Negative pressure suction system, control method and medium applied to surgical robot
By combining eye movement and image data, or eye movement, image and EEG data, the negative pressure attraction device is automatically controlled, which solves the problem that traditional negative pressure attraction systems requires additional operators, and realizes that doctors independently control surgical instruments and negative pressure attraction The function of the device reduces the difficulty of surgery and energy consumption.
Patent Information
- Application Number
- CN202410967352.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-18
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2044-07-18
AI Technical Summary
The control of traditional negative pressure attraction systems requires the coordination of additional operators, which increases the difficulty of the operation and makes it difficult to directly control it according to the doctor's ideas.
Using the combination of eye movement and image data, or eye movement, image and EEG data, the negative pressure attraction device is controlled to perform motion in the anatomical area through the processor, to identify the motion intention in the image partition and EEG data, and to realize the negative pressure attraction device. automated control.
Without additional staff assistance, doctors can independently control the surgical instruments and negative pressure suction devices, reducing the difficulty of surgery and operating energy consumption and improving surgical safety.
Smart Images

Figure CN118512271B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of medical devices, and in particular relates to a negative pressure suction system, a control method and a medium applied to a surgical robot. Background Art
[0002] The negative pressure suction device is used to suck out bleeding, exudate, pus, and contents of chest organs during surgery, making the surgery clearer and reducing contamination. The principle is to control the suction head to be in a negative pressure state. Under the action of negative pressure, the suction opening at the end of the suction head will suck out the substances that come into contact with the substance and extract the substances from the body.
[0003] Traditional vacuum suction systems often require manual operation. Doctors use the main hand of the surgical robot to perform surgical operations, so the control of the vacuum suction system usually requires additional operators. Since the doctor and the additional operator need to work together, it is difficult to control the vacuum suction system directly according to the doctor's ideas, which increases the difficulty of the surgery. Summary of the invention
[0004] In order to solve the problem that the control of the existing negative pressure suction system requires additional operators to perform, making the operation difficult, the present invention proposes the following technical solutions:
[0005] A first aspect of the present invention provides a negative pressure suction system applied to a surgical robot, the negative pressure suction system comprising: a main operating console, comprising a display device, an eye movement monitoring device and an electroencephalogram monitoring device, wherein the eye movement monitoring device is used to collect eye movement data, and the electroencephalogram monitoring device is used to collect electroencephalogram data; a slave operating device, comprising an image acquisition device and a negative pressure suction device, wherein the image acquisition device is used to acquire an image of an anatomical region and display it on the display device, and the image at least includes an image of the negative pressure suction device; a processor, configured to: control the negative pressure suction device to perform a first movement in the anatomical region according to the eye movement data and the image; or control the negative pressure suction device to perform a second movement in the anatomical region according to the electroencephalogram data, the eye movement data and the image; wherein the difference between the first movement and the second movement is that the movement range of the negative pressure suction device in the anatomical region is different.
[0006] Preferably, the anatomical region consists of an attraction region and a non-attraction region, and the processor is configured to: identify the image partition of at least one attraction region in the image by presetting a first model, identify the movement intention represented by the EEG data by presetting a second model, and identify the line of sight position of the eye movement data in the image by presetting a third model; if the movement intention does not include the preset first intention, then control the negative pressure suction device to perform the first movement in the attraction region according to the line of sight position and the image partition; if the movement intention includes the preset first intention, then control the negative pressure suction device to perform the second movement in the non-attraction region according to the first intention, the line of sight position and the image partition; wherein the movement range includes the attraction region and the non-attraction region.
[0007] Preferably, the image partition includes a first image partition of a first attraction area, and a second image partition of a second attraction area; when executing the step of controlling the negative pressure suction device to perform a second movement in the non-attraction area according to the first intention, the line of sight position and the image partition, the processor is also configured to: respond to an output instruction based on the first intention, determine a first path for the line of sight position to move from the first image partition to the second image partition; according to the first path, control the negative pressure suction device to move in the non-attraction area, so that the negative pressure suction device moves from the first attraction area to the second attraction area.
[0008] Preferably, the image partition includes a third image partition of a third attraction area, and a fourth image partition of a fourth attraction area; when executing the step of controlling the negative pressure suction device to perform a second movement in the non-attraction area according to the first intention, the line of sight position and the image partition, the processor is also configured to: respond to an output instruction based on the first intention, determine a second path for the line of sight position to move from the third image partition to outside the image partition; calculate the distance between the end point of the second path and each of the image partitions respectively, and determine a third path between the end point and the fourth image partition with the shortest distance; control the negative pressure suction device to move in the non-attraction area according to the second path and the third path, so that the negative pressure suction device moves from the third attraction area to the fourth attraction area.
[0009] Preferably, when executing the step of controlling the negative pressure suction device to perform the first movement in the suction area according to the line of sight position and the image partition, the processor is also configured to: determine the boundary line of the image partition when the line of sight position is outside the image partition; calculate the nearest point of the line of sight position on the boundary line, and control the negative pressure suction device to move along the edge of the suction area according to the nearest point.
[0010] Preferably, the processor is further configured to: when the sight line position is outside the image partition, mark the sight line position in the image and display it on the display device.
[0011] Preferably, the image acquisition device is provided with an image coordinate system, and the image coordinate system includes a first dimension and a second dimension, the first dimension is parallel to the plane where the image is located, and the second dimension is perpendicular to the plane where the image is located; the processor is configured to: if the movement intention represented by the EEG data does not include a preset second intention, then according to the line of sight position corresponding to the eye movement data, control the negative pressure suction device to perform a first movement along the first dimension; if the movement intention includes a preset second intention, then according to the second intention and the line of sight position, control the negative pressure suction device to perform a second movement along the second dimension; wherein the movement range includes different movement directions along the first dimension and the second dimension.
[0012] Preferably, the main operating console also includes an interaction device, and the processor is configured to: respond to a first interaction instruction input by the interaction device, and control the negative pressure suction device to perform a first movement in the anatomical region according to the eye movement data and the image; respond to a second interaction instruction input by the interaction device, and control the negative pressure suction device to perform a second movement in the anatomical region according to the EEG data, the eye movement data and the image.
