Energy control platform for bladder tumor electrocision based on obturator nerve electrophysiological feedback

Abstract

The present application relates to a kind of based on closed nerve electrophysiological feedback bladder tumor electrosection energy control platform, including operation field configuration unit, synchronous acquisition unit, interference stripping unit, risk control unit and recovery check unit.Platform according to the position of bladder tumor on lateral wall determines monitoring side, establishes unilateral or bilateral long adductor muscle monitoring channel, the electromyographic data is synchronously collected before and after energy output of resectoscope, and according to synchronous mark, stripping electric cautery artifact is extracted amplitude increment, burst wave density and side advantage feature, closed reflection risk grade is generated, then linkage restricts electrosection output power, single duration and interval of output again, and recovery setting output after risk falls back to meet safety condition, so as to improve energy control safety and stability in the process of bladder tumor electrosection.

Description

Technical Field

[0001] This invention relates to the field of medical technology, and in particular to an energy control platform for bladder tumor electroresection based on obturator nerve electrophysiological feedback. Background Technology

[0002] Transurethral resection of bladder tumors (TURP) is a commonly used endoscopic surgical procedure in the diagnosis and treatment of bladder tumors. It typically involves inserting a resectoscope through the urethra into the bladder cavity, and using high-frequency electroresection energy under direct visualization to remove, stop bleeding, or trim the bladder tumor tissue. For tumors located on the lateral wall of the bladder or near the course of the obturator nerve, the intraoperative electroresection energy may stimulate the obturator nerve, causing sudden contraction of the adductor muscles of the thigh, clinically known as the obturator nerve reflex. To reduce the surgical risks associated with this reflex, current techniques typically employ obturator nerve block, adjusting anesthesia methods, reducing electroresection power, changing the electroresection angle, shortening the single electroresection time, using bipolar electroresection equipment, or allowing the surgeon to pause the electroresection based on experience. Some procedures also attempt to use intraoperative electromyography or physiological signal monitoring to assist in assessing neuromuscular activity and provide guidance for the surgical procedure.

[0003] However, current methods primarily rely on preoperative blockade, surgeon's experience, or fixed power parameter settings, making it difficult to proactively adjust energy output based on real-time changes in the obturator nerve's electrophysiological state during electroresection. Conventional electromyography (EMG) monitoring signals are easily interfered with by the electrocautery artifacts generated when the electroresection endoscope outputs high-frequency energy, making it difficult to distinguish between genuine obturator nerve reflex auras and electroresection artifacts. Furthermore, current methods typically fail to comprehensively assess factors such as the tumor's location on the lateral wall, the obturator nerve's course, differences in unilateral or bilateral adductor longus muscle signals, and the timing of energy output, leading to delayed risk identification, misjudgments, missed judgments, or discontinuous control actions. When the risk increases, current equipment often requires manual power reduction or pausing, lacking a mechanism to control output power, single-session duration, and the interval between re-outputs. It also lacks a continuous verification process before resuming output after the risk subsides, making it difficult to balance electroresection efficiency, signal reliability, and surgical safety.

[0004] Therefore, it is necessary to provide a technical solution that can be applied to the scenario of bladder tumor resection and can improve the synergy between obturator nerve reflex risk identification and resection energy control. Summary of the Invention

[0005] This application provides a bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback to improve the safety of electroresection energy control.

[0006] This application provides a bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback, comprising: The surgical field configuration unit is used to determine the monitoring laterality related to the course of the obturator nerve based on the location of the bladder tumor on the lateral wall, and to generate unilateral or bilateral adductor longus monitoring channels and corresponding obturator risk lateralities based on the monitoring laterality. The synchronous acquisition unit is used to receive the trigger signal of the electroresection endoscope energy output, and establish the adductor longus baseline through the unilateral or bilateral adductor longus monitoring channel before energy output based on the trigger signal of the electroresection endoscope energy output, form a synchronous marker during energy output, and set a delayed detection window and acquire electromyographic data after energy output. The interference stripping unit is used to identify electrocautery artifacts that appear in phase with the high-frequency output of electromyography based on the synchronization marker, and to strip the electrocautery artifacts based on the long adductor baseline, so as to extract amplitude increment, burst wave density and lateral dominance features from the electromyography data within the delay detection window. The risk control unit is used to generate the pore reflection risk level based on the amplitude increment, burst wave density, side advantage characteristics and pore risk side, and to limit the electrical switching output power, single duration and re-output interval according to the pore reflection risk level. The recovery verification unit is used to continue to detect the risk level decline state based on the electromyography data within the delay detection window after the risk control unit limits the electrical switching output power, single duration and re-output interval. It allows the risk control unit to resume the set output only when the safety conditions are met in multiple consecutive delay detection windows.

[0007] This application has the following beneficial technical effects: (1) By determining the monitoring side based on the location of the bladder tumor on the lateral wall and establishing a unilateral or bilateral adductor longus muscle monitoring channel, the electromyography (EMG) acquisition corresponds to the high-risk area of ​​the obturator nerve, thereby improving the specificity of obturator reflex risk identification. (2) By receiving the trigger signal of the electroresection endoscope energy output and setting synchronous markers and delayed detection windows, the EMG acquisition and electroresection output timing can be linked, which is beneficial to distinguish between real neuroelectromyography changes and ordinary background fluctuations. (3) By stripping the electrocautery artifacts that appear in phase with the high-frequency output of electroresection based on the synchronous markers and extracting the amplitude increment, burst wave density and side advantage features, the accuracy and stability of obturator reflex risk level judgment can be improved. (4) By limiting the electroresection output power, single duration and re-output interval according to the obturator reflex risk level, and restoring the output after multiple consecutive delayed detection windows meet the safety conditions, the risk of false output is reduced and the safety of the electroresection process is improved. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of a bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback provided in the first embodiment of this application. Detailed Implementation

[0009] Many specific details are set forth in the following description to provide a full understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this application. Therefore, this application is not limited to the specific implementations disclosed below.

[0010] The first embodiment of this application provides a bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback. Please refer to... Figure 1 This figure is a schematic diagram of the first embodiment of this application. The following is in conjunction with... Figure 1 The first embodiment of this application provides a bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback.

[0011] The bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback includes a surgical field configuration unit 101, a synchronous acquisition unit 102, an interference removal unit 103, a risk control unit 104, and a recovery verification unit 105.

[0012] The surgical field configuration unit 101 is used to determine the monitoring laterality related to the course of the obturator nerve based on the location of the bladder tumor on the lateral wall, and to generate unilateral or bilateral adductor longus monitoring channels and corresponding obturator risk lateralities based on the monitoring laterality.

[0013] The surgical field configuration unit 101 is used to establish a monitoring basis that matches the surgical field location for subsequent electromyography (EMG) acquisition and risk control before the bladder tumor electroresection energy control platform begins operation. The lateral wall location of the bladder tumor refers to the spatial position of the tumor body or the edge to be electroresected relative to the left and right walls of the bladder and the area near the obturator nerve. This position is manually entered by the surgeon in the platform's human-computer interface based on cystoscopy observations, or can be imported from surgical recording equipment, endoscopic positioning equipment, or preoperative image marking information connected to the platform according to a DICOM structured report or a preset JSON format, and mapped to preset region codes by the platform's built-in parsing module. The obturator nerve area refers to the anatomically adjacent area where the obturator nerve passes through the pelvic lateral wall and is related to the innervation of the adductor muscles of the thigh. During bladder lateral wall tumor electroresection, this area is more likely to induce contraction of the adductor muscles of the thigh after being stimulated by high-frequency electroresection energy, thus serving as a basis for determining the monitoring laterality. The surgical field configuration unit 101 does not directly determine whether the obturator reflex has occurred, but rather first establishes the lateral correspondence required for subsequent monitoring and control based on the tumor location.

[0014] In one specific embodiment, the surgical field configuration unit 101 includes a location input interface, lateral determination logic, and a channel configuration output interface, which are sequentially connected to form a complete configuration link. Before the electrocautery begins, the surgeon selects the location of the tumor, such as the left side wall, right side wall, bilateral side walls, anterolateral side wall, posterolateral side wall, or near the midline of the bladder, via a touchscreen, foot control confirmation interface, or surgical host control interface. The platform divides the bladder cavity into several preset regions, such as a high-risk area for the left obturator, a high-risk area for the right obturator, bilateral risk areas, and a low-risk area for the obturator. The boundaries of these preset regions are pre-defined based on the anatomical proximity between the bladder sidewall and the obturator nerve pathway and are stored in a built-in bladder region-obturator risk comparison table. The lateralization determination logic matches the operator's input location or imported region code with the lookup table, outputting the corresponding monitoring lateralization and obturator risk lateralization: when the input location falls within the left obturator high-risk area, the monitoring lateralization is determined to be the left, and a left adductor longus monitoring channel is generated, while the left side is marked as the obturator risk lateralization; when the input location falls within the right obturator high-risk area, the monitoring lateralization is determined to be the right, and a right adductor longus monitoring channel is generated, while the right side is marked as the obturator risk lateralization; when the tumor is large, located near the junction of the two bladder walls, the operator has difficulty identifying the main risk lateralization, or it is expected that continuous electroresection from both left and right directions will be required during the operation, bilateral adductor longus monitoring channels are generated, and both the left and right sides are used as obturator risk lateralizations.

[0015] The adductor longus monitoring channel in this invention refers to a complete signal path for acquiring electromyographic signals related to the adductor longus muscle. It includes at least a monitoring electrode, a reference electrode or grounding electrode, a signal input port, a channel number, a lateral marker, and a data transmission relationship with the subsequent synchronous acquisition unit 102, all positioned on the corresponding side of the adductor longus muscle's surface projection area. The monitoring electrode is a surface disc electrode or a subcutaneous needle electrode, arranged on the adductor longus muscle's surface projection area according to conventional electromyographic monitoring methods in the art. The reference electrode is positioned near electrically inert bony landmarks (e.g., the patella or anterior superior iliac spine). The signal input port uses a standard interface of conventional multi-channel electromyographic acquisition equipment in the art. The channel number and electrode port establish a physical correspondence through a pre-stored mapping table when the platform starts. The surgical field configuration unit 101 generates a unilateral or bilateral adductor longus monitoring channel. This does not only refer to opening a certain electrode port, but also includes assigning a clear left or right label to the channel, determining whether the channel participates in the obturator reflex risk assessment, setting the correspondence between the channel and the obturator risk side, and sending the configuration to the synchronous acquisition unit 102, the interference removal unit 103, and the risk control unit 104, so that the electromyographic data subsequently acquired can be identified as coming from the left adductor longus, the right adductor longus, or the corresponding side of the bilateral adductor longus.

[0016] For example, if the surgeon confirms during cystoscopy that the tumor is mainly located in the lateral region of the left lateral wall of the bladder, and the electroresection path is close to the course of the left obturator nerve, the surgical field configuration unit 101, upon receiving this location information, sets the monitoring side to the left by looking up a table. This prompts the surgeon to place the monitoring electrodes on the corresponding surface location of the patient's left adductor longus muscle and generates a left adductor longus monitoring channel within the platform. Simultaneously, the left side is output as the obturator risk side. In this case, even if there are normal background electromyographic fluctuations on the right side, subsequent risk control will still prioritize the assessment based on the electromyographic changes in the left adductor longus monitoring channel. If the tumor is located on the left lateral wall of the bladder but close to the midline, or if the surgeon selects a bilateral risk mode, the surgical field configuration unit 101 will simultaneously activate both the left and right adductor longus monitoring channels and use both sides as the obturator risk sides. This allows subsequent units to compare the differences in electromyographic changes between the two sides, avoiding the omission of risk signals due to monitoring only one side.

