An intelligent control system and method for urinary dysfunction based on multi-modal diagnostic data

Abstract

The application discloses a micturition disorder intelligent regulation system and method based on multi-modal diagnostic data. The system comprises a sensing module, a central processing module and a collaborative execution module. The sensing module collects multi-modal physiological signals related to micturition control in real time; the central processing module is used for receiving and processing the multi-modal physiological signals, and collaboratively controls each unit of the collaborative execution module through instructions; the collaborative execution module comprises an artificial urethral sphincter device, a sacral nerve stimulator and a pudendal nerve stimulator, which respectively respond to instructions to adjust urethral closure pressure and adjust electrical stimulation parameters. The application deeply integrates mechanical closure and multi-target point nerve electrical regulation, realizes the collaborative control of an artificial urethral sphincter (AUS), a sacral nerve stimulator (SNM) and a pudendal nerve stimulator (PNS), and is especially suitable for mixed urinary incontinence treatment.

Description

Technical Field

[0001] This invention relates to the field of implantable medical device technology, specifically to an intelligent collaborative regulation system and method for voiding dysfunction guided by multimodal diagnostic data. Background Technology

[0002] The pathophysiological basis of urinary dysfunction is complex. For cases clinically presenting as mixed urinary incontinence, the challenge often lies in the coexistence of bladder detrusor overactivity (leading to storage dysfunction) and urethral sphincter insufficiency (leading to voiding dysfunction). This complex mixed pathological state makes the efficacy of any traditional treatment targeting only a single point often limited and unpredictable.

[0003] Currently, the mainstream implantable surgical treatment options are mainly divided into the following three categories. They are based on different mechanisms of action, but all have clear and inherent technical limitations.

[0004] The first type is the Artificial Urinary Sphincter (AUS), a surgically implanted hydraulic mechanical device. Its core mechanism utilizes an inflatable cuff surrounding the urethra to apply circumferential mechanical pressure, thereby generating static closing pressure to control urinary incontinence. Although this device is considered the "gold standard" for treating severe stress urinary incontinence, its design principles have led to a series of long-term complications: First, the applied pressure is constant or only passively adjustable, unable to respond in real time to sudden increases in intra-abdominal pressure caused by actions such as coughing or jumping, thus exhibiting inherent delay and ineffectiveness in dealing with stress urinary incontinence events; second, the continuous local pressure of the cuff on the urethral tissue is a major cause of urethral atrophy, mucosal erosion, and ultimately, the need for device removal; finally, its sophisticated hydraulic system (including the pump, reservoir, and connecting tubing) carries a long-term risk of fluid leakage and mechanical failure, resulting in a high rate of reoperation for patients.

[0005] The second type is sacral neuromodulation (SNM), which involves implanting electrodes to chronically electrically stimulate the sacral nerve roots (usually S3 or S4) to modulate the neural reflex circuits controlling bladder, sphincter, and pelvic floor function. Its main mechanism of action is to inhibit abnormal detrusor overactivity, thus showing significant efficacy for urge incontinence and some types of urinary retention. However, its limitations are: as a "neuromodulation" method, its direct excitation effect on the external urethral sphincter is weak and indirect, therefore its treatment effect is poor for stress incontinence, which is primarily caused by sphincter dysfunction; furthermore, while current mainstream commercial SNM systems possess open-loop or limited feedback-based (e.g., patient subjective feelings) programming capabilities, they generally lack real-time perception of multimodal physiological signals and adaptive, personalized adjustment capabilities based on complex algorithms. The setting of stimulation parameters still relies on physician experience and remains relatively fixed.

[0006] The third type is pudendal nerve stimulation (PNS), which directly stimulates the pudendal nerve that innervates the external urethral sphincter, triggering voluntary contraction of the sphincter and thus immediately increasing urethral closure pressure. This mechanism makes it theoretically well-suited for combating stress urinary incontinence. However, existing PNS techniques are mostly open-loop stimulation or controlled by simple triggers (such as patient manual switches). This approach fails to integrate stimulation with the real-time state of the bladder (such as degree of fullness, presence of involuntary contractions) and the broader physiological context (such as body position, activity level). Therefore, it is possible to have incorrect stimulation timing (e.g., incorrectly increasing urethral closure force when the bladder is undergoing uninhibited contractions, thus exacerbating urinary urgency) or stimulation conflicts, failing to achieve intelligent coordination with the body's actual physiological needs.

[0007] In summary, existing technologies offer solutions for both "exit" problems (AUS or PNS) and "storage" problems (SNM), but these solutions operate in isolation, are unidirectional, and lack intelligence. In the clinical treatment of complex mixed voiding dysfunctions, physicians often find themselves in a dilemma: for example, strengthening urethral closure (using AUS or PNS) may worsen bladder emptying disorders, while inhibiting overactive bladder (using SNM) has little effect on inherent sphincter defects.