[0013] Preferably, the second model includes: a preprocessing layer, a convolution layer, a memory layer and a prediction layer; when executing the step of identifying the movement intention represented by the EEG data through the second model, the processor is also configured to: modulate the EEG data according to multiple preset frequency bands through the preprocessing layer, and perform sliding window processing according to a preset time length to obtain multiple EEG signal sequences; perform convolution processing on each of the EEG signal sequences through the convolution layer to obtain a feature sequence, and calculate the global pooling result of each feature sequence; extract the context information of each feature sequence through the memory layer, and generate context-dependent features based on the context information; perform weighted processing on the corresponding context-dependent features according to the global pooling result of each feature sequence through the prediction layer, and predict the movement intention based on the weighted processing result.
[0014] Preferably, the negative pressure suction device includes a terminal suction opening; when executing the step of controlling the negative pressure suction device to perform a first movement in the anatomical region according to the eye movement data and the image, the processor is further configured to: determine the image coordinates of the line of sight position corresponding to the eye movement data in the image, and convert the image coordinates into surgical coordinates of the negative pressure suction device; control the terminal suction opening to move to a position corresponding to the surgical coordinates.
[0015] A second aspect of the present invention provides a control method for a negative pressure suction system, the control method comprising: controlling the negative pressure suction device to perform a first movement in the anatomical area based on the eye movement data and the image; or, controlling the negative pressure suction device to perform a second movement in the anatomical area based on the EEG data, the eye movement data and the image; wherein the difference between the first movement and the second movement is that the movement range of the negative pressure suction device in the anatomical area is different.
[0016] A third aspect of the present invention provides a readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the above method for controlling the negative system are implemented.
[0017] The beneficial effect of the present invention is that the doctor can control the movement of the negative pressure suction device through the combination of eye movements and images, or through the combination of eye movements, images and EEG, without the need for additional staff to assist in the operation. The doctor can also simultaneously control the main hand end of the surgical instrument and the negative pressure suction device. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 Schematic diagram of an embodiment of the electrode distribution of the EEG monitoring device of the present invention;
[0019] Figure 2 It is a schematic diagram of an embodiment of the overall structure of the surgical robot of the present invention;
[0020] Figure 3 It is a schematic diagram of an embodiment of the overall structure of the surgical instrument of the present invention;
[0021] Figure 4 It is a schematic diagram of an embodiment of a usage scenario of the surgical instrument of the present invention;
[0022] Figure 5 is a schematic diagram of another embodiment of the surgical instrument of the present invention in another usage scenario;
[0023] Figure 6 It is a simplified schematic diagram of the negative pressure suction system of the present invention;
[0024] Figure 7 is another simplified schematic diagram of the negative pressure suction system of the present invention;
[0025] Figure 8 Schematic diagram of the first embodiment of the mode switching of the negative pressure suction system of the present invention;
[0026] Fig. 9 is a schematic diagram of a second embodiment of mode switching of the negative pressure suction system of the present invention;
[0027] Fig.10 is a schematic diagram of a third embodiment of mode switching of the negative pressure suction system of the present invention;
[0028] Fig.11 Schematic diagram of the fourth embodiment of the mode switching of the negative pressure suction system of the present invention;
[0029] Fig.12 is a schematic diagram of a fifth embodiment of mode switching of a negative pressure suction system of the present invention;
[0030] Fig.13 It is a schematic diagram of an embodiment of a control method of a negative pressure suction system of the present invention. DETAILED DESCRIPTION
[0031] In order to facilitate the understanding of the present invention, the present invention is described in more detail below in conjunction with the accompanying drawings and specific embodiments. Preferred embodiments of the present invention are provided in the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described in this specification. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive.
[0032] It should be noted that, unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as those understood by a person skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For example, the term "plurality" includes two or more.
[0033] 1. Eye Tracking Technology
[0034] Eye Tracking refers to a technology that measures the position and movement of the eyeball based on optical principles. The principle is that when our eyes are illuminated by an infrared light source, the cornea of the eyeball will reflect a bright point (relative to the entire eyeball). The reflection point is identified by a high-speed camera, and the position of the pupil is also identified. Based on the relative position of the pupil and the reflection point, the direction of the eyeball movement can be determined.
[0035] Based on this principle, an eye movement monitoring device is designed, which includes at least a near-infrared light source and a camera. The near-infrared light source and the camera are pre-positioned, aiming at the user's eyes to emit near-infrared light and collect reflected light to obtain eye movement data. For wireless eye movement monitoring devices, a communication module is also provided to transmit the eye movement data to a processor for data analysis.
[0036] The basic measurement unit of eye movement data collected by the eye movement monitoring device is the fixation point, which is an original sample in the eye movement data. The subsequent eye movement data can be converted into a variety of eye movement indicators, such as: 1) fixation point, where the eyes stay and focus; 2) fixation duration, the length of time the eyes stay in a specific position; 3) scanning path, the path of the eyes moving from one fixation point to another.
[0037] Specifically, the gaze point can also be collected based on algorithms such as pupil corneal reflection method, retinal imaging method, eye modeling and visual center calculation method, retinal reflected light intensity method, corneal reflected light intensity method, etc.
[0038] 2. Brain-Computer Interface Technology
[0039] Brain Computer Interface (BCI) technology refers to the technology that creates a direct connection between the human or animal brain and external devices to achieve information exchange between the brain and external devices. It is a technology that does not rely on the user's normal peripheral nerves and muscle tissues for special information exchange and control channels, and can directly convert central nervous system activities into artificial outputs.
[0040] Based on BCI technology, an EEG monitoring device is set up to collect scalp EEG data (ElectroEncephaloGraphy, EEG, all EEG data below refer to scalp EEG data) as the input signal of the processor. After signal processing, the user's movement intention is identified in the signal and converted into a corresponding command signal to control the external device.
