Colon dialysis catheter system with intestinal wall tension and temperature sensing feedback

Abstract

The invention provides a colon dialysis catheter system with intestinal wall tension and temperature sensing feedback, which comprises a colon dialysis catheter body which is a multi-cavity flexible tube body capable of being inserted into a colon cavity through the anus, the catheter body is made of a medical flexible high polymer material, a supporting framework is arranged in the catheter wall along the axial direction, and the supporting framework is made of a medical flexible high polymer material. A plurality of chordae tendineae are arranged in the circumferential direction of the catheter, the number of the chordae tendineae is N, one end of each chordae tendineae is fixed to a flexible section at the far end of the catheter, the other end of each chordae tendineae is connected to a chordae tendineae driving mechanism at the near end of the catheter, and the chordae tendineae driving mechanism is used for adjusting the bending posture and local rigidity of the far end of the catheter in a colon cavity by pulling or loosening the chordae tendineae. An intestinal wall tension sensing array and multi-point temperature sensing are integrated, multi-source data such as tension and temperature are subjected to fusion analysis through a central control module, risk indexes are dynamically calculated, perfusion flow, negative pressure drainage and catheter postures are adjusted in a linkage mode, and feature weights and threshold values can be switched according to different clinical scenes.

Description

Technical Field

[0001] This invention belongs to the field of colonic dialysis, specifically relating to a colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback. Background Technology

[0002] Colonic irrigation and colonic dialysis, as techniques for injecting large amounts of fluid into the intestine to remove toxins and fecal load, have been applied in various scenarios such as chronic constipation, hepatic encephalopathy, and preoperative bowel preparation. Clinically, the common equipment configuration is a "colonic dialysis machine + disposable catheter" or an "intestinal irrigation system + enema tube." Warm water or dialysis fluid is injected into the colon through the anus using a pressure pump, and then drained through the catheter or naturally to achieve cleansing, deburden reduction, or localized treatment. Because the target organ is the thin-walled colon, and a large volume of fluid and a certain pressure are often used during irrigation, guidelines and health departments in various countries emphasize safety issues related to colonic irrigation and colonic hydrotherapy, especially the prevention of serious complications such as colonic perforation, infection, and electrolyte imbalance.

[0003] Existing technologies pose risks related to water pressure and mechanical damage. Literature reports that high-pressure, high-flow enemas or colonic irrigation can lead to hydrostatic colonic perforation, especially in anatomically vulnerable areas such as the sigmoid colon; there are also cases where the irrigation tip or catheter tip directly mechanically punctures the intestinal wall, causing perforation. In so-called "high-volume colonic irrigation," due to the large irrigation volume and inconsistent operating standards, early literature has already indicated risks such as rectal / colonic perforation, infection, water intoxication, and electrolyte imbalance. To reduce water pressure-related risks, some commercial equipment and regulations require the installation of pressure-limiting valves and back pressure release structures. Operators adjust the flow rate and pressure based on the patient's complaints and experience, observing subjective manifestations such as abdominal distension and pain. However, these safety measures often rely on tubing pressure or pump outlet pressure, making it difficult to reflect the true stress situation of a specific area of ​​the colonic wall in a timely and accurate manner, especially failing to identify stress concentration states where "the overall pressure is not high, but the local catheter is pressing against the mucosa."

[0004] On the other hand, temperature control of the irrigation medium and temperature control of the current technology are also inadequate. Most colonic irrigation devices only have simple temperature control or temperature limiting systems at the water source to keep the irrigation water in a range "close to body temperature," but heat loss of the irrigation fluid in the tubing and the temperature gradient caused by local retention in the colon often lack effective monitoring. Studies and clinical experience suggest that excessively cold irrigation fluid can induce intestinal spasms, discomfort, and even reflexive cardiovascular reactions, while excessively hot liquids or local heat accumulation may cause burns to the mucosa or aggravate inflammatory reactions. Existing irrigation devices generally only indirectly estimate the intestinal lumen temperature through a single-point temperature probe or outlet water temperature display, lacking real-time monitoring of the temperature of the mucosa near the catheter tip and active protection control based on the rate of temperature change.

[0005] Furthermore, regarding the structure of the catheter itself, flexible gastrointestinal catheters, colonoscopes, and other instruments have adopted multi-joint or flexible designs to improve their passage through the intestinal lumen. Tendon drive and end-effector contact force sensing technologies have been explored in some flexible endoscopic robots to guide insertion force and reduce the risk of perforation. However, these force sensors are mostly designed for "catheter guidance / navigation" scenarios, typically using end-effector contact force feedback to assist doctors in manually adjusting the operation. They rarely integrate with the conditions of colonic dialysis / irrigation, directly incorporating intestinal wall contact force signals into the multi-actuator linkage control of perfusion flow, negative pressure drainage, and catheter posture. They also lack a systematic safety strategy specifically for the colonic dialysis scenario. Existing colonic dialysis catheters are mostly single or double-lumen catheters with only simple perfusion and drainage functions. They lack tension sensor arrays arranged axially and circumferentially, failing to form a "spatial image" of the intestinal wall force field. Furthermore, there is no evidence of integrating multiple temperature sensors on the same catheter to distinguish between the mucosal temperature at the catheter tip and the dialysate temperature within the lumen, and to jointly determine local safety boundaries with intestinal wall stress information.

[0006] In terms of control methods, most existing colonic irrigation / dialysis equipment adopts a simple safety control mode of "single parameter threshold + manual intervention," such as stopping and alarming when the pump outlet pressure exceeds the upper limit or the water supply temperature exceeds the standard. Although some systems have multiple parameter monitoring interfaces, the parameters are mostly displayed independently, and true multimodal data fusion and risk index quantification have not yet been formed. There is also no systematic introduction of algorithms such as "comprehensive risk scoring + support vector machine classification" in the field of colonic dialysis to uniformly model and intelligently distinguish tension, pressure, temperature and their time characteristics; moreover, the differences in safety priorities under different clinical scenarios (such as preoperative cleaning, postoperative drainage, and local hyperthermic perfusion) have not been solidified into the equipment control strategy in the form of "scenario mode," so that the equipment can automatically adjust weights and thresholds according to the scenario.

[0007] From a clinical perspective, preoperative colonic dialysis needs to minimize intestinal distension and patient pain while ensuring effective cleansing; postoperative drainage requires maintaining catheter patency and preventing negative pressure from adsorbing the mucosa or pulling on the anastomosis; emerging treatments such as local hyperthermic perfusion chemotherapy require precise control of local temperature and perfusion pressure to provide sufficient heat dose and drug concentration while avoiding burns to normal tissues. However, most colonic irrigation / dialysis catheters and related equipment currently on the market are "universal," with parameter settings mainly relying on the experience of doctors or nurses for adjustment. They cannot automatically optimize control for different scenarios and patients, nor can they incorporate patient subjective discomfort (such as abdominal distension and pain scores) into the control loop.

[0008] In summary, existing colonic dialysis catheter systems still have significant shortcomings in terms of safety monitoring and intelligent control: First, they cannot monitor the spatial distribution of actual intestinal wall stress, and can only estimate it indirectly through tubing pressure, making it difficult to detect local stress concentration risks in a timely manner; second, temperature monitoring has limited dimensions, lacking multi-point monitoring of mucosal temperature and temperature rise rate near the tip and active overheat protection; third, there is a lack of collaborative control based on multimodal risk assessment among the actuators such as perfusion, drainage, and catheter posture, mainly relying on manual experience, resulting in significant differences in safety among different operators; fourth, there is a lack of parameter modes and algorithm adaptations for different clinical scenarios, making it impossible to achieve individualized and scenario-based colonic dialysis safety management. Therefore, it is necessary to propose a colonic dialysis catheter system that integrates an intestinal wall tension sensing array and multi-point temperature sensors on the catheter body, and uses a central control module to fuse and analyze multi-source data such as tension and temperature, dynamically calculate the risk index, and adjust the perfusion flow, negative pressure drainage, and catheter posture in a coordinated manner. Furthermore, it can switch feature weights and thresholds according to different clinical scenarios to improve the safety and intelligence of colonic dialysis and reduce the risk of mechanical perforation, temperature-related mucosal damage, and other complications. Summary of the Invention

[0009] The purpose of this invention is to provide a colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback. This invention can integrate an intestinal wall tension sensing array and multi-point temperature sensing, and perform fusion analysis on multi-source data such as tension and temperature through a central control module to dynamically calculate the risk index, and adjust the perfusion flow rate, negative pressure drainage and catheter posture in a coordinated manner. Furthermore, it can switch feature weights and thresholds according to different clinical scenarios.

[0010] To achieve the above objectives, the present invention provides the following technical solution, comprising: a colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback, comprising:

[0011] The colonic dialysis catheter body is a multi-lumen flexible tube that can be inserted into the colonic lumen through the anus. The catheter body is made of medical flexible polymer material. A support skeleton is set along the axial direction inside the catheter wall, and several tendon cords are arranged along the circumference of the catheter. The number of tendon cords is N, where N≥3. One end of each tendon cord is fixed to the flexible segment at the distal end of the catheter, and the other end is connected to the tendon cord driving mechanism at the proximal end of the catheter. The tendon cord driving mechanism is used to adjust the bending posture and local stiffness of the distal end of the catheter in the colonic lumen by pulling or relaxing the tendon cords.

