Nasointestinal tube system and positioning and safety monitoring method thereof

Abstract

The invention relates to the technical field of medical instruments, and discloses a nasointestinal tube system and a positioning and safety monitoring method thereof, and the system comprises a catheter body, a positioning beacon, a multi-mode positioning module, an integrated sensing module and an intelligent decision terminal. The positioning beacon emits an ultrasonic positioning signal under the excitation of an alternating magnetic field based on a magnetostriction or magnetoelectric effect; the multi-mode positioning module works cooperatively with an ultrasonic receiving array through a magnetic field excitation unit, and real-time three-dimensional coordinates of the tail end of the catheter are calculated by a positioning calculation processor; the integrated sensing module monitors the pressure and the pH value of fluid in the pipe cavity in real time; the intelligent decision-making terminal carries out fusion processing on the positioning data and the sensing data, and dynamic mapping of a catheter path, abnormal event recognition alarm and prospective early warning of complications are achieved in sequence. The positioning and safety monitoring method corresponds to the system. The real-time, accurate and integrated intelligent monitoring and safety management of the position and the function of the nasointestinal tube, especially the nasointestinal tube for children in the indwelling period are achieved.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, specifically to a nasoenteric tube system and its positioning and safety monitoring method. Background Technology

[0002] Nasoenteric tube placement is a key technique for pediatric enteral nutrition support. However, current techniques have significant limitations in achieving precise, real-time, and safe monitoring of nasoenteric tubes, especially in pediatric patients.

[0003] Traditional positioning methods (such as X-ray fluoroscopy) cannot achieve continuous tracking and pose radiation exposure risks. While some electromagnetic or ultrasound-based active navigation technologies can provide some guidance, they are often complex, and their positioning accuracy is easily affected by the internal environment, making passive, high-precision continuous positioning of inserted catheter tips particularly difficult. Furthermore, clinical reliance on nurses' regular observation and experience to identify complications such as catheter displacement, reflux, or blockage results in delayed warnings. For pediatric patients, whose digestive tracts are more delicate and have smaller lumens, the risk of complications is higher and earlier detection is more difficult; existing single-function catheters or discrete monitoring methods are insufficient to meet their unique clinical safety needs.

[0004] Although existing studies have attempted to integrate multiple sensors (such as pressure and pH sensors) onto catheters, or to use artificial intelligence algorithms to analyze a single type of data (such as ultrasound images) for auxiliary judgment, these solutions are mostly simple superposition of functional modules or post-analysis, and cannot achieve intelligent early warning through real-time high-precision in vivo positioning, simultaneous monitoring of multiple physiological parameters, and deep data fusion.

[0005] Therefore, how to design an intelligent nasoenteric tube system that can passively, accurately, and continuously track the three-dimensional position of the catheter tip and integrate multi-dimensional time-series data such as position, pressure, and pH for forward-looking risk prediction has become an urgent technical problem to be solved. Summary of the Invention

[0006] The purpose of this invention is to provide a nasoenteric tube system and its positioning and safety monitoring method to solve the technical problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention discloses the following technical solutions: In a first aspect, the present invention discloses a nasoenteric tube system, comprising: Catheter body; A miniature biocompatible positioning beacon is embedded at the distal end of the catheter body. The positioning beacon includes a magnetic core and a transducer coupled to the magnetic core. The transducer is configured to generate mechanical vibration based on magnetostrictive or magnetoelectric effects under the action of an external alternating magnetic field in order to emit an ultrasonic positioning signal. A multimodal positioning module, comprising: A magnetic field excitation unit is used to generate an alternating magnetic field to excite the positioning beacon in a non-contact manner; An ultrasonic receiving array is used to capture the ultrasonic positioning signal; The positioning calculation processor is configured to calculate the real-time three-dimensional coordinates of the positioning beacon based on the ultrasonic positioning signal; An integrated sensing module is installed on the catheter body to monitor the fluid pressure and pH level within the lumen; The intelligent decision-making terminal is communicatively connected to the multimodal positioning module and the integrated sensing module, and is configured to perform: path mapping based on the real-time three-dimensional coordinates, abnormal event identification and alarm based on the data of the integrated sensing module, and complication risk warning based on the risk prediction model.

[0008] Optionally, the positioning calculation processor calculates the real-time three-dimensional coordinates by: determining the spatial coordinates of the positioning beacon based on the time difference and / or phase difference between the arrival of the ultrasonic positioning signal at different array elements in the ultrasonic receiving array, and based on the intensity of the received signal, using a fusion algorithm, and using the spatial coordinates as the real-time three-dimensional coordinates.

[0009] Optionally, the risk prediction model is a time-series prediction model, whose input features include: the time-series changes of the real-time three-dimensional coordinates, the time-series data of pressure and pH monitored by the integrated sensing module, and the historical displacement records of the catheter; its output is used to characterize the risk index of the possibility of future displacement or blockage.

[0010] Optionally, the catheter body includes, along its axial direction from the end to the front end, a flexible end section, a transition section, and a rigid front end section, wherein the transition section connects the flexible end section and the rigid front end section and its stiffness gradually increases from the side closer to the flexible end section to the side closer to the rigid front end section.

[0011] Optionally, the Shore hardness of the flexible end segment is within a preset flexibility range, the Shore hardness of the rigid front end segment is within a preset rigidity range, and the preset rigidity range is higher than the preset flexibility range.

[0012] Optionally, the intelligent decision-making terminal is further configured to send a notification containing complication risk information and catheter location information to a preset terminal when an early warning or alarm is triggered.

[0013] Secondly, the present invention discloses a method for locating the nasoenteric tube system based on any one of the above claims, comprising: Step S1: Apply an alternating magnetic field through the magnetic field excitation unit to excite the positioning beacon to emit an ultrasonic positioning signal; Step S2: Capture the ultrasonic positioning signal through the ultrasonic receiving array; Step S3: Process the ultrasound positioning signal and calculate the real-time three-dimensional coordinates of the distal side of the catheter body; Step S4: Based on the deviation between the real-time 3D coordinates and the expected path, generate position status information or correction guidance.

