artificial intelligence-based ultrasonic scalpel energy control method. This method identifies tissue types through artificial intelligence, thereby controlling the energy output. This solves the problem that in actual use, the current cannot be adjusted at the same setting, which prevents dynamic control of energy output and affects the cutting and coagulation effect. Furthermore, it identifies tissue detachment through artificial intelligence, automatically controls the energy output, and issues prompts.

[0007] The technical solution of this invention is: An AI-based ultrasonic scalpel energy control method includes the following steps: Step S1: Acquire image or video data of the surgical area in real time; Step S2: Input the image or video data into a pre-trained artificial intelligence model, analyze the surgical image or video data, and obtain the tissue state parameter set output by the artificial intelligence model. The tissue state parameter set includes at least tissue state probability parameters, risk level parameters, surgical stage parameters, and model confidence parameters. Step S3: Based on the set of tissue state parameters, determine the target energy range according to the preset energy mapping rules. The target energy range is limited by the minimum target energy value and the maximum target energy value. Step S4: Control the energy output of the ultrasonic scalpel to confine it within the target energy range.

[0008] Preferably, in step S3, determining the target energy range based on the tissue state parameter set specifically includes: S31: Determine the main organizational state based on the organizational state probability parameters: Select the organizational state with the highest probability value among the organizational state probability parameters as the main organizational state; S32: Obtain the corresponding basic energy range based on the main tissue state; S33: The basic energy range is modified at least according to the surgical stage parameters and / or the risk level parameters to obtain a modified energy range; S34: Based on the comparison result between the model confidence parameter and the preset confidence threshold, select the corrected energy range or a preset safe energy range as the target energy range.

[0009] Preferably, step S33 includes: Based on the energy correction coefficients corresponding to the surgical stage parameters, the upper and lower limits of the basic energy range are scaled proportionally to obtain the first corrected energy range; Based on the risk level parameter, the upper limit of the first modified energy range is constrained to obtain a second modified energy range as the modified energy range.

[0010] Preferably, the method for determining the target energy range in step S3 further includes a weighted fusion method: Based on the probability values ​​in the organizational state probability parameters, the corresponding basic energy intervals are weighted and summed to obtain a comprehensive energy interval as the basis.

[0011] Preferably, in step S4, when controlling the energy output, a feedback value of historical energy output is also introduced. The energy setpoint calculated based on the target energy range is fused with the historical energy feedback value to smooth energy changes.

[0012] An energy control system for an ultrasonic scalpel includes: An image acquisition module is used to acquire image data of the surgical area in real time; an artificial intelligence recognition module is connected to the image acquisition module and is used to analyze the image data based on a pre-trained model and output the set of tissue state parameters. An energy range mapping module, connected to the artificial intelligence recognition module, is used to determine the target energy range based on the tissue state parameter set and a preset energy mapping rule; An energy control module, connected to the energy range mapping module, is used to generate control signals to constrain the energy output of the ultrasonic scalpel within the target energy range.

[0013] Preferably, it also includes a security monitoring and fallback module, which is configured to monitor the model confidence parameters; When the model confidence parameter is lower than the preset confidence threshold, the energy range mapping module is triggered to switch the target energy range to the safe energy range; and / or, the abnormal energy output status is monitored, and a safe rollback strategy is executed when an abnormality occurs.

[0014] Preferably, the energy control module specifically includes: The resonant frequency tracking unit is used to track and adjust the driving frequency of the ultrasonic transducer in real time to maintain the resonant state. The current closed-loop control unit, connected to the resonant frequency tracking unit and the energy range mapping module, is used to perform closed-loop regulation of the drive current according to the target energy range in the resonant state and limit the rate of change of the current.

[0015] The present invention also proposes an ultrasonic scalpel device, including the ultrasonic scalpel energy control system, ultrasonic transducer, and scalpel head.

[0016] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the ultrasonic scalpel energy control method.

[0017] Compared with the prior art, the technical solution of the present invention has the following significant advantages: 1. This invention uses AI visual recognition to determine tissue type, thickness, and surgical stage in real time, and dynamically adjusts the energy output range, achieving a leap from "fixed level" to "intelligent self-adaptation", which significantly improves cutting efficiency and coagulation effect, and reduces tissue carbonization and adhesion.