[0012] An intestinal wall tension sensor array is disposed on the outer wall of the distal flexible section of the catheter body. Several annular sensing bands are arranged at intervals along the axial direction of the catheter. Multiple force-sensitive sensing elements are disposed on each annular sensing band along the circumference of the catheter. The force-sensitive sensing elements are used to output tension signals related to the contact stress of the colon wall in the corresponding annular area.

[0013] A temperature sensor unit is disposed on the catheter body for monitoring the local temperature in the colon lumen and the temperature of the dialysate. The temperature sensor unit includes: a first temperature sensor attached to the outer wall of the distal end of the catheter for detecting the temperature of the mucosa near the catheter tip, and a second temperature sensor disposed in the catheter lumen near the dialysate outlet for detecting the temperature of the dialysate outlet.

[0014] The perfusion and drainage module is connected to the perfusion and drainage channels of the catheter body and includes a reversibly driven micro-perfusion pump and a negative pressure drainage unit. The micro-perfusion pump is used to deliver dialysate into the colonic lumen at an adjustable flow rate, and the negative pressure drainage unit is used to aspirate and depressurize gas and / or liquid in the colonic lumen when needed.

[0015] The central control module is electrically connected to the intestinal wall tension sensor array, temperature sensor unit, perfusion and drainage module, and chordae tendineae drive mechanism.

[0016] The central control module includes:

[0017] The data acquisition unit is used to acquire the tension signal of each annular sensing belt and the temperature signal of each temperature sensor at a preset sampling frequency, and to calculate the instantaneous values, rate of change and fluctuation statistics of tension and temperature within a sliding time window.

[0018] The risk assessment unit is used to normalize the tension and temperature characteristics within the sliding time window, calculate the joint risk index M-Risk according to the pre-stored feature weights, the joint risk index simultaneously characterizes the stress state of the colon wall and the local temperature state, and compare the joint risk index with at least two preset risk thresholds to determine the risk level of the current dialysis process.

[0019] The control unit is used to maintain the current perfusion flow rate and catheter posture when the risk level is safe, reduce the flow rate of the micro-perfusion pump and / or intermittently control the negative pressure drainage unit to slowly depressurize when the risk level is warning, and drive the chordae tendineae drive mechanism to adjust the bending posture of the distal end of the catheter according to the force distribution output by the intestinal wall tension sensor array; when the risk level is high risk, immediately stop the output of the micro-perfusion pump and continuously control the depressurization of the negative pressure drainage unit, while controlling the chordae tendineae drive mechanism to release or reverse the adjustment of each chordae tendineae to make the distal end of the catheter withdraw from the high-pressure area, and trigger the alarm system to output danger warning to medical staff, thereby forming a closed-loop safety control of colonic dialysis based on intestinal wall tension and temperature sensor feedback.

[0020] Furthermore, in the intestinal wall tension sensor array, the spacing of the annular sensing bands along the axial direction of the catheter is 2 to 10 mm, and each annular sensing band contains m force-sensitive sensing elements, 3 ≤ m ≤ 12. The central control module reconstructs the circumferential-axial intestinal wall contact stress distribution at the distal end of the catheter based on the tension signals of each annular sensing band and each circumferential position, which is used to identify local stress concentration areas.

[0021] Furthermore, the risk assessment unit calculates the joint risk index M-Risk using the following example formula: Among them, P~ max σ~P is the normalized value of the maximum contact pressure within the sliding time window, σ~P is the normalized value of the variance of the tension of the annular sensing belt, and ΔP~ ring T is the normalized value of the tension difference between adjacent annular sensing strips. mucosa Let dT be the normalized deviation of the mucosal temperature detected by the first temperature sensor relative to the target temperature, dT be the normalized value of the mucosal temperature change rate, and w1~w5 be the weighting coefficients of the corresponding features, and satisfy the following conditions: .

[0022] Furthermore, the central control module also includes a scenario mode management unit, which stores at least three sets of feature weights and risk threshold parameters for different clinical scenarios. The clinical scenarios include preoperative colonic irrigation, postoperative colonic drainage, and local hyperthermic perfusion therapy. The scenario mode management unit loads the corresponding parameter set according to the user's selection. In the preoperative colonic irrigation scenario, the weights of tension and pressure-related features are increased and the maximum contact pressure threshold is set to 3-5 kPa. In the local hyperthermic perfusion therapy scenario, the weights of temperature deviation and temperature rise rate features are increased and the temperature threshold is set to 42-43°C.

[0023] Furthermore, the central control module includes a state classification subunit. The state classification subunit uses a support vector machine model with radial basis function kernel function as input, and the feature vector composed of joint risk index M-Risk and the multimodal features to classify the colon dialysis status into three categories: safe, warning, and high risk. The support vector machine model is obtained by training on historical colon dialysis data containing safe, warning, and high risk samples, and the penalty coefficient and kernel parameters are optimized through cross-validation.

[0024] Furthermore, the central control module includes a dynamic threshold adjustment unit and a feature weight adjustment unit. The dynamic threshold adjustment unit is configured to statistically analyze the distribution characteristics of the joint risk index M-Risk within a preset time period, and fine-tune the warning threshold and high-risk threshold based on the mean and standard deviation or quantile information. The feature weight adjustment unit is configured to adjust the weights of tension-related features and temperature-related features by a limited range based on recent risk events and control effects, so as to reduce unnecessary perfusion interruptions while ensuring safety.

[0025] Furthermore, in the preoperative colonic irrigation scenario, when the combined risk index M-Risk is at the warning level, the central control module controls the micro-infusion pump to reduce the irrigation flow rate to 50% to 80% of the original set flow rate, and periodically activates the negative pressure drainage unit for short periods to slowly depressurize; when the combined risk index M-Risk is at the high risk level, the output of the micro-infusion pump is immediately stopped and the negative pressure drainage unit is continuously controlled to depressurize until the estimated internal pressure is lower than the set safety value, while driving the chordae tendineae drive mechanism to adjust the bending posture of the distal end of the catheter to avoid the local pressure area.

[0026] Furthermore, in the postoperative colonic drainage scenario, the central control module considers a slow increase in tension and a decrease in drainage volume as characteristics of drainage obstruction risk. When the catheter tension baseline is observed to increase and the drainage volume is lower than the preset lower limit within multiple consecutive sliding time windows, the module controls the micro-infusion pump to perform low-flow flushing and controls the negative pressure drainage unit to perform low-intensity pulse aspiration to attempt to clear the drainage path. If the combined risk index M-Risk remains high, the drainage is suspended and an alarm is triggered to prompt medical staff to check and handle the situation.

[0027] Furthermore, the central control module includes a subjective discomfort prediction unit, which is used to establish a prediction model based on the correspondence between the patient's subjective pain score in historical records and objective sensor parameters such as catheter tension and temperature. During colonic dialysis, the subjective discomfort prediction value is calculated based on the real-time tension peak, contact pressure and temperature deviation. When the subjective discomfort prediction value exceeds the preset comfort threshold, even if the combined risk index M-Risk is still below the warning threshold, the control unit is configured to reduce the perfusion flow rate in advance, extend the perfusion interval time or temporarily suspend dialysis, so as to reduce the patient's subjective discomfort and optimize the clinical operation experience.

[0028] Compared with the prior art, the present invention has at least the following beneficial effects.

[0029] This invention enables spatial distribution monitoring of intestinal wall stress, allowing for early prevention of mechanical complications such as perforation. By axially spaced multiple ring-shaped tension sensing bands at the distal end of the catheter, and uniformly arranging multiple force-sensitive sensing elements circumferentially within each ring, this invention does not simply monitor intra-tubular pressure but obtains intestinal wall contact stress distribution information in the "circumferential × axial" direction. The central control module can accurately identify which ring and which side experiences stress peaks, and through the chordae tendineae drive mechanism, it can selectively change the catheter's bending direction and wall adhesion, actively dispersing pressure in high-pressure areas. This allows for the detection and mitigation of localized stress concentrations before overall intra-tubular pressure exceeds limits, significantly reducing the risk of mucosal compression necrosis and colonic perforation.

[0030] This invention achieves dual-point monitoring of mucosal temperature and perfusion fluid temperature, along with rapid overheating linkage protection, improving thermal safety. A first temperature sensor is positioned on the outer surface of the catheter tip, and a second temperature sensor is positioned near the dialysate outlet within the catheter lumen. This allows for simultaneous monitoring of both the "actual outlet temperature of the perfusion fluid" and the "local temperature of the catheter near the mucosa." When calculating the combined risk index M-Risk, the risk assessment unit uses temperature deviation and temperature rise rate as key characteristics. When any temperature measurement point exceeds the threshold or the temperature rises too rapidly, the local overheat control unit intervenes without waiting for the overall risk index to exceed the limit. It rapidly cools the catheter by reducing heating power, decreasing flow rate, switching to room temperature solution, or briefly pausing the machine. Simultaneously, it can adjust the catheter's orientation to move it away from the hot spot area, effectively preventing burns and hyperthermia-related complications.

[0031] This invention improves the accuracy and sensitivity of risk identification by combining joint risk assessment and SVM classification based on multimodal features. Instead of relying on a single threshold, it normalizes multidimensional features such as maximum contact pressure, tension fluctuations, tension differences between turns, mucosal temperature deviation, and temperature change rate, then calculates a joint risk index (M-Risk) by weighting these features. This index is then combined with a support vector machine model to perform pattern recognition for safe, early warning, and high-risk states. This approach filters out transient, isolated anomalies, avoiding excessive false alarms. Furthermore, it identifies hidden risks arising from the superposition of minor anomalies in multiple features. For example, when tension is slightly higher or temperature is slightly increased, but the trend is similar to previous high-risk patterns, SVM can identify this as an early warning or high-risk situation, allowing the control system to intervene earlier and reducing edge dangers that cannot be captured by a single threshold.