[0014] Optionally, processing the ultrasound positioning signal and calculating the real-time three-dimensional coordinates of the distal side of the catheter body includes: The time difference and / or phase difference of the ultrasonic positioning signal arriving at different array elements in the ultrasonic receiving array are analyzed, and the spatial coordinates of the positioning beacon are determined by a fusion algorithm based on the intensity of the received signal, and the spatial coordinates are used as the real-time three-dimensional coordinates.

[0015] Thirdly, the present invention discloses a safety monitoring method for the nasoenteric tube system based on any one of the above claims, comprising: Step P1: Continuously acquire fluid pressure monitoring data and pH monitoring data output by the integrated sensing module; Step P2: Based on the real-time changes in the fluid pressure monitoring data and the pH monitoring data, determine whether a backflow event or a blockage event has occurred; Step P3: If a backflow event or a blockage event is determined to have occurred, trigger the corresponding local real-time alarm according to the determined event type; Step P4: Input the continuous real-time three-dimensional coordinates, the fluid pressure monitoring data, the pH monitoring data, and the catheter historical status records into the risk prediction model to obtain the predicted complication risk value for the future period. Step P5: If the predicted risk value of the complication exceeds the preset warning threshold, a forward-looking risk warning is triggered.

[0016] Optionally, in step P2: The conditions for determining that a backflow event has occurred are: the fluid pressure monitoring data drops sharply beyond a pressure threshold within a preset time window, and the pH monitoring data is simultaneously below a preset acidity threshold and remains below it for a set duration. The conditions for determining that a blockage event has occurred are: the fluid pressure monitoring data shows a step-like increase and its gradient exceeds a preset gradient threshold, and the fluid flow rate calculated based on the fluid pressure monitoring data is lower than a preset flow rate threshold.

[0017] Compared with the prior art, the nasoenteric tube system and its positioning and safety monitoring method of the present invention have the following beneficial effects: A miniature passive positioning beacon based on magnetostrictive / magnetoelectric effects, a fusion algorithm based on multimodal fusion positioning, an integrated sensing module based on pressure and pH, and a risk prediction model based on multidimensional time-series data are integrated into an organically coordinated intelligent system. This system first triggers the positioning beacon at the catheter tip to emit an ultrasonic positioning signal via magnetic field excitation. Utilizing acousto-magnetic fusion technology, it achieves radiation-free, high-precision, real-time three-dimensional positioning of the catheter tip within the body, solving the problem of continuous tracking in traditional methods. Based on this, the system simultaneously monitors changes in fluid pressure and pH within the lumen, deeply fusing these physiological parameters with the catheter's real-time coordinates and historical displacement records. The risk prediction model dynamically assesses the risk of displacement and blockage. This combination of technologies enables the system not only to immediately alarm when reflux or blockage occurs but also to provide early warnings of potential risks, achieving an upgrade from passive alarm to proactive prediction. The entire solution specifically considers the physiological characteristics of pediatric patients, improving safety through a flexible segmented design of the catheter body. Ultimately, it provides an integrated solution for pediatric enteral nutrition support that offers precise positioning, comprehensive monitoring, and intelligent early warning, enhancing the safety and reliability of clinical operations. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a structural block diagram of the nasoenteric tube system provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart of the positioning method provided in Embodiment 2 of the present invention; Figure 3 This is a flowchart of the safety monitoring method provided in Embodiment 3 of the present invention. Detailed Implementation

[0020] To facilitate understanding of the technical solutions provided in the embodiments of this application, the background technology involved in the embodiments of this application will be described below.

[0021] Nasoenteric tube placement in children is a crucial technique for pediatric enteral nutrition support, particularly for critically ill children or those with congenital digestive tract malformations. The core clinical challenge lies in achieving precise, real-time, and safe positioning and monitoring of the catheter tip within the complex digestive tract. Current technologies have developed several approaches to address this issue, but each has significant limitations, particularly in meeting the safety needs of this specific group of children: The first category is static confirmatory techniques based on medical imaging, represented by X-ray fluoroscopy. While this method is considered effective for localization, it is essentially a non-real-time, non-continuous static examination. It cannot provide continuous information on whether the catheter has shifted during its indwelling period (which can be up to several weeks). Frequent re-examinations mean that the child will be repeatedly exposed to ionizing radiation, which contradicts the fundamental principle of minimizing radiation damage in pediatric care. Furthermore, this method relies entirely on passive examination by healthcare professionals and cannot achieve active, continuous monitoring.

[0022] The second category is active navigation or monitoring technologies based on a single physical principle. For example, some solutions integrate a miniature electromagnetic coil at the end of the catheter, calculating its position by measuring changes in the magnetic field outside the body. However, in the dynamic human environment filled with fluids and tissues, this type of electromagnetic positioning technology is susceptible to interference, resulting in poor accuracy and stability, making it difficult to achieve millimeter-level precise tracking. Other solutions attempt to integrate a single type of sensor, such as a pH sensor at the end of the catheter, indirectly determining whether it has shifted into the stomach by detecting sudden changes in acidity or alkalinity. However, this single-parameter approach has a high false alarm rate, insufficient information dimensions, and cannot effectively distinguish between physiological fluctuations and actual complications, let alone provide early warnings of risks such as catheter blockage.

[0023] The third category involves the simple overlay of functional modules or post-hoc data analysis. With technological advancements, explorations have emerged that combine pressure and pH sensing, or attempt to use machine learning algorithms to analyze ultrasound images for assisted diagnosis. However, most of these attempts have failed to overcome the bottlenecks of functional discreteness and analytical lag. Pressure, pH, and other sensor data, along with the spatial location information of the catheter, have not achieved deep, real-time fusion in both time and space; algorithmic analysis is often limited to identifying anomalies that have already occurred (i.e., post-hoc alarms), rather than conducting proactive modeling and risk prediction based on multi-source time-series data.