[0018] 2. This invention introduces a confidence-based safety fallback mechanism and risk level parameters. When the AI ​​identifies high uncertainty or high risk, the system can automatically switch to a conservative safety mode, which greatly improves the safety of the surgery.

[0019] 3. By accurately judging tissue detachment and promptly reducing or stopping energy output, this invention effectively avoids friction between the blade tip and the tissue pad under no-load conditions, reduces the blade tip temperature, and thus extends the service life of the ultrasonic scalpel and its components.

[0020] 4. The vision-based AI recognition method of this invention does not rely on the specific electrical properties of tissues, and can adapt to more diverse tissue types and surgical scenarios, making it far more versatile than impedance-based methods. Attached Figure Description

[0021] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a flowchart of an artificial intelligence-based ultrasonic scalpel energy control method. Figure 2 This is a structural diagram of an AI-based ultrasonic scalpel energy control system.

[0022] Figure 3 This is a structural diagram of an ultrasonic scalpel device proposed in this invention. Detailed Implementation

[0023] like Figure 1 As shown, the present invention provides an artificial intelligence-based ultrasonic scalpel energy control method, which includes the following steps: 1. The organization type and the status of separation are marked.

[0024] Acquire surgical videos (endoscopic, real-time surgical recording, etc.) and use annotation software to label tissue types (small intestine, stomach, pleura, arteries, veins, etc.) and transection status (transcribed, not transcribed).

[0025] 2. The model is trained using the YOLO model.

[0026] The model labeled above was trained by setting appropriate IOU, confidence, loss function, etc., using NVIDIA's CUDA.

[0027] 3. Call the trained model.

[0028] In actual surgery, the host computer that acquires the video calls the large model trained in the previous step.

[0029] 4. Analyze the tissue type and control the current output.

[0030] The current tissue type is determined by using the model called in the previous step, and the output current level is determined accordingly.

[0031] 5. Analyze the tissue separation situation in real time and control subsequent operations.

[0032] The trained model is invoked to analyze the tissue detachment in real time. If detachment occurs, an alert is issued, and the current output is reduced until the ultrasonic scalpel is stopped.

[0033] This invention trains a model using the YOLO model, which is then used to determine whether tissue has severed. It can identify different types of tissue and has a wide range of applications. Based on the YOLO model, the type of tissue is determined, and the energy output is dynamically adjusted to ensure that the cutter head temperature does not become too high, thus extending the cutter head's lifespan.

[0034] This invention relates to an artificial intelligence-based ultrasonic scalpel energy control method. In specific implementation, an ultrasonic scalpel energy range mapping control method is adopted. This method combines the tissue state parameters output by the artificial intelligence model with preset energy mapping rules to determine the target energy range, and adjusts the energy output of the ultrasonic scalpel in real time within the energy range to achieve precise cutting and coagulation control under different tissue thicknesses and tissue types.

[0035] In this embodiment, the artificial intelligence model does not directly output specific energy values, but outputs parameters related to the organization state. The system constrains the energy according to the mapping rules, thereby improving the system's security and stability.

[0036] The specific method steps in this embodiment include: Step S1: Obtain tissue information of the surgical area.

[0037] The image acquisition module installed on the ultrasonic scalpel or surgical device can acquire surgical images or video data of the surgical area in real time, and input the image or video data into the artificial intelligence recognition module.

[0038] Step S2: The artificial intelligence model outputs tissue status parameters.

[0039] The artificial intelligence recognition module analyzes the surgical image or video data based on a pre-trained model, and outputs at least the following parameters: Tissue state probability parameter P: used to characterize the probability distribution of the current surgical area belonging to different tissue states; Risk level parameter R: used to characterize the risk level of bleeding, carbonization or damage to the current tissue during cutting or coagulation; Surgical stage parameter S: used to characterize whether the current surgery is in the cutting stage, coagulation stage, or mixed stage; Confidence parameter C: used to characterize the reliability of the artificial intelligence model for the above output results.

[0040] Step S3: Determine the main organization status.

[0041] The energy range mapping module determines the main tissue state corresponding to the current surgical area based on the tissue state probability parameter P.

