Smoke removing and lighting integrated system for neuroendoscopic surgery

Abstract

The invention discloses a smog removal and illumination integrated system for neuroendoscopic surgery, belongs to the technical field of neuroendoscopes, and can improve the observation effect of endoscopic images. The flow velocity monitoring module is used for collecting real-time flow velocity data of flushing fluid and transmitting the real-time flow velocity data to the central processing unit; the pressure monitoring module is used for collecting real-time pressure data in the cavity and transmitting the real-time pressure data to the central processing unit; the image acquisition module is used for capturing a real-time operation visual field image in the cavity and transmitting the real-time operation visual field image to the central processing unit; the central processing unit is used for analyzing the image data through an image processing algorithm, extracting smoke region feature data, and comparing the smoke region feature data with a preset smoke feature threshold to generate a smoke diffusion coefficient; performing multi-source data fusion based on the flow velocity data, the pressure data and the smoke diffusion coefficient to generate a dynamic regulation factor; and when the dynamic regulation factor exceeds a preset threshold value, the closed-loop control module is triggered to regulate the environment in the cavity.

Description

Technical Field

[0001] The present invention relates to the technical field of neuroendoscopy, and particularly to an integrated system for smoke removal and illumination in neuroendoscopic surgery. Background Art

[0002] Endoscopes are important medical devices in minimally invasive surgery. Compared with traditional surgeries, endoscopes can enter the human body through body cavities or small incisions. During neurosurgical operations, they facilitate doctors in observing the lesion sites, observing nerves with the aid of neuroendoscopes to determine the location of the lesions, and then cooperating with other surgical instruments to complete the surgery. Neuroendoscopes play an important role in neurosurgical operations, especially in the treatment of common clinical diseases such as hydrocephalus and cerebral hemorrhage. There is a large amount of body fluid in human tissues, and factors such as water vapor scattering and tissue reflection all affect the imaging of neuroendoscopes. Although the lens can be preheated with hot saline or wiped with an anti-fog reagent before inserting the endoscope deep into the human body, during operations such as microwave ablation, there will also be water vapor or smoke in the body cavity after heating, which also has a great impact on observing tissues through neuroendoscopes.

[0003] The disclosure of the above background art content is only used to assist in understanding the concept and technical solution of the present invention, and it does not necessarily belong to the prior art of this patent application. Without clear evidence indicating that the above content was publicly available on the filing date of this patent application, the above background art should not be used to evaluate the novelty and inventiveness of this application. Summary of the Invention

[0004] This application provides an integrated system for smoke removal and illumination in neuroendoscopic surgery, which can improve the observation effect of endoscopic images.

[0005] To achieve the above object, the embodiments of this application disclose the following technical solutions:

[0006] An integrated system for smoke removal and illumination in neuroendoscopic surgery, comprising: a flow rate monitoring module, a pressure monitoring module, an image acquisition module, a central processing unit, and a closed-loop control module. The flow rate monitoring module is used to collect real-time flow rate data of the flushing liquid and transmit it to the central processing unit;

[0007] The pressure monitoring module is used to collect real-time pressure data in the cavity and transmit it to the central processing unit;

[0008] The image acquisition module is used to capture real-time surgical field images in the cavity and transmit them to the central processing unit;

[0009] The central processing unit is configured to receive data from the flow rate monitoring module, the pressure monitoring module, and the image acquisition module; analyze the image data through an image processing algorithm, extract the smoke area feature data, compare it with a preset smoke feature threshold, and generate a smoke diffusion coefficient; perform multi-source data fusion based on the flow rate data, pressure data, and smoke diffusion coefficient to generate a dynamic adjustment factor; when the dynamic adjustment factor exceeds a preset threshold, trigger the closed-loop control module to adjust the environment inside the cavity;

[0010] The closed-loop control module includes a flushing adjustment sub-module, a suction adjustment sub-module, an optical compensation and image enhancement sub-module, and a feedback sub-module. The flushing adjustment sub-module is configured to dynamically adjust the opening degree of the flushing valve according to the difference between the flow rate data and the target flow rate; the suction adjustment sub-module is configured to monitor the negative pressure value of the suction channel in real time. If the negative pressure value is lower than the safety threshold, pressure compensation is performed to restore the negative pressure value to the safety range by increasing the power of the suction pump; the optical compensation and image enhancement sub-module is configured to perform optical compensation and enhancement processing on the real-time image based on the adjusted flushing and suction parameters; the feedback sub-module is configured to feedback the adjusted opening degree of the flushing valve and the power parameters of the suction pump to the central processing unit, and transmit the optimized image data to the display device.

[0011] In the embodiment of the present application, the flow rate monitoring module and the pressure monitoring module respectively collect the flow rate of the flushing liquid and the pressure inside the cavity in real time, providing basic data for fluid control; the image acquisition module captures the surgical field image, and the central processing unit generates a smoke diffusion coefficient by analyzing the smoke area features in the image to quantify the change trend of the smoke concentration.

[0012] Based on the multi-source data fusion of the flow rate, pressure, and smoke diffusion coefficient, the central processing unit calculates the dynamic adjustment factor. When the dynamic adjustment factor exceeds a preset threshold, the closed-loop control module is triggered to perform multi-dimensional adjustment. The flushing adjustment sub-module dynamically adjusts the opening degree of the flushing valve according to the flow rate difference to directly remove the smoke; the suction adjustment sub-module compensates the negative pressure of the suction channel in real time to prevent smoke from staying;

[0013] The optical compensation and image enhancement sub-module combines the value and the real-time image data to offset the interference of the residual smoke by enhancing the contrast and brightness. In this way, it ensures that the smoke removal intensity is dynamically matched with the real-time state of the cavity pressure and flow rate. At the same time, the image enhancement compensates for the short-term vision loss during the smoke removal process, improving the observation effect of the endoscope image, breaking through the limitations of the traditional method that relies on pre-operative anti-fog treatment or single-parameter adjustment, and achieving the triple goals of efficient smoke removal, stable cavity environment, and optimized surgical field during the operation, avoiding surgical risks caused by lagging manual intervention or operation conflicts.