[0024] In summary, the existing technological system is fragmented: positioning technology, sensing technology, and analysis technology are isolated from each other. Especially for pediatric patients with narrower lumens, more delicate mucous membranes, and weaker compensatory abilities, there is an urgent clinical need for an integrated solution that organically combines radiation-free, continuous, precise positioning, simultaneous sensing of multidimensional physiological parameters, and intelligent, proactive early warning based on data fusion. This would fundamentally transform the management of nasoenteric tube placement from rough estimation and post-treatment to real-time tracking and proactive prevention. This constitutes the core technical problem that this application aims to solve.

[0025] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application. Secondly, in this document, the term "comprising" is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements, but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0026] Furthermore, it should be specifically noted that the collection, transmission, storage, and processing of patient physiological data (such as pressure and pH levels) and location information involved in the clinical use of this invention system must be conducted with the explicit informed consent of the patient or their legal guardian. All data processing procedures must comply with relevant national laws and regulations on personal information protection and medical data security management, such as employing encrypted data transmission, anonymization or de-identification processing, and setting strict access control measures to ensure patient privacy and security.

[0027] Example 1 This embodiment addresses the problems of discontinuous positioning, limited monitoring dimensions, and delayed early warning in existing nasoenteric tube technologies, and provides an integrated system that combines real-time positioning, multi-parameter monitoring, and intelligent early warning. Specifically, a nasoenteric tube system, such as... Figure 1 As shown, it includes: Catheter body; A miniature biocompatible positioning beacon is embedded at the distal end of the catheter body. The positioning beacon includes a magnetic core and a transducer coupled to the magnetic core. The transducer is configured to generate mechanical vibration based on the magnetostrictive effect or the magnetoelectric effect under the action of an external alternating magnetic field in order to emit an ultrasonic positioning signal. A multimodal positioning module, comprising: A magnetic field excitation unit is used to generate an alternating magnetic field to non-contactly excite a positioning beacon; An ultrasonic receiver array is used to capture ultrasonic positioning signals; The positioning solver is configured to calculate the real-time three-dimensional coordinates of the positioning beacon based on ultrasonic positioning signals; An integrated sensing module, which is mounted on the catheter body, is used to monitor the fluid pressure and pH level within the lumen. The intelligent decision-making terminal is connected to a multimodal positioning module and an integrated sensing module, and is configured to perform: path mapping based on real-time three-dimensional coordinates, abnormal event identification and alarm based on data from the integrated sensing module, and complication risk warning based on a risk prediction model.

[0028] In its implementation, the catheter body is made of medical-grade polyurethane or silicone, with an outer diameter ranging from 1.8 to 2.5 mm and a length ranging from 80 to 120 cm to accommodate the anatomical pathway from the nasal cavity to the jejunum in pediatric patients of different ages. A miniature biocompatible positioning beacon is completely encapsulated within a pre-reserved cavity on the distal sidewall of the catheter body, with a maximum radial dimension not exceeding 2.0 mm. This positioning beacon comprises a cylindrical magnetic core made of soft magnetic ferrite material (such as manganese zinc ferrite), with a diameter of approximately 0.8 mm and a height of approximately 1.2 mm. Surrounding the magnetic core is a ring transducer made of piezoelectric ceramic material (such as lead zirconate titanate PZT-5A), whose inner diameter is interference-fitted with the magnetic core, with an outer diameter of approximately 1.5 mm and a thickness of approximately 0.3 mm. The electrodes of the transducer are connected via microwires to a miniature antenna circuit embedded in the catheter wall, which is used to receive external alternating magnetic field energy. When an alternating magnetic field with a frequency of 10kHz to 100kHz generated by an external magnetic field excitation unit acts on this region, the magnetic core undergoes periodic magnetostrictive deformation due to the alternating magnetic field. This deformation directly drives the coupled piezoelectric ceramic transducer, causing it to vibrate under forced conditions and emit ultrasonic positioning signals with a frequency range of 1MHz to 5MHz. In practical applications, the material and size of the positioning beacon can be other options, and this embodiment does not limit them.

[0029] The multimodal positioning module comprises a standalone hardware unit. The magnetic field excitation unit, consisting of a signal generator, power amplifier, and Helmholtz coil, generates a uniform alternating magnetic field in the target area (e.g., the patient's abdomen) with a frequency matched to the resonant frequency of the positioning beacon. The ultrasound receiving array consists of 16 to 64 elements arranged in a planar or curved surface, with a center frequency of 2.5 MHz. The array connects to the positioning calculation processor via a USB or Ethernet interface. The positioning calculation processor is an embedded computer or FPGA chip, running positioning calculation software internally. The integrated sensing module includes a miniature piezoresistive pressure sensor and a miniature solid-state pH sensor, integrated side-by-side on the outer wall of the catheter, 30 to 50 cm from the catheter tip, and connected to a connector at the proximal end via flexible printed circuitry. The pressure sensor has a range of -50 to 300 mmHg, and the pH sensor measures pH 1 to pH 9. The intelligent decision-making terminal is a tablet computer or bedside monitor with dedicated software installed, communicating with the positioning calculation processor and the interface circuitry of the integrated sensing module via Bluetooth or Wi-Fi.

[0030] Based on the above, in this system, the magnetic field excitation unit provides non-contact energy excitation, driving the positioning beacon to autonomously emit ultrasonic signals, achieving passive positioning without the need for a built-in power supply. After the ultrasonic receiving array captures this signal, the positioning calculation processor calculates its three-dimensional coordinates. Simultaneously, the integrated sensing module continuously collects physiological parameters. The intelligent decision-making terminal receives and integrates these two types of information to complete spatial path visualization, real-time anomaly detection, and long-term risk prediction. This integrated structure combines the previously discrete positioning, sensing, and decision-making functions into a collaborative system, solving the problem of functional fragmentation in existing technologies and providing a hardware foundation for continuous, accurate, and early warning-enabled nasoenteric tube safety management.