[0032] Multi-actuator coordinated control adjusts flow rate, posture, and pressure release, truly forming a closed-loop, safe, and smooth dialysis process. Unlike traditional equipment that only stops the pump or provides simple pressure limiting protection, the central control module of this invention can simultaneously issue coordinated control commands to the micro-infusion pump, negative pressure drainage unit, and chordae tendineae drive mechanism: in case of an early warning, it first smoothly reduces the flow rate and intermittently releases pressure, while fine-tuning the catheter's bending posture to disperse local stress; in case of high risk, it immediately stops the pump, continuously drains and reduces pressure, and releases the chordae tendineae to withdraw the distal end of the catheter from the danger zone. Multi-actuator linkage ensures that the control action is no longer abruptly shut down, but rather a graded and gradual intervention, ensuring both safety and treatment continuity while maintaining cleaning / drainage effectiveness.

[0033] The scenario-based parameter modes support various clinical applications, allowing a single system to cover multiple applications such as irrigation, drainage, and hyperthermic perfusion. This invention pre-configures multiple scenario modes in the central control module, including preoperative colonic cleaning and irrigation, postoperative colonic drainage, and local hyperthermic perfusion therapy. Each mode has an independent weight matrix and risk threshold: in irrigation mode, the weight of tension / pressure characteristics is enhanced, and the contact pressure threshold is tightened, focusing on preventing mechanical overload and perforation; in drainage mode, the combined characteristics of a slow increase in tension baseline and a decrease in drainage volume are emphasized for early identification of tube blockage and infection; in hyperthermic perfusion mode, the weight of temperature deviation and temperature rise rate is significantly increased, making it extremely sensitive to overheating. Doctors only need to select the mode according to the clinical scenario, and the system automatically adopts the corresponding assessment and control strategies, reducing frequent manual parameter adjustments and improving the equipment's adaptability to multiple scenarios.

[0034] By introducing subjective discomfort prediction and control, the system balances "safety" and "comfort," enhancing the patient experience. By collecting patients' VAS pain scores at different stages of the procedure and establishing a regression model with characteristics such as contact pressure, peak tension, and temperature deviation, the subjective discomfort prediction unit of this invention can estimate the current level of discomfort in real time without adding additional sensors. When the predicted subjective discomfort index exceeds a preset comfort threshold, even if the physical risk is still within a safe range, the control unit will proactively reduce the perfusion rate or extend the interval, and appropriately adjust the catheter posture, further elevating the focus from "avoiding as much danger as possible" to "minimizing pain and abdominal distension." This dual closed loop of "human feedback + physical safety" helps improve patient tolerance and compliance, making it particularly suitable for patients requiring repeated or prolonged colonic dialysis.

[0035] This invention exhibits excellent engineering feasibility and scalability, facilitating integration and upgrades within existing clinical systems. It provides clearly defined numerical ranges for catheter materials, sensor selection, pump flow rates, and threshold settings, such as catheter material thermal conductivity, tension threshold of 2N, contact pressure threshold of 5kPa, temperature threshold of 42℃, and overthreshold time. This ensures ease of selection and calibration in practical engineering while leaving room for subsequent parameter optimization and algorithm iteration based on real case data. The system structure employs a modular design. The central control module connects to the perfusion pump, drainage system, and chordae tendineae actuator via a standardized "actuator interface (parameter bus)," allowing for easy portability to pump and catheter platforms from different manufacturers. This decoupling of hardware and software facilitates subsequent product serialization and multi-departmental deployment.

[0036] In summary, this invention integrates multi-coil, multi-point intestinal wall tension sensing and dual-point electrical temperature monitoring on the colonic dialysis catheter, and combines it with a joint risk index, SVM classification, multi-execution end collaborative control, scenario-based parameter modes, and subjective discomfort prediction models to achieve refined, safe, and individualized management of the colonic dialysis process. Compared with existing technologies, it has significant technological advancements in preventing perforation, preventing burns, improving comfort, and adaptability to multiple scenarios. Attached Figure Description

[0037] Figure 1 This is a block diagram of the overall structure of the colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback of the present invention, showing the signal and control relationship between the catheter-side hardware and the control-side hardware. In the diagram: the catheter body 100, support frame 110, chordae tendineae assembly 120, intestinal wall tension sensor array 130, temperature sensor unit 140, and pressure sensor / one-way valve assembly 160 constitute the front-end sensing and execution part, which is connected to the central control module 200 via signal conditioning and analog-to-digital conversion unit 210; the historical data and patient parameter storage unit 230 provides individual baselines and parameters to the central control module 200; the central control module 200 includes a risk assessment unit 220 and an actuator interface / parameter bus 240, outputting control signals to drive the perfusion pump mechanism 300, negative pressure drainage unit 310, and chordae tendineae drive motor 320, etc., to achieve closed-loop control of the colonic dialysis process.

[0038] Figure 2This is a longitudinal structural diagram of the flexible catheter of the present invention, showing the structural composition of the catheter body 100 in the length direction and the arrangement of key components. In the figure: the proximal connecting section 102 of the catheter is used to connect with the infusion pump mechanism 300 and the negative pressure drainage unit 310; the distal flexible section 101 of the catheter is used to insert into the colonic lumen; the support frame 110 is embedded in the catheter wall along the catheter axis to provide shape retention and anti-bending support; multiple tendon chord assemblies 120 are arranged parallel to the catheter axis and are fixed to the distal flexible section 101 of the catheter by tendon chord fixation rings 122 at the distal end; a tension sensor band 130 is formed on the outer periphery of the distal flexible section, and several force-sensitive sensing elements 131 are arranged on the band; a tip temperature sensor 141 is provided on the outer side of the catheter tip, and a middle temperature sensor 142 is provided on the inner wall of the middle part of the catheter to measure the mucosal temperature and the infusion fluid temperature, respectively.

[0039] Figure 3 This is an enlarged structural schematic diagram of the distal intestinal wall tension sensor array of the catheter of the present invention, showing the specific arrangement of the distal flexible segment 101 and its peripheral tension sensors. In the figure: one or more annular sensing bands 130 are arranged around the periphery of the distal flexible segment 101, and multiple thin-film force-sensitive resistor force-sensitive sensing elements 131 are arranged at equal intervals along the circumference on each sensing band; a force transmission block 132 is disposed between the force-sensitive sensing element 131 and the outer surface of the catheter, used to transmit the contact force between the catheter and the intestinal wall to the force-sensitive sensing element; a chordae tendineae fixing ring 122 is arranged inside the sensing band 130 for fixing the chordae tendineae assembly 120; a tip temperature sensor 141 is disposed on the side of the catheter tip facing the intestinal wall for measuring the mucosal temperature adjacent to the tip. Figure 3 As can be seen, the present invention achieves multi-point detection of contact stress of the distal circumferential intestinal wall through the ring-shaped multi-point arrangement of force-sensitive sensing elements 131.

[0040] Figure 4 This is a schematic diagram of the cross-sectional structure of the catheter of the present invention, showing the multi-lumen structure of the catheter body 100 in cross-section and the spatial layout of the supporting skeleton and chordae tendineae channels. In the figure: the outermost ring structure of the catheter sheath 100 is a supporting skeleton layer 110 inside it; a central perfusion cavity 150 is provided in the center of the catheter for delivering dialysate; a lateral drainage cavity 151 is provided on one side of the central perfusion cavity 150 for negative pressure drainage and pressure relief; multiple chordae tendineae channels 121 are evenly arranged circumferentially inside the supporting skeleton layer 110, and a chordae tendineae assembly 120 passes through each channel; a sensor wire channel 152 is provided at a certain position inside the catheter wall to accommodate the leads of the tension sensor array 130 and the temperature sensor unit 140. Figure 4 This vividly illustrates the structural features of the present invention, which integrates multi-cavity arrangement, support frame and tendon cable channel, and sensing circuit channel within a limited cross-sectional space. Detailed Implementation

[0041] The flexible catheter infusion system of the present invention includes the following components:

[0042] The catheter body is made of medical-grade elastic polymer materials, such as silicone rubber or polyurethane, combining flexibility with sufficient mechanical strength. Silicone rubber has a Young's modulus of approximately 0.001–0.05 GPa (i.e., around 1–50 MPa), while polyurethane or nylon have Young's moduli ranging from several hundred MPa to several GPa. The catheter's outer diameter is, for example, 10–15 mm, and its length can reach 1 m to accommodate colonic anatomy. The material's thermal conductivity is approximately 0.2 W / (m·K) to ensure moderate heat conduction and avoid localized overcooling or overheating. A hydrophilic coating may be applied to the outer surface of the catheter to reduce frictional resistance, lowering the coefficient of friction in the mucosa from over 1.0 to 0.01–0.05, thus reducing insertion force and improving patient comfort.

[0043] Support skeleton: A flexible support structure embedded within the catheter body wall to enhance the catheter's shape retention and bending resistance. The skeleton can be made of stainless steel or nickel-titanium alloy braided mesh or spiral winding. The braided mesh provides axial support and torsional rigidity through the interlacing of fine steel wires, while the spiral winding improves bending and kinking resistance. Skeleton structural parameters (such as wire diameter, braiding density, and pitch) can be adjusted as needed in different sections of the catheter to achieve a gradient stiffness design with a harder proximal end and a softer distal end, balancing both pushing and flexibility.