[0017] To ensure reliable configuration results, the surgical field configuration unit 101 performs a channel confirmation operation after generating unilateral or bilateral adductor longus muscle monitoring channels. This channel confirmation operation includes detecting electrode connection status, confirming whether the left and right side inputs correspond to the electrode ports, prompting the surgeon to verify the tumor location and monitoring side, and displaying the enabled monitoring channels and corresponding obturator risk sides on the interface. If electrodes are not connected, left and right side inputs conflict, or the surgeon fails to complete confirmation, the surgical field configuration unit 101 first outputs an invalid configuration status to the synchronous acquisition unit 102 to block the acquisition process and displays the reason for the abnormality on the user interface. If the surgeon fails to resolve the abnormality within a preset time, further prohibition is imposed on subsequent energy control processes, thereby preventing subsequent units from performing risk assessments based on incorrect side data. Through the above settings, the surgical field configuration unit 101 can establish a continuous correspondence between the lateral wall location of the bladder tumor, the obturator nerve pathway, the monitoring side, the adductor longus muscle monitoring channel, and the obturator risk side, providing a clear, stable, and executable basic configuration for subsequent synchronous acquisition, interference removal, and obturator reflex risk control.

[0018] Furthermore, the surgical field configuration unit is specifically used for: Receive the surgical field coordinates established by connecting the center of the bladder neck and the left and right ureteral orifices in the cystoscopic surgical field, and record the bladder tumor edge points marked by the surgeon to generate tumor lateral wall position data. The tumor lateral wall location data is matched with the preset left obturator risk zone and right obturator risk zone to generate candidate results for risk side and the corresponding risk distance level; The monitoring side is determined based on the candidate risk side results and risk distance level. When only one obturator risk zone is hit, a unilateral adductor longus monitoring channel is generated. When both obturator risk zones are hit or the tumor edge point crosses the bladder midline, a bilateral adductor longus monitoring channel is generated. When generating the bilateral adductor longus monitoring channel, the risk side of the main obturator and the auxiliary control side are determined according to the risk distance level, and the risk side of the main obturator is configured as the risk feature extraction channel, and the auxiliary control side is configured as the common mode artifact reference channel. The unilateral or bilateral adductor longus monitoring channel is bound to the corresponding electrode port, channel side, obturator risk side, and channel purpose to generate a surgical field configuration package, which is then sent to the synchronous acquisition unit, interference removal unit, and risk control unit.

[0019] In this embodiment, the surgical field configuration unit is used to convert the location of the bladder tumor observed in the cystoscopic surgical field into lateral configuration data that the platform can recognize and call upon before the start of the electroresection operation or when repositioning is required during the operation. This allows the subsequent synchronous acquisition unit, interference stripping unit, and risk control unit to process the data based on the same set of obturator risk lateralization. Here, the cystoscopic surgical field refers to the visible area formed on the display interface after the electroresection scope or cystoscope enters the bladder cavity; the bladder neck center refers to the central position of the bladder neck in the surgical field image confirmed by the surgeon; the line connecting the left and right ureteral orifices refers to the reference line established by the platform between the left and right ureteral orifices after the surgeon identifies them under cystoscopy. Establishing surgical field coordinates using the bladder neck center and the line connecting the left and right ureteral orifices means that the platform uses the bladder neck center as a positioning reference point and the line connecting the left and right ureteral orifices as a left-right direction reference line, thereby forming coordinate references in the cystoscopic image that can be used to distinguish the left side wall, right side wall, the area near the bladder midline, and the lateral area of ​​the side wall. The surgical field coordinates do not need to reach the accuracy of three-dimensional navigation. As long as the platform can stably map the tumor location marked by the surgeon to the left obturator risk zone, the right obturator risk zone, or the area across the midline, the implementation requirements can be met.

[0020] The bladder tumor margin points marked by the surgeon refer to several location points selected by the surgeon on the visible boundary of the tumor during cystoscopy observation via a touchscreen, mouse, foot pedal confirmation button, surgical host input interface, or image annotation interface. Tumor margin points can include the left, right, upper, and lower margins of the tumor, as well as points closest to the outermost lateral wall, or can be formed by the surgeon continuously marking along the tumor contour. After receiving these tumor margin points, the surgical field configuration unit converts them into surgical field coordinates to generate tumor lateral wall position data. This data can include the position of each tumor margin point relative to the center of the bladder neck, its left-right offset direction relative to the line connecting the left and right ureteral orifices, whether the tumor margin point is close to the left or right lateral wall, and whether the tumor contour crosses the bladder midline. For example, if multiple tumor margin points are all located in the lateral region of the left lateral wall in the surgical field coordinates and are close to the preset left obturator risk zone, the surgical field configuration unit records the tumor lateral wall position data as left lateral wall related data; if the tumor margin points are distributed on both sides and cross the midline, they are recorded as bilateral or midline-crossing related data.

[0021] The left and right obturator risk zones are pre-defined risk areas on the surgical field coordinates corresponding to the areas adjacent to the obturator nerve on the left and right lateral walls of the bladder. Because the obturator nerve is located near the pelvic lateral wall, it is more susceptible to high-frequency electrosurgical energy stimulation during bladder lateral wall tumor resection. Therefore, the platform does not simply determine whether the tumor is on the left or right side, but further assesses the relationship between the tumor margin and the left or right obturator risk zone. Risk lateral candidate results refer to the side or both sides that may have obturator reflex risk after matching, such as left candidate, right candidate, or bilateral candidate. Risk distance level refers to the grading result of the proximity between the tumor lateral wall location data and the corresponding obturator risk zone, indicating the distance of the tumor margin from the obturator risk area. This level can be preset by the platform as near, intermediate, and far, or it can be set as high-risk, intermediate, and low-risk distance levels. In practice, if the tumor edge falls within the obturator risk zone, or the distance between the tumor edge and the boundary of the obturator risk zone is less than a first distance threshold, a higher risk distance level is generated; if the tumor edge is close to but does not enter the obturator risk zone, a medium risk distance level is generated; if the tumor edge is significantly far from the obturator risk zone, a lower risk distance level is generated. The distance here can be the pixel distance in the surgical field image coordinates, the normalized coordinate distance, or the relative distance calculated by the platform based on the surgeon's calibration ratio, as long as a consistent judgment standard is used within the same platform.

[0022] The surgical field configuration unit determines the monitoring side based on the candidate risk sides and risk distance levels. Monitoring side refers to the side or both sides where the adductor longus monitoring channel needs to be activated. When only the left obturator risk zone is hit, and the right obturator risk zone is not hit, the surgical field configuration unit generates a left-sided unilateral adductor longus monitoring channel; when only the right obturator risk zone is hit, a right-sided unilateral adductor longus monitoring channel is generated. When both obturator risk zones are hit, or when the tumor edge crosses the bladder midline, the surgical field configuration unit generates bilateral adductor longus monitoring channels. Here, "hitting" can be understood as the tumor edge falling within the corresponding obturator risk zone, or the distance between the tumor edge and the corresponding obturator risk zone meeting the platform's preset risk distance level condition. A tumor edge crossing the bladder midline means that the tumor edge is simultaneously distributed on both sides of the midline in the surgical field coordinates, or a portion of the tumor outline extends from one side to the other. Using this configuration method avoids missing obturator nerve-related electromyographic changes by only activating one channel when the tumor location is unclear or both sides are at risk.

[0023] When generating bilateral adductor longus muscle monitoring channels, the surgical field configuration unit further determines the primary obturator risk side and the auxiliary control side based on the risk distance level. The primary obturator risk side refers to the side that is closer to the obturator risk zone or has a higher risk distance level, as determined by the tumor lateral wall location data; the auxiliary control side refers to the side with a lower risk distance level compared to the primary obturator risk side. If the tumor edge point on the left enters the left obturator risk zone, while only a few edge points on the right are close to the midline, then the left side is determined as the primary obturator risk side, and the right side is determined as the auxiliary control side; if the risk distance levels on both sides are the same, the platform can configure both sides as obturator risk sides, or it can prompt the surgeon to confirm the primary obturator risk side based on the actual electrosurgical resection path. Configuring the primary obturator risk side as the risk feature extraction channel means that the subsequent interference stripping unit and risk control unit preferentially use the amplitude increment, burst wave density, and lateral advantage features in this channel as the basis for judging the obturator reflection risk. Configuring the auxiliary control side as a common-mode artifact reference channel means that this channel is primarily used to compare with the main obturator risk side channel to identify bilaterally occurring electrocautery artifacts, wire-induced interference, or other common-mode interference, rather than directly as the primary risk side for control assessment. In this way, the two channels do not simply acquire signals simultaneously, but are assigned different data uses, enabling continuous coordination between surgical field location determination and subsequent artifact removal and risk control.

[0024] The surgical field configuration unit then binds the unilateral or bilateral adductor longus monitoring channel with the corresponding electrode port, channel side, obturator risk side, and channel purpose to generate a surgical field configuration package. The electrode port refers to the specific input interface on the platform connected to the adductor longus monitoring electrode, such as the left and right adductor longus electrode ports. The channel side indicates whether the data acquired by this port comes from the left or right adductor longus. The obturator risk side indicates whether the platform identifies this side as a risk side for the obturator nerve reflex. The channel purpose indicates the channel's use in subsequent processing, such as a risk feature extraction channel, a common-mode artifact reference channel, or a bilateral risk monitoring channel. The surgical field configuration package may include tumor lateral wall location data, risk side candidate results, risk distance level, monitoring side, main obturator risk side, auxiliary control side, electrode port number, channel side, and channel purpose. After the surgical field configuration package is generated, it is sent to the synchronous acquisition unit, interference removal unit, and risk control unit. The synchronous acquisition unit activates the corresponding unilateral or bilateral adductor longus muscle monitoring channel according to the surgical field configuration package, and writes the channel side identification when acquiring electromyographic data; the interference stripping unit identifies which side is the risk feature extraction channel and which side is the common mode artifact reference channel according to the surgical field configuration package; the risk control unit determines the obturator risk side according to the surgical field configuration package, and uses the amplitude increment, burst wave density and side advantage feature corresponding to that side as important basis for generating the obturator reflex risk level.

[0025] The synchronous acquisition unit 102 is used to receive the energy output trigger signal of the electroresection endoscope, and establish the adductor longus baseline through the unilateral or bilateral adductor longus monitoring channel before energy output according to the energy output trigger signal of the electroresection endoscope, form a synchronous marker during energy output, set a delay detection window after energy output and acquire electromyographic data.

[0026] The synchronous acquisition unit 102 is used to establish the energy output process of the electroresection endoscope and the electromyography (EMG) acquisition process of the adductor longus muscle under the same time reference after the surgical field configuration unit 101 has generated unilateral or bilateral adductor longus muscle monitoring channels and corresponding obturator risk lateral identification. This allows the subsequent interference removal unit 103 to determine whether a certain EMG change occurs before, during, or after energy output. The electroresection endoscope energy output trigger signal here refers to a control signal that characterizes whether the electroresection endoscope is about to start outputting, is outputting, or has stopped outputting high-frequency electroresection energy. This signal comes from the output control port of the electroresection host, the foot switch status detection port, the energy output relay status port, the current sampling port, or a communication interface (such as RS-232 serial port, CAN bus, or a proprietary status push protocol based on Ethernet) open to the platform. This signal does not need to carry a specific power value; as long as the synchronous acquisition unit 102 can identify the start time, duration, and end time of the electroresection energy output, it can be used as the electroresection endoscope energy output trigger signal. In one embodiment, when the foot switch is pressed and the power cutter host confirms that output is allowed, the power cutter host sends a high-level signal to the synchronization acquisition unit 102, and the synchronization acquisition unit 102 records the rising edge of the high level as the start time of energy output; when the foot switch is released or the power cutter host stops outputting, the signal returns to a low level, and the synchronization acquisition unit 102 records the falling edge as the end time of energy output.