[0008] Therefore, a long-standing technological gap exists in this field: there is an urgent need for an integrated implantable system that can deeply integrate the two mechanisms of mechanical closure (AUS) and neuroelectric modulation (SNM and PNS), and can use individualized diagnostic data as a treatment baseline. This system should be able to sense multimodal physiological signals in real time (such as bladder pressure, urethral sphincter electromyography, intra-abdominal pressure, and body movement status), utilize embedded intelligent algorithms for comprehensive decision-making, and ultimately achieve closed-loop, adaptive, and synergistic regulation of the artificial urethral sphincter (AUS), sacral nerve stimulator (SNM), and pudendal nerve stimulator (PNS). This system aims to simulate and restore near-physiological urine storage and voiding control logic, thereby achieving precise and personalized treatment for complex voiding dysfunctions, especially mixed urinary incontinence. Summary of the Invention

[0009] To overcome the shortcomings of existing technologies, this invention proposes an intelligent regulation system for urinary disorders based on multimodal diagnostic data, characterized by comprising: The sensing module is used to collect multimodal physiological signals related to urination control in real time; the central processing module is communicatively connected to the sensing module and is used to receive and process the multimodal physiological signals and coordinate the control of each unit of the collaborative execution module; the collaborative execution module is communicatively connected to the central processing module and is used to control the urethral closure pressure and stimulate the sacral nerve and pudendal nerve. The sensing module includes: an intra-abdominal pressure sensing unit for capturing changes in intra-abdominal pressure; a bladder status sensing unit for monitoring bladder pressure and / or bladder volume; a urethral pressure sensing unit for measuring urethral closure pressure; and an electromyography and nerve electrical signal sensing unit, including pelvic floor electromyography electrodes, sacral nerve root electrophysiological recording electrodes, and pudendal nerve electrophysiological recording electrodes, for decoding the patient's intention to urinate voluntarily and assessing the pelvic floor muscle tone status, recording the afferent and efferent nerve electrical activity signals of the sacral nerve, and monitoring the sensory and motor fiber activity of the pudendal nerve. The collaborative execution module includes an independently controllable artificial urethral sphincter device, a sacral nerve stimulator, and a pudendal nerve stimulator; The central processing module includes a signal receiving unit for receiving raw information from the sensing module; a signal processing unit for processing the signals to obtain multimodal physiological signal-related data; a personalized baseline model storage unit for storing personalized baseline models; and a collaborative control algorithm unit, including a collaborative working mode determination unit and a collaborative instruction generation unit. The collaborative working mode determination unit compares the multimodal physiological signal-related data with the parameters in the personalized baseline model to determine the collaborative working mode to be invoked. The collaborative instruction generation unit generates a corresponding collaborative control instruction set based on the determined collaborative working mode and sends the instructions to the artificial urethral sphincter device, sacral nerve stimulator, and pudendal nerve stimulator of the collaborative execution module according to the timing relationship.

[0010] To overcome the shortcomings of existing technologies, this invention proposes an intelligent regulation method for urinary disorders based on multimodal diagnostic data, characterized by the following steps: S1: Establish a personalized baseline model; S2: Real-time acquisition and signal processing of multimodal physiological signals to obtain multimodal physiological signal related data; the multimodal physiological signals include intra-abdominal pressure signals, bladder status signals, urethral closure pressure signals, electromyography and nerve electrical signals; the electromyography and nerve electrical signals include pelvic floor electromyography signals, afferent and efferent nerve electrical activity signals of the sacral nerve, and pudendal nerve electrical signals; S3: Compare and analyze the multimodal physiological signal-related data with the parameters in the personalized baseline model to determine the patient's current physiological state; S4: Based on the physiological state obtained in S3, a set of control instructions for synergistic regulation of urination function is generated. The instruction set includes urethral closure pressure regulation instructions and nerve electrical stimulation parameter instructions, and the instructions have a predefined temporal relationship. The instruction set performs synergistic intervention on urethral closure mechanism target points, sacral nerve regulation target points and pudendal nerve regulation target points. S5: Perform coordinated control operations in sequence according to the control instruction set, including adjusting the closure pressure acting on the urethra according to the urethral closure pressure adjustment instruction, and generating and applying a nerve electrical stimulation signal according to the nerve electrical stimulation parameter instruction.

[0011] Compared with the prior art, the present invention has the following significant advantages: First, it achieves true multi-target synergistic treatment: through a central "brain" (central processing module) to uniformly command the three "actuators" AUS, SNM, and PNS, it achieves synchronous and coordinated intervention on the bladder (urine storage organ) and urethra (outlet control), fundamentally overcoming the problem of the limited efficacy of single therapy, and is especially suitable for mixed urinary incontinence.

[0012] Second, it achieves closed-loop and physiological control: by sensing physiological state in real time (such as detecting a cough), the system can proactively and preemptively trigger defensive coordinated responses (such as instantly increasing PNS and AUS pressure). This closed-loop feedback mode greatly improves the reliability of urinary control and is closer to the body's physiological protective reflexes.

[0013] Third, it achieves a high degree of personalization and adaptive optimization: the system is initialized with individualized diagnostic data to optimize the starting point of treatment. At the same time, the system can dynamically fine-tune parameters based on the sensed data, realizing "treatment parameters evolve with the patient's condition," which, while improving efficacy, is expected to minimize the side effects of long-term tissue compression and nerve stimulation.

[0014] Fourth, it enhances the system's intelligence and energy efficiency: the intelligent algorithm enables "on-demand treatment," which means maintaining a low-energy-consumption baseline during most quiet urine storage periods and only initiating high-intensity collaborative work when the risk of incontinence or the urination command is detected, thereby significantly extending the overall battery life of the implantable device. Attached Figure Description

[0015] Appendix Figure 1 This is a schematic diagram of the overall system architecture of Embodiment 1 of the present invention.

[0016] Appendix Figure 2 This is a schematic diagram of the overall system architecture in Embodiment 2 of the present invention.

[0017] Appendix Figure 3 This is a flowchart of Embodiment 3 of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in further detail below with reference to the accompanying drawings. The following description is merely illustrative of the principles and preferred embodiments of the invention and should not be construed as limiting the scope of protection.