[0044] Chord actuation mechanism: Multiple thin steel wires or polymer cables (similar to biological tendons, hence the name chordae) are arranged axially parallel within the catheter body, typically 3-4 symmetrically distributed evenly along the circumference of the catheter cross-section. One end of each chord is fixed to a connecting ring at the distal end of the catheter, and the other end is connected to an actuator at the proximal end of the catheter. By controlling the traction length of each chord, multi-degree-of-freedom bending and orientation of the distal end of the catheter can be achieved. A low-friction bushing (such as PTFE tubing) is used to cover the chordae and the outer sheath of the catheter to form a tendon sheath structure, reducing frictional loss during long-distance force transmission. This actuation mechanism can consist of a stepper motor and a winding reel, enabling precise control of the tension of each chord and adjustment of the catheter's posture.

[0045] Tension Sensor Array: Multiple tension / compression sensors are integrated near the distal end of the catheter to form a sensor array for real-time monitoring of the tension of each chordae tendineae and the external force generated by the catheter's contact with the intestinal wall. Thin-film force-sensitive resistors (FSRs) are preferred due to their small size, thin thickness (approximately 0.2 mm), and ease of integration. Six FSRs are evenly arranged in a ring around the distal end of the catheter. When the chordae tendineae are under tension or the catheter is under external pressure, the pressure is transmitted to the corresponding FSR via a force-transmitting block, generating an electrical signal proportional to the magnitude of the force. This array is installed near the distal end of the catheter, reducing measurement errors caused by chordae tendineae friction due to proximity to the point of action. The array has an outer diameter of approximately 16 mm and a length of 20 mm, making it very compact and suitable for passage through the colon (the adult colon's internal diameter is typically 26–45 mm). The sensor sensitivity range covers tension variations of 0–4 N, with a force measurement accuracy error of less than 0.2 N, enabling precise capture of minute changes in chordae tendineae tension and catheter contact force.

[0046] Temperature sensors: Temperature sensors are placed at the distal and middle sections of the catheter to monitor the temperature of the local environment and the perfusion fluid. Digital thermistors or miniature thermocouples are preferred, with a measurement range of 0–100 °C and an accuracy of ±0.1 °C. The distal temperature sensor is close to the outer surface of the catheter tip and reflects temperature changes in the area where the catheter contacts the intestinal wall; the middle sensor monitors the temperature of the delivered fluid. The sensors are waterproof and integrally molded with the catheter wall, without affecting the catheter's flexibility. Temperature data can be used to assess the risk of local overheating and the degree of patient discomfort.

[0047] Irrigation pump mechanism: Connected to the proximal end of the catheter, it is used to pump perfusion fluid (such as warm saline, normal saline, etc.) into the catheter at a controlled flow rate. The pump mechanism can employ a reversible peristaltic pump or an injection pump to achieve constant flow rate perfusion and, when necessary, aspiration. The pump flow rate is adjustable within the range of 0–500 mL / min to adapt to different clinical needs (irrigation or drainage). A pressure sensor and a one-way check valve are connected to the pump head outlet to monitor the hydraulic pressure in the tubing in real time and prevent backflow. The pump speed can be precisely controlled via a central control module to adjust the injection fluid rate and pressure gradient, achieving gentle perfusion.

[0048] Central Control Module: The brain of the system, employing an embedded microprocessor or industrial control computer, it realizes sensor signal acquisition, data processing, control algorithm calculation, and actuator drive. The module includes signal conditioning circuitry and an analog-to-digital converter, converting analog signals from the tension sensor array and temperature sensor into digital data. Built-in memory stores control algorithms and parameter thresholds, and calculates the M-Risk index and control output in real time. The module controls the chordae tendineae drive motor and infusion pump via drive circuitry, forming a closed-loop control system. The central control module also provides a human-machine interface for parameter setting and status display, with wired or wireless communication capabilities for convenient monitoring and intervention by physicians. The module is equipped with a safety power supply and backup battery to ensure a safe and orderly system shutdown in the event of a sudden power outage.

[0049] The above components together constitute a closed-loop perfusion system that integrates perception, decision-making, and execution. Its structural features include flexibility and controllability, sensor fusion, and high module integration, enabling it to adapt to the complex environment inside the colon and ensure operational safety.

[0050] Control closed loop and algorithm details

[0051] This system employs a closed-loop control strategy, utilizing multi-sensor information fusion and intelligent algorithms to achieve autonomous adjustment and risk control of the infusion process. Its control flow and key algorithm points are as follows:

[0052] Sensor signal acquisition and preprocessing: The central control module polls and reads the values ​​from each channel of the tension sensor array and the temperature sensor at a high frequency (e.g., 100 Hz). The acquired signals are denoised, filtered, and zero-drift calibrated to obtain accurate real-time tension, pressure, and temperature data. Simultaneously, based on the catheter structure model, the multi-point tension values ​​in the array can be converted into estimates of the magnitude and direction of the force at the catheter tip. The risk assessment unit calculates the combined risk index M-Risk using the following example formula: Among them, P~ max σ~P is the normalized value of the maximum contact pressure within the sliding time window, σ~P is the normalized value of the variance of the tension of the annular sensing belt, and ΔP~ ring T is the normalized value of the tension difference between adjacent annular sensing strips. mucosa Let dT be the normalized deviation of the mucosal temperature detected by the first temperature sensor relative to the target temperature, dT be the normalized value of the mucosal temperature change rate, and w1~w5 be the weighting coefficients of the corresponding features, and satisfy the following conditions: .

[0053] SVM Classification and Recognition: To further improve the accuracy and relevance of risk assessment, this system introduces a Support Vector Machine (SVM) model to classify sensor features into patterns. The SVM model uses tension distribution, temperature distribution, and the M-Risk index as input features and outputs the risk category of the current state. For example, the state can be divided into three levels: safe state, warning state, and dangerous state. The SVM model is trained offline and can identify state patterns under complex nonlinear relationships. When the system is running, the SVM classifies the feature vector at each new moment: if the classification result is a safe state, normal perfusion continues; if it is a warning state, preventative control measures are implemented; if it is a dangerous state, emergency protection actions are triggered. SVM classification complements the continuous M-Risk index to avoid missed or false alarms that may occur with a single threshold judgment. For example, when the M-Risk approaches the threshold, the SVM can issue an early warning based on the feature trend, thereby improving the timeliness of the response.

[0054] Dynamic Threshold Adjustment: Considering individual patient differences and changes in the procedure's progression, the system employs a dynamic threshold adjustment mechanism. The central control module adaptively modifies certain control threshold parameters based on real-time data and SVM classification results. For example, in the initial stage of preoperative irrigation, when patients are not yet accustomed to catheter stimulation, the M-Risk trigger threshold can be set lower for greater caution. As the procedure progresses, if the system detects no abnormalities for a period (frequent occurrences of safe states), certain thresholds can be appropriately increased to improve efficiency. Furthermore, in cases where the tension sensor baseline drifts over time, the system can dynamically reset the tension zero point and adjust reference values ​​for risk criteria such as 5 kPa contact pressure. Dynamic threshold adjustment makes the control strategy adaptive, avoiding unnecessary interruptions or missed risks caused by rigid threshold settings, and ensuring that the entire irrigation process remains in an optimized balance of safety and efficiency.

[0055] Feature-weighted strategy: In the risk index calculation, a feature-weighted strategy is introduced to dynamically adjust the contribution weight of each feature in the M-Risk based on different risk scenarios. When the SVM determines that the current main risk factor is mechanical stress (such as excessive local pressure on the conduit), the system automatically increases the weight of features related to tension and pressure and decreases the weight of temperature features, making the M-Risk more sensitive to mechanical risks. Conversely, if a rapid temperature rise is detected and the SVM identifies it as a thermal risk, the weight of temperature features is increased. This feature-weighted strategy ensures that the M-Risk index responds more sensitively to the main issues and reduces the interference of secondary factors. In addition, the system can continuously adjust the weights based on historical control effects, gradually optimizing the risk assessment model. For example, in the initial stage of use, the tension feature weight w1 can be higher to ensure safety, while after multiple runs without pressure problems, w1 can be slightly reduced to reduce unnecessary downtime caused by over-protection. Feature weight adjustments are performed in the central control module according to a set cycle (such as every 5 minutes) or specific events (such as risk state transitions), with an upper limit on the adjustment range to ensure a smooth transition.

[0056] Local Overheat Control Module: To prevent tissue burns or thermal discomfort, a dedicated local overheat monitoring and control module is installed within the control system. This module reads data from multiple temperature sensors in real time, monitoring the temperature of various parts of the catheter and surrounding tissues. When any point is detected to have a temperature exceeding a safe threshold (e.g., 42 °C) or an abnormally high rate of temperature rise, the module immediately intervenes. Its algorithm logic includes: calculating an overheat risk score based on the degree and duration of temperature exceedance, and combining this with the M-Risk index to determine whether a thermal risk event is confirmed; once confirmed, it executes predetermined cooling measures, such as automatically reducing the perfusion fluid temperature (if the perfusion fluid has a heating function, lowering the heater setting; otherwise, injecting room temperature saline for heat absorption and cooling), temporarily reducing or stopping the perfusion flow rate to reduce heat input, and notifying the chordae tendineae drive module to slightly move the catheter away from the high-temperature area to promote heat dissipation. Local overheat control is a fast-track channel in a closed loop; its triggering does not depend on the overall M-Risk threshold, but rather responds immediately to any temperature anomaly, prioritizing ensuring that tissues are not damaged by heat. The module also features a temperature hysteresis recovery function: after the temperature returns to normal and remains at a safe level for a period of time, the cooling measures are automatically lifted, and the original injection parameters are restored, thereby minimizing the impact on the normal process.