[0014] In some embodiments, the dynamic adjustment factor is calculated by the following formula:

[0015]

[0016] Among them, w 1 , w 2 , w 3 are weight coefficients, satisfying w 1 + w 2 + w 3 = 1 and w 1 , w 2 , w 3 > 0;

[0017] Q max is the preset maximum allowable flushing flow rate, unit: mL / s;

[0018] P max is the preset maximum allowable cavity pressure, unit: mmHg;

[0019] α is the smoke diffusion coefficient, dimensionless. When α > α threshold , w 3 = 0.5, w 1 = w 2 = 0.25, α threshold is the smoke diffusion emergency threshold; When Q 1 ≥ Q max or P 1 ≥ P max , the central processing unit triggers an alarm and pauses the weight adjustment until the flow rate or pressure returns to the safe range.

[0020] In this way, through the collaborative design of dynamic weight allocation and emergency priority rules, the balance between smoke clearance efficiency and cavity safety is achieved in the complex and changeable intraoperative environment. By fusing three types of heterogeneous parameters, namely flow rate, pressure, and smoke diffusion coefficient, into a single control variable, the system can adaptively adjust the intensity of flushing, suction, and optical compensation. At the same time, the normalization process in the formula (such as ) eliminates the dimensional differences of the flow rate (mL / s), pressure, and dimensionless smoke coefficient, enabling multi-source data to be quantitatively compared within the same mathematical framework and providing consistent decision-making inputs for closed-loop control. In addition, the system has a built-in safety constraint mechanism: if the real-time flow rate Q 1 or pressure P 1 exceeds the preset safety threshold Q max or P max , an alarm is immediately triggered and the weight adjustment is paused to prevent further increase in the flushing intensity in the over-limit state, thus avoiding the risk of tissue damage or cavity edema caused by fluid impact.

[0021] In some embodiments, the smoke diffusion coefficient α is calculated by the following formula:

[0022]

[0023] Among them, V 1 is the gray variance of the smoke area in the surgical field image, dimensionless, and is obtained by calculating the variance of the pixel gray values in the image area;

[0024] V 0 is the reference value of the gray variance of the surgical field image in the smoke-free state, dimensionless;

[0025] T 1 is the texture feature value of the smoke area, dimensionless, and is obtained by calculating the variance of the local binary pattern (LBP) in the image area;

[0026] M 1 is the motion feature value of the smoke area, dimensionless, and is obtained by calculating the mean value of the optical flow change between consecutive frames;

[0027] w v 、w t 、w m are weight coefficients, satisfying w v +w t +w m = 1 and are all greater than 0; the value range of α is α ≥ 0.

[0028] In this way, when the sudden change of illumination causes the abnormal decrease of the gray variance V 1 , the texture feature T 1 can identify non-smoke areas (such as specular reflection points) to avoid misjudgment. On the contrary, in a low-light environment, the gray variance is more sensitive to smoke than the texture feature, and the two complement each other to reduce missed detection. The short-term appearance of specular reflection or lens jitter may cause abnormal gray-scale / texture in a single frame, but the motion feature can verify whether it has the continuous diffusion characteristic of smoke. When the initial diffusion of smoke (V 1 ) does not decrease significantly, the motion feature M 1 can trigger an early warning in advance to make up for the insufficient sensitivity of the gray-scale / texture feature.

[0029] In some embodiments, the adjustment amount Δθ of the opening degree of the flushing valve is adjusted by the following formula:

[0030]

[0031] Among them, Δθ is the adjustment amount of the opening degree of the flushing valve, and K p 、K i 、K d are the proportional coefficient, integral coefficient, and differential coefficient in the PID control algorithm respectively, and β is a dynamic adjustment factor. When the opening degree of the flushing valve reaches the preset limit, K i· The integration of the ∫βdt term is paused. In this way, the dynamic adjustment factor β can be converted into a flushing valve opening control signal with high precision and high stability, and at the same time, it is linked with the multi-source data fusion and safety constraint mechanism in the foregoing solution to achieve triple optimization of the smoke removal efficiency, system response speed, and fluid safety. Specifically: If the flow rate Q 1 or the pressure P 1 exceeds the safety threshold (Q max or P max ), claim 2 triggers an alarm and pauses the weight adjustment. At this time, the β value is restricted, and the Δθ output by the PID controller is synchronously controlled to prevent the flushing intensity from further increasing in the over-limit state. As the input of the PID, β integrates the information of the flow rate, pressure, and smoke concentration, enabling the adjustment of the flushing valve opening to take into account both

[0032] In some embodiments, the central processing unit calculates the power increment ΔW of the suction pump through the following formula for pressure compensation:

[0033]

[0034] where ΔW is the power increment of the suction pump, unit: W;

[0035] P 0 is the safety threshold negative pressure of the suction channel, unit: mmHg;

[0036] P 2 is the actually monitored negative pressure of the suction channel, unit: mmHg;

[0037] C is the preset pressure compensation coefficient, unit: mmHg / W.

[0038] In this way, the deficiency of the suction efficiency can be reflected in real time through the negative pressure difference (|P 0 |-|P 2 |). The linear model ensures that the power compensation is proportional to the difference. For example, when the actual negative pressure is severely insufficient (the difference is large), the power compensation amount ΔW increases significantly, quickly restoring the negative pressure to the safe range. In addition, the linear coefficient C maps the difference to the power increment, preventing sudden power increases or oscillations caused by non-linear relationships (such as exponential compensation).