[0042] In this embodiment, the main organizational state is determined as follows: The organization state with the highest probability value among the organization state probability parameters is selected as the primary organization state.

[0043] Step S4: Obtain the basic energy range.

[0044] The system pre-stores multiple basic energy ranges, each corresponding to a different organizational state.

[0045] The energy range mapping module selects the corresponding basic energy range from the basic energy range based on the main organization state, as the initial energy control range.

[0046] Step S5: Adjust the energy range according to the stage of the surgery.

[0047] The energy range mapping module modifies the basic energy range according to the surgical stage parameter S to obtain the first modified energy range.

[0048] In this embodiment, different energy correction coefficients correspond to different surgical stages. By multiplying the upper and lower limits of the basic energy range by the corresponding correction coefficients, the cutting or coagulation requirements can be adapted.

[0049] Step S6: Set energy upper limit constraints based on risk level.

[0050] The energy range mapping module limits the maximum energy value of the first modified energy range according to the risk level parameter R, and obtains the second modified energy range.

[0051] When the risk level is high, the system automatically reduces the maximum permissible energy output to reduce the risk of tissue damage or abnormalities.

[0052] Step S7: Make a safety fallback judgment based on the confidence level.

[0053] The energy range mapping module determines whether the preset confidence threshold is met based on the confidence parameter C.

[0054] When the confidence parameter is greater than or equal to the preset threshold, the second corrected energy range is used as the target energy range; When the confidence parameter is lower than the preset threshold, the output of the artificial intelligence model is ignored, and the preset safe energy range is used as the target energy range.

[0055] Step S8: Perform energy control within the target energy range.

[0056] Under the constraints of the target energy range, the energy control module achieves real-time output control of ultrasonic scalpel energy by adjusting parameters such as ultrasonic drive current, drive voltage, or duty cycle.

[0057] In this embodiment, the energy control module does not allow energy output to exceed the target energy range. Energy range = AI output × rule mapping table; Organizational state probability vector: P = [P1, P2, P3], where P1: Probability of the first organizational state (low damping / thin). P2: Probability of the second organizational state (moderate); P3: Probability of the third organizational state (high damping / thickness).

[0058] Risk level parameter R ∈ {low, medium, high}; Surgical stage parameter S ∈ {cutting, coagulation, mixing}; The model confidence parameter C ∈ [0, 1].

[0059] Based on the tissue state parameters, surgical stage parameters, risk level parameters, and confidence level parameters output by the artificial intelligence model, the corresponding target energy range is determined through preset energy mapping rules, and the target energy range is used as a constraint input to the energy control module.

[0060] Multiple energy ranges are weighted and fused based on the probability of organizational state, E = Σ (Pi × E_base_i); Historical energy feedback E_final = f(E_AI, E_feedback) is introduced to enhance stability.

[0061] like Figure 2 As shown, the present invention also proposes an energy control system for an ultrasonic scalpel, comprising: An image acquisition module is used to acquire image data of the surgical area in real time; an artificial intelligence recognition module is connected to the image acquisition module and is used to analyze the image data based on a pre-trained model and output the set of tissue state parameters. An energy range mapping module, connected to the artificial intelligence recognition module, is used to determine the target energy range based on the tissue state parameter set and a preset energy mapping rule; An energy control module, connected to the energy range mapping module, is used to generate control signals to constrain the energy output of the ultrasonic scalpel within the target energy range.

[0062] A safety monitoring and fallback module, configured to monitor the model confidence parameters; When the model confidence parameter is lower than the preset confidence threshold, the energy range mapping module is triggered to switch the target energy range to the safe energy range; monitor abnormal energy output status, and execute a safe rollback strategy when an abnormality occurs.

[0063] The energy control module specifically includes: The resonant frequency tracking unit is used to track and adjust the driving frequency of the ultrasonic transducer in real time to maintain the resonant state. The current closed-loop control unit, connected to the resonant frequency tracking unit and the energy range mapping module, is used to perform closed-loop regulation of the drive current according to the target energy range in the resonant state and limit the rate of change of the current.

[0064] The modules of this invention can be integrated inside the ultrasonic scalpel host or deployed in a distributed manner through the communication interface.

[0065] like Figure 3 As shown, the present invention also proposes an ultrasonic scalpel device, including the ultrasonic scalpel energy control system, ultrasonic transducer, and scalpel head.