[0041] Motor imagery (MI) is a BCI paradigm that represents motor intention. It can transform "motor intention" into practical "action" by driving functional electrical stimulation, intelligent prostheses or mechanical exoskeletons through subjective motor intention. In addition to expressing the intention to choose a limb, motor intention also involves the motor intention containing kinematics and kinetics information during the implementation of the action.
[0042] Kinematic intention refers to the parameters that describe the spatiotemporal form of movement, such as direction and speed; dynamic intention refers to the causal force that produces muscle activity, including the dynamic force that causes movement and the static force required to maintain a given posture. The MI-BCI paradigm can induce different EEG patterns in the brain by identifying the user's imaginary movement of a certain part of the body (such as left and right hands, left and right feet, etc.), which can be collected by EEG monitoring equipment to form different EEG data and convert limb activities into corresponding control instructions.
[0043] When the EEG monitoring device collects EEG data generated by the brain, it can refer to Figure 1 The electrode distribution (top of the head direction) in the international 10-20 standardized electrode system shown in the figure includes multiple monitoring electrodes, including central electrodes (C3, C4, Cz), temporal electrodes (T3, T4, T5, T6), frontal electrodes (F3, F4, F7, F8, Fz), prefrontal electrodes (Fp1, Fp2), parietal electrodes (P3, P4, Pz), and occipital electrodes (not marked in the figure).
[0044] In addition, according to research, the EEG signal amplitudes in the δ band (0.5-4Hz), θ band (4-8Hz), α band (8-13Hz), β band (13-30), and γ band (>30Hz) can specifically reflect the brain's EEG activity levels in different states and movement patterns. Combined with the multiple groups of EEG data monitored by the previous multiple monitoring electrodes, the specific expression of each group of EEG data is analyzed in different frequency bands to identify the user's movement intention.
[0045] Event Related Potential (ERP) is a potential change in the brain caused by intentional stimulation when the user is awake. The changes in event-related potentials in the aforementioned α and β bands can be described by Event Related Synchronization (ERS) and Event Related Desynchronization (ERD). When the user's brain is performing motor imagery, the potential changes in the α and β bands are attenuated in amplitude, which is event-related desynchronization; when the user's brain is not performing motor imagery, the potential changes in the α and β bands are enhanced in amplitude, which is event-related synchronization.
[0046] Specifically, motor imagery such as left hand movement, right hand movement, leg movement, and tongue movement will all cause significant event-related desynchronization. Among them, left hand motor imagery is easily detected by the C4 electrode, right hand movement is detected by the C3 electrode, and foot movement is detected by the Cz electrode. Based on the characteristic bands of event-related desynchronization in the EEG data, the recognition of motor intention is completed.
[0047] 3. Communication system between eye movement monitoring device and EEG monitoring device
[0048] The communication system for linking the eye movement monitoring device and the electroencephalogram monitoring device includes: an image host, an eye movement monitoring device and an electroencephalogram monitoring device. The image host is connected to the eye movement monitoring device in communication, and the eye movement monitoring device is connected to the electroencephalogram monitoring device in communication. The image host sends a synchronization time signal to the eye movement monitoring device, and the eye movement monitoring device generates a synchronization time mark signal according to the message and sends it to the electroencephalogram monitoring device. The eye movement data generated by the eye movement monitoring device and the electroencephalogram data generated by the electroencephalogram monitoring device achieve clock synchronization through the matching relationship between the synchronization time and the synchronization time mark.
[0049] In addition, the eye movement data and EEG data will also be synchronized with the image in the image host. Finally, when the image, eye movement data, and EEG data are transmitted to the processor respectively, the matching is completed according to the synchronization time and synchronization time mark carried, and the successfully matched data can continue to perform motion control analysis.
[0050] 4. Surgical Robotic System
[0051] like Figure 2 , Figure 3 As shown, the surgical robot includes a main operation console 1 and a slave operation device 2. The main operation console 1 is used to send control commands to the slave operation device 2 according to the doctor's operation to control the slave operation device 2, and is also used to display the image obtained from the slave operation device 2. The slave operation device 2 is used to respond to the control command sent by the main operation console 1 and perform corresponding operations, and the slave operation device 2 is also used to obtain images in the body.
[0052] The slave operation device 2 includes a mechanical arm 21, a power mechanism 22 disposed on the mechanical arm 21, a surgical instrument 100 disposed on the power mechanism 22, and a sleeve 23 on which the long axis 110 of the surgical instrument 100 is sleeved. The mechanical arm 21 is used to adjust the position of the surgical instrument 100; the power mechanism 22 is used to drive the surgical instrument 100 to perform corresponding operations, and the end effector 111 of the surgical instrument 100 is used to extend into the body and perform surgical operations and / or obtain in-body images through the end instrument located at the distal end. The surgical instrument used by the user to obtain in-body images is an image acquisition device.
[0053] Specifically, Figure 4 , Figure 5 As shown, the long axis 110 of the surgical instrument 100 passes through the sleeve 23, and the end effector 111 thereof extends out of the sleeve 23 and is driven by the power mechanism 22 to perform operations. Figure 4 In the figure, the area where the long axis 110 of the surgical instrument 100 is located inside the sleeve 23 is a rigid area; Figure 5In the embodiment, the area where the long axis 110 of the surgical instrument 100 is located inside the sleeve 23 is a flexible area, and the sleeve bends along with the flexible area. The sleeve 23 may also be omitted.
[0054] In order to provide a satisfactory aseptic environment during the surgical robot operation, it is necessary to isolate the sterile instruments from the sterile instruments. Generally, the robot arm 21 and the power mechanism 22 of the operating device 2 are sterile, and the surgical instrument 100 needs to be sterile. A sterile joint device needs to be configured between the sterile power mechanism 22 and the sterile surgical instrument 100 to isolate the sterile power mechanism 22 from the sterile surgical instrument 100. The image acquisition device 170 also extends out of the cannula 23 like the conventional surgical instrument 100, and acquires the operation image of the surgical instrument / negative pressure suction device in the anatomical area in real time.