[0031] High-precision real-time three-dimensional coordinate calculation is a prerequisite for achieving accurate path mapping and risk warning. Therefore, as an optional implementation method in this embodiment, the positioning calculation processor calculates the real-time three-dimensional coordinates by: determining the spatial coordinates of the positioning beacon based on the time difference and / or phase difference of the ultrasonic positioning signal arriving at different array elements in the ultrasonic receiving array, and based on the intensity of the received signal, using a fusion algorithm, and using the spatial coordinates as the real-time three-dimensional coordinates.

[0032] In its implementation, the localization processor executes a fusion algorithm that combines two measurement information: signal arrival time difference and signal arrival angle. It then uses signal strength indicators for weighted optimization to combat interference from multiple reflections within the body. The specific calculation steps are as follows: ①Time difference and phase difference extraction: First, the signals acquired by each channel of the ultrasonic receiving array are cross-correlated to calculate the time difference between the arrival of the signals at different array element pairs. Simultaneously, the instantaneous phase of the signal is extracted using Hilbert transform, and the phase difference between array elements is calculated. Among them, subscript and These represent different array element numbers.

[0033] ② Initial coordinate estimation: based on time difference The Chan Taylor series expansion method or particle filter algorithm is used to solve for a set of possible target source locations. ,in Represents a three-dimensional coordinate point. For collection index.

[0034] ③ Angle Constraint and Fusion: Based on Phase Difference Calculate the direction of arrival (AOA) of the signal at different array elements to obtain the AOA estimate, including the elevation angle. and azimuth Construct a constraint vector pointing from the array to the source using the AOA estimate. Then, use the constraint vector for each candidate point obtained in step ②. A consistency check is performed with the constraint vector to remove points that significantly deviate from the constraint direction, resulting in a refined set of positions. .

[0035] ④ Signal strength weighted optimization: Read the signal strength received by each array element. Calculate its normalized intensity weight ,in This represents the total number of array elements. The array element coordinates are... The coordinates of the candidate points are Define the objective function The weighted sum of squared distance errors: , In the formula, This is an initial distance estimate calculated based on the arrival time. The average speed of sound in the organization, The arrival time is minimized using an iterative algorithm (such as the Gauss-Newton method). From the set Determine the point that minimizes the objective function. , which are the spatial coordinates of the final solution.

[0036] In the above algorithm, It can be set to a fixed value (e.g., 1540 m / s) based on clinical experience, or calibrated online using dual-frequency transmission signals. In practical applications, other multi-source information fusion algorithms can also be used, but this embodiment does not limit this approach.

[0037] Based on the above, this fusion algorithm comprehensively utilizes the spatiotemporal characteristics (time difference, phase difference) and energy characteristics (signal strength) of the signal. The time difference provides basic ranging information, while the phase difference provides precise angular constraints; the combination of the two effectively reduces the computational space. Signal strength weighting reduces the impact of ranging errors caused by signal attenuation and reflection. Through this series of calculation steps, the algorithm can stably output high-precision three-dimensional coordinates in complex in vivo acoustic environments, achieving millimeter-level positioning accuracy, providing a reliable data foundation for subsequent precise path mapping and location-based trend analysis.

[0038] After obtaining high-precision positioning and multi-parameter monitoring data, the key to proactive prevention lies in how to proactively identify risks from this massive amount of time-series data. Therefore, as an optional implementation method in this embodiment, the risk prediction model is a time-series prediction model. Its input features include: real-time temporal changes in three-dimensional coordinates, time-series data of pressure and pH monitored by the integrated sensing module, and historical displacement records of the catheter; its output is used to characterize a risk index representing the likelihood of future displacement or blockage.

[0039] In its implementation, this risk prediction model is a bidirectional long short-term memory network model based on an attention mechanism. The model's input is a multivariate time series window. ,in For the current moment, The time window length (e.g., for data from the past 30 minutes, sampled at 1Hz, T=1800). Feature vector at each time point. It includes the following sub-features: ① Spatial characteristics: Current three-dimensional coordinates And its difference from the coordinates of the previous time step. Together, they constitute the instantaneous displacement vector.

[0040] ② Sensing characteristics: lumen pressure value pH value .

[0041] ③Derived characteristic: Rate of change of pressure within a short time window And the curvature of the motion trajectory fitted by the coordinate sequence over a slightly longer window (such as the past 5 minutes). .

[0042] The model architecture consists of an input layer, a bidirectional LSTM layer, an attention layer, and a fully connected output layer. The bidirectional LSTM layer is used to learn the input sequence from both forward and backward directions. Long-term dependencies, whose hidden state is denoted as The calculation process for the attention layer is as follows: First, calculate the last time step. Hidden state Hidden states at all time steps in the sequence correlation score : , in, , , , These are trainable parameters. Then, the relevance scores are normalized into attention weights using the Softmax function. : , Finally, we obtain the weighted context vector. As a summary of the sequence: , context vector The input to the fully connected layer, after passing through the sigmoid activation function, outputs a scalar value between 0 and 1, which is the risk index. The closer the index is to 1, the higher the risk of catheter displacement or blockage in the near future (e.g., within the next hour). The model is trained using historical clinical data, where each data sample is an input sequence for a time window. The labels are binary values ​​defined based on subsequent actual clinical events (displacement or blockage occurring / not occurring). The training objective is to minimize the binary cross-entropy loss function.

[0043] In practical applications, the network structure, hyperparameters, and feature engineering of the model can be designed in other ways, and this embodiment does not limit this. Model training and operation are both completed on the intelligent decision-making terminal or the connected cloud server.

[0044] Based on the above, this model deeply integrates spatial positioning data (coordinates and displacement) with physiological sensor data (pressure, pH) over time. A bidirectional LSTM captures temporal dynamics, while an attention mechanism enables the model to autonomously focus on historical segments most relevant to the current state (e.g., an abnormal displacement or pressure fluctuation). In this way, the model learns complex, subtle early multidimensional patterns leading to complications, thus outputting a quantified risk index before traditional threshold alarms are triggered. This represents a shift from ex-post alarms based on a single threshold to early risk quantification assessment based on multi-source fusion data.