[0057] Control Decision and Execution: Based on the aforementioned risk index, SVM classification, and the outputs of various specialized monitoring modules, the central control module generates final control commands and sends them to the infusion pump and chordae tendineae actuators. Control decisions follow a "safety first" principle: maintaining operational continuity as much as possible when risks are controllable, and prioritizing protective actions when risks increase (see the next section, "Control Feedback Strategy," for details). For example, if the M-Risk remains low and the SVM determines it is safe, the system maintains the current pump speed and catheter posture; if a warning state is entered, the system may preemptively reduce the infusion rate and adjust the catheter bend to alleviate potential pressure concentration; if a dangerous state is reached or a specific criterion (such as a pressure or temperature threshold) is triggered, a series of safety actions are immediately executed, including pausing infusion, pressure relief, and alarm notification. The entire closed-loop control cycle repeats every very short period (e.g., 10 ms) to ensure timely capture of changes and rapid response, achieving real-time intelligent control of the catheter infusion process.

[0058] Contact force and temperature rise model and tissue damage criteria

[0059] To achieve quantitative risk assessment and safety control, this system establishes a model of catheter-intestinal wall interaction and the impact of perfusion on tissues, and determines the threshold for tissue damage based on literature and experiments.

[0060] Catheter-Intestinal Wall Contact Force Model: This invention employs elasticity and geometric nonlinear models to describe the force relationship between a flexible catheter and the colonic wall. The distal end of the catheter is approximated as a compliant beam. When the tendineae pull, causing the catheter to bend and press against the intestinal wall, a normal pressure P and tangential frictional force are generated in the contact area. The model assumes that the intestinal wall can be macroscopically considered as an elastic support, and its compliance is determined by the equivalent elastic modulus E of the intestinal wall. colon The force is determined by both the duct and the intestinal wall thickness. Under the simplified planar beam contact theory, the normal contact force F between the duct and the intestinal wall can be expressed as... n The tension T of the tendineae can be approximately expressed as:

[0061]

[0062] Where θ is the conduit bending angle at the contact point, and L c Let P be the effective contact length between the catheter and the intestinal wall. When the contact surface is subjected to approximately uniform force, the local average pressure P can be expressed as:

[0063]

[0064] Among them, A c Let be the area of ​​contact between the catheter and the intestinal wall. Considering the influence of friction between the tendineae and tendon sheath on the tension of the tendineae during transmission along the catheter, a modified Capstan formula can be introduced to correct the tension of the tendineae, that is, to model the attenuation of the tension of the tendineae along the path length l as... For the proximal driving end tension, T eff Let μ be the effective tension at the distal end, μ be the coefficient of friction, and φ be the chordae tendineae bending angle. By calibrating the parameters in the above model using data from isolated porcine intestine experiments, the contact force F can be estimated under known conditions of duct bending radius r and chordae tendineae tension T. n On the one hand, the tension sensor array on the distal outer wall of the catheter provides multi-point contact pressure information. By comparing the differences in the outputs of each annular sensing band and each circumferential sensing element, the force field of the intestinal wall can be "inverted" to obtain the magnitude and distribution of the intestinal wall force under different contact postures. In summary, the key variables of the catheter-intestinal wall contact force model include: chordae tendineae tension T, catheter radius of curvature r, and equivalent intestinal wall stiffness k. colon (can be made by E) colon The model includes parameters such as wall thickness conversion, friction coefficient μ, and contact length Lc. The control module of this invention utilizes this model to predict the pressure exerted by the catheter on the colonic wall based on given drive inputs (tendon tension and bending angle) and real-time tension sensing data. The predicted contact pressure P is used as an important source of tension-related features in the joint risk index M-Risk for risk assessment and safety control triggering.

[0065] Perfusion Temperature Rise Prediction Model: This invention further establishes a mathematical model for heat transfer between the perfusion fluid and colonic tissue in vivo, used to predict the local temperature change trend of the colon, providing a basis for overheating risk assessment and temperature control strategies. The model is based on the principles of energy conservation and biological heat transfer: the initial temperature Tin and volumetric flow rate V˙ of the perfusion fluid are known. It enters the colonic lumen through the catheter's outlet, contacts the intestinal wall mucosa for heat exchange, and gradually approaches thermal equilibrium. Assuming that a certain local intestinal wall region can be equivalently represented as a flat plate of tissue with blood perfusion, its temperature field evolution can be expressed by the Pennes biological heat transfer equation: Where ρ and c are the tissue density and specific heat capacity, respectively, k is the thermal conductivity of the tissue, and ω b ρ is the blood perfusion rate. b and c b These are blood density and specific heat capacity, respectively, T b Q represents blood temperature (approximately 37°C). perf This represents the equivalent internal heat source term of the perfusion fluid in this region. For ease of real-time calculation, in the engineering implementation of this invention, the intestinal wall within a certain depth range around the catheter is considered as a lumped "heat capacity block," establishing a one-dimensional lumped heat transfer model:

[0066] Among them, T tissue T represents the average temperature of the intestinal wall in that region. fluid Here, is the temperature of the perfusion fluid, h is the convective heat transfer coefficient, and A is the effective contact area between the perfusion fluid and the intestinal wall (m² / c²). t f is the total heat capacity of this local tissue. blood(⋅) is a simplified blood flow cooling correction term, used to characterize the effect of blood flow carrying away heat. Model parameters: h, A, mc t and f blood The specific form can be obtained through literature references and in vitro experimental calibration. Using the above model, when the current inlet temperature T of the perfusion fluid is input... in Real-time measured outlet liquid temperature T of the conduit fluid Perfusion flow rate V˙ and current tissue temperature T tissue When parameters are equal, the central control module can solve for the predicted curve of intestinal wall temperature over a short period of time in discrete time steps, thereby determining whether the tissue temperature will exceed a preset safety threshold (e.g., 42℃) under the current perfusion conditions. If the model predicts that at a certain time point t+Δt, T... tissue If the temperature exceeds the set temperature threshold, the control module can issue adjustment commands in advance, such as lowering the set temperature of the perfusion fluid, reducing the perfusion flow rate, or temporarily stopping the perfusion, so that the actual tissue temperature is always kept within a safe range. The key variables in the temperature rise prediction model of this invention include: perfusion fluid temperature T. fluid Inlet temperature T in Perfusion flow rate V˙, current tissue temperature T tissue Tissue density ρ, tissue specific heat c, tissue thermal conductivity k, and blood perfusion-related parameters ω b By introducing this model, the system can predict risks before tissue temperature actually rises to dangerous levels, transforming "overheating" risk from post-event detection to pre-event prediction and proactive control.

[0067] Tissue damage criteria: Combining medical literature and experimental results, multiple criteria were developed to determine whether a tissue is at risk of injury, and these criteria serve as triggering conditions for control strategies.

[0068] Pressure threshold criterion: If the contact pressure of the catheter against the intestinal wall consistently exceeds 5 kPa (approximately 37.5 mmHg), a risk of local ischemia is considered, and decompression measures should be taken immediately. Studies have shown that in colonic dilatation experiments, subjects experience significant pain and discomfort when the internal pressure reaches 5–6 kPa; sustained pressure exceeding this level can lead to obstruction of blood supply to the intestinal mucosa, potentially causing edema and erosion over time. Therefore, 5 kPa was selected as the pressure risk threshold. When the contact pressure calculated by the tension sensor exceeds this value and persists for several seconds, M-Risk increases significantly and triggers corresponding control actions.

[0069] Edema Criteria: If the local pressure, while not extremely high, remains at a moderately high level for an extended period (e.g., >3 kPa for several minutes), accompanied by an abnormal increase in temperature in the area or a slow, upward trend in the catheter stress curve, then tissue edema may be suspected. Edema causes tissue swelling and decreased elasticity, resulting in a gradual increase in tension sensor readings under the same pressure. The control module identifies edema by detecting changes in the tension-displacement curve (e.g., a gradual increase in tension under the same chordae tendineae displacement). Once the criteria are met, the system reduces the perfusion rate and extends the pause time to allow tissue recovery. Furthermore, edema may indicate an inflammatory response, and the control module may prompt the application of local anti-inflammatory treatment.

[0070] Perforation Criteria: Colonic perforation is one of the most serious risks. Literature reports that animal colons rupture at internal pressures of approximately 16.7–21.3 kPa, with an average critical pressure of approximately 19.8 kPa. To provide sufficient margin to avoid approaching the rupture point, this system defines a perforation risk index P. erf :P erf = Current intestinal pressure / 20 kPa. When P erf A pressure >0.8 (equivalent to an internal pressure >16 kPa) is considered a high-risk area for perforation. Since this system does not directly install an intestinal pressure sensor, P... erf It can be estimated by combining the tension sensor and the pump pressure sensor: using the hydraulic pressure P at the conduit outlet. pump And the contact pressure estimated by the tension model infers the trend of intraluminal pressure. When the continuously estimated intraluminal pressure approaches 15 kPa (P erf (Approximately 0.75) The system will issue an alarm and forcibly stop infusion, initiating an automatic pressure relief procedure. If the pressure does not drop within a predetermined time, it is assumed that perforation or severe overpressure may have occurred, the system enters an emergency state, completely stops all pumping and movement, and notifies medical personnel for immediate examination and treatment. The introduction of perforation indicators ensures that the system can take timely measures when approaching the danger zone, preventing actual tissue rupture accidents.