[0039] In addition, the power compensation is triggered only when the actual negative pressure |P 2 | is lower than the safety threshold |P 0 |. Otherwise, ΔW = 0. When the negative pressure is sufficient (|P 2 |≥|P 0 |), the power compensation can be stopped to avoid energy waste and equipment wear caused by the continuous high-power operation of the suction pump, and prevent tissue adsorption damage caused by excessive negative pressure (such as too high suction pump power) or cavity pressure imbalance caused by excessive suction of the flushing fluid.

[0040] In some embodiments, the central processing unit is provided with a monitoring module. When the smoke diffusion coefficient values in three consecutive sampling periods all exceed a preset second threshold, the central processing unit will control the flow rate monitoring module to increase the flow rate of the flushing liquid to the maximum safe value. In this way, the smoke discharge efficiency can be effectively improved.

[0041] In some embodiments, the integrated system for endoscopic surgery smoke removal and illumination further includes an illumination module. The central processing unit dynamically adjusts the brightness distribution of the surgical field by combining the image data obtained through the image acquisition module with the optical parameters of the illumination module, including the following steps:

[0042] Obtain real-time surgical field image data through the image acquisition module, and extract the brightness distribution characteristic value L of the image 1 ;

[0043] Compare the brightness distribution characteristic value L 1 with a preset brightness target value L 0 to calculate the brightness deviation ΔL, and its formula is:

[0044] ΔL = L 0 - L 1

[0045] Based on the brightness deviation ΔL, use the dynamic adjustment algorithm of the illumination module to calculate the light source power adjustment amount ΔP;

[0046] Fuse the light source power adjustment amount ΔP with the smoke diffusion coefficient α calculated in real time during the smoke removal process to generate a light source adjustment factor λ, and its formula is:

[0047]

[0048] where w λ is a dynamic weight coefficient, 0 ≤ w λ ≤ 1; K is a smoke compensation coefficient, unit: W, α max is the maximum calibration value of the smoke diffusion coefficient;

[0049] The illumination module dynamically adjusts the output power of the light source according to the light source adjustment factor λ, the brightness deviation ΔL, and the smoke diffusion coefficient α. In this way, when the smoke concentration is high: the α value increases, the smoke compensation weight (1 - w) is increased, and the light source power increment is more used to penetrate the smoke rather than simply increasing the overall brightness, avoiding the waste of light energy and image overexposure caused by smoke scattering. When the smoke concentration is low: the weight is biased towards ΔP, and the brightness uniformity is finely adjusted to ensure the illumination optimization of the core area of the surgery (such as blood vessels and nerves).

[0050] In some embodiments, the central processing unit obtains the luminance distribution characteristic value L through the image acquisition module 1 The specific method includes the following steps:

[0051] Divide the surgical field image data I captured by the image acquisition module 1 into sub-regions to generate multiple image region blocks R 1 、R 2 、...、Rn;

[0052] Calculate the average gray value L of each image region block Ri , and its formula is:

[0053]

[0054] where L Ri is the average gray value of the i-th region block, G j is the gray value of the j-th pixel, and m is the total number of pixels in this region block;

[0055] Perform a weighted sum of the average gray values of all image region blocks to generate the overall luminance distribution characteristic value L 1 , and its formula is:

[0056]

[0057] where L 1 is the luminance distribution characteristic value, w Li is the weight coefficient of the i-th region block, and n is the total number of region blocks. In this way, priority perception of the core region can be achieved, and the driving light source can preferentially compensate the key region. At the same time, the luminance fluctuations caused by instrument reflection or liquid flow in the edge region can be filtered with low weights, avoiding false triggering of luminance adjustment.

[0058] In addition, calculating the average gray value can, on the one hand, avoid the influence of region size or pixel density, accurately reflect the true luminance level of the region, and prevent large regions from overly affecting the overall calculation due to a large number of pixels; on the other hand, when the surgical instrument moves or the smoke spreads, the region luminance change can be captured in real time, providing dynamic data support for the priority of the core region weight.

[0059] In addition, the weighted sum of the high weight of the core region and the low weight of the edge region generates the overall luminance characteristic value L 1 , which can, while giving priority to ensuring the luminance of the core region, still partially incorporate the luminance change of the edge region into the calculation, preventing the overall image contrast from being unbalanced due to completely ignoring the edge. For example, if the luminance of the core region meets the standard but the edge is too dark, L 1 will still reflect a slight deviation and trigger appropriate supplementary lighting to maintain a natural transition of the field of view.

[0060] In some embodiments, the calculation of the light source power adjustment amount ΔP of the lighting module is based on the following formula:

[0061] ΔP = K p ·ΔL + K i ·∫ΔLdt

[0062] where ΔP is the light source power adjustment amount, ΔL is the brightness deviation, and K p , K i are the proportional coefficient and the integral coefficient. In this way, complementary control can be achieved through the proportional term (K p ·ΔL) and the integral term (K i ·∫ΔL dt). Specifically, when the brightness drops suddenly due to a sudden burst of smoke or the movement of the instrument (ΔL increases instantaneously), the proportional term immediately increases the light source power by K p ·ΔL. For example, at the moment of electrocoagulation hemostasis, the proportional term can quickly compensate for a brightness loss of 10 - 20 lux, avoiding blurred vision and affecting the operation. Under the continuous interference of low-concentration smoke (ΔL deviates slightly from the target value for a long time), the integral term gradually increases the power by accumulating historical deviations. For example, when a 5 lux deviation lasts for 10 seconds, the integral term contributes a power increment of K i ·5·10 = 50K i to completely eliminate the residual brightness error. The proportional term and the integral term form a two-layer control of "quick compensation - fine calibration", which not only avoids the steady-state error of pure proportional control (such as the brightness is always 2 lux lower) but also prevents the response lag of pure integral control (such as acting 3 seconds after the sudden burst of smoke).