[0045] Children's digestive tract mucosa is fragile and their anatomical path is tortuous and varied. Uniform catheter rigidity can easily cause damage or make catheter placement difficult. Therefore, as an optional implementation method in this embodiment, the catheter body includes, along its axial direction from the end to the front end, a flexible end segment, a transition segment, and a rigid front end segment, wherein the transition segment connects the flexible end segment and the rigid front end segment and its rigidity gradually increases from the side closer to the flexible end segment to the side closer to the rigid front end segment.

[0046] In practice, the catheter body has a single-lumen or double-lumen structure. The flexible distal segment originates at the catheter tip, is 15 to 25 cm long, and is made of medical-grade silicone with a Shore A hardness of 20A to 30A. The wall thickness of this segment is uniform, approximately 0.4 to 0.6 mm, and its soft material effectively reduces the risk of scraping and irritation to the digestive tract mucosa when passing through narrow or tortuous areas such as the pylorus and duodenal ligament.

[0047] The rigid proximal segment, located in the main body of the catheter outside the body, is 50 to 80 centimeters in length and is made of medical-grade polyurethane with a Shore A hardness of 50A to 60A. The wall thickness of this segment is approximately 0.5 to 0.7 millimeters. This high rigidity ensures the straightness of the catheter outside the body, preventing bending or compression outside the nasal cavity due to patient movement or the weight of the tubing, thus ensuring unobstructed enteral nutrition infusion.

[0048] The transition section connects the flexible distal section and the rigid proximal section, with a length of 5 to 10 centimeters. This section is manufactured using a co-extrusion molding process, and its material is a gradient composite of silicone and polyurethane. Microstructurally, starting from the connection point with the flexible distal section, the proportion of polyurethane material increases linearly along the axial direction, while the proportion of silicone material decreases linearly accordingly. This achieves a continuous and smooth increase in Shore hardness from approximately 25A (distal side) to approximately 55A (proximal side). This continuous gradual change in stiffness avoids stress concentration points caused by abrupt changes in hardness at the material interface, reducing the risk of fatigue fracture at these points during long-term use of the conduit.

[0049] If the three sections are manufactured separately, they are bonded using medical adhesive or by hot-melt welding, ensuring that the inner wall of the joint is smooth, the outer diameter is consistent, and there are no steps. In practical applications, the length, hardness, and manufacturing process of each section can be selected in other ways, which are not limited in this embodiment.

[0050] Based on the above, this segmented design endows the catheter body with differentiated mechanical properties. The flexible distal segment ensures passability and biocompatibility in complex in vivo environments; the rigid proximal segment maintains the morphology and functional reliability of the external portion of the tubing; and the transition segment achieves a smooth transition of mechanical properties between the two. This structural design is directly tailored to the characteristics of pediatric patients, ensuring that the catheter can be successfully inserted and reach the target intestinal segment while minimizing the risk of mucosal damage and catheter structural failure, thus adapting to the physiological structure of children and balancing passability and safety.

[0051] Furthermore, the flexibility and rigidity of each segment need to be defined by objective quantitative standards to ensure controllable and reproducible performance. Therefore, as a further optional implementation of this embodiment, the Shore hardness of the flexible end segment is within a preset flexibility range, the Shore hardness of the rigid front segment is within a preset rigidity range, and the preset rigidity range is higher than the preset flexibility range.

[0052] In this embodiment, the preset flexibility range is specifically a Shore hardness of 20A to 35A. Preferably, the Shore hardness of the flexible distal segment is 25A ± 3A. This hardness range is achieved by selecting a medical-grade silicone with a specific formulation, whose elastic modulus is approximately between 1.0MPa and 3.0MPa. Materials within this range feel soft to the touch at a body temperature of 37°C. In an experimental model simulating intestinal curvature, when passing through an intestinal segment with a curvature radius of less than 2 cm, the lateral pressure exerted by the intestinal wall on the simulated mucosa is less than 15 mmHg, which is below the critical pressure that may cause mucosal ischemia.

[0053] The preset rigidity range is specifically a Shore hardness of 50A to 65A. Preferably, the Shore hardness of the rigid proximal segment is 55A ± 3A. This hardness range is achieved by selecting a specific grade of medical polyurethane, whose elastic modulus is approximately between 10MPa and 30MPa. Materials within this range have sufficient bending stiffness. In standard testing, a 20cm long, freely suspended section of this catheter requires a force greater than 0.5N to deflect 1cm downwards at its center point under external force, indicating good resistance to sagging deformation and effective maintenance of the infusion pathway patency.

[0054] Shore hardness was measured according to standard GB / T531.1-2008 "Test Method for Indentation Hardness of Vulcanized Rubber or Thermoplastic Rubber - Part 1: Shore Hardness Tester Method (Shore Hardness)". In practical applications, the specific boundary values ​​of the preset flexibility range and preset rigidity range can be fine-tuned based on clinical feedback; this embodiment does not limit this.

[0055] Based on the above, by quantifying flexibility and rigidity into measurable Shore hardness ranges, clear technical indicators are provided for catheter material selection and product quality control. The preset hardness ranges are established based on extensive biomechanical experiments and clinical feedback, ensuring that the flexible distal segment achieves the expected damage reduction effect, while the rigid proximal segment meets the functional requirements for bending resistance. This quantitative limitation transforms the technical solution from a qualitative description into a specific standard for industrial production and quality inspection, enhancing the feasibility and repeatability of the solution.

[0056] When an early warning or alarm occurs, timely transmission of critical information to medical staff or guardians is crucial for rapid intervention and ensuring patient safety. Therefore, as an optional implementation method in this embodiment, the intelligent decision-making terminal is also configured to send a notification containing complication risk information and catheter location information to a preset terminal when an early warning or alarm occurs.