[0066] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the ultrasonic scalpel energy control method.

[0067] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All modifications made according to the spirit and essence of the main technical solution of the present invention should be covered within the scope of protection of the present invention.

Claims

1. An artificial intelligence-based ultrasonic scalpel energy control method, characterized in that, Includes the following steps: Step S1: Acquire image or video data of the surgical area in real time; Step S2: Input the image or video data into a pre-trained artificial intelligence model, analyze the surgical image or video data, and obtain the tissue state parameter set output by the artificial intelligence model. The tissue state parameter set includes at least tissue state probability parameters, risk level parameters, surgical stage parameters, and model confidence parameters. Step S3: Based on the set of tissue state parameters, determine the target energy range according to the preset energy mapping rules. The target energy range is limited by the minimum target energy value and the maximum target energy value. Step S4: Control the energy output of the ultrasonic scalpel to confine it within the target energy range.

2. The ultrasonic scalpel energy control method according to claim 1, characterized in that, In step S3, determining the target energy range based on the tissue state parameter set specifically includes: S31: Determine the main organizational state based on the organizational state probability parameters: Select the organizational state with the highest probability value among the organizational state probability parameters as the main organizational state; S32: Obtain the corresponding basic energy range based on the main tissue state; S33: The basic energy range is modified at least according to the surgical stage parameters and / or the risk level parameters to obtain a modified energy range; S34: Based on the comparison result between the model confidence parameter and the preset confidence threshold, select the corrected energy range or a preset safe energy range as the target energy range.

3. The ultrasonic scalpel energy control method according to claim 2, characterized in that, Step S33 includes: Based on the energy correction coefficients corresponding to the surgical stage parameters, the upper and lower limits of the basic energy range are scaled proportionally to obtain the first corrected energy range; Based on the risk level parameter, the upper limit of the first modified energy range is constrained to obtain a second modified energy range as the modified energy range.

4. The ultrasonic scalpel energy control method according to claim 2, characterized in that, The method for determining the target energy range in step S3 also includes a weighted fusion method: Based on the probability values ​​in the organizational state probability parameters, the corresponding basic energy intervals are weighted and summed to obtain a comprehensive energy interval as the basis.

5. The ultrasonic scalpel energy control method according to claim 1, characterized in that, In step S4, when controlling the energy output, a feedback value of historical energy output is also introduced. The energy setpoint calculated based on the target energy range is fused with the historical energy feedback value to smooth energy changes.

6. An ultrasonic scalpel energy control system for implementing the ultrasonic scalpel energy control method as described in any one of claims 1-5, characterized in that, include: The image acquisition module is used to acquire image data of the surgical area in real time; An artificial intelligence recognition module, connected to the image acquisition module, is used to analyze the image data based on a pre-trained model and output the set of tissue state parameters. An energy range mapping module, connected to the artificial intelligence recognition module, is used to determine the target energy range based on the tissue state parameter set and a preset energy mapping rule; An energy control module, connected to the energy range mapping module, is used to generate control signals to constrain the energy output of the ultrasonic scalpel within the target energy range.

7. The ultrasonic scalpel energy control system according to claim 6, characterized in that, It also includes a security monitoring and fallback module, which is configured to monitor the model confidence parameters; When the model confidence parameter is lower than the preset confidence threshold, the energy range mapping module is triggered to switch the target energy range to the safe energy range; and / or, the abnormal energy output status is monitored, and a safe rollback strategy is executed when an abnormality occurs.

8. The ultrasonic scalpel energy control system according to claim 6, characterized in that, The energy control module specifically includes: The resonant frequency tracking unit is used to track and adjust the driving frequency of the ultrasonic transducer in real time to maintain the resonant state. The current closed-loop control unit, connected to the resonant frequency tracking unit and the energy range mapping module, is used to perform closed-loop regulation of the drive current according to the target energy range in the resonant state and limit the rate of change of the current.

9. An ultrasonic scalpel device, characterized in that, It includes the ultrasonic scalpel energy control system, ultrasonic transducer, and scalpel head as described in any one of claims 6-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the ultrasonic scalpel energy control method as described in any one of claims 1-5.

Images