[0027] After receiving the trigger signal for the electroresection endoscope energy output, the synchronous acquisition unit 102 first establishes the adductor longus baseline via a unilateral or bilateral adductor longus monitoring channel before the energy output. The adductor longus baseline refers to the background electromyographic state acquired by the corresponding adductor longus monitoring channel during a relatively stable period before the current electroresection energy output stimulation occurs. It is used to represent the basic electrophysiological level of the adductor longus on that side under the current patient position, anesthesia status, electrode attachment status, and environmental interference conditions. When establishing the adductor longus baseline, the synchronous acquisition unit 102 continuously acquires electromyographic data within a preset time period before the occurrence of the electroresection endoscope energy output trigger signal, for example, acquiring data within one to three seconds before the occurrence of the trigger signal. It also uses conventional motion artifact recognition methods in the art to exclude obvious abnormal segments: when the electromyographic amplitude exceeds a preset threshold or the low-frequency components of the signal are abnormally enhanced, the corresponding segment is identified as an abnormal segment caused by patient position movement, electrode loosening, or instrument collision and is removed. The remaining stable segment is then used as the adductor longus baseline. If it is a unilateral adductor longus monitoring channel, then the adductor longus baseline on that side is established; if it is a bilateral adductor longus monitoring channel, then the left and right adductor longus baselines are established separately, and their lateral markings are retained to avoid mixing the data from both sides later. During a single surgery, the adductor longus baseline is re-established before each trigger signal of the electrosurgical energy output, or it can remain valid after the initial establishment until an abnormal status marker appears or after a preset valid duration (e.g., five minutes) is elapsed before being updated again.

[0028] During energy output, the synchronous acquisition unit 102 generates a synchronization marker. The synchronization marker is an output status identifier recorded on the electromyography (EMG) data timeline, used to indicate the start time, duration, and end time of the high-frequency energy output from the electroresection endoscope. In one embodiment, the synchronization marker is written into the acquired data using sampling point numbers combined with interval labels, enabling subsequent units to locate the output interval through these numbers. For example, if the synchronous acquisition unit 102 acquires EMG data at a frequency of 2,000 times per second, and the electroresection endoscope energy output trigger signal shows a rising edge at the 6,000th sampling point and a falling edge at the 7,200th sampling point, the synchronous acquisition unit 102 marks the period from the 6,000th to the 7,200th sampling point as the energy output period and writes the synchronization marker into the corresponding data frame. Through this synchronization marker, the subsequent interference removal unit 103 can identify which EMG data points occur in phase with the high-frequency electroresection output, thus providing a basis for identifying and removing electrocautery artifacts.

[0029] After energy output, the synchronous acquisition unit 102 sets a delay detection window and acquires electromyographic (EMG) data. The delay detection window is a time interval for EMG observation calculated from the end of the current energy output. It is used to acquire EMG data that may reflect the adductor longus response after obturator nerve stimulation. The reason for setting a delay detection window is that the EMG channel is easily affected by electrocautery artifacts during high-frequency electrosurgery, and the adductor longus response after obturator nerve stimulation may still show sudden waves or amplitude increases for a short period after the energy output ends. Therefore, a separate observation interval needs to be set after energy output. The delay detection window is preset according to the device sampling rate and clinical control needs, for example, starting 10, 20, or 50 milliseconds after the energy output ends, and lasting from 200 milliseconds to 1 second. Alternatively, the platform can automatically adjust it based on the previous acquisition results. For example, if a significant sudden wave appears within the previous delay window, the platform will advance the start time of the next delay window by 10 milliseconds and extend the duration by 100 milliseconds to more fully capture residual reflection signals. The adjustment range does not exceed the preset upper and lower limits. For example, if an electrical switching output ends at the 7,200th sampling point, with a sampling rate of 2,000 times per second, the synchronous acquisition unit 102 sets the 40th to 1,000th sampling points after the end as the delay detection window, corresponding to an electromyography observation range of approximately 20 to 500 milliseconds, and continuously acquires electromyography data within this range.

[0030] The electromyographic data output by the synchronous acquisition unit 102 should include at least the acquisition time, channel side, sampling amplitude, baseline correspondence of the adductor longus muscle, synchronization marker, and delay detection window identifier. For bilateral adductor longus muscle monitoring channels, the synchronous acquisition unit 102 should maintain time alignment of the data on both sides so that the electromyographic changes of the left and right adductor longus muscles at the same time can be compared by subsequent units. If electrode detachment, trigger signal loss, channel saturation, or data interruption occurs during the acquisition process, the synchronous acquisition unit 102 outputs an abnormal status marker and stops transmitting that segment of data as valid electromyographic data, or requires the re-establishment of the adductor longus muscle baseline before proceeding to the subsequent processing flow. In this way, the synchronous acquisition unit 102 can form a continuous data chain from the electroresection endoscope energy output trigger signal, adductor longus muscle baseline, synchronization marker, delay detection window, and electromyographic data, enabling the interference removal unit 103 to process the electromyographic data on a clear temporal basis, and also providing the risk control unit 104 with a reliable data source for subsequently generating the obturator reflex risk level.

[0031] Furthermore, the synchronous acquisition unit is specifically used for: The system receives the energy output trigger signal from the electroresection endoscope, identifies its rising and falling edges, and generates an output timing record for each continuous energy output, including the start time, end time, and output event number. Based on the output timing record, it selects stable sampling segments from the rolling buffer electromyography data before the rising edge that do not reach the upper or lower sampling limits, do not exhibit continuous unidirectional shifts, do not exceed the preset candidate mutation threshold for short-term amplitude mutations, and are not within the delay detection window range of the previous energy output. It then establishes the adductor longus baseline for the corresponding unilateral or bilateral adductor longus monitoring channel, generates an adductor longus baseline number for the baseline, and binds the baseline and baseline number to the output event number. Between the rising and falling edges, the output event number and energy output... During the output period, the identifier and channel side are written into the corresponding electromyography (EMG) data frame to form a synchronization marker. After the falling edge, a sliding window is used to detect the amplitude change of each adductor longus monitoring channel point by point. After the saturation or high-amplitude oscillation state ends, the first sampling point that meets the preset continuous sampling point condition that does not exceed the baseline fluctuation range of the corresponding adductor longus baseline is identified as the first valid sampling point. After the first valid sampling point, a preset protection sampling number positively correlated with the duration of the energy output is added to generate the starting point of the delay detection window. EMG data is collected from the starting point of the delay detection window, and the output event number, channel side condition, adductor longus baseline number, and delay detection window identifier are added to the collected EMG data so that each continuous energy output corresponds to an independent EMG analysis cycle.

[0032] In this embodiment, the synchronous acquisition unit establishes a strict one-to-one correspondence between each continuous energy output of the electroresection endoscope and the corresponding adductor longus electromyography (EMG) data. This allows the platform to clearly distinguish whether a segment of EMG data belongs to the baseline phase before energy output, the high-frequency interference phase during energy output, or the obturator nerve reflex risk observation phase after energy output. The electroresection endoscope energy output trigger signal can originate from the output control port, foot switch detection port, energy output confirmation port, or data interface connected to the electroresection endoscope, as long as the signal reflects the start and stop of high-frequency energy output by the electroresection endoscope. The rising edge refers to the moment when the trigger signal switches from the non-output state to the output state, and the falling edge refers to the moment when the trigger signal switches from the output state to the stop output state. After identifying a rising edge, the synchronous acquisition unit records it as the start time of this continuous energy output; after identifying the corresponding falling edge, it records it as the end time of this continuous energy output, and generates a unique output event number for this continuous energy output. The output event number can be an incrementing sequence number, a timestamp sequence number, or a number formed by combining the surgery number and the number of outputs. This number is used to keep the energy output, the corresponding adductor longus baseline, the synchronization marker, the delay detection window, and subsequent electromyographic data bound together, so as to avoid data confusion between multiple consecutive electroresection operations.

[0033] After generating the output timing record, the synchronous acquisition unit does not directly use any segment of EMG data before the rising edge as the baseline for the adductor longus muscle. Instead, it selects a stable sampling segment from the rolling buffer EMG data before the rising edge. The rolling buffer EMG data refers to a segment of the most recent EMG data that the synchronous acquisition unit continuously and temporarily saves before the arrival of the electroresection endoscope energy output trigger signal. For example, it can save unilateral or bilateral adductor longus muscle monitoring channel data within one, two, or three seconds before the rising edge. A stable sampling segment should simultaneously meet several conditions: it should not reach the upper or lower sampling limits, indicating that the data in this segment is not distorted due to amplifier saturation, upper or lower limits of the analog-to-digital converter; it should not show continuous unidirectional shift, indicating that the data in this segment does not continuously drift upward or downward, thus eliminating slow-varying interference caused by electrode loosening, body position movement, or changes in contact impedance; the short-term amplitude mutation amount should not exceed the preset candidate mutation threshold, indicating that the amplitude change in adjacent short time intervals does not suddenly jump, thus eliminating spikes caused by device collision, circuit disturbance, or non-target muscle activity; and the data in this segment should not be within the delayed detection window of the previous energy output, avoiding the incorrect use of residual electromyographic response or risk observation data after the previous electrocautery stimulation as the current baseline.

[0034] The preset candidate mutation threshold here can be set based on the device's sampling sensitivity, the noise level of the adductor longus monitoring channel, and the normal fluctuations before energy output. For example, before the start of surgery or before the current output event, the platform can first obtain a segment of adductor longus background data in a quiet state. If this background data usually only fluctuates slightly in a short period of time, the change value that is significantly higher than this normal short-term fluctuation amplitude can be used as the candidate mutation threshold. If one or more mutation points in a candidate segment exceed this threshold, it indicates that the segment may be subject to transient interference and is not suitable as the adductor longus baseline. For bilateral adductor longus monitoring channels, the synchronous acquisition unit selects stable sampling segments that meet the above conditions in the left and right channels respectively, and establishes corresponding adductor longus baselines respectively; for unilateral adductor longus monitoring channels, only the adductor longus baseline of that side is established. The adductor longus baseline is used to represent the basic electromyographic state of the corresponding channel before the start of this energy output, and can include the representative amplitude of the stable sampling segment, the baseline fluctuation range, and the channel side. The synchronous acquisition unit generates a long adductor baseline number for the established long adductor baseline and binds the long adductor baseline and its number to the current output event number, so that the subsequent interference stripping unit and risk control unit can determine which output event and which baseline a certain electromyographic segment should be compared with.

[0035] During the continuous output of the electroresection endoscope energy output trigger signal between the rising and falling edges, the synchronous acquisition unit writes the output event number, energy output period identifier, and channel side identification into the corresponding EMG data frame, forming a synchronization marker. An EMG data frame refers to a data unit formed by the synchronous acquisition unit according to the sampling time sequence, and may include information such as sampling time, sampling amplitude, channel number, and channel side identification. The energy output period identifier indicates that the data frame was acquired during the period when the electroresection endoscope is outputting high-frequency energy; therefore, EMG data during this period is more likely to contain electrocautery artifacts. Through the synchronization marker, the subsequent interference removal unit can identify which data overlaps with the high-frequency output of the electroresection in time, and thus identify spikes, saturation, or continuous oscillation data that occur synchronously with the start and end times of the synchronization marker.