[0019] Example 1:

[0020] This invention provides an intelligent control system for urinary disorders based on multimodal diagnostic data, including a sensing module, a central processing module, and a collaborative execution module.

[0021] The sensing module is used to collect multimodal physiological signals related to urination control in real time and is communicatively connected to the central processing module. The sensing module includes: An intra-abdominal pressure sensing unit is used to capture rapid fluctuations in intra-abdominal pressure caused by actions such as coughing, sneezing, and jumping. One embodiment is a standalone, miniaturized implantable pressure sensing device that can be implanted within the patient's abdominal cavity, such as in the rectouterine pouch or outside the rectal serosa, to directly and accurately measure real intra-abdominal pressure changes. Preferably, the sampling frequency is not less than 20 Hz. Alternatively, a pressure sensor implanted in the perineal tissue can be used to indirectly estimate intra-abdominal pressure changes through signal processing algorithms (e.g., adaptive noise cancellation).

[0022] A bladder status sensing unit is used for long-term, continuous, and stable monitoring of the human bladder status. One embodiment is a miniature wireless sensing capsule that can be implanted in the bladder for a long time. It can be placed in the bladder fornix through minimally invasive surgery. The wireless sensing capsule integrates a high-precision pressure sensing element and a volume measurement unit based on the principle of bioimpedance, and can exchange data and energy with the central processing module wirelessly.

[0023] A urethral pressure sensing unit is used to measure urethral closure pressure. One embodiment is a pressure sensor located near or inside the compression cuff of an artificial urethral sphincter device.

[0024] An electromyography (EMG) and neuroelectric signal sensing unit is used to directly acquire the activity potentials of the pelvic floor muscles and the compound action potentials of the target nerves, providing the system with direct state input from the neuromuscular system. This unit is implemented through a multi-channel, long-term implantable bioelectric signal acquisition electrode array, comprising: The pelvic floor electromyographic electrode consists of at least one pair of long-term implantable fine wire electrodes. The implantation location should meet the requirements of signal stability, clear anatomical landmarks, ease of surgical implantation, and low risk of long-term displacement. Preferably, it can be implanted within the belly of the levator ani muscle. The acquired electromyographic signals are primarily used to decode the patient's voluntary urination intention and assess the overall tension of the pelvic floor muscles.

[0025] The sacral nerve root electrophysiological recording electrode is a set of high-impedance, highly selective sleeve-type or hook-shaped electrodes placed on the anterior root of the S3 sacral nerve. This electrode is used to record the afferent and efferent electrical activity signals of the sacral nerve, for example, recording afferent nerve impulses from the bladder for system monitoring of bladder fullness; and recording retrograde conduction response signals generated under sacral nerve modulation stimulation, providing feedback for closed-loop adjustment of stimulation parameters. Structurally, this recording electrode can be integrated with or separately from the stimulation electrodes of the sacral nerve stimulator in the co-execution module.

[0026] The pudendal nerve electrophysiological recording electrode is precisely implanted near the target point of the pudendal nerve to monitor the sensory and motor fiber activity of the pudendal nerve.

[0027] Preferably, the sensing module further includes a motion state sensing unit for monitoring changes in the patient's body position and specific activities (such as coughing and walking), which can be integrated into the housing of the central processing module for detecting changes in body position and acceleration.

[0028] The central processing module, communicatively connected to the sensing module, receives and processes multimodal physiological signals and runs intelligent collaborative control algorithms. Preferably, the central processing module is encapsulated in an implantable titanium alloy shell, similar to existing cardiac pacemakers.

[0029] The central processing module includes a receiving unit, a signal processing unit, a personalized baseline model storage unit, and a cooperative control algorithm unit.

[0030] The signal receiving unit is used to receive raw information from the sensing module, i.e., multimodal physiological signals. Preferably, the system can be externally configured with a remote control, allowing the patient to send signals via the external remote. In this case, the signal receiving unit can receive not only the raw information from the sensing module but also external signals.

[0031] The signal processing unit performs filtering, amplification, analog-to-digital conversion, and other operations on the signals received by the signal receiving unit, and calculates relevant key derived parameters in real time, namely multimodal physiological signal related data, such as intra-abdominal pressure change rate (dp / dt), bladder compliance curve, urethral pressure profile, etc.

[0032] The personalized baseline model storage unit is used to store personalized baseline models. The personalized baseline model is established as follows: Before system implantation, key personalized anatomical and physiological parameters of the patient are obtained through urodynamic examination and pelvic floor electromyography assessment, including but not limited to: maximum urethral closure pressure, abdominal pressure leakage point pressure, bladder safe capacity, detrusor overactivity threshold, and pelvic floor muscle electromyography characteristics. These parameters constitute the patient's individualized diagnostic baseline data. Based on this, the system generates a series of individualized threshold parameters for collaborative control according to preset clinical conversion rules, which include the following principles: Bladder safety pressure restraint: To avoid urinary retention and damage to the upper urinary tract, the intrabladder pressure should always be kept below the safe upper limit of 40 cmH2O.

[0033] The principle of residual urine control: The amount of residual urine in the bladder after urination should be less than 200ml to reduce the risk of urinary tract infection and bladder stones.

[0034] Baseline closure pressure setting: For artificial urethral sphincter devices, the baseline closure pressure is set to 80% of the maximum urethral closure pressure as the initial value; for patients with stress urinary incontinence, it can be set slightly lower than the detrusor leakage point pressure to minimize the risk of long-term compression of urethral tissues without causing involuntary urinary leakage.