[0055] When the doctor is operating the surgical instrument at the main console, he cannot operate the negative pressure suction device at the same time and can only operate it with the assistance of other medical staff. The doctor wears an eye movement monitoring device and an EEG monitoring device. The main console includes a display device for displaying in vivo images, which collects the doctor's eye movement data when he looks at the display device, and collects the EEG data when the doctor imagines the movement of body parts. The negative pressure suction device is controlled by eye movement and movement imagination, so that the doctor can independently complete the control of the surgical instrument and the negative pressure suction device.
[0056] 5. Negative pressure suction system for surgical robots
[0057] The above describes the various technical components of the negative pressure suction system used in the surgical robot. Next, the negative pressure suction system in the present invention is described as a whole. Figure 6 , an embodiment of the negative pressure suction system of the present invention is described below.
[0058] The negative pressure suction system 600 includes: a main operating console (not shown in the figure), including a display device 610, an eye movement monitoring device 620 and an EEG monitoring device 630, wherein the eye movement monitoring device 620 is used to collect eye movement data, and the EEG monitoring device 630 is used to collect EEG data; a slave operating device (not shown in the figure), including an image acquisition device 650 and a negative pressure suction device 660 placed in the anatomical area 640, wherein the image acquisition device 650 is used to acquire an image of the anatomical area 640 and display it on the display device 610, and the image at least includes an image 611 of the negative pressure suction device 660.
[0059] The image, eye movement data and EEG data are transmitted to the processor 670 for data analysis, so as to output control instructions for controlling the negative pressure suction device 660. Specifically, according to the eye movement data and the image, the negative pressure suction device 660 is controlled to perform a first movement in the anatomical region 640; or, according to the EEG data, the eye movement data and the image, the negative pressure suction device 660 is controlled to perform a second movement in the anatomical region 640; wherein the difference between the first movement and the second movement is that the movement range of the negative pressure suction device 660 in the anatomical region 640 is different.
[0060] In this embodiment, the switching conditions between the first movement and the second movement are preset. When the preset first switching conditions are met, the negative pressure suction device is controlled to perform the first movement. When the preset second switching conditions are met, the negative pressure suction device is controlled to perform the second movement.
[0061] Since eye movement data can be more easily and accurately analyzed for relevant movement parameters, such as movement direction, position, etc., the first movement and the second movement are both controlled based on the eye movement data. The switching condition can be set by: whether the EEG data can analyze a valid movement intention; a valid movement intention is a recorded target movement intention.
[0062] For example, when the target movement intention is analyzed from the EEG data, it is determined that the preset second switching condition is met, otherwise the preset first switching condition is met. In addition, the first movement or the second movement of the negative pressure suction device can be manually switched. Other switching methods can be set based on specific needs, so that the movement ranges of the first movement and the second movement are different, which will not be repeated here.
[0063] In summary, the first movement can be used to perform the basic movement operation of the negative pressure suction device, and the second movement can be used to perform the extended movement operation of the negative pressure suction device. It is easier for doctors to control eye movements, but more difficult to control brain waves. By switching the first and second movements, the doctor's operation difficulty and energy consumption are reduced, and the safety of the operation can be further improved.
[0064] In addition, as a surgical instrument, the negative pressure suction device at least includes a negative pressure drainage tube, which is a part used to extend into the anatomical part of the body, including a proximal part and a distal part, the distal part provides a suction opening for performing a suction operation, the proximal part is connected to a driving device of a slave operating device, outputs a control instruction to the slave operating device, and the slave operating device controls the first movement or the second movement of the negative pressure suction device through the driving device, and the first movement or the second movement can include at least one degree of freedom in the six degrees of freedom of the distal part. The first movement has the same or fewer degrees of freedom of movement than the second movement, and the first movement has the same or smaller range of movement than the second movement.
[0065] In one embodiment, the negative pressure suction device includes a terminal suction opening; the image coordinates of the sight line position corresponding to the eye movement data in the image are determined, and the image coordinates are converted into the surgical coordinates of the negative pressure suction device; the terminal suction opening is controlled to move to the position corresponding to the surgical coordinates. The doctor's sight line position in the display device is mapped to the terminal suction opening, and as the sight line position moves, the terminal suction opening is controlled to move accordingly.
[0066] Specifically, an image coordinate system is provided with respect to the image acquisition device, and a surgical coordinate system is provided with respect to the negative pressure suction device, and a coordinate conversion relationship is provided between the two. According to the conversion relationship, the image coordinates are converted into surgical coordinates, and the control parameters for the terminal suction opening to reach the specified position are calculated using inverse kinematics to drive the movement of the surgical robot and drive the movement of the terminal suction opening.
[0067] In this embodiment, the doctor wears an eye movement monitoring device to watch the display device and perform surgical operations. When it is necessary to control the movement of the negative pressure suction device, the doctor's gaze position in the display device can be measured through the eye movement monitoring device, and as the doctor's gaze position moves, eye movement data including a gaze point sequence is dynamically generated; the processor determines multiple eye movement indicators, such as fixation indicators, scan indicators, blink indicators, etc., by analyzing the gaze point sequence, and generates a control instruction based on at least one eye movement indicator to control the first movement of the negative pressure suction device, so that the negative pressure suction device moves according to the gaze point position.
[0068] In this embodiment, the doctor wears an EEG monitoring device to perform surgical operations. When it is necessary to control the movement of the negative pressure suction device, different EEG patterns generated by the brain are induced by motor imagination, which are collected by the EEG monitoring device as voltage changes over time to generate EEG data; the processor analyzes the EEG data to determine the movement intention, such as the movement imagination of one of the limbs can determine the intention to select the limb, and the movement imagination of a specific movement of a single limb can determine the intention to select fine movements (such as shaking hands, letting go, etc.). Based on the movement intention and combined with the aforementioned eye movement indicators, a control instruction is generated to control the second movement of the negative pressure suction device.
[0069] 6. Switching of the first and second motion control of the negative pressure suction system
[0070] like Figure 7As shown, the anatomical region 710 is composed of an attraction region 711 and a non-attraction region 712, and the processor 720 is configured to: identify an image partition 731 of at least one attraction region 711 in the image 730 by presetting a first model 721, identify the movement intention represented by the EEG data by presetting a second model 722, and identify the line of sight position of the eye movement data in the image 730 by presetting a third model 723.