[0071] The aforementioned models and criteria provide quantitative basis for the control module. For example, the contact force model and pressure threshold criterion can determine when it is necessary to relieve the pressure of the catheter on the intestinal wall; the temperature rise model and temperature threshold can prevent thermal damage; and the perforation index can prevent fatal damage caused by overfilling. These criteria are directly related to the control strategy, and their application will be explained in detail in the feedback control section of the next section.

[0072] Control Feedback Strategy

[0073] Based on the risk assessment results obtained through real-time calculation, the central control module adopts a hierarchical feedback control strategy, dynamically executing a series of protective actions to ensure safety. The main control actions and their triggering logic are as follows:

[0074] Perfusion Suspension: When a risk indicator reaches a dangerous threshold or a sudden abnormality occurs, the system first and foremost stops the perfusion pump to prevent further fluid delivery or pressure buildup. This is a basic safety response. For example, if the catheter tip tension is ≥2 N and lasts for more than 10 seconds, the central control module determines that the tissue may be under excessive stretching load and will immediately suspend perfusion. Similarly, if the temperature sensor reading exceeds a preset upper limit (e.g., 42 °C), it will also trigger pump shutdown. Suspending perfusion buys time for subsequent measures and prevents the situation from worsening. During the suspension, the system continues monitoring until the risk index drops to a safe range and the cause is identified and eliminated, at which point perfusion can be resumed with the confirmation of medical personnel.

[0075] Catheter Posture Adjustment: The central control module adjusts the catheter's posture and position within the intestine based on force direction information provided by tension sensors to alleviate localized high pressure or adverse contact. When the chordae tendineae tension in a certain direction is persistently excessive and accompanied by high sensor readings on that side, the system determines that the catheter may be pressing against the intestinal wall at an unfavorable angle. In this case, the control module coordinates the chordae tendineae drive, slightly altering the catheter's bending direction or curvature to move the catheter tip away from the high-pressure area. For example, if excessive pressure is generated on the left side of the intestinal wall at the catheter tip, the left chordae tendineae are relaxed, and the opposite chordae tendineae are tightened, causing the catheter tip to turn slightly to the right, thereby reducing the contact force on the left side. Similarly, in postoperative drainage situations, if changes in patient position cause increased torsional tension on the catheter, the system can notify the patient or automatically adjust the external catheter suspension to restore a neutral posture. Posture adjustment is typically triggered when the risk index is moderate (warning state) as a preventative intervention to avoid escalating to the point where emergency cessation is required. The entire adjustment process is limited in scope and slow and smooth to prevent sudden movement from causing new discomfort or risks.

[0076] Chord-driven deformation: When local pressure is too high or the catheter shape needs to be changed to fit the lumen, the system activates the chord-driven module to actively control the deformation of the catheter shape. This is similar to posture adjustment, but focuses on changing the catheter's local bending radius and wall adhesion, rather than its overall direction. For example, during colonic irrigation, if excessive dilation of a segment of the intestine is detected and the catheter is applying significant pressure, the control module can appropriately reduce the catheter's bending curvature, allowing the catheter to adhere more gently to the wall and distribute the pressure more evenly. Chord-driven deformation can also be used to reduce the catheter diameter when necessary: ​​by uniformly tightening all chordae, the catheter is slightly thinned to increase the gap between the catheter and the intestinal wall, releasing stagnant fluid or gas and achieving a dynamic pressure relief effect. Conversely, to increase the support area of ​​the catheter tip against the intestinal wall and reduce pressure per unit area, the system can also briefly tighten specific chordae to form a slight "S" shaped bend at the tip, thereby increasing the contact area to distribute the force. All deformation adjustments caused by chord-driven deformation are calculated by the central control module and are automated, subtle actions that patients often do not notice but can effectively reduce local stress peaks.

[0077] Dynamic decompression: When a gradual increase in intraluminal pressure is detected and approaches a warning value (e.g., P), erf (If the pressure index rises significantly), the system will proactively perform a slow decompression process instead of passively waiting for high pressure to trigger an emergency stop. Dynamic decompression can be achieved in two ways: reducing the infusion rate or opening the drainage channel. First, the peristaltic pump can be controlled to reduce its speed or even reverse, drawing back some of the infusion fluid and gradually reducing the intestinal pressure. Second, if the catheter is designed with a dedicated drainage or pressure relief channel, the solenoid valve can be opened to allow the intestinal contents to flow out slowly until the pressure drops below a safe level. Compared to instantaneous emergency stop, dynamic decompression can release pressure smoothly, avoiding discomfort or other risks caused by a sudden drop in pressure (e.g., rapid pressure relief may cause intestinal spasms). Dynamic decompression is usually initiated in a warning state; for example, the system gradually decompresses when the pressure threshold of 5 kPa is just reached, without waiting for higher risks to occur. The central control module automatically adjusts the decompression amplitude according to the pressure drop rate, ensuring sufficient decompression while preventing excessive venting from affecting operational continuity. If the risk is eliminated after decompression, the system will resume the normal infusion process; if decompression cannot control the pressure from continuing to rise, it will escalate to suspending infusion or even emergency shutdown.

[0078] The strategies described above can be used in combination. For example, during preoperative colonic irrigation, the system may first detect a gradual increase in pressure and reduce the pump rate to ease the pressure. If the pressure still approaches the threshold, irrigation is paused and some fluid is drained. During this process, the catheter's posture and shape are simultaneously adjusted to optimize force distribution. All actions are triggered based on real-time risk output and are automatically decided by the central control module. Importantly, these feedback strategies all have built-in safety verification mechanisms: after each action is executed, the system continuously monitors the trend of risk index changes, evaluates the effect of the action, and takes further measures or restores the original state if necessary. For example, if the risk is quickly resolved and there is no recurrence trend after pausing irrigation, the system can orderly resume irrigation; conversely, if the risk persists, and an alarm is triggered, the system remains paused awaiting manual intervention.

[0079] In addition, the system is designed with redundant safety protections: in the event of automatic control failure or unforeseen circumstances, such as sensor malfunctions preventing timely protection activation, the equipment is equipped with hardware pressure relief valves and over-temperature circuit breakers as a final barrier. Once the internal pressure or temperature exceeds the absolute safety limit, these hardware protections will automatically activate (such as mechanical depressurization or power cut-off) to ensure patient safety.

[0080] Typical clinical scenarios and usage procedures

[0081] The following section uses two typical clinical application scenarios to explain how to use this system, the characteristics of sensor data changes, the control logic, and the safety measures.

[0082] Scenario 1: Preoperative colonic irrigation

[0083] Background: Before colorectal surgery, thorough colonic irrigation is often necessary to empty the intestinal lumen and reduce the risk of infection. This system's flexible catheter is inserted preoperatively through the anus into the colon to automatically perform the enema process. Patients are typically in the left lateral decubitus or lithotomy position, and may be awake and aware of discomfort.

[0084] Procedure: Under image guidance, medical staff slowly insert the soft tip of the catheter into the colon through the anus, reaching the predetermined depth (e.g., the transverse colon). The central control module's preset "colon irrigation mode" is activated: a peristaltic pump begins injecting warm saline (approximately 37°C) at a low speed (e.g., 100 mL / min), while tension and temperature sensors monitor the catheter's status in real time. For the first few minutes, sensor data remain stable: the catheter advances along the intestinal curves, and the tension array readings are low and uniform, indicating that no significant pressure is being applied by the catheter; the temperature is close to body temperature, approximately 36–37°C.

[0085] As saline solution was continuously infused, the colon gradually filled and expanded, and the internal pressure slowly increased. Sensor changes were observed: increased pressure in front of the catheter tip led to a slight increase in the contact force with the intestinal wall, manifested as a slow rise in readings of several channels of the tension sensor array to the level of several hundred millinewtons; if a segment of the intestine began to expand and resist the catheter, a more significant increase in the chordae tendineae tension was observed in the corresponding direction. Regarding temperature, because the saline temperature was close to body temperature and the infusion rate was moderate, the intestinal wall temperature remained essentially constant, with occasional fluctuations of ±1 ℃, indicating no risk of overheating.

[0086] Control Logic: The central control module continuously calculates the M-Risk index. Initially, due to low values ​​of various characteristics, the M-Risk remains at a safe level below 0.2. Following a preset protocol, the system pauses briefly after injecting a certain volume (e.g., 500 mL) to allow fluid and excretion to drain through the catheter or via intestinal peristalsis, simulating the steps of segmented irrigation in clinical practice. As irrigation continues, if the M-Risk rises to near the warning threshold of 0.5 (mainly due to increased internal pressure and tension), the system automatically takes measures: first, it dynamically depressurizes by slightly reducing the pump rate to 50 mL / min, while simultaneously partially opening the drain valve on the catheter to slowly release excess fluid and gas from the intestinal lumen. This slightly reduces the intestinal pressure, and the tension sensor reading stops rising or even falls back. Subsequently, the system adjusts the catheter posture; if it detects high pressure on the left side of the sigmoid colon, the chordae tendineae drive rotate the catheter tip a few degrees to the right to reduce stress on the left intestinal wall.