[0063] In some embodiments, the steps for the lighting module to dynamically adjust the output power of the light source according to the light source adjustment factor λ, the brightness deviation ΔL, and the smoke diffusion coefficient α include:

[0064] Calculating the smoke compensation amount ΔP according to the smoke diffusion coefficient α smoke , and its formula is:

[0065]

[0066] where K is the smoke compensation coefficient;

[0067] Fusing ΔP and ΔP smoke with a dynamic weight to generate the final light source output power P′, and its formula is:

[0068] P′ = min(P′ max , P 0 + w p ·ΔP + (1 - w)·ΔP smoke )

[0069] where w pFor the dynamic weight coefficient \(0\leq w\) p \(\leq1\), which is calculated and determined by the central processing unit according to the real-time surgical scenario; \(P\) 0 is the current output power of the light source; \(P'\) max is the dynamic power upper limit, which is calculated according to the average brightness value of the real-time image: if the average brightness exceeds the preset overexposure threshold, then \(P'\) max \(=P\) max \(-K\) exp \(\cdot(L\) avg \(-L\) th )), where \(K\) exp is the overexposure suppression coefficient, \(L\) th is the overexposure brightness threshold, otherwise \(P'\) max \(=P\) max .

[0070] In this way, when the average image brightness exceeds the threshold \(L\) th , the dynamic power upper limit is reduced to prevent detail loss caused by over-strong light source (such as overexposure of blood vessel texture); in addition, the weight \(w\) p is dynamically calculated by the central processing unit according to the smoke diffusion coefficient \(\alpha\) and the surgical stage, and \(P'\) max is adjusted through image brightness feedback. The two jointly restrict the final power output. Specifically, when the smoke concentration \(\alpha\) is extremely high and the image brightness \(L\) avg is extremely low, \(w\) approaches 0 and \(P'\) max remains at the maximum value, allowing the light source to penetrate the smoke at the maximum power; when the smoke clears and \(\alpha\) returns to zero and the brightness recovers, \(w\) approaches 1 and \(P'\) max is finely adjusted according to the real-time brightness to prevent power surplus. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 FIG. is a schematic structural diagram of an integrated system for smoke removal and illumination in neuroendoscopic surgery provided by some embodiments of the present application.

[0072] Figure 2 is Figure 1 a schematic structural diagram of the closed-loop control module in the integrated system for smoke removal and illumination in neuroendoscopic surgery shown in FIG. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0073] Specific embodiments of the present invention will now be described in detail. Although the present invention is described in connection with these specific embodiments, it should be understood that the intention is not to limit the present invention to these specific embodiments. On the contrary, these embodiments are intended to cover alternatives, modifications, or equivalent embodiments that may be included within the spirit and scope of the invention as defined by the claims. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The present invention may be practiced without some or all of these specific details. In other instances, well-known process operations have not been described in detail in order not to unnecessarily obscure the present invention.

[0074] As used in conjunction with the terms "comprising," "method comprising," or similar language in this specification and the appended claims, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise. All technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains, unless otherwise defined.

[0075] Please refer to Figure 1 and in conjunction with Figure 2 , an embodiment of the present application provides a neuroendoscopic surgery smoke removal and illumination integrated system, comprising: a flow rate monitoring module 2, a pressure monitoring module 3, an image acquisition module 4, a central processing unit 1, and a closed-loop control module 5. The flow rate monitoring module 2 is configured to collect real-time flow rate data of the flushing liquid and transmit it to the central processing unit 1;

[0076] The pressure monitoring module 3 is configured to collect real-time pressure data in the cavity and transmit it to the central processing unit 1;

[0077] The image acquisition module 4 is configured to capture real-time surgical field images in the cavity and transmit them to the central processing unit 1;

[0078] The central processing unit 1 is configured to receive data from the flow rate monitoring module 2, the pressure monitoring module 3, and the image acquisition module 4; analyze the image data through an image processing algorithm, extract smoke area feature data, and compare it with a preset smoke feature threshold to generate a smoke diffusion coefficient; perform multi-source data fusion based on the flow rate data, pressure data, and smoke diffusion coefficient to generate a dynamic adjustment factor; when the dynamic adjustment factor exceeds a preset threshold, trigger the closed-loop control module 5 to adjust the internal environment of the cavity;

[0079] The closed-loop control module 5 includes a flushing adjustment submodule 51, a suction adjustment submodule 52, an optical compensation and image enhancement submodule 53 and a feedback submodule 54. The flushing adjustment submodule 51 is used to dynamically adjust the opening of the flushing valve according to the difference between the flow rate data and the target flow rate; the suction adjustment submodule 52 is used to monitor the negative pressure value of the suction channel in real time. If the negative pressure value is lower than the safety threshold, pressure compensation is performed to restore the negative pressure value to a safe range by increasing the suction pump power; the optical compensation and image enhancement submodule 53 is used to perform optical compensation and enhancement processing on the real-time image based on the adjusted flushing and suction parameters; the feedback submodule 54 is used to feed back the adjusted flushing valve opening and suction pump power parameters to the central processing unit 1, and transmit the optimized image data to the display device.

[0080] In the embodiment of the present application, the flow rate monitoring module 2 and the pressure monitoring module 3 respectively collect the flushing fluid flow rate and the pressure in the cavity in real time to provide basic data for fluid control; the image acquisition module 4 captures the surgical field image, and the central processing unit 1 generates the smoke diffusion coefficient by analyzing the smoke area characteristics in the image to quantify the trend of smoke concentration changes.