[0071] Among them, the eye movement monitoring device 740 includes an eye movement monitor 741 worn by the user, and the eye movement monitor 741 is used to collect raw eye movement data, and obtain, temporarily store, pre-process, and preliminarily analyze the raw eye movement data through a processor 742 connected to the eye movement monitor to obtain the final eye movement data. The EEG monitoring device 750 includes an EEG monitor 751 worn by the user, and the EEG monitor 751 is used to collect raw EEG data, and obtain, temporarily store, pre-process, and preliminarily analyze the raw EEG data through a processor 752 connected to the EEG monitor to obtain the final EEG data.
[0072] In this embodiment, during the operation, because the anatomical area 710 contains the location areas of bleeding, exudate, pus, and the contents of the chest organs, it is the suction area 711, indicating that the negative pressure suction device is required to perform the suction operation from these suction areas. The corresponding image partition 731 can be identified through the image 730. The non-suction area 712 does not have the location areas of bleeding, exudate, pus, and the contents of the chest organs, which is the non-suction area 732, and is automatically divided into the area outside the image partition in the image 730.
[0073] like Figure 7 As shown, the anatomical region 710 includes a suction region 711 and a non-suction region 712, and the negative pressure suction device 713 performs a suction operation in the suction region 711 or other suction regions. The suction region 711 is identified and divided into image partitions 731 in the image 730, and the non-suction region 712 is a non-suction region 732 outside the image partitions.
[0074] Specifically, the first model can be an algorithm model based on image threshold segmentation, image edge segmentation, image morphology segmentation, image region segmentation, etc., for specific detection of attraction areas. Specifically, it can be a model using algorithms such as FCN (Fully Convolutional Networks), SegNet (A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation), and Mask R-CNN (Mask Region-based Convolutional Neural Network, a region-based deep learning target detection algorithm). The specific algorithm and execution process will not be repeated here.
[0075] In this embodiment, the movement intention includes at least one of the following intentions based on motor imagination: intention to select a limb, intention to select a limb based on kinematics and dynamics, intention to select fine limb movements, etc. The characteristic bands of event-related desynchronization in the EEG data are mainly extracted through the second model; the category of the movement intention is detected according to the characteristic bands.
[0076] Specifically, the second model can be a combination model of algorithms such as CWT (Continuous Wavelet Transform) + MixedCNN (Mixed- Convolutional Neural Network), CSP (Common Spatial Patterns) + SVM (Support Vector Machine), AAR (Adaptive Autoregressive Model) + LDA (Linear Discriminant Analysis), DWT (Discrete Wavelet Transform) + LSTM (Long Short-Term Memory), STFT (Short-Time Fourier Transform) + SAE (Stacked Auto-Encoders). The specific algorithms and execution processes are not described here.
[0077] The second model may also include algorithms for removing noise, artifacts, and impedance from EEG data. When removing noise, the EEG data may be filtered by a bandpass filter / notch filter to remove DC drift and high-frequency noise; artifacts caused by electrode failure or EEG drift may be removed by bad track detection and interpolation processing algorithms; and impedance may be corrected by checking and adjusting electrode impedance. That is, the function of the processor 752 may be replaced.
[0078] In this embodiment, identifying the line of sight position of the eye movement data in the image by presetting a third model means that based on the originally extracted eye movement data such as the gaze point, the position of the gaze point projected into the image is analyzed by the third model, that is, the line of sight position; further combining multiple gaze points that change over time to determine the moving path of the line of sight position; further combining the gaze duration of the gaze point to determine the line of sight position of the gaze point; further combining other parameters related to the gaze point to determine the line of sight position of other related indicators.
[0079] Specifically, the third model includes a fixation point extraction algorithm, a fixation feature extraction algorithm, a glance path analysis algorithm, an eye movement behavior modeling algorithm, etc. Among them, the fixation point extraction algorithm specifically includes a clustering algorithm, a threshold algorithm, and a model fitting algorithm, etc.; the fixation feature extraction algorithm is based on the aforementioned fixation points, including calculation algorithms for fixation duration, number of fixations, number of fixation jumps, line of sight movement speed, and fixation deviation; the glance path analysis algorithm specifically includes algorithms such as a glance path graph, a vector field, and a probability distribution; and the eye movement behavior modeling algorithm specifically includes a Markov model, a hidden Markov model, and a Bayesian network, etc.
[0080] The third model also includes a preprocessing algorithm, which performs preprocessing such as filtering, noise reduction and interpolation on the eye movement data (such as gaze point data) to maintain the smoothness of the eye movement data, reduce noise and maintain the continuity of the eye movement data, that is, replace part of the functions of the processor 742.
[0081] 1) The first switching method
[0082] If the movement intention does not include the preset first intention, the negative pressure suction device is controlled to perform the first movement in the suction area according to the line of sight position and the image partition; if the movement intention includes the preset first intention, the negative pressure suction device is controlled to perform the second movement in the non-attraction area according to the first intention, the line of sight position and the image partition; wherein the movement range includes the attraction area and the non-attraction area.
[0083] In this embodiment, the preset first intention is to perform the movement control of the negative pressure suction device by combining with the sight position through motion imagination; when the first intention is not included, the motion data does not affect the movement control of the negative pressure suction device by the sight position. That is, the sight position is the basic continuous control of the negative pressure suction device, while the motion intention is the intermittent control of the adjustment of the negative pressure suction device.
[0084] For example, if you imagine the left hand moving, the right hand moving, the foot moving, the left hand making a fist or the right hand making a fist, or imagine the end of the handheld negative pressure suction device moving in a certain direction, a preset first intention is generated and set. Then, when the movement imagination triggers the first intention, the sight line position is adjusted to control the negative pressure suction device to move in the non-attraction area. Otherwise, the sight line position can only control the negative pressure suction device to move in the attraction area. That is, the triggering of the first intention increases the movement range of the sight line position.