[0087] Throughout the irrigation process, the sensor data exhibits a periodic variation: pressure and tension gradually increase during each injection phase, triggering a drainage phase when approaching the threshold, followed by a decrease in readings, and this cycle repeats. The M-Risk index fluctuates accordingly, but is always kept below the threshold. Patients may report mild abdominal distension and discomfort during injection, which is relieved after the system drains and depressurizes. This closed-loop control ensures thorough cleaning while avoiding over-distension that could lead to severe pain or other risks. If the M-Risk unexpectedly spikes close to 1 during a cycle (e.g., due to a sudden increase in intestinal pressure caused by the patient holding their breath or straining), the system immediately pauses irrigation and performs emergency depressurization: the pump stops, the drainage valve opens to rapidly depressurize to a safe level, and an audible and visual alarm alerts healthcare professionals. Once the risk has been eliminated and confirmed manually, subsequent steps can continue.

[0088] Safety Measures: In the preoperative irrigation scenario, the system's safety measures include: 1) Multiple threshold control: It features both continuous M-Risk monitoring and hard thresholds such as 5 kPa pressure and 42 ℃ temperature, providing double protection for timely response; 2) Automatic pressure relief function: A built-in mechanical pressure relief valve automatically opens at approximately 13 kPa pressure to relieve pressure if abnormal pressure is not handled by the software, preventing excessive pressure accumulation; 3) Integration of patient subjective feedback: If the patient cries out in pain or exhibits significant discomfort, medical staff can quickly press the "Emergency Stop" button through the control interface. The system immediately stops and empties, prioritizing patient feedback (see the subjective discomfort index below). Through these measures, the irrigation process achieves the goal of cleansing the intestines while minimizing discomfort and risks.

[0089] Scenario 2: Postoperative Drainage

[0090] Background: Following colorectal surgery, such as anastomotic surgery or laparoscopic bowel resection, catheters are often placed to drain fluid and blood from the surgical site, reduce intestinal tension, and prevent anastomotic leakage and abdominal infection. This system can be used as a postoperative internal drainage tube, maintaining drainage patency and safety through sensing and control. Postoperatively, patients are often not fully awake from sedation or anesthesia, have reduced pain sensation, and cannot promptly respond to discomfort; therefore, automated monitoring is even more important.

[0091] Procedure: Before the end of the surgery, the surgeon inserts a flexible catheter into the drainage site. For example, this can be done transrectally to the proximal intestinal lumen at the anastomosis, or via an abdominal drainage port to the pelvic fluid accumulation area. The catheter is connected to the "postoperative drainage mode" of the central control module. Initially, the infusion pump stops, and the catheter is in open drainage mode, allowing fluid to drain naturally by gravity or tissue pressure differential. The tension sensor primarily monitors the pressure on the catheter: when there is no obstruction, the sensor reading is close to zero or has only minimal static tension (due to the catheter's own weight or slight contact with tissue), and the signals from each channel are stable. The temperature sensor reading reflects the body cavity temperature, approximately 36–37 °C, and can be used as one of the references for the patient's core body temperature.

[0092] As time progresses post-surgery, drainage problems or increased fluid accumulation may occur: if the drainage tube is partially blocked by blood clots or viscous fluid, fluid outflow is impeded, and pressure at the drainage site (such as the intestinal lumen) will rise. The system will detect this sign—abnormal patterns in the tension sensor array readings: the tension in some channels slowly increases, indicating that the catheter is being pushed by internal pressure or compressed by tissue expansion. For example, gas and fluid accumulation in the intestinal segment near the anastomosis can expand, pushing against the catheter and creating additional pressure, resulting in an overall increase in readings at multiple sensor points distributed in a ring on the array. At the same time, the temperature sensor may detect a local temperature increase of 1–2 °C, which may be due to an inflammatory response or increased blood flow. The M-Risk index thus rises from near 0 to 0.4–0.6, indicating the need for a warning.

[0093] Control Logic: The central control module executes a series of unblocking and relief measures for cases of poor drainage. First, it initiates mild negative pressure aspiration: a low-speed reverse peristaltic pump generates a small negative pressure (e.g., -50 mmHg, lasting several seconds) to attempt to suction out the blockage from the catheter. If the tension sensor reading rapidly decreases and returns to near zero with the application of negative pressure, it indicates that the blockage has been cleared and the pressure has been relieved. The system records the intervention as successful, M-Risk returns to a safe value, and monitoring continues. If the tension does not improve after negative pressure or even continues to rise, the system determines that the blockage has not been relieved or there is a significant increase in fluid accumulation. At this point, the control strategy is upgraded—small-dose irrigation flushing: a small amount of physiological saline (e.g., 20 mL each time) is injected forward with the pump, intermittently and pulsatiatingly impacting the inner wall of the catheter to mechanically dislodge the blockage. Simultaneously, the chordae tendineae drive slight vibrations of the catheter, using the movement of the catheter tip to disturb the blockage. During flushing, the system closely monitors tension and pressure feedback: once a sudden decrease in resistance (a sharp drop in tension reading) and fluid outflow are detected, it indicates successful unblocking, and flushing is immediately stopped, switching to drainage mode to allow the accumulated fluid to drain from the body. The entire unblocking process is completed automatically under closed-loop control, which can solve most problems of poor drainage without human intervention.

[0094] During normal continuous drainage, changes in sensor data are closely related to the patient's condition: for example, changes in patient position or mild coughing may cause traction on the catheter, resulting in a brief peak in tension. However, the system absorbs the impact by relaxing the chordae tendineae, causing the tension peak to decrease and quickly subside. Similarly, if the patient develops a fever or a worsening local infection, the temperature sensor reading may gradually rise to 38–39 °C. Based on this, the central control module alerts medical staff to check the wound and may adjust treatment accordingly. The M-Risk index generally remains low (<0.3) under stable drainage conditions, only increasing and triggering corresponding controls when abnormalities such as blockage or drastic changes in position occur.

[0095] Safety Assurance: In postoperative drainage mode, the system employs multiple measures to ensure patient safety: 1) Tension Limit Protection: Prevents the catheter from being accidentally pulled out or tearing tissue. If the tension suddenly increases beyond a certain upper limit (e.g., 2 N), the system immediately locks the catheter at its current length and sounds an alarm to prevent further pulling that could tear the wound. 2) Negative Pressure Limitation: Negative pressure suction has a strict upper limit (e.g., not exceeding -100 mmHg) and a limited duration to prevent excessive suction from damaging newly sutured tissue or mucosa. 3) Redundant Channels: The catheter is designed with multiple side holes, allowing partial drainage even if one is blocked; simultaneously, the system automatically fine-tunes the catheter position periodically to prevent it from adhering to the wall and creating a vacuum. 4) Continuous Monitoring and Alarms: Due to the poor subjective perception of postoperative patients, the system continuously monitors sensor data 24 hours a day, immediately issuing an alarm to notify nursing staff for inspection if any abnormal trend is detected. Through these mechanisms, the drainage tube can operate safely even without human supervision, preventing harm to the patient due to blockage or displacement.

[0096] Subjective discomfort index prediction model

[0097] To further enhance the system's intelligence and user-friendliness, this invention incorporates the prediction and control of patients' subjective discomfort levels. The subjective discomfort index quantifies the pain or discomfort experienced by patients during catheter procedures, and is primarily based on a model constructed according to the correlation between patients' subjective VAS score data and objective sensory characteristics.

[0098] Data Acquisition and Modeling: The Visual Analogue Scale (VAS, 0 for no discomfort and 10 for unbearable pain) is commonly used to evaluate the intensity of pain and discomfort perceived by patients. In the system's clinical trials, we recruited volunteers or patients who periodically reported their subjective VAS scores during colonic irrigation or drainage procedures. Simultaneously, the system recorded sensor data such as tension and temperature, as well as control parameters, at the corresponding times. Statistical analysis revealed a correlation between patient VAS scores and objective physical quantities: for example, higher colonic pressure or catheter tension often correlated with stronger reported abdominal distension and pain; and higher catheter temperature may increase burning discomfort. Specifically, in the colonic dilation experiment, the subjective pain score of the VAS increased significantly with increasing intestinal pressure. Based on the collected historical dataset, we used machine learning regression to build a VAS prediction model. Input features included: average contact pressure, maximum tension of the chordae tendineae, rate of tension change, local temperature and its changes, etc.; the output was the predicted VAS score. Support vector regression (SVR) or multiple linear regression were used to fit the model, both yielding good results. For example, an empirical formula was obtained through fitting. Where P avg F represents the average pressure at which the conduit contacts. peakΔT represents the peak tension of the chordae tendineae, ΔT is the difference between the current temperature and body temperature, and a, b, c, and d are the fitting coefficients. The model achieved a high R² value on the validation data, indicating that it can accurately estimate VAS. Given individual tolerance differences, the model can be calibrated for patients before practical application: record some initial patient responses and adjust the model output to ensure the predictions are consistent with their subjective reports.

[0099] Decision Fusion: The introduction of the Subjective Discomfort Index (VAS) allows the control module to consider patient comfort in addition to physical safety. During control decisions, the central control module considers VAS alongside the M-Risk. For example, the system can set a target comfort threshold, such as a VAS not exceeding 4 (mild discomfort is tolerable). When the predicted VAS approaches or exceeds this threshold, even if the M-Risk is still within a safe range, the system will take preventative measures in advance. These measures include: reducing the perfusion rate, pausing the procedure to allow the patient to rest, or adjusting the chordae tendineae to reduce stimulation of sensitive areas. For instance, during preoperative irrigation, even if pressure and other parameters are not yet excessive, if the model predicts a VAS of 5, the system will temporarily halt irrigation and increase drainage to alleviate the patient's distending pain, continuing only when the estimated VAS drops to an acceptable level. This discomfort index-based regulation effectively reduces patient subjective suffering and improves procedural tolerability.