[0035] Urinary management for special populations: For patients with disabilities or dementia such as Alzheimer's disease, it is necessary to refer to the patient's self-care ability assessment results and set a trigger threshold for the autonomous urination state, namely the "physiological assisted autonomous urination mode".

[0036] The individualized threshold parameters generated based on the above principles include, but are not limited to: trigger thresholds for each collaborative working mode (such as the intra-abdominal pressure change rate threshold ΔP, bladder fullness threshold), initial values ​​of basic stimulation parameters for each neurostimulator, baseline values ​​of AUS dynamic compensation pressure and nighttime baseline pressure levels, and temporal intervals between each collaborative command. The individualized diagnostic baseline data and the individualized threshold parameters are jointly stored in the personalized baseline model storage unit, collectively referred to as the personalized baseline model. The establishment of the personalized baseline model is crucial for this system to achieve its core feature of "based on multimodal diagnostic data."

[0037] The collaborative control algorithm unit is the intelligent decision-making core of the central processing module. This unit is programmed to make dynamic decisions based on multimodal sensing signals and a personalized baseline model, determining the collaborative working mode that the system should invoke, and issuing precise control commands to the collaborative execution module. Its core logic lies in the timing and parameter coordination of AUS, SNM, and PNS, rather than independent control. The collaborative control algorithm unit should include a collaborative working mode determination unit and a collaborative command generation unit.

[0038] The collaborative working mode determination unit can receive multimodal physiological signal related data (e.g., intra-abdominal pressure, bladder volume, intra-bladder pressure, urine flow rate, urethral sphincter electromyographic activity, pelvic floor muscle electromyographic activity, intra-abdominal pressure change rate, etc.) from the signal processing unit in real time, and compare and analyze them with the parameters in the personalized baseline model to determine which collaborative working mode should be invoked. The collaborative working modes include the following three: “Dynamic leak prevention collaborative mode”: When the abdominal pressure is detected to reach the threshold parameter (e.g., rises above the threshold ΔP>20cmH2O within 200 milliseconds), and the bladder status sensing unit (102) shows that the bladder is in a full state (i.e., reaches the threshold parameter, e.g., volume>150ml), the patient is determined to be in a high-risk leak state, that is, the “dynamic leak prevention collaborative mode” should be called.

[0039] "Physiological Assisted Voiding Mode": When the signal receiving unit receives a "preparing to urinate" signal from the patient via an external device (e.g., a remote control), or when the signal processing unit monitors the "preparing to urinate" conscious behavior in real time (e.g., continuous, rapid contraction of the pelvic floor muscles), and the bladder status sensing unit confirms that the bladder fullness has reached the safe voiding threshold (e.g., an individualized threshold parameter of 150ml), the patient is determined to be in a state of voluntary voiding request, and the "Physiological Assisted Voiding Mode" should be activated.

[0040] "Night or Long-Term Protection Mode": Based on the built-in clock of the central processing module (the central processing module is configured to have a clock) or the motion state sensing unit (which is included in the sensing module at this time), if it is determined that the patient is in a long-term resting or sleeping state, then it is determined that "Night or Long-Term Protection Mode" should be activated.

[0041] The collaborative instruction generation unit generates a specific, multi-target control instruction set based on the determination result of the collaborative working mode determination unit, i.e., the collaborative working mode that should be invoked. The instruction set precisely specifies which units / units in the collaborative execution module should receive the instructions, the content of the instructions, and the timing relationships between the instructions.

[0042] The following describes the instruction sets corresponding to the three collaborative working modes. The various preset parameters involved in the instruction sets (including preset time, preset duration, preset short duration, preset frequency, and preset pressure) are all individualized threshold parameters that are pre-set based on the patient's individualized diagnostic baseline data and stored in the personalized baseline model.

[0043] When the collaborative work mode determination unit determines that the "dynamic anti-leakage collaborative mode" should be invoked, the collaborative instruction generation unit generates a set of instructions, which includes: The first instruction is sent to the artificial urethral sphincter device, instructing it to add a preset dynamic pressure compensation amount above the basic closing pressure within a preset time (a very short time, such as 10 milliseconds). The second instruction is simultaneously sent to the pudendal nerve stimulator and / or sacral nerve stimulator, instructing them to immediately trigger a preset short-duration, preset high-intensity tetanic stimulation pulse sequence to further enhance the contractile force of the urethral sphincter and pelvic floor muscles in an electrophysiological manner. The third instruction is that after the sensing module detects that the intra-abdominal pressure has returned to a stable state, the collaborative control algorithm unit will further generate and issue a third instruction to make the parameters of each actuator gradually return to the baseline state before the event within a preset time (usually within a few seconds, such as 3 seconds).

[0044] When the collaborative working mode determination unit determines that the "physiological assisted spontaneous urination mode" should be invoked, the collaborative instruction generation unit generates a set of instructions, which includes: The first instruction is generated and sent to the artificial urethral sphincter device, instructing the pressure of the compression cuff to linearly decrease to a preset minimum pressure or zero pressure within a preset time (e.g., 5 seconds) to first relieve the mechanical obstruction of the urethra. The second instruction is generated and issued after a preset time interval (e.g., 2 seconds) following the issuance of the first instruction, or based on confirmation of a pressure drop. The second instruction is sent to the sacral nerve stimulator to adjust its stimulation parameters to preset voiding assist parameters, aiming to relax the bladder neck and inhibit unstable contractions of the detrusor muscle. The voiding assist includes continuous stimulation at a preset frequency (e.g., 10 Hz). The third instruction, sent synchronously with the second instruction, is sent to the pudendal nerve stimulator, instructing it to provide inhibitory or relaxation stimulation according to preset parameters to assist the pelvic floor muscles in achieving a state of complete relaxation. The fourth command, upon receiving a confirmation signal from the patient that urination has ended (such as via an external remote control or a specific physiological signal), generates and issues a fourth command to slowly restore the pressure of the artificial urethral sphincter device to the normal baseline closing pressure under controlled conditions, thereby completing the entire urination assistance cycle and resetting the system.