[0085] In actual application, compared with the movement intention, it is easier for the operator to control the negative pressure suction device based on the sight position, and the eye movement monitoring device can also identify more accurately. At the same time, in order to effectively attract bleeding, exudate, pus, and the contents of chest organs, the control of the negative pressure suction device based on the sight position is limited to the image partition. For other undetected bleeding, exudate, pus, and the contents of chest organs, the sight position is further combined with the triggering of the first intention, so that the sight position can control the negative pressure suction device to perform free suction operations in non-image partitions. Ultimately, the difficulty of operation is reduced, while the suction efficiency is improved.
[0086] like Figure 7 As shown, the image 733 of the negative pressure suction device in the image 730 is located in the image partition 731. If the negative pressure suction device is controlled only based on the line of sight position, the mapping of the line of sight position on the image is limited to the image partition 731, and the end position contained in the generated control instruction is limited to the image partition 731; when the negative pressure suction device 713 is controlled to move in the anatomical area 710 based on the control instruction, it will not exceed the suction area 711.
[0087] In one embodiment, Figure 8 As shown, in the anatomical region 810 and the image 820, the image partitions include a first image partition 821 of a first attraction region 811, and a second image partition 822 of a second attraction region 812, and the processor is further configured to: respond to an output instruction based on the first intention, determine a first path 823 for the line of sight position to move from the first image partition 821 to the second image partition 822; and according to the first path 823, control the negative pressure suction device to move in the non-attraction region, so that the negative pressure suction device moves along the actual path 813 from the first attraction region 811 to the second attraction region 812.
[0088] That is, the user imagines the preset action in the brain, and the EEG data will be detected by the first intention, triggering the output command, unlocking the sight position to control the negative pressure suction device to move in the non-negative pressure suction device. The specific actual path 813 is still determined only based on the first path 823 of the sight position scanning in the image. The EEG data / first intention does not participate in the specific movement direction control of the negative pressure suction device, and only unlocks the movement range of the negative pressure suction device.
[0089] In one embodiment, Fig. 9 As shown, in the anatomical region 910 and the image 920, the image partitions include a third image partition 921 of a third attraction region 911, and a fourth image partition 922 of a fourth attraction region 912; the processor is further configured to: in response to the output instruction based on the first intention, determine the second path 931 where the sight line position moves from the third image partition 921 to outside the image partition; calculate the distances D1, D2, and D3 between the end point 932 of the second path 931 and each of the image partitions, and determine a third path 933 between the end point 932 and the fourth image partition 922 with the shortest distance D3; according to the second path 931 and the third path 933, control the negative pressure suction device to move in the non-attraction region, so that the negative pressure suction device moves along the actual path 934 and the actual path 935, and moves from the third attraction region 911 to the fourth attraction region 912. Wherein, the distances D1, D2, and D3 are the minimum distances between the end point 932 and each image partition.
[0090] That is, when the first intention cannot be monitored due to incorrect user movement imagination or other reasons, if the line of sight is in the non-image partition, it needs to be mapped to the nearest image partition so that the negative pressure suction device is automatically moved to the nearest partition to be attracted.
[0091] Among them, according to the first path, the negative pressure suction device is controlled to move from the first suction area to the second suction area, or according to the second path and the third path, the negative pressure suction device is controlled to move from the third suction area to the fourth suction area, specifically, the negative pressure suction device follows the sight line position in real time for movement control. That is, the first path is gradually generated as the sight line position increases over time. During the generation process, the extended position of the first path is periodically updated according to the preset time granularity, and the negative pressure suction device is controlled to move in real time with the extended position.
[0092] like Fig.10As shown, in image 1010, when the line of sight position 1011 is outside the image partition 1012, the boundary line 1013 of the image partition 1012 is determined; the nearest point 1014 of the line of sight position 1011 on the boundary line 1013 is calculated, and in the anatomical region 1020, the negative pressure suction device is controlled to move along the edge 1022 of the suction region 1021 according to the nearest point 1014 until the edge point 1023 corresponding to the nearest point 1014 is reached; the line of sight position is marked in the image and displayed on the display device.
[0093] Among them, if the line of sight position 1011 is previously in the image partition 1012, the nearest point 1014 between the current line of sight position 1011 and the corresponding boundary line 1013 is directly determined; if the line of sight position 1011 is not previously in any image partition, the distance between the current line of sight position 1011 and each image partition is determined to determine the image partition 1012 with the shortest distance.
[0094] 2) The second switching method
[0095] like Fig.11 As shown, in a different embodiment, in image 1110, when the line of sight position 1111 is outside the image partition, the distance between the line of sight position 1111 and each image partition is calculated, and the boundary line of the image partition 1112 with the shortest distance and the corresponding nearest point 1113 are determined; at this time, in the anatomical area 1120, the negative pressure suction device is controlled to move to the edge point 1122 of the corresponding suction area 1121.
[0096] When the sight line position 1111 moves to the sight line position 1114, the image partition 1115 with the shortest distance to the sight line position 1114 is determined, and the nearest point 1116 of the boundary line between the sight line position 1114 and the image partition 1115 is determined; at this time, the negative pressure suction device is controlled to move from the edge point 1122 to the edge point 1124 of the suction area 1123.
[0097] In this embodiment, when the first intention triggers the output instruction, the control of the negative pressure suction device based on the line of sight position is unlocked, and it can move freely in the non-suction area.
[0098] 3) The third switching method
[0099] like Fig.12As shown, in a different embodiment, in the anatomical region 1210, the image acquisition device is provided with an image coordinate system AxAyAz, and the image coordinate system includes a first dimension AxAy and a second dimension Az, wherein the first dimension AxAy is parallel to the plane where the image 1220 is located, and the second dimension Az is perpendicular to the plane where the image 1220 is located; wherein the first dimension AxAy corresponds to the plane coordinate ByBx of the image 1220.
[0100] The processor is configured to: if the movement intention represented by the EEG data does not include a preset second intention, then control the negative pressure suction device to perform a first movement along the first dimension according to the line of sight position corresponding to the eye movement data; if the movement intention includes a preset second intention, then control the negative pressure suction device to perform a second movement along the second dimension according to the second intention and the line of sight position; wherein the movement range includes different movement directions along the first dimension and the second dimension.