[0081] Based on the multi-source data fusion of flow rate, pressure and smoke diffusion coefficient, the central processing unit 1 calculates the dynamic adjustment factor. When the dynamic adjustment factor exceeds the preset threshold, the closed-loop control module 5 is triggered to perform multi-dimensional adjustment. The flushing adjustment submodule 51 dynamically adjusts the flushing valve opening according to the flow rate difference to directly remove the smoke; the suction adjustment submodule 52 compensates for the negative pressure in the suction channel in real time to prevent smoke from being retained; specifically, increase the flushing valve opening → increase the impact force of the water flow → peel off the smoke particles from the tissue surface, enhance the suction negative pressure → form a directional fluid channel → quickly export the suspended smoke.

[0082] The optical compensation and image enhancement submodule 53 combines the value and real-time image data to offset the interference of residual smoke by enhancing contrast and brightness. In this way, the dynamic matching of smoke removal intensity with the real-time state of cavity pressure and flow rate is ensured, and at the same time, image enhancement is used to compensate for the temporary loss of visual field during smoke removal, thereby improving the observation effect of endoscopic images, breaking through the limitations of traditional methods that rely on preoperative anti-fog treatment or single parameter adjustment, and achieving the triple goals of efficient smoke removal, cavity environment stability and surgical field optimization during surgery, avoiding surgical risks caused by delayed manual intervention or operation conflicts.

[0083] In some embodiments, the dynamic adjustment factor is calculated by the following formula:

[0084]

[0085] Among them, w 1 、w 2 、w3 is a weight coefficient, satisfying w 1 + w 2 + w 3 = 1 and w 1 , w 2 , w 3 > 0;

[0086] Q max is the preset maximum allowable flushing flow rate, unit: mL / s;

[0087] P max is the preset maximum allowable cavity pressure, unit: mmHg;

[0088] α is the smoke diffusion coefficient, dimensionless. When α > α threshold , w 3 = 0.5, w 1 = w 2 = 0.25, α threshold is the smoke diffusion emergency threshold; When Q 1 ≥ Q max or P 1 ≥ P max , the central processing unit 1 triggers an alarm and pauses weight adjustment until the flow rate or pressure returns to the safe range.

[0089] In this way, through the co - design of dynamic weight allocation and emergency priority rules, the balance between smoke clearance efficiency and cavity safety is achieved in the complex and changeable intraoperative environment. By fusing three types of heterogeneous parameters, namely flow rate, pressure, and smoke diffusion coefficient, into a single control variable, the system can adaptively adjust the intensity of flushing, suction, and optical compensation. At the same time, the normalization process in the formula (such as ) eliminates the dimensional differences of the flow rate (mL / s), pressure, and dimensionless smoke coefficient, enabling multi - source data to be quantitatively compared within the same mathematical framework and providing consistent decision - making inputs for closed - loop control. In addition, the system has a built - in safety constraint mechanism: if the real - time flow rate Q 1 or pressure P 1 exceeds the preset safety threshold Q max or P max , an alarm is immediately triggered and weight adjustment is paused to prevent further increase in flushing intensity in the over - limit state, thus avoiding the risk of tissue damage or cavity edema caused by fluid impact.

[0090] In some embodiments, the smoke diffusion coefficient α is calculated by the following formula:

[0091]

[0092] where V 1is the gray variance of the smoke area in the surgical field image, dimensionless, and is obtained by calculating the variance of the pixel gray values in the image area;

[0093] V 0 is the reference value of the gray variance of the surgical field image in the smoke-free state, dimensionless;

[0094] T 1 is the texture feature value of the smoke area, dimensionless, and is obtained by calculating the variance of the local binary pattern LBP in the image area;

[0095] M 1 is the motion feature value of the smoke area, dimensionless, and is obtained by calculating the mean value of the optical flow change between consecutive frames;

[0096] w v 、w t 、w m are weight coefficients, satisfying w v +w t +w m = 1 and are all greater than 0; the value range of α is α≥0.

[0097] In this way, when the sudden change of illumination causes the abnormal decrease of the gray variance V 1 , the texture feature T 1 can identify the non-smoke area (such as the reflection point) to avoid misjudgment. On the contrary, in a low-light environment, the gray variance is more sensitive to smoke than the texture feature, and the two complement each other to reduce missed detection. The transient reflection or lens shake may cause abnormal gray / texture in a single frame, but the motion feature can verify whether it has the continuous diffusion characteristic of smoke. When the initial diffusion of smoke (V 1 ) does not decrease significantly, the motion feature M 1 can trigger an early warning in advance to make up for the insufficient sensitivity of the gray / texture feature.

[0098] In some embodiments, the adjustment amount Δθ of the opening degree of the flushing valve is adjusted by the following formula:

[0099]

[0100] where Δθ is the adjustment amount of the opening degree of the flushing valve, and K p 、K i 、K d are the proportional coefficient, integral coefficient and differential coefficient in the PID control algorithm respectively, and β is a dynamic adjustment factor. When the opening degree of the flushing valve reaches the preset limit, the integral term K i ·∫βdt pauses to accumulate. Among them, the proportional term (K p ·β): quickly responds to the smoke diffusion coefficient α, flow rate Q 1 and pressure P 1Real-time changes. For example, when sudden smoke appears during the operation (α suddenly rises), the β value rapidly increases, and the proportional term directly increases the opening degree Δθ of the flushing valve, immediately enhancing the flushing intensity to remove the smoke. The integral term (K i ·∫βdt): Accumulates the historical β value to eliminate the steady-state error. For example, in a continuous low-concentration smoke environment (α fluctuates slightly), the integral term gradually adjusts the opening degree to the optimal value to avoid frequent oscillations under proportional term control. The differential term Predicts the change trend of β and suppresses overshoot. For example, when β starts to decline after the smoke is cleared, the differential term reduces the opening degree in advance to prevent excessive flushing caused by inertia.