[0045] When the collaborative work mode determination unit determines that the "night or long-term protection mode" should be invoked, the collaborative instruction generation unit generates a set of synchronization instructions, which includes: The first instruction generates and issues a first instruction to the artificial urethral sphincter device to adjust the closure pressure it maintains to a preset nighttime baseline pressure that is lower than the daytime baseline value (e.g., reducing the daytime baseline pressure by 10 cmH2O), aiming to reduce long-term pressure on the urethral tissue. The second instruction generates and issues a second instruction to the sacral nerve stimulator, switching its stimulation program to a preset nighttime program that focuses on inhibiting excessive detrusor muscle activity that may occur at night. The third instruction generates and issues a third instruction to the pudendal nerve stimulator to pause its regular stimulation or reduce its stimulation frequency to a preset minimum level to maintain nerve activity, in order to further save system energy.

[0046] Furthermore, when the collaborative working mode is in "night or long-term protection mode," the system continues to monitor key risk signals from the sensing module (such as a sharp increase in intra-abdominal pressure or bladder volume exceeding a safety threshold). Once such a risk event is identified, the collaborative working mode determination unit will immediately interrupt the current night mode and prioritize the highest-priority "dynamic leak-proof collaborative mode" to ensure treatment safety.

[0047] This system also includes a collaborative execution module, which comprises three independently controllable devices: Artificial urethral sphincter device (AUS): Its compression cuff is wrapped around the urethra, and its pressure regulating pump is controlled by a central processing unit via wireless commands to achieve dynamic and rapid adjustment of urethral closure pressure.

[0048] Sacral nerve stimulator (SNM): Its stimulating electrode is implanted near the S3 sacral foramen. Its pulse generator is controlled by a central processing unit to output adjustable pulse frequency, amplitude and pulse width.

[0049] Puddle nerve stimulator (PNS): Its stimulation electrodes are precisely implanted into the puddle nerve target point and are independently controlled by the central processing unit to output optimized stimulation parameters.

[0050] Example 2:

[0051] The individualized threshold parameters in the personalized baseline model are no longer a static set of data that is completely fixed at the time of implantation, but can be continuously and gradually adapted to the patient's real-time physiological changes during long-term operation after implantation, thereby achieving the adaptive optimization capability of "treatment parameters evolving with the patient's state".

[0052] The collaborative control algorithm unit further includes a data update unit. The data update unit is configured to: continuously and adaptively update the individualized threshold parameters stored in the personalized baseline model storage unit based on real-time physiological feedback signals from the perception module after the collaborative execution module applies neural stimulation or adjusts pressure in response to a collaborative instruction.

[0053] The data update unit includes the following three sub-units: The stimulation effect feedback acquisition subunit is configured to immediately receive corresponding physiological feedback signals from the sensing module after the system applies any neural stimulation (such as SNM or PNS stimulation) or adjusts AUS pressure according to the coordinated instruction. These physiological feedback signals include, but are not limited to: the amplitude and latency of stimulation-evoked compound muscle action potentials (CMAPs) recorded by the pudendal nerve electrophysiological recording electrodes to quantify neural recruitment efficiency; the intensity of evoked electromyographic responses of the pelvic floor muscles and external urethral sphincter recorded by the electromyography and nerve electrical signal sensing unit; the actual urethral closure pressure change after AUS pressure adjustment detected by the urethral pressure sensing unit; and the degree of detrusor muscle activity inhibition fed back by the bladder status sensing unit.

[0054] The feedback quality assessment subunit is configured with built-in preset assessment criteria related to efficacy and safety. These criteria are used to assess the quality of physiological feedback signals received by the stimulation effect feedback acquisition subunit and generate assessment results. For example, when the "dynamic anti-leakage synergistic mode" is triggered, if the actual increase in urethral closure pressure does not reach the preset effective lower limit of sealing, or the induced sphincter electromyographic response intensity is lower than the effective threshold, the synergistic stimulation response is determined to be "insufficient efficacy." If the induced electromyographic response intensity exceeds the safe upper limit, or is accompanied by preset abnormal autonomic nerve response indicators, or a patient's discomfort report is received via external device, the synergistic stimulation response is determined to be "overstimulated." If all feedback indicators are within the preset effective and safe range, the stimulation is determined to be "appropriate."

[0055] The model parameter adaptive update subunit is configured to: based on the evaluation results of the feedback quality assessment subunit, use a preset recursive optimization algorithm to perform small, gradual adjustments to the individualized threshold parameters. The recursive optimization algorithm can be a recursive least squares algorithm with an adaptive forgetting factor or an online learning algorithm based on gradient descent. The individualized threshold parameters that can be updated include: trigger thresholds for each collaborative working mode (such as the intra-abdominal pressure change rate threshold ΔP, bladder safe capacity threshold); baseline values ​​of the optimal stimulation pulse amplitude, frequency, and pulse width for the sacral nerve stimulator and pudendal nerve stimulator in each mode; baseline values ​​of the dynamic compensation pressure of the artificial urethral sphincter device and the level of the nighttime baseline closure pressure; and the precise timing interval between each collaborative command.