[0101] In this embodiment, the preset second intention may also be an intention monitored such as imagining left hand movement, right hand movement, foot movement, left hand fisting or right hand fisting, or imagining the end of the handheld negative pressure suction device moving in a certain direction. The preset second intention includes two types of intentions, one for controlling the negative pressure suction device to move along Az, and the other for controlling the negative pressure suction device to move along -Az.
[0102] 4) The fourth switching method
[0103] In a different embodiment, the main operating console also includes an interaction device, and the processor is configured to: in response to a first interaction instruction input by the interaction device, control the negative pressure suction device to perform a first movement in the anatomical region according to the eye movement data and the image; and in response to a second interaction instruction input by the interaction device, control the negative pressure suction device to perform a second movement in the anatomical region according to the EEG data, the eye movement data and the image.
[0104] In this embodiment, the interactive device may be a pedal device, and according to each pressing of the pedal device, at least the first interactive instruction and the second interactive instruction are alternately output. In addition, with each pressing, the start instruction of the negative pressure suction device-the first interactive instruction-the second interactive instruction-the stop instruction of the negative pressure suction device may be alternately output.
[0105] In addition, the interactive device can also be other forms of control devices; for example, a button-type device, in which a single button is set to be pressed alternately to switch different interactive instructions, or multiple buttons are set to control different interactive instruction outputs; for example, a touch device, in which a touch button is added to the operating touch screen to switch / control different interactive instruction outputs.
[0106] 7. Deep Learning Network for Negative Pressure Suction System
[0107] The second model includes: a preprocessing layer, a convolution layer, a memory layer and a prediction layer; through the preprocessing layer, the EEG data is modulated according to multiple preset frequency bands, and sliding window processing is performed according to a preset time length to obtain multiple EEG signal sequences; through the convolution layer, each of the EEG signal sequences is convolved to obtain a feature sequence, and the global pooling result of each feature sequence is calculated; through the memory layer, the context information of each feature sequence is extracted respectively, and context-dependent features are generated based on the context information; through the prediction layer, according to the global pooling result of each feature sequence, the corresponding context-dependent features are weighted, and based on the result of the weighted processing, the movement intention is predicted.
[0108] In this embodiment, the preset frequency bands may include: δ band (0.5-4Hz), θ band (4-8Hz), α band (8-13Hz), β band (13-30), γ band (>30Hz), 0-40 Hz, 0-70 Hz, etc.; for the EEG data collected by n electrodes in the EEG monitoring device, some or all of the m frequency bands are selected to modulate the EEG data respectively to obtain n*m EEG modulation signals.
[0109] If the duration of the EEG modulation signal is K, the preset duration may be k, and a% (preferably 50%) repetition rate is set. The n*m EEG modulation signals are subjected to sliding window processing to obtain (K*2 / k-1)*n*m EEG signal sequences.
[0110] Convolution processing is performed on (K*2 / k-1)*n*m one-dimensional EEG signal sequences, preferably through CNN, to obtain a feature sequence. In addition, global pooling processing is performed on the feature sequence to determine the importance of each feature sequence in the current EEG data, and dynamically adjust the subsequent prediction results of the motion intention based on the feature sequence. The global pooling result can be a global average pooling result or a global maximum pooling result, etc., preferably a global average pooling result.
[0111] In this embodiment, the context information of the feature sequence refers to the relationship between each feature in the feature sequence and the generation of other features. The closer the features are in the feature sequence, the greater the influence, and the farther they are apart, the smaller the influence. Based on the influence of other features (i.e., context information), the context-dependent features of all the features are output.
[0112] The memory layer can be RNN (Recurrent Neural Network), such as LSTM, GRU (Gated Recurrent Unit), and Bi-RNN (Bi-directional Recurrent Neural Network). Taking the LSTM network as an example, the context information is the state output of each memory unit in the LSTM network, and the context-dependent feature is the result output of each memory unit.
[0113] In this embodiment, the importance of each context-dependent feature is determined based on the global pooling result, and the more important context-dependent features are relatively increased and the less important context-dependent features are relatively decreased after weighted processing. Finally, the motion intention category is determined based on the increased / decreased features to obtain the prediction result.
[0114] 8. Control Method of Negative Pressure Suction System
[0115] like Fig.13 As shown, an embodiment of a control method of a negative pressure suction system in the present invention is provided, which is specifically as follows:
[0116] S131, in response to a start instruction for the negative pressure suction device, detecting in real time that the negative pressure suction device meets a preset first switching condition or a preset second switching condition;
[0117] S132: If the negative pressure suction device satisfies a preset first switching condition, controlling the negative pressure suction device to perform a first movement in the anatomical region according to the eye movement data and the image; or
[0118] S133. If the negative pressure suction device meets a preset second switching condition, the negative pressure suction device is controlled to perform a second movement in the anatomical region according to the EEG data, the eye movement data and the image.
[0119] The difference between the first movement and the second movement is that the movement range of the negative pressure suction device in the anatomical area is different.
[0120] In this embodiment, the switching conditions between the first movement and the second movement are preset. When the preset first switching conditions are met, the negative pressure suction device is controlled to perform the first movement. When the preset second switching conditions are met, the negative pressure suction device is controlled to perform the second movement.
[0121] Since eye movement data can be more easily and accurately analyzed for relevant movement parameters, such as movement direction, position, etc., the first movement and the second movement are both controlled based on the eye movement data. The switching condition can be set by: whether the EEG data can analyze a valid movement intention; a valid movement intention is a recorded target movement intention.
[0122] For example, when the target movement intention is analyzed from the EEG data, it is determined that the preset second switching condition is met, otherwise the preset first switching condition is met. In addition, the first movement or the second movement of the negative pressure suction device can be manually switched. Other switching methods can be set based on specific needs, so that the movement ranges of the first movement and the second movement are different, which will not be repeated here.
[0123] It should be noted that other sorting schemes that can be easily thought of by those skilled in the art within the technical scope disclosed in the present invention should also be within the protection scope of the present invention and will not be described in detail here.