[0101] In this way, the dynamic adjustment factor β can be converted into a flushing valve opening control signal with high precision and high stability, and at the same time, it is linked with the multi-source data fusion and safety constraint mechanism in the foregoing solution to achieve triple optimization of smoke removal efficiency, system response speed, and fluid safety. Specifically: If the flow rate Q 1 or the pressure P 1 exceeds the safety threshold (Q max or P max ), Claim 2 triggers an alarm and pauses the weight adjustment. At this time, the β value is restricted, and the Δθ output by the PID controller is synchronously controlled to avoid further increasing the flushing intensity in the over-limit state. As the input of the PID, β integrates the flow rate, pressure, and smoke concentration information, enabling the adjustment of the flushing valve opening to take all into account.

[0102] In some embodiments, the central processing unit 1 calculates the power increment ΔW of the suction pump through the following formula for pressure compensation:

[0103]

[0104] where ΔW is the power increment of the suction pump, unit: W;

[0105] P 0 is the safety threshold negative pressure of the suction channel, unit: mmHg;

[0106] P 2 is the actually monitored negative pressure of the suction channel, unit: mmHg;

[0107] C is the preset pressure compensation coefficient, unit: mmHg / W.

[0108] In this way, through the negative pressure difference (|P 0 |-|P 2|) Reflect the deficiency of the suction efficiency in real time. The linear model ensures that the power compensation is proportional to the difference. For example, when the actual negative pressure is severely insufficient (large difference), the power compensation amount ΔW increases significantly, quickly restoring the negative pressure to the safe range. Additionally, the linear coefficient C maps the difference to the power increment, preventing sudden power surges or oscillations caused by non-linear relationships (such as exponential compensation).

[0109] In addition, only when the actual negative pressure |P 2 | is lower than the safety threshold |P 0 | does the power compensation get triggered, otherwise ΔW = 0. When the negative pressure is sufficient (|P 2 | ≥ |P 0 |), the power compensation stops, avoiding energy waste and equipment wear caused by the continuous high-power operation of the suction pump, and preventing tissue adsorption damage caused by excessive negative pressure (such as too high suction pump power) or cavity pressure imbalance caused by excessive aspiration of the flushing fluid.

[0110] In some embodiments, the central processing unit 1 is provided with a monitoring module. When the smoke diffusion coefficient values in three consecutive sampling periods all exceed the preset second threshold, the central processing unit 1 will control the flow rate monitoring module 2 to increase the flow rate of the flushing fluid to the maximum safe value. In this way, the smoke discharge efficiency can be effectively improved.

[0111] In some embodiments, the neuroendoscopic surgery smoke removal and illumination integrated system further includes an illumination module. The central processing unit 1 dynamically adjusts the brightness distribution of the surgical field by combining the image data obtained through the image acquisition module 4 with the optical parameters of the illumination module, including the following steps:

[0112] Obtain the real-time surgical field image data through the image acquisition module 4, and extract the brightness distribution characteristic value L of the image 1 ;

[0113] Compare the brightness distribution characteristic value L 1 with the preset brightness target value L 0 and calculate the brightness deviation ΔL. The formula is:

[0114] ΔL = L 0 - L 1

[0115] Based on the brightness deviation ΔL, calculate the light source power adjustment amount ΔP using the dynamic adjustment algorithm of the illumination module;

[0116] Fuse the light source power adjustment amount ΔP with the smoke diffusion coefficient α calculated in real time during the smoke removal process to generate a light source adjustment factor λ. The formula is:

[0117]

[0118] Among them, w λ is the dynamic weight coefficient, where 0 ≤ w λ ≤ 1; K is the smoke compensation coefficient, unit: W, and α max is the maximum calibration value of the smoke diffusion coefficient;

[0119] The lighting module dynamically adjusts the output power of the light source according to the light source adjustment factor λ, the brightness deviation ΔL, and the smoke diffusion coefficient α. In this way, when the smoke concentration is high: the value of α increases, enhancing the smoke compensation weight (1 - w), and more of the light source power increment is used to penetrate the smoke rather than simply increasing the overall brightness, avoiding the waste of light energy and overexposure of the image caused by smoke scattering. When the smoke concentration is low: the weight biases towards ΔP, finely adjusting the brightness uniformity to ensure the lighting optimization in the core area of the surgery (such as blood vessels and nerves).

[0120] In some embodiments, the specific method for the central processing unit 1 to obtain the brightness distribution feature value L 1 through the image acquisition module 4 includes the following steps:

[0121] Divide the surgical field image data I 1 captured by the image acquisition module 4 into multiple regions to generate multiple image region blocks R 1 , R 2 ,..., Rn;

[0122] Calculate the average gray value L Ri of each image region block, and its formula is:

[0123]

[0124] where L Ri is the average gray value of the i-th region block, G j is the gray value of the j-th pixel, and m is the total number of pixels in this region block;

[0125] Perform weighted summation on the average gray values of all image region blocks to generate the overall brightness distribution feature value L 1 , and its formula is:

[0126]

[0127] where L 1 is the brightness distribution feature value, w Li is the weight coefficient of the i-th region block, and n is the total number of region blocks. In this way, priority perception of the core area can be achieved, driving the light source to preferentially compensate the key area. At the same time, the brightness fluctuations caused by instrument reflection or liquid flow in the edge area can be filtered by low weights, avoiding false triggering of brightness adjustment.

[0128] In addition, calculating the average gray value can, on the one hand, avoid the influence of the region size or pixel density, accurately reflect the true brightness level of the region, and prevent a large region from overly affecting the overall calculation due to a large number of pixels. On the other hand, when the surgical instrument moves or the smoke spreads, the change in the region brightness can be captured in real time, providing dynamic data support for the priority of the core region weight.

[0129] In addition, the weighted sum of the high weight of the core region and the low weight of the edge region generates the overall brightness feature value L 1 , which can, while giving priority to ensuring the brightness of the core region, still partially incorporate the brightness change of the edge region into the calculation, preventing the overall image contrast from being unbalanced due to completely ignoring the edge. For example, if the brightness of the core region meets the standard but the edge is too dark, L 1 will still reflect a slight deviation, triggering appropriate supplementary lighting to maintain the natural transition of the field of view.