[0057] In its implementation, the intelligent decision-making terminal integrates a communication service module. When the system triggers a local real-time alarm (corresponding to step P3 described later) or a forward-looking risk warning (corresponding to step P5 described later), the communication service module is automatically invoked. This module generates a structured notification message according to a predefined protocol template.

[0058] The notification message includes the following core information: 1) Event type: such as "reflux alarm", "blockage warning", "high-risk displacement warning"; 2) Risk level: such as "high risk", "medium risk", "low risk", which is classified according to the risk index or the severity of the event; 3) Catheter location information: a text description of the most likely anatomical location of the current catheter tip (e.g., "descending duodenum"), accompanied by a simplified two-dimensional or three-dimensional catheter path diagram or an accessible link against the background of the patient's abdominal outline; 4) Timestamp: the specific time the event occurred; 5) Patient identification: a de-identified bed number or medical record number.

[0059] The generated encrypted notification messages are sent via the hospital's internal Wi-Fi network or cellular mobile network (4G / 5G) to one or more pre-set terminals in the form of push notifications or SMS messages. These terminals include the central monitoring screen at the nurse station, the dedicated mobile terminal (such as a PDA or mobile phone) of the on-duty doctor, and the smartphone of the child's caregiver in home care mode. The sending action can be implemented using any message queue or instant messaging interface in the prior art.

[0060] In practical applications, the content format, sending protocol, and receiving terminal type of the notification can be configured in other ways, and this embodiment does not limit this.

[0061] Based on the above, this configuration extends the perception and decision-making capabilities of the intelligent decision terminal, making it not limited to local alarms. By automatically generating and sending structured notifications, it disseminates critical alarm and warning information from a single point (the device itself) to relevant personnel throughout the entire nursing network. This ensures that, whether in a ward, doctor's office, or home environment, the relevant nursing staff can be informed of the risk situation and catheter location immediately, gaining valuable time to activate emergency plans (such as adjusting the patient's position, suspending feeding, flushing the catheter, or going to the hospital). This forms a complete closed loop from risk identification to information dissemination to clinical intervention, improving the timeliness and effectiveness of the entire system in responding to emergencies.

[0062] Example 2 Example 1 describes the static configuration of the system, while how to operate the system to achieve precise catheter positioning is one of its core functions. Therefore, in this example, a positioning method for the nasoenteric tube system proposed in Example 1 is described, such as... Figure 2 As shown, it includes: Step S1: Apply an alternating magnetic field through the magnetic field excitation unit to excite the positioning beacon to emit ultrasonic positioning signals; Step S2: Capture the ultrasonic positioning signal through the ultrasonic receiving array; Step S3: Process the ultrasound positioning signal and calculate the real-time three-dimensional coordinates of the distal side of the catheter body; Step S4: Based on the deviation between the real-time 3D coordinates and the expected path, generate position status information or correction guidance.

[0063] In practice, in step S1, the operator places the coil of the magnetic field excitation unit above the estimated area of ​​the catheter tip (the distal end of the catheter body) on the patient's abdomen. After the device is started, the internal circuit of the magnetic field excitation unit generates an alternating sinusoidal magnetic field with a frequency of 28kHz (matching the resonant frequency of the positioning beacon) and an intensity of about 10Gauss, which is applied continuously or intermittently according to a specific duty cycle.

[0064] In step S2, after the probe of the ultrasound receiving array is coated with coupling agent, it is fixed to the patient's abdominal surface, covering a larger area including the central region of the coil disc. Each piezoelectric element in the array continuously receives acoustic signals and buffers the analog signals after converting them into digital signals.

[0065] In step S3, the fusion algorithm flow detailed in Example 1 of the localization and resolution processor is used to process the cached multi-channel digital signals and finally output a three-dimensional coordinate value. This coordinate is a value in a coordinate system with the geometric center of the ultrasound receiving array as the origin. It can be converted to a clinical coordinate system with reference to the patient's anatomical landmarks through pre-calibration.

[0066] Step S4 is executed on the intelligent decision-making terminal. The terminal software has a pre-stored expected three-dimensional path model for standard nasoenteric tube placement, or a reference path learned from the correct location points confirmed in the initial placement. The software compares the real-time coordinates calculated in step S3 with the expected path and calculates the Euclidean distance deviation. : , in This represents the coordinates of the point on the expected path closest to the current measured point. If... If the deviation exceeds the set threshold (e.g., 5 mm), it is determined to be a positional offset. The terminal will generate positional status information, such as highlighting the offset point in red on the screen map, and at the same time generate a text correction guide, such as "The tip of the duct is suspected to have shifted backward, it is recommended to pull back 1 cm".

[0067] Based on the above, this method integrates the system's hardware functions into a coherent operation and calculation process. Through the steps of "magnetic excitation - acoustic emission - acoustic reception - algorithm solution," it achieves dynamic tracking of tiny passive beacons within the body. Finally, by comparing the coordinates with the expected path, the abstract coordinate data is transformed into intuitive, clinically guiding status information and action suggestions. This allows medical staff to monitor the precise location and status of the catheter within the body in real time, much like using GPS navigation, solving the problems of blind insertion or reliance on intermittent imaging confirmation, and achieving continuous, precise, and visualized positioning.

[0068] Furthermore, as an optional implementation of this embodiment, processing the ultrasound positioning signal and calculating the real-time three-dimensional coordinates of the distal side of the catheter body includes: The time difference and / or phase difference of the ultrasonic positioning signal arriving at different array elements in the ultrasonic receiving array are analyzed, and the spatial coordinates of the positioning beacon are determined by a fusion algorithm based on the intensity of the received signal, and the spatial coordinates are used as real-time three-dimensional coordinates.