[0056] Example 3:

[0057] This invention provides an intelligent regulation method for urination disorders based on multimodal diagnostic data. Its core lies in determining the patient's physiological state based on real-time acquired multimodal physiological signals and a personalized baseline model, and generating coordinated instructions with specific timing to coordinately regulate urination function.

[0058] The method includes the following steps: S1: Establish a personalized baseline model.

[0059] Key individualized anatomical and physiological parameters of the patient are obtained, including but not limited to: maximum urethral closure pressure, intra-abdominal pressure leakage point pressure, bladder safe capacity, detrusor overactivity threshold, and pelvic floor electromyographic characteristics. These parameters constitute the individualized diagnostic baseline data.

[0060] Based on this, according to preset clinical transition rules, a series of individualized threshold parameters for coordinated commands are generated. The principles included in the clinical transition rules are as described in Example 1. The individualized threshold parameters include, but are not limited to: intra-abdominal pressure change rate threshold and bladder fullness threshold corresponding to high-risk urinary leakage state, bladder fullness threshold corresponding to spontaneous urination request state, judgment time window or motor state conditions corresponding to long-term or sleep state, initial values ​​of basic stimulation parameters for each neural stimulation, basic values ​​of urethral dynamic compensation pressure and nighttime basic pressure level, and time intervals between coordinated commands, etc.

[0061] The individualized diagnostic baseline data and the individualized threshold parameters are collectively referred to as the personalized baseline model.

[0062] S2: Real-time acquisition and processing of multimodal physiological information.

[0063] S2-1: Information Acquisition. Real-time acquisition of multimodal physiological signals from the target patient, as well as any external signals (e.g., signals emitted by the patient via an external remote control). These physiological signals include intra-abdominal pressure signals, bladder status signals (including bladder pressure and bladder capacity signals), urethral closure pressure signals, and electromyographic and neuroelectrical signals. The electromyographic and neuroelectrical signals include pelvic floor electromyographic signals, afferent and efferent electrical activity signals of the sacral nerves, and pudendal nerve electrical signals. Preferably, the information acquired in step S2-1 also includes information on the patient's movement status, used to determine changes in the patient's posture and specific activities (such as coughing or walking).

[0064] S2-2: Signal Processing. The acquired raw signals are filtered, amplified, and converted from analog to digital, and key derived features are calculated in real time, such as intra-abdominal pressure change rate (dp / dt), bladder compliance curve, urethral pressure profile, etc., to obtain multimodal physiological signal related data.

[0065] S3: Physiological state assessment.

[0066] Based on the multimodal physiological signal correlation data obtained in step S2-2, a comparison and fusion analysis is performed with the individualized threshold parameters in the personalized baseline model to determine the patient's current physiological state. The patient's physiological state includes: High-risk urinary leakage state: if the real-time intra-abdominal pressure change rate exceeds the intra-abdominal pressure change rate threshold, and the bladder fullness exceeds the bladder fullness threshold; Voluntary urination request status: If a clear urination command is received from an external device, or if the electromyography signal is analyzed to identify a “preparation to urinate” characteristic pattern (such as continuous and rapid contraction of the pelvic floor muscles), and the bladder fullness exceeds the bladder fullness threshold. Long-term or sleep state: If the current time information meets the judgment time window, or if the patient's movement status information indicates that the patient is in a long-term resting or sleep state.

[0067] S4: Generate cooperative execution control instructions.

[0068] Based on the physiological state determined in step S3, a specific set of control instructions is generated for the coordinated regulation of urination function. This set of instructions includes pressure regulation instructions and neural electrical stimulation parameter instructions, and these instructions have a predefined temporal relationship.

[0069] The control instruction set targets three points for coordinated intervention in urination control: the urethral closure mechanism target, the sacral nerve modulation target, and the pudendal nerve modulation target. The core of the instruction set lies in defining the synergistic relationship that must be followed between the intervention instructions targeting these three points, rather than controlling them independently.

[0070] The instruction sets corresponding to the three physiological states are as follows: When a "high-risk urinary leakage state" is determined, a first type of collaborative instruction set (dynamic urinary leakage prevention collaborative instruction set) is generated, including: a first instruction, instructing to add a preset dynamic pressure compensation amount to the basic closure pressure applied to the urethra within a preset time (a very short time, such as 10 milliseconds); a second instruction, synchronously instructing to immediately trigger a preset short-term, preset high-intensity tonic electrical stimulation pulse sequence applied to the pudendal nerve and / or sacral nerve; and a third instruction, after monitoring that the intra-abdominal pressure has returned to a stable state, instructing to gradually restore the dynamic pressure compensation amount and the tonic electrical stimulation pulse to the baseline state before the event within a preset duration (usually within a few seconds, such as 3 seconds).

[0071] When the state is determined to be "voluntary urination request state", a second type of coordinated instruction set (physiological assisted voluntary urination instruction set) is generated, including: a first instruction, instructing the closure pressure applied to the urethra to be linearly reduced to a preset minimum pressure or zero pressure within a preset time period to relieve the mechanical obstruction of the urethra; a second instruction, issued after the first instruction, delayed by a preset time interval or based on the confirmation of the pressure reduction state, instructing the electrical stimulation parameters applied to the sacral nerve to be adjusted to a preset urination assistance state to relax the bladder neck and inhibit the unstable contraction of the detrusor muscle; a third instruction, issued synchronously with the second instruction, applying preset inhibitory or relaxation electrical stimulation to the pudendal nerve to assist the pelvic floor muscles to achieve a complete relaxation state; and a fourth instruction, issued after receiving the urination completion confirmation signal, instructing the closure pressure applied to the urethra to be slowly restored to the daily baseline closure pressure under controlled conditions.