[0036] After the falling edge, the synchronous acquisition unit does not immediately begin the delayed detection window. Instead, it uses a sliding window to detect amplitude changes in each adductor longus monitoring channel point by point. The sliding window refers to a short observation interval that moves sequentially across the electromyography sampling sequence according to the sampling points. After moving one or more sampling points, the amplitude within that observation interval is reassessed to determine if it is stable. Saturation refers to the sampling value reaching the upper or lower limit of the sampling system, resulting in the inability to fully represent the true signal changes. High-amplitude oscillation refers to the channel exhibiting rapid oscillations significantly higher than the adductor longus baseline after the electroresection output ends. This usually originates from residual high-frequency electroresection interference, wire induction, or short-term coupling interference after tissue electrocautery. After detecting saturation or high-amplitude oscillation, the synchronous acquisition unit continues to determine whether subsequent sampling points continuously fall back to the baseline fluctuation range of the corresponding adductor longus baseline. The baseline fluctuation range refers to the normal upper and lower fluctuation range determined by the adductor longus baseline bound to the current output event, used to represent the stable background level of the corresponding adductor longus when not affected by the current electroresection output.

[0037] The first valid sampling point refers to the sampling point that, after the end of a saturated or high-amplitude oscillation state, first satisfies the condition that none of the preset consecutive sampling points exceed the baseline fluctuation range of the corresponding adductor longus baseline. For example, if the preset number of consecutive sampling points is ten, and the first few sampling points after the falling edge are still significantly higher than the baseline fluctuation range, but ten consecutive sampling points starting from a certain sampling point all fall within the baseline fluctuation range of the corresponding adductor longus baseline, then the first sampling point in this continuous interval can be identified as the first valid sampling point. This setting can avoid mistakenly believing that the interference has disappeared just because a single sampling point accidentally falls into the baseline range, thereby improving the reliability of the starting point of the delay detection window.

[0038] After identifying the first valid sampling point, the synchronous acquisition unit adds a preset number of protection samples that are positively correlated with the duration of the energy output, generating the starting point of the delay detection window. Here, "positive correlation" means that the longer the duration of the continuous energy output, the more protection samples are added; conversely, the shorter the duration, the fewer protection samples are added. This is because longer-duration high-frequency electrical switching outputs are more likely to cause electromyographic channel saturation, residual oscillations, or electrical burn artifacts, thus requiring a longer protection sampling interval to minimize residual interference in the delay detection window. For example, if the duration of an energy output is short, the platform can add fewer protection samples after the first valid sampling point; if the duration is long, the platform adds more protection samples after the first valid sampling point, and then uses the sampling point after the protection sampling ends as the starting point of the delay detection window. The preset number of protection samples can be determined by a device preset table, or different numbers of samples can correspond to different duration intervals. For example, a short continuous output corresponds to a smaller number of protection samples, a medium continuous output corresponds to a medium number of protection samples, and a long continuous output corresponds to a larger number of protection samples.

[0039] Starting from the beginning of the delay detection window, the synchronous acquisition unit collects electromyographic (EMG) data and adds an output event number, channel lateralization, adductor longus baseline number, and delay detection window identifier to the collected EMG data. The delay detection window identifier indicates that the data segment belongs to the risk observation interval after the end of this energy output, and is suitable for the interference stripping unit to extract amplitude increment, burst density, and lateralization dominance features. Since each segment of delay detection window data carries an output event number and an adductor longus baseline number, subsequent processing can clearly determine that this segment of data should be compared with the adductor longus baseline established before this output, without incorrectly referencing the baseline of the previous or subsequent energy output. Through the above settings, each continuous energy output corresponds to an independent EMG analysis cycle. This independent EMG analysis cycle includes output timing records, stable sampling segments, adductor longus baseline, synchronization markers, delay detection window start point, and EMG data with event binding information. This reduces baseline contamination, artifact tailing, and signal aliasing between different output events during continuous electroresection operations, providing a clear, stable, and repeatable data basis for determining the risk level of obturator reflex.

[0040] Interference stripping unit 103 is used to identify electrocautery artifacts that appear in phase with the high-frequency output of electromyography based on synchronization markers, and to strip the electrocautery artifacts based on the long adductor baseline, so as to extract amplitude increment, burst wave density and lateral dominance features from the electromyography data within the delay detection window.

[0041] The interference removal unit 103 is used to identify and remove non-physiological interference introduced by the high-frequency output of electromyography (EMG) in the EMG data after the synchronous acquisition unit 102 has formed a synchronization mark and acquired the EMG data within the delayed detection window. It then converts the removed effective EMG changes into feature data usable by the risk control unit 104. Here, "electrocautery artifacts" refer to abnormal fluctuation signals formed in the adductor longus monitoring channel by factors such as high-frequency current, tissue electrocautery, changes in instrument contact, the conductive path of patient body fluids, and electrode wire induction when the electroretractor outputs high-frequency energy. These signals are usually highly time-synchronized with the high-frequency output of the electroretractor, manifesting as sudden high-amplitude spikes, channel saturation, continuous dense oscillations, or common-mode fluctuations occurring simultaneously in both channels. However, their origin is not the actual contraction activity of the adductor longus induced by the obturator nerve. Therefore, directly using such signals as a basis for obturator nerve reflex risk can easily lead to misjudgment.

[0042] In practice, the interference removal unit 103 first determines the time interval of the high-frequency output of electrocautery based on the synchronization marker, and then searches for wave segments in the electromyography data that appear in phase with that time interval. Phase occurrence means that the start time, duration, or peak time of a wave segment corresponds to the energy output period indicated by the synchronization marker. In one embodiment, a wave segment is considered to appear in phase when its start time is within five milliseconds before or after the start time of the synchronization marker, or when the main energy of the wave is concentrated within the coverage area of ​​the synchronization marker. For example, a sharp increase in amplitude immediately after the energy output begins, which quickly disappears after the energy output stops, or a similar peak pattern repeats during each electrocautery output. For such segments, the interference removal unit 103 marks them as suspected electrocautery artifacts and confirms them in conjunction with the adductor longus baseline. The long adductor baseline is used to represent the basic electromyographic level of the current channel when it is not affected by the current high-frequency output of the electrocautery. If a fluctuation is much higher than the long adductor baseline (for example, the fluctuation amplitude is more than five times the root mean square value of the long adductor baseline), and similar waveforms appear in the left and right channels at the same time, or if the fluctuation only appears during the energy output covered by the synchronization marker and does not continue within the delay detection window, it can be determined that it is mainly an electrocautery artifact.

[0043] When the interference removal unit 103 removes electrocautery artifacts based on the long adductor muscle baseline, it employs methods such as fragment rejection, amplitude limiting, baseline backfilling, or artifact labeling exclusion. Fragment rejection refers to excluding sampling fragments identified as electrocautery artifacts from the subsequent feature extraction range; amplitude limiting refers to limiting a peak to an acceptable range when a short-term spike significantly exceeds the allowable variation range of the long adductor muscle baseline for that channel; baseline backfilling refers to replacing transiently saturated or invalid fragments with stable values ​​near the long adductor muscle baseline of that channel; artifact labeling exclusion refers to retaining the original electromyographic data but adding invalid labels to the corresponding fragments so that the risk control unit 104 does not use the fragments when calculating the obturator reflex risk level. In one embodiment, obvious electrocautery artifacts within the synchronous label coverage area are preferentially removed by fragment rejection; short-term channel saturation fragments are backfilled by baseline; fragments with transiently excessive amplitude but identifiable morphology are amplitude-limited; and fragments whose nature is difficult to determine are excluded by artifact labeling and handed over to the risk control unit 104 for judgment on whether to include them in the calculation. In practical applications, priority is given to retaining the electromyographic data within the delayed detection window that is not marked as an electrocautery artifact, because this part of the data can better reflect the true response that the adductor longus muscle may have after electrocautery stimulation.

[0044] After completing the electrocautery artifact removal, the interference removal unit 103 extracts amplitude increment, burst density, and lateral dominance features from the electromyography (EMG) data within the delay detection window. Amplitude increment refers to the increase in effective EMG amplitude within the delay detection window relative to the baseline of the corresponding adductor longus muscle. In one embodiment, the amplitude increment is represented by the ratio of the root mean square (RMS) value of the effective EMG signal within the delay detection window to the RMS value of the baseline of the corresponding adductor longus muscle. A larger ratio indicates a larger amplitude increment. Alternatively, the difference between the maximum instantaneous amplitude within the window and the baseline mean can be used. For example, if the baseline of the adductor longus muscle on one side is at a low stable level, and EMG fluctuations significantly higher than this baseline level continuously appear within the delay detection window after electrocautery output, it indicates an abnormal enhancement trend in the adductor longus muscle on that side. The interference removal unit 103 outputs this increase as the amplitude increment. Burst density refers to the number or density of effective bursts occurring per unit time within the delay detection window. A burst wave refers to a short-duration wave that suddenly rises relative to the adductor longus baseline and exhibits electromyographic (EMG) activity. In one embodiment, a burst wave is defined as an EMG activity segment with a duration between 20 and 200 milliseconds, an ascent slope exceeding a preset threshold, and a spectral principal component located between 20 and 500 Hz. For example, a single short wave occurring within a 500-millisecond delay detection window results in a low burst wave density; conversely, multiple short waves occurring consecutively with short intervals result in a high burst wave density. Lateral dominance refers to the degree of dominance of effective EMG changes on one side relative to the other side when both adductor longus monitoring channels are simultaneously activated. It characterizes whether EMG responses are concentrated on the side corresponding to the obturator risk side. In one embodiment, lateral dominance is characterized by the ratio of the amplitude increment of the channel corresponding to the obturator risk side to the amplitude increment of the contralateral channel. When the ratio is greater than a preset multiple (e.g., twice), that side is considered dominant. For example, if the obturation risk is on the left side, the amplitude increment and burst density in the delayed detection window on the left side are significantly higher than those on the right side after electrocautery, while the right side only has background fluctuations close to the long adductor baseline. In this case, the interference stripping unit 103 outputs the left-side dominant side advantage feature. If fluctuations with similar morphology and high overlap with the synchronization marker appear on both sides at the same time, it is more likely to be common-mode interference, and the effectiveness of this side advantage feature should be reduced.

[0045] Through the above processing, the amplitude increment, burst wave density, and lateral advantage characteristics output by the interference removal unit 103 all originate from the electromyographic data within the delayed detection window after removing the electrocautery artifact, and are consistent with the corresponding obturator reflex risk laterality. This avoids mistaking the electrocautery artifact directly caused by the high-frequency output of electrocautery as a precursor to the obturator reflex, while preserving the real electromyographic changes of the adductor longus muscle after electrocautery stimulation, providing a stable, clear, and repeatable data basis for the risk control unit 104 to generate the obturator reflex risk level.

[0046] Furthermore, the interference stripping unit is specifically used for: Based on the synchronization marker, the electromyographic data is divided into energy output period segments and delayed detection window segments. The spike, saturation or continuous oscillation data that appear synchronously with the start and end times of the synchronization marker in the energy output period segment are generated as electrocautery artifact time index. The electrocautery artifact time index includes the start and end times, peak time and duration of each artifact. Based on the long adductor baseline and the electrocautery artifact time index, sampling points in the delayed detection window segment that are continuous or adjacent to the electrocautery artifact time index, whose amplitude change direction is consistent with the amplitude change direction before the end of the electrocautery artifact, and which have not fallen back to the baseline fluctuation range of the corresponding long adductor baseline are marked as invalid to obtain candidate valid electromyographic segments. When the bilateral adductor longus monitoring channel is enabled, data in the candidate effective electromyography segments with synchronously increasing obturator risk side and non-obturator risk side, peak time difference not exceeding the preset sampling interval, and waveform morphology similarity between the two sides obtained by normalized cross-correlation higher than the preset threshold are marked as common mode artifacts. The common mode artifacts are not included in subsequent feature extraction, thus obtaining effective segments of obturator risk side and effective segments of non-obturator risk side. Based on the effective fragments of the obturator risk side, the amplitude increment relative to the corresponding adductor longus baseline is calculated, and the number of burst waves that simultaneously meet the following conditions are counted: exceeding the baseline fluctuation range, duration within the preset electromyographic wave width range, rising slope exceeding the preset slope threshold, and the spectral principal component located within the preset frequency band. Burst wave density is then generated. The side advantage feature is generated based on the ratio of the amplitude increment between the effective fragments of closed-pore risk side discrimination and the effective fragments of non-closed-pore risk side discrimination, as well as the proportion of effective sampling points retained by the effective fragments of closed-pore risk side discrimination after common-mode artifact labeling.