[0069] The specific processing procedure here is an implementation of the fusion algorithm proposed in Example 1. After receiving a frame of multi-channel ultrasound data, the localization and resolution processor executes the following sub-steps in sequence: ① Signal preprocessing: Bandpass filtering (passband 1.5MHz-3.5MHz) is performed on the data of each channel to remove noise, and envelope detection is performed.

[0070] ② Feature extraction: On the envelope signal, the generalized cross-correlation (GCC-PHAT) method is used to calculate the signal arrival time difference between each pair of elements. Simultaneously, the phase of the original radio frequency signal is analyzed, and the phase difference between the same array element and different frequency components is calculated to assist in calibration.

[0071] ③ Initial solution space generation: based on time difference set By using the least squares method to solve the overdetermined system of equations, an initial set of spatial coordinates that may have multiple fuzzy solutions is obtained.

[0072] ④ Intensity-weighted iterative optimization: The peak value of each channel signal is read as the intensity value. Construct the objective function as proposed in Example 1. Starting with the initial solution, the Levenberg-Marquardt nonlinear optimization algorithm is used iteratively to find the solution that satisfies the initial condition. Minimize the optimal coordinates .

[0073] ⑤ Coordinate output: After coordinate system transformation, the real-time three-dimensional coordinates of the distal end of the catheter body at this moment are output to the intelligent decision-making terminal.

[0074] The above processing steps are executed multiple times per second (e.g., 10 times), thereby achieving real-time coordinate updates. In a specific implementation, steps 3 and 4 can be integrated into an Extended Kalman Filter (EKF) framework to utilize historical coordinate information for smoothing and prediction, further improving robustness.

[0075] Based on the above, this method transforms the corresponding algorithm principle proposed in Example 1 into a set of clearly defined, sequentially executable operations. Through a chain of "preprocessing-feature extraction-solution-optimization," the original ultrasonic simulation signal is progressively transformed into precise spatial coordinates. The specific content of the algorithm processing is clearly defined, ensuring that the positioning method is not a black box operation; each step is supported by explicit technical means, and the final result undergoes multi-information fusion optimization, thereby guaranteeing the reliability, accuracy, and anti-interference capability of the positioning result at the methodological level.

[0076] Example 3 Example 1 describes the static configuration of the system, and Example 2 describes the precise positioning of the catheter. However, real-time monitoring and early warning of the catheter's functional status (whether there is reflux or blockage) is another core value of this system. Therefore, in this example, a safety monitoring method for the nasoenteric tube system proposed in Example 1 is described, such as... Figure 3 As shown, it includes: Step P1: Continuously acquire fluid pressure monitoring data and pH monitoring data output by the integrated sensing module; Step P2: Based on the real-time changes in fluid pressure monitoring data and pH monitoring data, determine whether a backflow event or a blockage event has occurred; Step P3: If a backflow event or a blockage event is determined to have occurred, trigger the corresponding local real-time alarm according to the determined event type; Step P4: Input the continuous real-time three-dimensional coordinates, fluid pressure monitoring data, pH monitoring data, and catheter historical status records into the risk prediction model to obtain the predicted complication risk value for the future period. Step P5: If the predicted risk of complications exceeds the preset warning threshold, a prospective risk warning will be triggered.

[0077] In specific implementation, in step P1, the intelligent decision-making terminal continuously reads the digitized pressure values ​​from the integrated sensing module through its communication interface at a sampling rate of no less than 10Hz. and pH value ,in This is the index for the sampling points.

[0078] In step P2, the decision-making logic is executed by the real-time analysis module within the terminal. This module maintains two parallel detectors: a backflow event detector and a congestion event detector. The analysis module performs sliding window analysis on the input data stream, and immediately marks the event as occurring once the conditions for any event are met.

[0079] In step P3, the local real-time alarm is triggered by the terminal's hardware and software in concert. If it's a backflow event, the terminal controls its built-in buzzer to emit intermittent beeps (e.g., once per second), while the screen flashes a yellow warning box and displays "Backflow Alarm". If it's a congestion event, a more rapid, continuous beeping sound is triggered, the screen flashes a red warning box and displays "Congestion Alarm". The correspondence between alarm types and event types is configurable.

[0080] Step P4 is an asynchronous process that runs in parallel with steps P1 / P2. A separate data preprocessing thread is responsible for continuously collecting time-aligned coordinate sequences, pressure sequences, and pH sequences, and extracting the features mentioned in Example 1 to construct a time window sample. This sample is periodically (e.g., every minute) submitted to the risk prediction model running in the background for inference, and the model outputs the current risk index. .

[0081] In step P5, the intelligent decision-making terminal will output the risk index from the model. Compared with preset warning thresholds (e.g., displacement risk thresholds) Blockage risk threshold Compare them. If If any threshold is exceeded, a forward-looking risk warning will be triggered. At this time, the terminal screen will display "High-risk displacement warning" or "High-risk congestion warning" in a conspicuous manner (such as a purple flashing bar), the buzzer may emit a different prompt tone than the real-time alarm, and the notification sending process proposed in Example 1 will be automatically executed.

[0082] Based on the above, this method constructs a two-layer monitoring and early warning system. The first layer (P1-P3) is real-time event monitoring based on fixed thresholds, used for rapid response to clearly identified anomalies. The second layer (P4-P5) is a proactive risk assessment based on an artificial intelligence model, used to identify potential risks that have not yet reached the alarm threshold but show a dangerous trend. The two layers operate in parallel and complement each other. Real-time alarms ensure immediate safety, while proactive early warnings provide a time window for preventative intervention. This method elevates simple status monitoring to a comprehensive safety protection system encompassing immediate response and risk prediction, fully demonstrating the core intelligent advantages of this system.

[0083] Furthermore, as an optional implementation of this embodiment, in step P2: The conditions for determining that a backflow event has occurred are: the fluid pressure monitoring data drops sharply beyond the pressure threshold within a preset time window, and the acidity / alkalinity monitoring data is simultaneously below the preset acidity threshold and remains below it for a set duration. The conditions for determining that a blockage event has occurred are: the fluid pressure monitoring data shows a step-like increase and its gradient exceeds a preset gradient threshold, and the fluid flow rate calculated based on the fluid pressure monitoring data is lower than a preset flow rate threshold.