[0072] When the state is determined to be "long-term or sleep state", a third type of collaborative instruction set (nighttime or long-term protection instruction set) is generated, including: a first instruction, instructing the closure pressure applied to the urethra to be adjusted to a preset nighttime baseline pressure lower than the daytime baseline value; a second instruction, instructing the electrical stimulation program applied to the sacral nerve to be switched to a preset nighttime program that focuses on inhibiting excessive detrusor muscle activity; and a third instruction, instructing the suspension of electrical stimulation applied to the pudendal nerve or reducing its electrical stimulation frequency to a preset minimum level to maintain nerve activity. In this state, key risk signals are continuously monitored, and once a high-risk urinary leakage event is identified, the current instruction set is immediately interrupted and the first type of collaborative instruction set is executed first.

[0073] Among them, the preset parameters (including preset duration, preset time, and preset pressure) involved in each instruction set are all individualized threshold parameters that are pre-set based on the patient's individualized diagnostic baseline data and stored in the personalized baseline model.

[0074] S5: Perform coordinated control.

[0075] According to the control instruction set generated in step S4, corresponding coordinated control operations are performed, including: adjusting the closing pressure acting on the urethra according to the pressure adjustment instruction; generating and applying a nerve electrical stimulation signal with specific parameters according to the nerve electrical stimulation parameter instruction to regulate or stimulate the sacral nerve and / or pudendal nerve; wherein the operation of adjusting pressure and applying electrical stimulation are performed in coordination according to the timing relationship specified in the instruction set.

[0076] S6 (Preferred Step): Adaptively update the individualized threshold parameters in the personalized baseline model.

[0077] As a further enhancement to the above steps, this method may also include an adaptive update step for individualized threshold parameters, that is, based on real-time acquired physiological feedback signals, the individualized threshold parameters in the personalized baseline model are continuously and adaptively updated.

[0078] This step specifically includes: S6-1: Stimulation effect feedback acquisition. After applying nerve stimulation or adjusting pressure, the corresponding physiological feedback signals are received immediately, including but not limited to: the amplitude and latency of the complex muscle action potential induced by stimulation, the intensity of the induced electromyographic response of the pelvic floor muscles and external urethral sphincter, the actual change in urethral closure pressure after pressure adjustment, and the degree of detrusor muscle activity inhibition.

[0079] S6-2: Feedback Quality Assessment. Based on pre-defined assessment criteria related to efficacy and safety, the collected physiological feedback signals are assessed for quality: if the actual increase in urethral closure pressure does not reach the pre-defined effective lower limit of sealing, or the induced sphincter electromyographic response intensity is below the effective threshold, it is judged as "insufficient efficacy"; if the induced electromyographic response intensity exceeds the safe upper limit, or is accompanied by pre-defined abnormal autonomic nervous system response indicators, or if a patient reports discomfort via external device, it is judged as "overstimulation"; if all feedback indicators are within the pre-defined effective and safe range, it is judged as "appropriate stimulation".

[0080] S6-3: Adaptive Update of Model Parameters. Based on the feedback quality assessment results, a pre-defined recursive optimization algorithm (such as a recursive least squares algorithm with an adaptive forgetting factor or an online learning algorithm based on gradient descent) is used to perform small, gradual adjustments to the individualized threshold parameters. The individualized threshold parameters that can be updated include: trigger thresholds corresponding to different physiological states, the baseline values ​​of the optimal stimulation pulse amplitude, frequency, and pulse width for the sacral and pudendal nerves under different physiological states, the baseline value of the dynamic urethral compensation pressure and the level of the nighttime baseline pressure, and the precise temporal interval between each coordination command.

[0081] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A smart regulation system for voiding disorders based on multimodal diagnostic data, characterized in that, include: The sensing module is used to collect multimodal physiological signals related to urination control in real time; The central processing module, which is communicatively connected to the sensing module, is used to receive and process the multimodal physiological signals and to coordinate the control of each unit of the collaborative execution module; the collaborative execution module, which is communicatively connected to the central processing module, is used to control the urethral closure pressure and stimulate the sacral nerve and pudendal nerve. The sensing module includes an intra-abdominal pressure sensing unit for capturing changes in intra-abdominal pressure; A bladder status sensing unit is used to monitor bladder pressure and / or bladder volume; a urethral pressure sensing unit is used to measure urethral closure pressure. The electromyography and neuroelectric signal sensing unit includes pelvic floor electromyography electrodes, sacral nerve root electrophysiological recording electrodes, and pudendal nerve electrophysiological recording electrodes. It is used to decode the patient's intention to urinate voluntarily and assess the pelvic floor muscle tone, record the afferent and efferent nerve electrical activity signals of the sacral nerve, and monitor the sensory and motor fiber activity of the pudendal nerve. The collaborative execution module includes an independently controllable artificial urethral sphincter device, a sacral nerve stimulator, and a pudendal nerve stimulator; The central processing module includes a signal receiving unit for receiving raw information from the sensing module; a signal processing unit for processing the signal to obtain multimodal physiological signal related data; and a personalized baseline model storage unit for storing personalized baseline models. The collaborative control algorithm unit includes a collaborative working mode determination unit and a collaborative instruction generation unit. The collaborative working mode determination unit is used to compare the relevant data of multimodal physiological signals with the parameters in the personalized baseline model to determine the collaborative working mode to be invoked. The collaborative instruction generation unit is used to generate a corresponding collaborative control instruction set according to the determined collaborative working mode, and send the instructions to the artificial urethral sphincter device, sacral nerve stimulator and pudendal nerve stimulator of the collaborative execution module according to the timing relationship.