[0124] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example for illustration. In actual applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the mobile terminal can be divided into different functional units or modules to complete all or part of the functions described above. The functional modules in the embodiment can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the modules in the above-mentioned mobile terminal can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.
[0125] The embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.
[0126] An embodiment of the present application provides a computer program product. When the computer program product runs on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned method embodiments when executing the computer program product.
[0127] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.
[0128] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.
[0129] As used in the specification of this application and the appended claims, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting" depending on the context. Similarly, the phrase "if it is determined" or "if the described condition or event is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce the described condition or event is detected" or "in response to detecting the described condition or event" depending on the context.
[0130] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.
[0131] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.
[0132] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.
[0133] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0134] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0135] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0136] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electric carrier signals and telecommunication signals.
[0137] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.
Claims
1. A negative pressure suction system for a surgical robot. It is characterized in that The negative pressure suction system comprises: A main operating console, comprising a display device, an eye movement monitoring device and an electroencephalogram monitoring device, wherein the eye movement monitoring device is used to collect eye movement data, and the electroencephalogram monitoring device is used to collect electroencephalogram data; A slave operating device includes an image acquisition device and a negative pressure suction device, wherein the image acquisition device is used to acquire an image of an anatomical region and display it on the display device, the image at least includes an image of the negative pressure suction device, and the anatomical region consists of a suction region and a non-suction region; The processor is configured as: By presetting a first model, an image partition of at least one attraction area in the image is identified; by presetting a second model, a movement intention represented by the EEG data is identified; and by presetting a third model, a sight line position of the eye movement data in the image is identified; If the movement intention does not include the preset first intention, controlling the negative pressure suction device to perform a first movement in the suction area according to the sight line position and the image partition; If the movement intention includes a preset first intention, the negative pressure suction device is controlled to perform a second movement in the non-suction area according to the first intention, the sight line position and the image partition.
2. The negative pressure suction system according to claim 1, It is characterized in that The image partitions include a first image partition of a first attraction area, and a second image partition of a second attraction area; When executing the step of controlling the negative pressure suction device to perform a second movement in the non-suction area according to the first intention, the sight line position and the image partition, the processor is further configured to: In response to an output instruction based on the first intention, determining a first path for the sight line position to move from the first image subarea to the second image subarea; According to the first path, the negative pressure suction device is controlled to move in the non-suction area, so that the negative pressure suction device moves from the first suction area to the second suction area.
3. The negative pressure suction system according to claim 1, It is characterized in that The image subareas include a third image subarea of a third attraction area, and a fourth image subarea of a fourth attraction area; When executing the step of controlling the negative pressure suction device to perform a second movement in the non-suction area according to the first intention, the sight line position and the image partition, the processor is further configured to: In response to an output instruction based on the first intention, determining a second path for the sight line position to move from the third image partition to outside the image partition; Calculating the distance between the end point of the second path and each of the image partitions, and determining a third path between the end point and a fourth image partition with the shortest distance; According to the second path and the third path, the negative pressure suction device is controlled to move in the non-suction area, so that the negative pressure suction device moves from the third suction area to the fourth suction area.
4. The negative pressure suction system according to claim 1, It is characterized in that When executing the step of controlling the negative pressure suction device to perform a first movement in the suction area according to the sight line position and the image partition, the processor is further configured to: When the sight line position is outside the image partition, determining a boundary line of the image partition; The nearest point of the sight line position on the boundary line is calculated, and the negative pressure suction device is controlled to move along the edge of the suction area according to the nearest point.
5. The negative pressure suction system according to claim 4, It is characterized in that The processor is further configured to: when the sight line position is outside the image partition, mark the sight line position in the image and display it on the display device.
6. The negative pressure suction system according to claim 1, It is characterized in that The image acquisition device is provided with an image coordinate system, the image coordinate system includes a first dimension and a second dimension, the first dimension is parallel to the plane where the image is located, and the second dimension is perpendicular to the plane where the image is located; The processor is configured to: If the movement intention represented by the EEG data does not include the preset second intention, controlling the negative pressure suction device to perform the first movement along the first dimension according to the sight line position corresponding to the eye movement data; If the movement intention includes a preset second intention, the negative pressure suction device is controlled to perform a second movement along the second dimension according to the second intention and the line of sight position.
7. The negative pressure suction system according to claim 1, It is characterized in that The main operating console also includes an interaction device, and the processor is configured to: In response to a first interaction instruction input by the interaction device, the negative pressure suction device is controlled to perform a first movement in the anatomical region according to the eye movement data and the image; In response to a second interaction instruction input by the interaction device, the negative pressure suction device is controlled to perform a second movement in the anatomical region according to the EEG data, the eye movement data and the image.
8. The negative pressure suction system according to any one of claims 1 to 5, It is characterized in that The second model includes: a preprocessing layer, a convolution layer, a memory layer and a prediction layer; When executing the step of identifying the movement intention represented by the EEG data by presetting the second model, the processor is further configured to: The EEG data is modulated according to a plurality of preset frequency bands by the preprocessing layer, and sliding window processing is performed according to a preset time length to obtain a plurality of EEG signal sequences; Performing convolution processing on each of the EEG signal sequences respectively through the convolution layer to obtain a feature sequence, and calculating a global pooling result of each of the feature sequences; Extracting context information of each of the feature sequences through the memory layer, and generating context-dependent features based on the context information; Through the prediction layer, the corresponding context-dependent features are weighted according to the global pooling results of each feature sequence, and the motion intention is predicted based on the weighted processing results.
9. The negative pressure suction system according to any one of claims 1 to 7, It is characterized in that The negative pressure suction device comprises a terminal suction opening; When executing the step of controlling the negative pressure suction device to perform a first movement in the suction area according to the sight line position and the image partition, the processor is further configured to: Determine the image coordinates of the sight line position corresponding to the eye movement data in the image, and convert the image coordinates into the surgical coordinates of the negative pressure suction device; The terminal suction opening is controlled to move to a position corresponding to the surgical coordinates.