[0047] In this embodiment, the interference stripping unit performs specialized processing on the electromyographic data output by the synchronous acquisition unit for the electroresection scenario, ensuring that the amplitude increment, burst density, and lateral advantage features used by the subsequent risk control unit are derived as much as possible from the actual electromyographic changes of the adductor longus muscle corresponding to the obturator risk lateral. Because the electroresection endoscope outputs high-frequency energy during bladder tumor resection, the adductor longus monitoring channel is prone to simultaneously receiving abnormal fluctuations caused by high-frequency current coupling, tissue electrocautery, changes in instrument contact, or lead induction. These abnormal fluctuations are referred to as electrocautery artifacts in this invention. Electrocautery artifacts typically appear suddenly near the start or end of energy output and may manifest as spikes, channel saturation, or continuous oscillations. Their temporal position is highly correlated with the synchronization marker. Therefore, the interference stripping unit first divides the electromyographic data into energy output period segments and delayed detection window segments based on the synchronization marker. The energy output period segment refers to the electromyographic data interval during which the electroresection endoscope is outputting high-frequency energy, corresponding to the synchronization marker. The delayed detection window segment refers to the electromyographic data interval used to observe the adductor longus response after the energy output ends. The interference stripping unit retrieves spikes, saturations, or continuous oscillations that occur synchronously with the start and end times of the synchronization markers within the energy output segment and generates an electrocautery artifact time index. This electrocautery artifact time index does not simply record a single artifact label, but rather records at least the start and end times, peak times, and duration of each artifact. This allows subsequent processing to determine when the artifact begins, when it reaches its maximum impact, when it ends, and whether it may have residual effects on the delayed detection window segment.

[0048] After establishing the electrocautery artifact time index, the interference stripping unit, in conjunction with the adductor longus baseline, determines which sampling points within the delayed detection window are still affected by residual electrocautery artifacts. The baseline fluctuation range refers to the permissible fluctuation interval formed during the stable acquisition phase of the adductor longus baseline before energy output, which can be determined by the normal up-and-down fluctuations of the electromyographic data within this phase. For example, if the adductor longus signal on a certain side fluctuates stably within a small range before energy output, this range can be used as the baseline fluctuation range for that side. If, after energy output, some sampling points have fallen into the delayed detection window, but their time is continuous or adjacent to the electrocautery artifact time index, and the direction of amplitude change is still consistent with that before the end of the electrocautery artifact, and they have not yet fallen back to the baseline fluctuation range of the corresponding adductor longus baseline, then these sampling points are more likely to be residual tails of the electrocautery artifact. Here, "continuous" or "adjacent" can be understood as the segment containing the sampling point immediately following the end of the electrocautery artifact, or separated from the end of the electrocautery artifact by only a very short sampling gap; "consistent amplitude change direction" means that when the signal is still in an upward or downward trend before the end of the electrocautery artifact, the corresponding sampling point in the delayed detection window segment continues to change in the same direction. The interference stripping unit marks these sampling points as invalid. The invalidated data can be retained in the original record but does not participate in the subsequent extraction of amplitude increment, burst density, and lateral dominance features. After this processing, the remaining segments that are not invalidated are considered as candidate valid electromyographic segments.

[0049] When the bilateral adductor longus monitoring channels are activated, the interference stripping unit further identifies common-mode artifacts by utilizing the correlation between the obturator risk side and the non-obturator risk side. The obturator risk side is the high-risk side determined by the surgical field configuration unit based on the location of the bladder tumor on the lateral wall, while the non-obturator risk side is the opposite side. Common-mode artifacts refer to similar fluctuations that occur almost simultaneously in the left and right adductor longus monitoring channels due to high-frequency output of electroresection or external electromagnetic interference. These fluctuations usually do not have a clear obturator risk side orientation. The interference stripping unit judges whether the obturator risk side and the non-obturator risk side increase synchronously in the candidate valid electromyography segments. If the peak time difference between the two sides does not exceed the preset sampling interval, and the similarity of the waveform morphology between the two sides obtained by normalized cross-correlation is higher than the preset threshold, then this part of the data is marked as a common-mode artifact. Normalized cross-correlation is a conventional signal comparison method used to compare whether the waveform morphology of two segments is similar. Its processing idea is to first eliminate the influence of the difference in the amplitude of the signals on the two sides on the comparison results, and then judge whether the fluctuation morphology of the two waveforms in the same time range is similar. For example, if both the left and right sides exhibit nearly identical peaks at the same time, and these peaks do not conform to the characteristic of unilateral enhancement in closure risk lateral identification, they can be considered common-mode artifacts. The preset sampling interval can be determined according to the sampling frequency; for example, under high sampling rate conditions, it can be set to a very short time corresponding to several sampling points, the purpose of which is to confirm whether the fluctuations on both sides occur approximately simultaneously. Data marked as common-mode artifacts are not included in subsequent feature extraction, thus obtaining effective fragments for closure risk lateral identification and effective fragments for non-closure risk lateral identification.

[0050] Subsequently, the interference stripping unit calculates the amplitude increment relative to the corresponding adductor longus baseline based on the effective segment of the obturator risk lateral segment. The amplitude increment represents the degree of increase of the effective electromyographic signal of the obturator risk lateral segment relative to its baseline state, and can be obtained by comparing the representative amplitude in the effective segment with the representative amplitude of the corresponding adductor longus baseline. The representative amplitude can be selected as the peak value, average effective amplitude, or short-time stable amplitude, which can be uniformly determined according to the device settings. To avoid mistaking occasional noise for burst waves, the interference stripping unit requires candidate burst waves to simultaneously meet several conditions when counting the number of burst waves: exceeding the baseline fluctuation range, duration within the preset electromyographic wave width range, rising edge slope exceeding the preset slope threshold, and the principal component of the spectrum located within the preset frequency band. The preset electromyographic wave width range is used to exclude electrocautery spikes with too short a duration and body motion interference with too long a duration; the rising edge slope is used to characterize the degree of rapid increase of the burst wave from near the baseline; the principal component of the spectrum refers to the frequency component or frequency band with the most concentrated energy in the burst wave analysis segment. For example, if multiple fluctuations exceeding the baseline fluctuation range appear within a certain delay detection window, but some of these fluctuations have extremely short durations and are similar to electrocautery artifacts, they are not counted in the burst wave count. Only fluctuations whose durations match the characteristics of electromyographic activity, whose rise is sufficiently obvious, and whose frequency distribution falls within a preset frequency band are counted as valid burst waves. Burst wave density can be understood as the density of valid burst waves appearing within the delay detection window; for example, the more valid burst waves within a delay detection window of the same length, the higher the burst wave density.

[0051] Finally, the interference stripping unit generates a lateral dominance feature based on the amplitude increment ratio between the effective segments of the obturator risk side and the effective segments of the non-obturator risk side, as well as the proportion of valid sampling points retained in the effective segments of the obturator risk side after common-mode artifact labeling. The amplitude increment ratio indicates the degree of enhancement of the obturator risk side relative to the non-obturator risk side. If the amplitude increment of the obturator risk side is significantly greater than that of the non-obturator risk side, it indicates that the electromyographic abnormalities are more concentrated on the risk side corresponding to the tumor lateral wall. The proportion of valid sampling points indicates how much data in the obturator risk side segments can still be used as valid electromyography after common-mode artifact removal. A higher proportion indicates better data reliability of the lateral dominance feature; a lower proportion indicates that the time period is heavily contaminated by common-mode artifacts, and the reliability of the lateral dominance feature should be reduced.

[0052] The risk control unit 104 is used to generate a closed-hole reflection risk level based on the amplitude increment, burst wave density, side advantage characteristics and closed-hole risk side, and to restrict the electrical switching output power, single duration and re-output interval according to the closed-hole reflection risk level. The higher the closed-hole reflection risk level, the stronger the restriction on the electrical switching output power, single duration and re-output interval.

[0053] The risk control unit 104 receives the amplitude increment, burst wave density, lateral dominance characteristics, and obturator risk laterality determined by the surgical field configuration unit 101 from the interference stripping unit 103, and converts this information into an obturator reflex risk level that can directly control the energy output behavior of the electroresection endoscope. Here, the obturator reflex risk level refers to a control parameter that grades the degree of risk of sudden contraction of the adductor longus muscle after stimulation of the obturator nerve, based on the effective electromyographic changes in the adductor longus monitoring channel corresponding to the obturator risk laterality. This risk level can be set to a safety level, a warning level, a power limiting level, and an interruption level. Alternatively, it can be set to low risk, medium risk, high risk, and extremely high risk depending on the equipment's control precision, as long as each level can clearly correspond to the limiting strength of subsequent electroresection output power, single-shot duration, and re-output interval.

[0054] In practice, the risk control unit 104 first determines the effective lateralization for the assessment. When the obturator risk lateralization output by the surgical field configuration unit 101 is left-sided, the risk control unit 104 prioritizes the amplitude increment and burst wave density corresponding to the left adductor longus monitoring channel, and combines this with the lateralization advantage characteristics of the left side relative to the right side for assessment. When the obturator risk lateralization is right-sided, the corresponding data from the right side is prioritized. When the obturator risk lateralization is bilateral, the risk control unit 104 generates preliminary risk levels for the left and right sides according to the following threshold grading method, and takes the side with the higher preliminary risk level as the basis for this control. The above lateralization processing enables the risk assessment to match the location of the bladder tumor on the lateral wall, avoiding the direct use of ordinary background fluctuations from non-risk sides as the main control basis.

[0055] The amplitude increment reflects the increase in effective EMG amplitude relative to the adductor longus baseline within the delayed detection window; the burst wave density reflects the density of effective EMG burst waves within a short period; and the lateral dominance feature reflects whether EMG abnormalities are concentrated on the side corresponding to the obturator reflex risk side. In one embodiment, the risk control unit 104 generates an obturator reflex risk level through a threshold grading method: when the amplitude increment is less than a first threshold (e.g., twice the root mean square value of the adductor longus baseline), the burst wave density is less than once per second, and the lateral dominance ratio is less than twice, a safety level is output; when the amplitude increment is between the first and second thresholds (e.g., two to five times), or the burst wave density reaches one to three times per second, a warning level is output; when the amplitude increment exceeds the second threshold, or the burst wave density exceeds three times per second and the lateral dominance ratio is greater than twice, an energy-limiting level is output; when the amplitude increment exceeds a third threshold (e.g., ten times the root mean square value of the adductor longus baseline), the burst wave density is continuously higher than five times per second, and the lateral dominance ratio is continuously greater than three times, an interruption level is output. In another embodiment, the risk control unit 104 may also generate the obturator reflex risk level by using a lookup table or a rule combination grading method. Those skilled in the art can achieve this by using a preset threshold table or preoperative calibration parameters, depending on the different electrosurgical unit, sampling sensitivity, and electrode arrangement. The preoperative calibration parameters are obtained by collecting the resting electromyographic signal of the adductor longus muscle of the patient before the start of electrosurgical resection, recording its root mean square value as the individualized baseline of the adductor longus muscle of the patient, and converting the judgment threshold of each risk level by multiples of the root mean square value of the baseline.