[0130] In some embodiments, the calculation of the light source power adjustment amount ΔP of the lighting module is based on the following formula:

[0131] ΔP = K p ·ΔL + K i ·∫ΔLdt

[0132] where ΔP is the light source power adjustment amount, ΔL is the brightness deviation, and K p , K i are the proportional coefficient and the integral coefficient. In this way, complementary control can be achieved through the proportional term (K p ·ΔL) and the integral term (K i ·∫ΔL dt). Specifically, when the brightness suddenly drops due to a sudden burst of smoke or the movement of the instrument (ΔL instantaneously increases), the proportional term immediately increases the light source power by K p ·ΔL. For example, at the moment of electrocoagulation hemostasis, the proportional term can quickly compensate for a brightness loss of 10 - 20 lux, avoiding blurred vision and affecting the operation. Under the continuous interference of low-concentration smoke (ΔL deviates from the target value slightly for a long time), the integral term gradually increases the power by accumulating the historical deviation. For example, when a 5 lux deviation lasts for 10 seconds, the integral term contributes K i ·5·10 = 50K i of power increment to completely eliminate the residual brightness error. The proportional term and the integral term form a two-layer control of "quick compensation - fine calibration", which not only avoids the steady-state error of pure proportional control (such as the brightness is always 2 lux lower) but also prevents the response lag of pure integral control (such as taking 3 seconds to act after a sudden burst of smoke).

[0133] In some embodiments, the steps for the lighting module to dynamically adjust the output power of the light source according to the light source adjustment factor λ, the brightness deviation ΔL, and the smoke diffusion coefficient α include:

[0134] Calculate the smoke compensation amount ΔP according to the smoke diffusion coefficient α smoke , and its formula is:

[0135]

[0136] where K is the smoke compensation coefficient;

[0137] Fuse ΔP and ΔP smoke according to the dynamic weight to generate the final light source output power P′, and its formula is:

[0138] P′ = min(P′ max , P 0 + w p ·ΔP + (1 - w)·ΔP smoke )

[0139] where w p is the dynamic weight coefficient, 0 ≤ w p ≤ 1, which is calculated and determined by the central processing unit 1 according to the real-time surgical scenario; P 0 is the current output power of the light source; P′ max is the dynamic power upper limit, which is calculated according to the average brightness value of the real-time image: if the average brightness exceeds the preset overexposure threshold, then P′ max = P max - K exp ·(L avg - L th ), where K exp is the overexposure suppression coefficient, L th is the overexposure brightness threshold, otherwise P′ max = P max .

[0140] In this way, when the average image brightness exceeds the threshold L th , the power upper limit is dynamically reduced to prevent details from being lost due to excessive light source (such as overexposure of blood vessel textures); in addition, the weight w p is dynamically calculated by the central processing unit 1 according to the smoke diffusion coefficient α and the surgical stage, and P′ max is adjusted through image brightness feedback. The two jointly constrain the final power output. Specifically, when the smoke concentration α is extremely high and the image brightness L avg is extremely low, w approaches 0 and P′ max remains at the maximum value, allowing the light source to penetrate the smoke at the maximum power; when the smoke is cleared and α returns to zero and the brightness recovers, w approaches 1 and P′ max is finely adjusted according to the real-time brightness to prevent power surplus.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that it is still possible to modify the specific implementation manners of the present invention or perform equivalent replacements on some technical features. Without departing from the spirit of the technical solutions of the present invention, they should all be covered within the scope of the technical solutions claimed by the present invention.

Claims

1. A neuroendoscopic surgery smoke removal and lighting integrated system, characterized in that: include: A flow rate monitoring module, a pressure monitoring module, an image acquisition module, a central processing unit and a closed-loop control module, wherein the flow rate monitoring module is used to collect real-time flow rate data of the flushing liquid and transmit it to the central processing unit; The pressure monitoring module is used to collect real-time pressure data in the cavity and transmit it to the central processing unit; The image acquisition module is used to capture real-time surgical field images in the cavity and transmit them to the central processing unit; The central processing unit is used to receive data from the flow rate monitoring module, the pressure monitoring module and the image acquisition module; analyze the image data through the image processing algorithm, extract the smoke area feature data, and compare it with the preset smoke feature threshold to generate the smoke diffusion coefficient; perform multi-source data fusion based on the flow rate data, pressure data and smoke diffusion coefficient to generate a dynamic adjustment factor; When the dynamic adjustment factor exceeds a preset threshold, the closed-loop control module is triggered to adjust the environment inside the cavity; The closed-loop control module includes a flushing adjustment submodule, a suction adjustment submodule, an optical compensation and image enhancement submodule, and a feedback submodule, wherein the flushing adjustment submodule is used to dynamically adjust the opening of the flushing valve according to the difference between the flow rate data and the target flow rate; The suction adjustment submodule is used to monitor the negative pressure value of the suction channel in real time. If the negative pressure value is lower than the safety threshold, pressure compensation is performed to restore the negative pressure value to a safe range by increasing the suction pump power; the optical compensation and image enhancement submodule is used to perform optical compensation and enhancement processing on the real-time image based on the adjusted flushing and suction parameters; The feedback submodule is used to feed back the adjusted flushing valve opening and suction pump power parameters to the central processing unit, and transmit the optimized image data to the display device.