[0100] Furthermore, VAS can be used for personalized control strategies: for patients with high pain sensitivity, the system will dynamically lower various risk thresholds (e.g., lower the M-Risk trigger value), while for those with high tolerance, the thresholds can be appropriately increased to improve efficiency. In the long term, the system can continuously optimize the VAS prediction model through the accumulation of a large amount of case data, making it more accurate for different populations. Even in the future, combining physiological feedback (such as heart rate and blood pressure) can further improve the "subjective discomfort index," achieving more comprehensive patient status monitoring. In summary, with subjective discomfort prediction and control, this system not only focuses on "non-invasiveness" but also strives to achieve the intelligent medical goal of "minimally painful" treatment.

[0101] In summary, the flexible catheter irrigation system provided by this invention, through its ingenious structural design, comprehensive sensor arrangement, and advanced closed-loop control algorithm, achieves intelligent and safe control of preoperative colonic irrigation and postoperative drainage. The detailed implementation of the system discloses the composition, parameters, and working principles of each module, effectively reducing clinical operational risks and improving patient safety and comfort, thus possessing significant practical application value. The components and methods are not limited to the above embodiments; various equivalent substitutions or modifications can be made by those skilled in the art without departing from the spirit of the invention. The scope of protection of this invention is defined by the appended claims.

Claims

1. A colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback, characterized in that, include: The colonic dialysis catheter body is a multi-lumen flexible tube that can be inserted into the colonic lumen through the anus. The catheter body is made of medical flexible polymer material. A support skeleton is set along the axial direction inside the catheter wall, and several tendon cords are arranged along the circumference of the catheter. The number of tendon cords is N, where N≥3. One end of each tendon cord is fixed to the flexible segment at the distal end of the catheter, and the other end is connected to the tendon cord driving mechanism at the proximal end of the catheter. The tendon cord driving mechanism is used to adjust the bending posture and local stiffness of the distal end of the catheter in the colonic lumen by pulling or relaxing the tendon cords. An intestinal wall tension sensor array is disposed on the outer wall of the distal flexible section of the catheter body. Several annular sensing bands are arranged at intervals along the axial direction of the catheter. Multiple force-sensitive sensing elements are disposed on each annular sensing band along the circumference of the catheter. The force-sensitive sensing elements are used to output tension signals related to the contact stress of the colon wall in the corresponding annular area. A temperature sensor unit is disposed on the catheter body for monitoring the local temperature in the colon lumen and the temperature of the dialysate. The temperature sensor unit includes: a first temperature sensor attached to the outer wall of the distal end of the catheter for detecting the temperature of the mucosa near the catheter tip, and a second temperature sensor disposed in the catheter lumen near the dialysate outlet for detecting the temperature of the dialysate outlet. The perfusion and drainage module is connected to the perfusion and drainage channels of the catheter body and includes a reversibly driven micro-perfusion pump and a negative pressure drainage unit. The micro-perfusion pump is used to deliver dialysate into the colonic lumen at an adjustable flow rate, and the negative pressure drainage unit is used to aspirate and depressurize gas and / or liquid in the colonic lumen when needed. The central control module is electrically connected to the intestinal wall tension sensor array, temperature sensor unit, perfusion and drainage module, and chordae tendineae drive mechanism. The central control module includes: The data acquisition unit is used to acquire the tension signal of each annular sensing belt and the temperature signal of each temperature sensor at a preset sampling frequency, and to calculate the instantaneous values, rate of change and fluctuation statistics of tension and temperature within a sliding time window. The risk assessment unit is used to normalize the tension and temperature characteristics within the sliding time window, calculate the joint risk index M-Risk according to the pre-stored feature weights, the joint risk index simultaneously characterizes the stress state of the colon wall and the local temperature state, and compare the joint risk index with at least two preset risk thresholds to determine the risk level of the current dialysis process. The control unit is used to maintain the current perfusion flow rate and catheter posture when the risk level is safe, reduce the flow rate of the micro-perfusion pump and / or intermittently control the negative pressure drainage unit to slowly depressurize when the risk level is warning, and drive the chordae tendineae drive mechanism to adjust the bending posture of the distal end of the catheter according to the force distribution output by the intestinal wall tension sensor array; when the risk level is high risk, immediately stop the output of the micro-perfusion pump and continuously control the depressurization of the negative pressure drainage unit, while controlling the chordae tendineae drive mechanism to release or reverse the adjustment of each chordae tendineae to make the distal end of the catheter withdraw from the high-pressure area, and trigger the alarm system to output danger warning to medical staff, thereby forming a closed-loop safety control of colonic dialysis based on intestinal wall tension and temperature sensor feedback.

2. The colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback according to claim 1, characterized in that, In the intestinal wall tension sensor array, the spacing of the annular sensing bands along the axial direction of the catheter is 2 to 10 mm. Each annular sensing band contains m force-sensitive sensing elements, where 3 ≤ m ≤ 12. The central control module reconstructs the circumferential-axial intestinal wall contact stress distribution at the distal end of the catheter based on the tension signals of each annular sensing band and each circumferential position, which is used to identify local stress concentration areas.

3. A colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback according to claim 1, characterized in that, The risk assessment unit calculates the joint risk index M-Risk using the following example formula: Among them, P~ max σ~P is the normalized value of the maximum contact pressure within the sliding time window, σ~P is the normalized value of the variance of the tension of the annular sensing belt, and ΔP~ ring T is the normalized value of the tension difference between adjacent annular sensing strips. mucosa Let dT be the normalized deviation of the mucosal temperature detected by the first temperature sensor relative to the target temperature, dT be the normalized value of the mucosal temperature change rate, and w1~w5 be the weighting coefficients of the corresponding features, and satisfy the following conditions: .

4. A colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback according to claim 1, characterized in that, The central control module also includes a scenario mode management unit, which stores at least three sets of feature weights and risk threshold parameters for different clinical scenarios. The clinical scenarios include preoperative colonic irrigation, postoperative colonic drainage, and local hyperthermic perfusion therapy. The scenario mode management unit loads the corresponding parameter set according to the user's selection. In the preoperative colonic irrigation scenario, the weights of tension and pressure-related features are increased and the maximum contact pressure threshold is set to 3-5 kPa. In the local hyperthermic perfusion therapy scenario, the weights of temperature deviation and temperature rise rate features are increased and the temperature threshold is set to 42-43°C.

5. A colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback according to claim 4, characterized in that, The central control module includes a state classification subunit. The state classification subunit uses a support vector machine model with radial basis function kernel function as input, and the feature vector composed of joint risk index M-Risk and the multimodal features to classify the colon dialysis status into three categories: safe, warning and high risk. The support vector machine model is obtained by training on historical colon dialysis data containing safe, warning and high risk samples, and the penalty coefficient and kernel parameters are optimized by cross-validation.

6. A colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback according to claim 5, characterized in that, The central control module includes a dynamic threshold adjustment unit and a feature weight adjustment unit. The dynamic threshold adjustment unit is configured to statistically analyze the distribution characteristics of the joint risk index M-Risk within a preset time period, and fine-tune the warning threshold and high-risk threshold based on the mean and standard deviation or quantile information. The feature weight adjustment unit is configured to adjust the weights of tension-related and temperature-related features by a limited margin based on recent risk events and control effects, in order to reduce unnecessary perfusion interruptions while ensuring safety.

7. A colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback according to claim 4, characterized in that, In the preoperative colonic irrigation scenario, when the combined risk index M-Risk is at the warning level, the central control module controls the micro-infusion pump to reduce the irrigation flow rate to 50% to 80% of the original set flow rate, and periodically activates the negative pressure drainage unit for short periods to slowly depressurize; when the combined risk index M-Risk is at the high risk level, the micro-infusion pump output is immediately stopped and the negative pressure drainage unit is continuously controlled to depressurize until the estimated internal pressure is lower than the set safety value, while driving the chordae tendineae drive mechanism to adjust the bending posture of the distal end of the catheter to avoid the local pressure area.

8. A colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback according to claim 7, characterized in that, In the postoperative colonic drainage scenario, the central control module considers a slow increase in tension and a decrease in drainage volume as characteristics of drainage obstruction risk. When the catheter tension baseline is observed to rise and the drainage volume is lower than the preset lower limit within multiple consecutive sliding time windows, the module controls the micro-infusion pump to perform low-flow flushing and controls the negative pressure drainage unit to perform low-intensity pulse aspiration to attempt to clear the drainage path. If the combined risk index M-Risk remains high, the drainage is suspended and an alarm is triggered to prompt medical staff to check and handle the situation.

9. A colonic dialysis catheter system with intestinal wall tension and temperature sensing feedback according to claim 5, characterized in that, The central control module includes a subjective discomfort prediction unit, which is used to establish a prediction model based on the correspondence between the patient's subjective pain score in historical records and objective sensor parameters such as catheter tension and temperature. During colonic dialysis, the subjective discomfort prediction value is calculated based on the real-time tension peak, contact pressure and temperature deviation. When the subjective discomfort prediction value exceeds the preset comfort threshold, even if the combined risk index M-Risk is still below the warning threshold, the control unit is configured to reduce the perfusion flow rate in advance, extend the perfusion interval time or temporarily suspend dialysis, so as to reduce the patient's subjective discomfort and optimize the clinical operation experience.

Images