[0056] After generating a risk level for obturator reflex, the risk control unit 104 limits the electrosurgical output power, single-cycle duration, and re-output interval according to the risk level. Electrosurgical output power refers to the energy intensity parameter output by the electrosurgical unit to the electrodes during tissue cutting or coagulation; single-cycle duration refers to the maximum allowed duration of a single continuous energy output; and re-output interval refers to the shortest waiting time between the end of the current output and the next allowed output. The linkage limitation means that the risk control unit 104 simultaneously constrains these three parameters, rather than adjusting only one parameter individually. Specifically, for electrosurgical output power, a higher limit corresponds to a lower allowed power upper limit; for single-cycle duration, a higher limit corresponds to a shorter allowed continuous output duration; and for re-output interval, a higher limit corresponds to a longer required waiting time between two outputs. This allows for simultaneous reduction of stimulation intensity, shortening of continuous stimulation time, and extension of tissue and nerve recovery observation time when the risk of obturator reflex increases, thereby improving control stability.

[0057] In one implementation, the limiting parameters corresponding to each risk level are as follows: When the obturator reflex risk level is at the safe level, the risk control unit 104 allows the electrosurgical endoscope to operate according to the power (e.g., 100% of the set value) and output rhythm set by the surgeon, without additional restrictions on the duration of a single output or the interval between outputs; when the risk level rises to the warning level, the risk control unit 104 limits the power to 80% of the surgeon's set value, the duration of a single output to three seconds, and the interval between outputs to at least one second; when the risk level rises to the energy-limiting level, the risk control unit 104 limits the power to 50% of the surgeon's set value, the duration of a single output to one second, and the interval between outputs to at least three seconds; when the risk level reaches the interruption level, the risk control unit 104 sends a power-off or output-prohibition command to the electrosurgical endoscope, causing the electrosurgical endoscope to pause energy output until the recovery verification unit 105 confirms that the risk level has fallen back and meets the safety conditions. Therefore, the higher the obturator reflex risk level, the stronger the restrictions on the electrosurgical output power, duration of a single output, and interval between outputs, and the control actions have a clear progressive relationship.

[0058] The risk control unit 104 implements the aforementioned restrictions through communication interfaces, power control interfaces, relay control interfaces, or foot pedal output interception interfaces with the electroresection host. When the electroresection host supports digital communication, the risk control unit 104 directly issues the target power, maximum duration, and minimum re-output interval. When the electroresection host only supports external enable control, the risk control unit 104 achieves equivalent control through allow output, time-limited output, and prohibit output signals. If the risk control unit 104 detects missing data, unclear obturator risk lateralization, abnormal electromyographic channels, or invalid interference stripping results, it treats the current state as a conservative risk state, which is equivalent to an energy-limiting level. The output of the electroresection endoscope is constrained according to the limiting parameters corresponding to the energy-limiting level until the data state returns to normal or the risk level is redefined, to avoid continuing to maintain high-intensity continuous output when the risk cannot be reliably determined. In this way, the risk control unit 104 can convert electromyographic characteristics and obturator risk lateralization into specific, executable electroresection energy control actions, enabling the platform to form a closed-loop control basis from risk identification to energy constraint during bladder tumor electroresection.

[0059] Furthermore, the risk control unit is specifically used for: Based on the closed-hole risk side, risk feature groups corresponding to the closed-hole risk side are extracted from amplitude increment, burst wave density, and side-discrimination advantage characteristics, and the features corresponding to the non-closed-hole risk side are used as control features. Based on risk feature groups and control features, a risk determination vector is generated that includes amplitude anomaly markers, burst density markers, side pointing markers, and continuous event enhancement markers. The continuous event enhancement marker is used to indicate that the amplitude increment or burst wave density of the closed-hole risk side in the delay detection window corresponding to multiple adjacent output event numbers increases continuously. The risk level of closed-pore reflection is determined based on the risk judgment vector. Among them, an early warning level is generated when the amplitude anomaly marker is established alone, an energy limitation level is generated when the amplitude anomaly marker and the sudden dense marker are established at the same time, and an interruption level is generated when the amplitude anomaly marker, the sudden dense marker, the lateral pointing marker, and the continuous event enhancement marker are established at the same time. An energy limiting parameter set is generated based on the closed-hole reflection risk level. The energy limiting parameter set includes a power limit value, a single duration upper limit, and a re-output interval lower limit. When the continuous event enhancement flag is set, the power limit value is further reduced, the single duration upper limit is shortened, and the re-output interval lower limit is extended. The energy limiting parameter set is sent to the energy output control terminal of the electrosurgical cutting mirror to limit the electrosurgical cutting output power, single-cycle duration, and re-output interval in a coordinated manner.

[0060] In this embodiment, the risk control unit converts the amplitude increment, burst density, and lateral dominance characteristics output by the interference stripping unit into control parameters that can directly constrain the energy output of the electroresection endoscope. Since the location of the bladder tumor on the lateral wall has been determined as the obturator risk side by the surgical field configuration unit, the risk control unit first extracts the corresponding risk feature set according to the obturator risk side. Here, the risk feature set refers to the combination of amplitude increment, burst density, and lateral dominance characteristics corresponding to the obturator risk side, used to represent the degree of abnormal electromyographic activity of the adductor longus muscle on that side after this or several adjacent electroresection outputs. Features corresponding to non-obturator risk sides serve as control features to determine whether the current abnormality has a clear obturator risk side orientation. For example, if the tumor is located on the left wall of the bladder and the left side is identified as the obturator risk side, then the amplitude increment, burst density, and side advantage characteristics corresponding to the effective segment of the left adductor longus muscle constitute the risk feature group, while the amplitude increment and burst density corresponding to the effective segment of the right adductor longus muscle serve as control characteristics; if both sides are identified as obturator risk sides, then the risk control unit can form a left risk feature group and a right risk feature group respectively, and use the side with the higher risk level as the basis for energy limitation control.

[0061] After obtaining the risk feature group and control features, the risk control unit generates a risk determination vector. This risk determination vector is not an abstract mathematical model, but rather structured control data used to record whether multiple judgment markers are valid. It includes at least amplitude anomaly markers, burst density markers, lateral orientation markers, and continuous event enhancement markers. The amplitude anomaly marker indicates that the amplitude increment of the obturator risk lateral has exceeded the safe variation range allowed by the corresponding adductor longus baseline. Specifically, if the effective EMG amplitude of the obturator risk lateral within the delay detection window is significantly higher than the adductor longus baseline bound to the current output event, and this increase is not caused by electrocautery artifacts or common-mode artifacts, the risk control unit establishes the amplitude anomaly marker. The burst density marker indicates that the number of effective burst waves appearing on the obturator risk lateral within the delay detection window reaches a level requiring control of energy output. For example, if multiple burst waves appear within the same delay detection window that meet preset EMG wave width, rise slope, and spectral principal component conditions, then the burst density marker is established. Lateral pointing markers are used to indicate that abnormal electromyographic activity is mainly concentrated on the obturator risk side, rather than ordinary interference that occurs simultaneously on both sides; for example, if the amplitude increment of the obturator risk side is significantly greater than that of the non-obturator risk side, and the lateral dominance feature shows that the obturator risk side is dominant, then the lateral pointing marker is valid.

[0062] The continuous event enhancement flag indicates that the risk is not a single, accidental fluctuation, but rather exhibits a continuous enhancement trend within the delay detection windows corresponding to multiple adjacent output event numbers. These adjacent output event numbers refer to the consecutive numbers generated by the synchronous acquisition unit for several consecutive electrical switching outputs, such as the fifteenth, sixteenth, and seventeenth output events. When comparing the delay detection windows corresponding to these output events, the risk control unit can establish the continuous event enhancement flag if it finds that the amplitude increment of the closure risk side increases successively, or the burst wave density increases successively, or at least one of the two continuously increases. For example, if a certain left-side closure risk side shows only a slight amplitude increase in the delay detection window after three adjacent electrical switching outputs, a higher amplitude with a few burst waves in the second, and an even higher amplitude and denser burst waves in the third, it indicates that continuing to output with the original parameters may further induce closure reflection risk, and the risk control unit can set the continuous event enhancement flag to active. If the amplitude increment after any output has decreased and the burst wave density has not continued to increase, it is not considered a continuous event enhancement.

[0063] The risk control unit determines the obturator reflex risk level based on the risk assessment vector. This level can be divided into a warning level, an energy-limiting level, and an interruption level, or more levels can be set in the device, but each level should have a clear correspondence with subsequent energy-limiting actions. When only the amplitude abnormality marker is established, while the sudden dense marker, lateral pointing marker, and continuous event enhancement marker are not yet established, it indicates that there is an initial increase in electromyography (EMG) on the obturator risk side, but it cannot yet be identified as a strong precursor to the obturator reflex; in this case, a warning level is generated. When both the amplitude abnormality marker and the sudden dense marker are established, it indicates that not only is the amplitude increased, but effective burst waves have also appeared densely; the risk control unit generates an energy-limiting level. When the amplitude abnormality marker, the sudden dense marker, the lateral pointing marker, and the continuous event enhancement marker are established simultaneously, it indicates that abnormal EMG activity is concentrated on the obturator risk side and shows an increasing trend in multiple adjacent output events; in this case, an interruption level is generated. Through the above-mentioned classification method, the risk control unit avoids relying solely on a single threshold for rough shutdown. Instead, it incorporates amplitude, sudden wave, lateral orientation, and enhancement trend in consecutive transurethral resection events into the judgment, making the obturator reflex risk level more consistent with the actual risk changes of transurethral resection of bladder tumors.

[0064] After determining the risk level of the obturator reflex, the risk control unit generates a set of energy limiting parameters. These parameters include a power limit, an upper limit for single-session duration, and a lower limit for the re-output interval. The power limit restricts the maximum allowed output power of the electroretractor in the next or current cycle, ensuring the output power does not exceed this limit. The upper limit for single-session duration restricts the maximum allowed duration of a single continuous energy output. The lower limit for the re-output interval restricts the minimum waiting time between two consecutive energy outputs. These three parameters work together to simultaneously reduce the electrical stimulation intensity, shorten the continuous stimulation time, and prolong the observation and recovery time of the obturator nerve-related electromyographic state. For example, at the warning level, the risk control unit can primarily shorten the upper limit for single-session duration and slightly reduce the power limit; at the energy-limiting level, it can significantly reduce the power limit, shorten the upper limit for single-session duration, and prolong the lower limit for the re-output interval; at the interruption level, the power limit can be set to disallow output, or a prohibition status can be sent to the electroretractor's energy output control terminal until the recovery verification unit confirms that the recovery conditions are met.

[0065] When the continuous event enhancement flag is triggered, the risk control unit further increases the limiting intensity. This further reduction in power limit value means setting the maximum allowable output power lower than when no continuous enhancement trend is observed, under the same pore reflection risk level; further shortening the upper limit of single-output duration means reducing the allowable duration of each continuous output; and further extending the lower limit of the re-output interval means requiring a longer waiting time between adjacent outputs. The reason for this setting is that the continuous event enhancement flag indicates that the risk has accumulated or worsened across multiple output events. If a standard energy limiting strategy is still used, it may not be able to suppress the pore reflection risk in time. Therefore, the risk control unit tightens the energy limiting parameter set a second time based on this flag, so that the control action can be enhanced in response to the increasing risk trend during continuous electrical switching operations.