2. The integrated neuroendoscopic surgery smoke removal and lighting system according to claim 1, characterized in that: The dynamic adjustment factor is calculated by the following formula: Among them, w1, w2, w3 are weight coefficients, satisfying w1+w2+w3=1 and w1,w2,w3>0; Q max is the preset maximum permissible flushing flow rate; P max is the preset maximum allowable cavity pressure; α is the smoke diffusion coefficient, when α>α threshold When w3=0.5, w1=w2=0.25, α threshold is the emergency threshold of smoke diffusion; when Q1≥Q max Or P1 ≥ P max When the flow rate or pressure returns to a safe range, the central processing unit triggers an alarm and suspends weight adjustment.

3. The integrated neuroendoscopic surgery smoke removal and lighting system according to claim 2, characterized in that: The smoke diffusion coefficient α is calculated by the following formula: Among them, V1 is the grayscale variance of the smoke area in the surgical field image, which is obtained by calculating the variance of the pixel grayscale values ​​in the image area; V0 is the grayscale variance reference value of the surgical field image in the smoke-free state; T1 is the texture feature value of the smoke area, which is obtained by calculating the variance of the local binary pattern LBP of the image area; M1 is the motion feature value of the smoke area, which is obtained by calculating the mean value of the optical flow change between consecutive frames; w v 、w t 、w m is the weight coefficient, satisfying w v +w t +w m =1 and are both greater than 0; the value range of α is α≥0.

4. The integrated neuroendoscopic surgery smoke removal and lighting system according to claim 3, characterized in that: The adjustment amount Δθ of the flushing valve opening is adjusted by the following formula: Among them, Δθ is the opening adjustment of the flushing valve, K p , K i , K d are the proportional coefficient, integral coefficient and differential coefficient in the PID control algorithm respectively, β is the dynamic adjustment factor, when the flushing valve opening reaches the preset limit, K i ·The accumulation of the integral term ∫βdt is suspended.

5. The integrated neuroendoscopic surgery smoke removal and lighting system according to claim 4, characterized in that: The central processing unit calculates the power increment ΔW of the suction pump by the following formula to perform pressure compensation: Among them, ΔW is the suction pump power increment; P0 is the safety threshold negative pressure of the suction channel; P2 is the actual monitored negative pressure of the suction channel; C is the preset pressure compensation coefficient.

6. The integrated neuroendoscopic surgery smoke removal and lighting system according to claim 5, characterized in that: The central processing unit is provided with a monitoring module. When the smoke diffusion coefficient values ​​of three consecutive sampling cycles exceed the preset second threshold value, the central processing unit will control the flow rate monitoring module to increase the flow rate of the flushing liquid to the maximum safe value.

7. The integrated neuroendoscopic surgery smoke removal and lighting system according to claim 6, characterized in that: It also includes an illumination module, wherein the central processing unit dynamically adjusts the brightness distribution of the surgical field of view by combining the image data acquired by the image acquisition module with the optical parameters of the illumination module, including the following steps: Acquire real-time surgical field image data through the image acquisition module and extract the brightness distribution characteristic value L1 in the image; Compare the brightness distribution characteristic value L1 with the preset brightness target value L0 to calculate the brightness deviation ΔL, the formula is: ΔL=L0-L1 Based on the brightness deviation ΔL, the light source power adjustment amount ΔP is calculated using the dynamic adjustment algorithm of the lighting module; The light source power adjustment ΔP is combined with the smoke diffusion coefficient α calculated in real time during the smoke removal process to generate the light source adjustment factor λ, whose formula is: Among them, w λ is the dynamic weight coefficient, 0≤w λ ≤1; K is the smoke compensation coefficient, α max is the maximum calibration value of the smoke diffusion coefficient; The lighting module dynamically adjusts the output power of the light source according to the light source adjustment factor λ, the brightness deviation ΔL and the smoke diffusion coefficient α.

8. The integrated neuroendoscopic surgery smoke removal and lighting system according to claim 7, characterized in that: The specific method for the central processing unit to obtain the brightness distribution characteristic value L1 through the image acquisition module includes the following steps: Divide the surgical field image data I1 captured by the image acquisition module into regions to generate a plurality of image region blocks R1, R2, ..., Rn; Calculate the average gray value L of each image area block Ri , the formula is: Among them, L Ri is the average gray value of the i-th region block, G j is the gray value of the jth pixel, and m is the total number of pixels in the area block; The weighted sum of the average grayscale values ​​of all image area blocks is used to generate the overall brightness distribution characteristic value L1, which is: Among them, L1 is the characteristic value of brightness distribution, w Li is the weight coefficient of the ith area block, and n is the total number of area blocks.

9. The integrated neuroendoscopic surgery smoke removal and lighting system according to claim 8, characterized in that: The calculation of the light source power adjustment amount ΔP of the lighting module is based on the following formula: ΔP=K p ·ΔL+K i ·∫ΔLdt Among them, ΔP is the light source power adjustment, ΔL is the brightness deviation, K p , K i are the proportional coefficient and the integral coefficient.

10. The integrated neuroendoscopic surgery smoke removal and lighting system according to claim 9, characterized in that: The step of dynamically adjusting the output power of the light source according to the light source adjustment factor λ, the brightness deviation ΔL and the smoke diffusion coefficient α comprises: Calculate the smoke compensation amount ΔP according to the smoke diffusion coefficient α smoke , the formula is: Wherein, K is the smoke compensation coefficient; Compare ΔP with ΔP smoke According to the dynamic weight fusion, the final light source output power P′ is generated, and the formula is: P′=min(P′ max ,P0+w p ·ΔP+(1-w)·ΔP smoke ) Among them, w p is the dynamic weight coefficient (0≤w p ≤1), which is calculated and determined by the central processing unit according to the real-time surgical scene; P0 is the current output power of the light source; P′ max is the dynamic power upper limit, which is calculated based on the average brightness value of the real-time image: If the average brightness exceeds the preset overexposure threshold, then P′ max =P max -K exp ·(L avg -L th ), where K exp is the overexposure suppression coefficient, L th is the overexposure brightness threshold, otherwise P′ max =P max .

Images