2. The system according to claim 1, characterized in that, The sensing module also includes a motion state sensing unit for monitoring changes in the patient's body position and activity status.

3. The system according to claim 1 or 2, characterized in that, The personalized baseline model includes individualized diagnostic baseline data and individualized threshold parameters; The individualized diagnostic baseline data includes maximum urethral closure pressure, abdominal pressure leakage point pressure, bladder safe capacity, detrusor overactivity threshold, and pelvic floor muscle electromyographic characteristics. The individualized threshold parameters include the trigger threshold for each collaborative working mode, the initial value of the basic stimulation parameters for each neural stimulation, the basic value of the dynamic compensation pressure of the urethra and the nighttime basic pressure level, and the time interval between each collaborative command.

4. The system according to claim 3, characterized in that, The collaborative control algorithm unit further includes a data update unit, which is configured to update the parameters in the personalized baseline model based on the real-time physiological feedback signal from the sensing module after the collaborative execution module applies neural stimulation or adjusts the urethral closure pressure in response to the collaborative instruction.

5. The system according to claim 4, characterized in that, The data update unit includes a stimulus effect feedback acquisition subunit, a feedback quality assessment subunit, and a model parameter adaptive update subunit. The stimulation effect feedback acquisition subunit is configured to receive physiological feedback signals from the sensing module immediately after the system applies nerve stimulation or adjusts the urethral closure pressure. The feedback quality assessment subunit is configured to assess the quality of the physiological feedback signal and generate an assessment result. The model parameter adaptive update subunit is configured to update the parameters in the personalized baseline model based on the evaluation results of the feedback quality assessment subunit using a preset recursive optimization algorithm.

6. A method for intelligent regulation of voiding disorders based on multimodal diagnostic data, characterized in that, Includes the following steps: S1: Establish a personalized baseline model; S2: Real-time acquisition and signal processing of multimodal physiological signals to obtain multimodal physiological signal related data; the multimodal physiological signals include intra-abdominal pressure signals, bladder status signals, urethral closure pressure signals, electromyography and nerve electrical signals; the electromyography and nerve electrical signals include pelvic floor electromyography signals, afferent and efferent nerve electrical activity signals of the sacral nerve, and pudendal nerve electrical signals; S3: Compare and analyze the multimodal physiological signal-related data with the parameters in the personalized baseline model to determine the patient's current physiological state; S4: Based on the physiological state obtained in S3, a set of control instructions for synergistic regulation of urination function is generated. The instruction set includes urethral closure pressure regulation instructions and nerve electrical stimulation parameter instructions, and the instructions have a predefined temporal relationship. The instruction set performs synergistic intervention on urethral closure mechanism target points, sacral nerve regulation target points and pudendal nerve regulation target points. S5: Perform coordinated control operations in sequence according to the control instruction set, including adjusting the closure pressure acting on the urethra according to the urethral closure pressure adjustment instruction, and generating and applying a nerve electrical stimulation signal according to the nerve electrical stimulation parameter instruction.

7. The method according to claim 6, characterized in that, The multimodal physiological signals mentioned in step S2 also include information on the patient's movement status, which is used to determine changes in the patient's body position and specific activities.

8. The method according to claim 6 or 7, characterized in that, The physiological states include high-risk urinary incontinence state, voluntary urination request state, and long-term or sleep state; The high-risk urinary leakage state is determined by the following conditions: the real-time intra-abdominal pressure change rate exceeds the intra-abdominal pressure change rate threshold, and the bladder fullness exceeds the bladder fullness threshold. The conditions for determining the autonomous urination request state are: receiving a clear urination command from an external device, or analyzing electromyographic signals to identify a urination preparation pattern, while the bladder fullness exceeds the bladder fullness threshold. The criteria for determining the long-term or sleep state are: the current time information meets the determination time window, or the patient's movement status information indicates that they are in a long-term resting or sleep period.

9. The method according to claim 8, characterized in that, In step S1, the personalized baseline model includes individualized diagnostic baseline data and individualized threshold parameters; The individualized diagnostic baseline data includes maximum urethral closure pressure, abdominal pressure leakage point pressure, bladder safe capacity, detrusor overactivity threshold, and pelvic floor muscle electromyographic characteristics. The individualized threshold parameters include the intra-abdominal pressure change rate threshold and bladder fullness threshold corresponding to the high-risk urinary leakage state, the bladder fullness threshold corresponding to the voluntary urination request state, the judgment time window or motor state conditions corresponding to the long-term or sleep state, the initial values ​​of the basic stimulation parameters of each neural stimulation, the basic value of the dynamic compensation pressure of the urethra and the nighttime basic pressure level, and the time interval between each coordination command.

10. The method according to claim 9, characterized in that, Also includes: S6: Adaptive update of individualized threshold parameters: After applying nerve stimulation or adjusting urethral closure pressure, the parameters in the personalized baseline model are updated based on real-time collected physiological feedback signals. Step S6 specifically includes: S6-1: Stimulation effect feedback acquisition, which receives corresponding physiological feedback signals immediately after the application of nerve stimulation or pressure adjustment; S6-2: Feedback quality assessment, assessing the quality of the physiological feedback signal and issuing the assessment results; S6-3: Based on the evaluation results of the feedback quality assessment subunit, the parameters in the personalized baseline model are updated using a recursive optimization algorithm.

Images