[0066] Finally, the risk control unit sends the energy limiting parameter set to the electroresection endoscope's energy output control terminal. The electroresection endoscope's energy output control terminal can be the digital control interface, power setting interface, output enable interface, foot signal interception interface, or other interfaces that can affect the electroresection output state of the electroresection host. When the electroresection host supports parameter control, the risk control unit can directly send the power limit value, the upper limit of the single-session duration, and the lower limit of the re-output interval; when the electroresection host only supports output enable or disable control, the risk control unit can achieve equivalent limitation by controlling the output enable time and output interval. In this way, the risk control unit can translate the abnormal electromyographic changes on the obturator risk side into specific electroresection energy control actions, making the electroresection output power, single-session duration, and re-output interval interconnected and limiting, thereby reducing the risk of obturator nerve reflex induction while maintaining the controllability of the electroresection operation.

[0067] The recovery verification unit 105 is used to continue to detect the risk level decline state based on the electromyographic data within the delay detection window after the risk control unit limits the electrical cutting output power, single duration and re-output interval, and allows the risk control unit to restore the set output only when the safety conditions are met in multiple consecutive delay detection windows.

[0068] The recovery verification unit 105 is used to continuously confirm whether the risk of occluded reflection has truly decreased after the risk control unit 104 has limited the output power, single duration, and re-output interval of the electrical switching operation. This prevents the platform from prematurely resuming the set output before the electromyographic abnormality has stabilized and disappeared. Here, the risk level decrease refers to the risk level of occluded reflection determined by the risk control unit 104 based on the amplitude increment, burst wave density, lateral dominance characteristics, and occluded risk lateral identification continuously output by the interference stripping unit 103, gradually decreasing from a high-risk state to a safe state that allows continued electrical switching operation. For example, if the previous delay detection window determined it to be at the power-limited or interruption level, and subsequent delay detection windows show a significant decrease in amplitude increment, a decrease in burst wave density, and the occluded risk lateral identification channel no longer continuously dominating, then the risk level can be considered to have a decreasing trend. If this trend remains stable across multiple consecutive delay detection windows, then the risk level can be further considered to have decreased to within the safe range.

[0069] The recovery verification unit 105 receives electromyographic data within the delay detection window formed by the synchronous acquisition unit 102 after each energy output, and uses the valid data after the interference removal unit 103 removes electrocautery artifacts as the verification basis. Multiple consecutive delay detection windows refer to two or more delay detection windows that are temporally adjacent, generated by the synchronous acquisition unit 102, and whose data is valid. In one embodiment, when the risk level is at the energy-limited level, the platform allows the surgeon to continue low-power electrocautery output according to the limiting parameters corresponding to the energy-limited level, forming a delay detection window after each output; when the risk level is at the interruption level, the platform prohibits energy output, and the synchronous acquisition unit 102 continuously acquires electromyographic data at a preset rhythm (e.g., once per second) to form an observation window, but does not output energy to the electrocautery electrode. If three consecutive delay detection windows all show a risk level at the safe level or low risk level, the continuous verification requirement can be considered met. The number of multiple consecutive delay detection windows can be preset by the platform or configured according to the degree of surgical risk, usually set to two, three, or more windows to reduce the possibility of false recovery caused by a single accidental fluctuation.

[0070] Safety conditions refer to the electromyographic stability requirements and channel status requirements that must be met simultaneously before the risk control unit 104 can resume its set output. In one embodiment, the electromyographic stability requirements include: the amplitude increment corresponding to the obturator risk lateral is less than the first threshold corresponding to the safety level in the risk control unit 104 (e.g., twice the root mean square value of the adductor longus baseline), the burst wave density is less than once per second, the lateral dominance ratio is less than twice, and there is no recurrence of continuous abnormal electromyographic activity related to the obturator reflex. The channel status requirements include: the unilateral or bilateral adductor longus monitoring channel is connected normally, the electromyographic data does not show saturation, dropout, long-term interruption, or strong interference that cannot be peeled off, and the data volume of the delay detection window meets the judgment requirements; where long-term interruption refers to the cumulative interruption time of the electromyographic data in a single delay detection window exceeding 10% of the duration of that window (e.g., when the delay detection window lasts for 500 milliseconds, the cumulative interruption time exceeds 50 milliseconds). If any of these conditions are not met, the recovery verification unit 105 continues to maintain the power-limited or power-off state and outputs a verification result indicating that recovery is not allowed to proceed to the risk control unit 104.

[0071] For example, during a certain electromyography (EMG) procedure, the risk control unit 104 pauses energy output due to a high amplitude increase and dense burst waves on the left obturator risk side. The recovery verification unit 105 resumes acquiring EMG data within the delayed detection window of the left adductor longus monitoring channel after the pause. In the first delayed detection window, short burst waves still exist on the left side, although the amplitude has decreased, but the safety conditions are not yet met. In the second delayed detection window, the burst waves are significantly reduced, but the left side is still higher than the right side. In the third, fourth, and fifth delayed detection windows, the amplitude increase on the left side is close to the adductor longus baseline, the burst wave density remains low, the difference between the left and right sides is no longer significant, and the channel data is continuous and valid. At this point, the recovery verification unit 105 determines that the safety conditions are met in multiple consecutive delayed detection windows and sends a verification result to the risk control unit 104 allowing the resumption of the set output.

[0072] Resumption of the set output means allowing the risk control unit 104 to remove the previous restrictions on the electrosurgical output power, single-shot duration, and re-output interval, returning the electrosurgical endoscope to the operating parameters preset by the surgeon or currently set by the platform. To ensure safety, the recovery verification unit 105 implements the recovery process in a progressive manner: in one embodiment, the progressive recovery uses a two-stage transition. The first stage restores the output to 50% of the surgeon's set power and limits the single-shot duration to one second (equivalent to the power-limiting level). After two consecutive delay detection windows continue to meet safety conditions, the second stage restores the output to 80% of the surgeon's set power (equivalent to the warning level). After the same number of verification windows, the output is finally restored to the complete set output. If an increase in risk level is detected again during the recovery process, the recovery verification unit 105 immediately stops the recovery permission and determines the fallback state based on the increased risk level: if it increases again to the warning level or power-limiting level, the risk control unit 104 enters the corresponding power-limiting control state; if it increases again to the interruption level, the risk control unit 104 enters the power-off control state. In this way, the recovery verification unit 105 enables the platform to have verifiable recovery conditions after limiting the output, avoiding premature recovery of electrocautery energy output due to the brief stability of a single window, thereby improving the continuity, reliability and surgical safety of obturator nerve electrophysiological feedback control.

[0073] Furthermore, the recovery verification unit is specifically used for: After the risk control unit performs power limiting or power shutdown, the amplitude increment, burst wave density and side dominance characteristics in the subsequent delay detection window are continuously read according to the output event number to generate a risk fallback sequence. Based on the risk fallback sequence, select windows that have at least three consecutive delayed detection windows that meet the following criteria: amplitude increment does not exceed the baseline fluctuation range of the corresponding adductor longus baseline, burst wave density does not exceed the preset low-risk density, and lateral dominance features do not point to the obturator risk side, and generate recovery candidate markers. Based on the recovery candidate flag and the current energy limit parameter set, a trial output parameter set is generated, wherein the power limit value of the trial output parameter set is lower than the set output power and the single duration is shorter than the set duration. After executing the trial output according to the trial output parameter group, the delay detection window corresponding to the output event number is re-acquired and the verification fallback result is generated; When the verification fallback result still meets the conditions corresponding to the recovery candidate flag, a recovery permission flag is generated and the risk control unit is allowed to resume the set output; otherwise, the recovery candidate flag is revoked and the power switching output is restricted by at least one of reducing the power limit value, shortening the upper limit of the single duration, or extending the lower limit of the re-output interval.

[0074] In this embodiment, the recovery verification unit is used to confirm whether the risk of obturator reflex has stabilized and decreased after the risk control unit has limited or disabled the electromyography output power, single duration, and re-output interval. It prevents the electromyography endoscope from directly reverting to the set output before confirming safety. Here, "limiting power" means that the risk control unit has not completely prohibited energy output, but has limited at least one of the power limit, single duration upper limit, or re-output interval lower limit. "Disabling power" means that the risk control unit temporarily prohibits the electromyography endoscope from continuing to output electromyography energy. The recovery verification unit does not use the recovery of a single electromyography signal as the basis for recovery. Instead, it uses the output event number and delay detection window formed by the synchronous acquisition unit to compare multiple consecutive observations after limiting or disabling power, thereby avoiding premature restoration of the set output due to accidental fluctuations, brief periods of stability, or temporary disappearance of electrocautery artifacts.

[0075] In practice, after the risk control unit performs power limiting or shutdown, the recovery verification unit continuously reads the amplitude increment, burst density, and lateral dominance characteristics within subsequent delay detection windows according to the order of the output event numbers, and generates a risk fallback sequence. The risk fallback sequence refers to multiple delay detection window risk status records arranged chronologically. Each record includes at least the output event number corresponding to that window, the amplitude increment of the closure risk lateral dominance, the burst density, and the lateral dominance characteristics. For example, if the risk control unit enters a power limiting state after the twelfth output event, the recovery verification unit can sequentially read the delay detection windows corresponding to the thirteenth, fourteenth, and fifteenth output events, or, while in a shutdown observation state, read the continuous delay detection windows formed by the platform's continued acquisition, and arrange the risk characteristics of each window in sequence to form a risk fallback sequence. Through this sequence, the recovery verification unit can determine whether the risk is continuously decreasing, fluctuating repeatedly, or increasing again.

[0076] The recovery verification unit filters whether at least three consecutive delayed detection windows meet safety conditions based on the risk decline sequence. Here, "at least three consecutive delayed detection windows" refers to three or more delayed detection windows that are temporally adjacent and have valid data; intermediate abnormal windows cannot be skipped in favor of splicing scattered windows. The safety conditions include: amplitude increment not exceeding the baseline fluctuation range of the corresponding adductor longus baseline, indicating that the EMG amplitude of the obturator risk side has fallen back to near the baseline EMG level of the current or corresponding event; burst wave density not exceeding the preset low-risk density, indicating that no dense, valid burst waves appear within the delayed detection window; and lateral dominance characteristics not pointing to the obturator risk side, indicating that the current EMG change does not continue to show unilateral dominance on the obturator risk side. The preset low-risk density can be understood as the upper limit of the number or density of low-level burst waves allowed within a delayed detection window. This upper limit can be determined by the device's factory parameters, background EMG activity in a preoperative stable state, or physician configuration. For example, within a delayed detection window, if only a very small number of brief burst waves with low amplitude appear and no unilateral enhancement is formed on the closed-hole risk side, it can be determined that the burst wave density does not exceed the preset low-risk density; if multiple valid burst waves appear consecutively, this condition is not met.

[0077] When at least three consecutive delayed detection windows in the risk fallback sequence meet the above conditions, the recovery verification unit generates a recovery candidate marker. A recovery candidate marker indicates that the current electromyographic state is ready to attempt to restore output, but this does not equate to immediately restoring the set output. The recovery verification unit then generates a trial output parameter set based on the recovery candidate marker and the current energy limit parameter set. The current energy limit parameter set includes the power limit value, single-output duration upper limit, and re-output interval lower limit currently being executed by the risk control unit. The trial output parameter set is a low-intensity verification output parameter set generated based on these limits. Its power limit value is lower than the output power originally set by the surgeon or system, and its single-output duration is shorter than the set duration. The purpose of this design is to verify whether the obturator risk lateralization remains stable through a controlled low-intensity output, rather than directly restoring the complete electroresection output after the conditions are met solely through static observation. For example, if the set output power is a certain conventional electroresection power and the set duration is the surgeon-allowed continuous output time, the trial output parameter set can select a lower power and a shorter duration, used only to observe whether this low-intensity output re-induces obturator risk lateralization electromyographic abnormalities.

[0078] After the trial output is executed according to the trial output parameter set, the synchronous acquisition unit will regenerate the output event number and delay detection window for this trial output. The interference removal unit continues to extract the amplitude increment, burst wave density, and lateral dominance feature from the delay detection window, and the recovery verification unit generates the verification fallback result accordingly. The verification fallback result refers to the judgment result of whether the electromyographic state of the obturator risk side still meets the corresponding conditions of the recovery candidate marker after the trial output. If, after the trial output, the amplitude increment in the corresponding delay detection window still does not exceed the baseline fluctuation range of the adductor longus baseline, the burst wave density still does not exceed the preset low-risk density, and the lateral dominance feature still does not point to the obturator risk side, then the verification fallback result is passed, the recovery verification unit generates a recovery permission marker, and allows the risk control unit to resume the set output. Here, resuming the set output means removing the previous power limiting or power stopping state, so that the electroresection endoscope can work according to the set output power, set duration, and set output interval; in a more conservative implementation, it can also be restored to an intermediate state close to the set output first, and then fully restored after the conditions are met in subsequent delay detection windows.

[0079] If the verification results no longer meet the conditions corresponding to the recovery candidate marker, such as the amplitude increment of the obturator risk side exceeding the baseline fluctuation range again after the trial output, the burst wave density increasing, or the lateral dominance feature reverting to the obturator risk side, the recovery verification unit cancels the recovery candidate marker and notifies the risk control unit to continue limiting the electroresection output. Continuing to limit the electroresection output can be achieved by reducing the power limit value, shortening the upper limit of the single-output duration, or extending the lower limit of the re-output interval. For example, if only a slight amplitude increase occurs after the trial output, the lower limit of the re-output interval can be extended to allow the platform a longer observation time; if a burst wave density increase occurs simultaneously, the upper limit of the single-output duration can be further shortened; if the obturator risk side shows a significant lateral dominance again, the power limit value can be reduced, or even the system can re-enter the shutdown state. Through the continuous process of recovery candidate, trial output, and re-verification, the recovery verification unit provides a clear verification basis for the platform's recovery action after risk decline, avoiding a direct jump from the power-limited or shutdown state back to the set output, thereby improving the stability and safety of the obturator neurophysiological feedback control during bladder tumor electroresection.

[0080] Although this application discloses preferred embodiments as described above, it is not intended to limit this application. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of this application. Therefore, the scope of protection of this application should be determined by the scope defined in the claims of this application.

Claims

1. A bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback, characterized in that, include: The surgical field configuration unit is used to determine the monitoring laterality related to the course of the obturator nerve based on the location of the bladder tumor on the lateral wall, and to generate unilateral or bilateral adductor longus monitoring channels and corresponding obturator risk lateralities based on the monitoring laterality. The synchronous acquisition unit is used to receive the trigger signal of the electroresection endoscope energy output, and establish the adductor longus baseline through the unilateral or bilateral adductor longus monitoring channel before energy output based on the trigger signal of the electroresection endoscope energy output, form a synchronous marker during energy output, and set a delayed detection window and acquire electromyographic data after energy output. The interference stripping unit is used to identify electrocautery artifacts that appear in phase with the high-frequency output of electromyography based on the synchronization marker, and to strip the electrocautery artifacts based on the long adductor baseline, so as to extract amplitude increment, burst wave density and lateral dominance features from the electromyography data within the delay detection window. The risk control unit is used to generate the pore reflection risk level based on the amplitude increment, burst wave density, side advantage characteristics and pore risk side, and to limit the electrical switching output power, single duration and re-output interval according to the pore reflection risk level. The recovery verification unit is used to continue to detect the risk level decline state based on the electromyography data within the delay detection window after the risk control unit limits the electrical switching output power, single duration and re-output interval. It allows the risk control unit to resume the set output only when the safety conditions are met in multiple consecutive delay detection windows.

2. The bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback according to claim 1, characterized in that, The interference stripping unit is specifically used for: Based on the synchronization marker, the electromyographic data is divided into energy output period segments and delayed detection window segments. The spike, saturation or continuous oscillation data that appear synchronously with the start and end times of the synchronization marker in the energy output period segment are generated as electrocautery artifact time index. The electrocautery artifact time index includes the start and end times, peak time and duration of each artifact. Based on the long adductor baseline and the electrocautery artifact time index, sampling points in the delayed detection window segment that are continuous or adjacent to the electrocautery artifact time index, whose amplitude change direction is consistent with the amplitude change direction before the end of the electrocautery artifact, and which have not fallen back to the baseline fluctuation range of the corresponding long adductor baseline are marked as invalid to obtain candidate valid electromyographic segments. When the bilateral adductor longus monitoring channel is enabled, data in the candidate effective electromyography segments with synchronously increasing obturator risk side and non-obturator risk side, peak time difference not exceeding the preset sampling interval, and waveform morphology similarity between the two sides obtained by normalized cross-correlation higher than the preset threshold are marked as common mode artifacts. The common mode artifacts are not included in subsequent feature extraction, thus obtaining effective segments of obturator risk side and effective segments of non-obturator risk side. Based on the effective fragments of the obturator risk side, the amplitude increment relative to the corresponding adductor longus baseline is calculated, and the number of burst waves that simultaneously meet the following conditions are counted: exceeding the baseline fluctuation range, duration within the preset electromyographic wave width range, rising slope exceeding the preset slope threshold, and the spectral principal component located within the preset frequency band. Burst wave density is then generated. The side advantage feature is generated based on the ratio of the amplitude increment between the effective fragments of closed-pore risk side discrimination and the effective fragments of non-closed-pore risk side discrimination, as well as the proportion of effective sampling points retained by the effective fragments of closed-pore risk side discrimination after common-mode artifact labeling.

3. The bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback according to claim 1, characterized in that, The synchronous acquisition unit is specifically used for: Receives the energy output trigger signal of the electrosurgical resection mirror, identifies its rising and falling edges, and generates an output timing record containing the start time, end time and output event number for each continuous energy output; Based on the output timing record, a stable sampling segment is selected from the rolling buffer electromyography data before the rising edge. The segment does not reach the upper or lower limit of sampling, does not show continuous same-direction offset, the short-term amplitude mutation amount does not exceed the preset candidate mutation threshold, and is not within the delay detection window of the previous energy output. The long adductor muscle baseline corresponding to the unilateral or bilateral adductor longus monitoring channel is established. A long adductor muscle baseline number is generated for the long adductor muscle baseline, and the long adductor muscle baseline and the long adductor muscle baseline number are bound to the output event number. Between the rising edge and the falling edge, the output event number, energy output period identifier, and channel side are written into the corresponding electromyographic data frame to form a synchronization marker; After the falling edge, a sliding window is used to detect the amplitude change of each long adductor muscle monitoring channel point by point. After the saturation or high amplitude oscillation state ends, the sampling point that first satisfies the preset continuous sampling point that does not exceed the baseline fluctuation range of the corresponding long adductor muscle baseline is identified as the first valid sampling point. After the first valid sampling point, a preset protection sampling number positively correlated with the duration of the energy output is added to generate the starting point of the delay detection window. Electromyography (EMG) data is collected starting from the beginning of the delayed detection window, and the collected EMG data is appended with the output event number, channel side, adductor longus baseline number, and delayed detection window identifier, so that each continuous energy output corresponds to an independent EMG analysis cycle.

4. The bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback according to claim 1, characterized in that, The risk control unit is specifically used for: Based on the closed-hole risk side, risk feature groups corresponding to the closed-hole risk side are extracted from amplitude increment, burst wave density, and side-discrimination advantage characteristics, and the features corresponding to the non-closed-hole risk side are used as control features. Based on risk feature groups and control features, a risk determination vector is generated that includes amplitude anomaly markers, burst density markers, side pointing markers, and continuous event enhancement markers. The continuous event enhancement marker is used to indicate that the amplitude increment or burst wave density of the closed-hole risk side in the delay detection window corresponding to multiple adjacent output event numbers increases continuously. The risk level of closed-pore reflection is determined based on the risk judgment vector. Among them, an early warning level is generated when the amplitude anomaly marker is established alone, an energy limitation level is generated when the amplitude anomaly marker and the sudden dense marker are established at the same time, and an interruption level is generated when the amplitude anomaly marker, the sudden dense marker, the lateral pointing marker, and the continuous event enhancement marker are established at the same time. An energy limiting parameter set is generated based on the closed-hole reflection risk level. The energy limiting parameter set includes a power limit value, a single duration upper limit, and a re-output interval lower limit. When the continuous event enhancement flag is set, the power limit value is further reduced, the single duration upper limit is shortened, and the re-output interval lower limit is extended. The energy limiting parameter set is sent to the energy output control terminal of the electrosurgical cutting mirror to limit the electrosurgical cutting output power, single-cycle duration, and re-output interval in a coordinated manner.

5. The bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback according to claim 1, characterized in that, The recovery verification unit is specifically used for: After the risk control unit performs power limiting or power shutdown, the amplitude increment, burst wave density and side dominance characteristics in the subsequent delay detection window are continuously read according to the output event number to generate a risk fallback sequence. Based on the risk fallback sequence, select windows that have at least three consecutive delayed detection windows that meet the following criteria: amplitude increment does not exceed the baseline fluctuation range of the corresponding adductor longus baseline, burst wave density does not exceed the preset low-risk density, and lateral dominance features do not point to the obturator risk side, and generate recovery candidate markers. Based on the recovery candidate flag and the current energy limit parameter set, a trial output parameter set is generated, wherein the power limit value of the trial output parameter set is lower than the set output power and the single duration is shorter than the set duration. After executing the trial output according to the trial output parameter group, the delay detection window corresponding to the output event number is re-acquired and the verification fallback result is generated; When the verification fallback result still meets the conditions corresponding to the recovery candidate flag, a recovery permission flag is generated and the risk control unit is allowed to resume the set output; otherwise, the recovery candidate flag is revoked and the power switching output is restricted by at least one of reducing the power limit value, shortening the upper limit of the single duration, or extending the lower limit of the re-output interval.

6. The bladder tumor electroresection energy control platform based on obturator nerve electrophysiological feedback according to claim 1, characterized in that, The surgical field configuration unit is specifically used for: Receive the surgical field coordinates established by connecting the center of the bladder neck and the left and right ureteral orifices in the cystoscopic surgical field, and record the bladder tumor edge points marked by the surgeon to generate tumor lateral wall position data. The tumor lateral wall location data is matched with the preset left obturator risk zone and right obturator risk zone to generate candidate results for risk side and the corresponding risk distance level; The monitoring side is determined based on the candidate risk side results and risk distance level. When only one obturator risk zone is hit, a unilateral adductor longus monitoring channel is generated. When both obturator risk zones are hit or the tumor edge point crosses the bladder midline, a bilateral adductor longus monitoring channel is generated. When generating the bilateral adductor longus monitoring channel, the risk side of the main obturator and the auxiliary control side are determined according to the risk distance level, and the risk side of the main obturator is configured as the risk feature extraction channel, and the auxiliary control side is configured as the common mode artifact reference channel. The unilateral or bilateral adductor longus monitoring channel is bound to the corresponding electrode port, channel side, obturator risk side, and channel purpose to generate a surgical field configuration package, which is then sent to the synchronous acquisition unit, interference removal unit, and risk control unit.

Images