[0084] In the specific implementation, a preset time window is set for backflow events. Seconds. Calculate the maximum decrease in pressure data within any consecutive 5-second window. .like ( If a pressure threshold is set (e.g., 30% of the current stable perfusion pressure), a pressure drop condition is triggered. Simultaneously, all pH samples taken within this 5-second window and for the following 2 seconds are checked. If more than 90% of the pH values ​​remain below [a certain threshold], a pressure drop condition is triggered. ( If a preset acid threshold is set (usually 4.0), then the acid condition is triggered. Only when both the pressure drop condition and the acid condition are met simultaneously is it ultimately determined to be a backflow event.

[0085] For congestion events, monitor pressure data changes over time. Defined in a length of... On a time series (e.g., data from the most recent 60 seconds), stress values It exhibits a "step-like ascending" pattern. This pattern is detected using the following algorithm: First, the sequence is segmented into linear approximations to find multiple line segments with significantly positive slopes. Second, a positive jump in pressure is required at the connection point between adjacent line segments. ,and The pressure gradient must exceed a preset step change threshold (e.g., 10 mmHg). Finally, calculate the average pressure gradient (in mmHg / s) over the entire monitoring period; this gradient must exceed a preset gradient threshold. (e.g., 0.15 mmHg / s). Simultaneously, the current flow velocity is estimated based on a pressure-velocity model. The model can be simplified to ,in Given the known infusion pump pressure or high potential energy, For the measured pressure at the tip of the catheter, This is a coefficient related to pipe resistance. It is used when estimating flow velocity. Continuously below the preset flow rate threshold (e.g., 1 ml / h) exceeding If the flow rate condition is met (e.g., 10 seconds), then the flow rate condition is considered satisfied. Only when both the "step-up" pattern condition and the flow rate condition are met simultaneously is it ultimately determined to be a complete blockage event.

[0086] Among these, all threshold parameters ( , , , All parameters (such as age and feeding formula) can be fine-tuned by medical staff in the configuration interface of the intelligent decision-making terminal according to the patient's specific situation. In practical applications, the logic and threshold for event judgment can also adopt other combinations or algorithms, which are not limited in this embodiment.

[0087] Based on the above, these specific conditions provide clear, multi-dimensional, and adjustable rules for automated event judgment. The reflux judgment combines two strongly correlated features: "sudden pressure drop" (intestinal contents reflux causing a momentary release of luminal pressure) and "persistent acidity" (refluxed material is usually gastric acid), greatly improving the specificity of the judgment and avoiding false alarms caused solely by pressure fluctuations. The blockage judgment combines the fluid dynamics principles of "gradual pressure increase" (pressure accumulation at the front of the blockage point) and "flow velocity returning to zero," reliably distinguishing between normal increased infusion resistance and true complete blockage. This design based on multi-condition joint judgment improves the accuracy and reliability of system alarms, allowing clinicians to trust and rely more on the system's automated monitoring results, thereby enabling efficient decision-making.

[0088] In the embodiments provided by this invention, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any suitable combination thereof. For hardware implementation, the processor can be implemented in one or more of the following: application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, other electronic units designed to implement the functions described herein, or combinations thereof. For software implementation, some or all of the processes of the embodiments can be performed by a computer program instructing the associated hardware. During implementation, the program can be stored in a computer-readable storage medium or transmitted as one or more instructions or code on a computer-readable storage medium. Computer-readable storage media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of a computer program from one place to another. Storage media can be any available medium accessible to a computer. Computer-readable storage media can include, but are not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code having the form of instructions or data structures and accessible to a computer.

[0089] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A nasojejunal tube system, characterized in that, The application relates to a multi-modal positioning system for a catheter, comprising: a catheter body; a micro-biocompatible positioning beacon embedded at a distal end of the catheter body, the positioning beacon comprising a magnetic core and a transducer coupled to the magnetic core, the transducer being configured to generate mechanical vibrations based on a magnetostrictive effect or a magnetoelectric effect under the action of an external alternating magnetic field to emit an ultrasonic positioning signal; a multi-modal positioning module comprising: a magnetic field excitation unit for generating an alternating magnetic field to non-contact excite the positioning beacon; an ultrasonic receiving array for capturing the ultrasonic positioning signal; a positioning solution processor configured to solve real-time three-dimensional coordinates of the positioning beacon based on the ultrasonic positioning signal; an integrated sensing module arranged on the catheter body for monitoring fluid pressure and pH in a lumen; and an intelligent decision terminal communicatively connected to the multi-modal positioning module and the integrated sensing module and configured to perform path mapping based on the real-time three-dimensional coordinates, abnormal event identification and alarm based on data of the integrated sensing module, and complication risk early warning based on a risk prediction model. The positioning solution processor solving the real-time three-dimensional coordinates comprises: determining spatial coordinates of the positioning beacon by a fusion algorithm based on time difference and / or phase difference of the ultrasonic positioning signal arriving at different array elements of the ultrasonic receiving array and based on intensity of the received signal, and taking the spatial coordinates as the real-time three-dimensional coordinates. The risk prediction model is a time series prediction model, input features of which include time series changes of the real-time three-dimensional coordinates, time series data of pressure and pH monitored by the integrated sensing module, and historical displacement records of the catheter, and an output thereof is a risk index for representing possibility of future displacement or occlusion. The catheter body comprises, in sequence from the distal end to the proximal end along an axial direction, a flexible distal end segment, a transition segment and a rigid proximal end segment, wherein the transition segment is connected between the flexible distal end segment and the rigid proximal end segment and gradually increases in rigidity from a side close to the flexible distal end segment to a side close to the rigid proximal end segment. The flexible distal end segment has a Shore hardness within a preset flexible range, the rigid proximal end segment has a Shore hardness within a preset rigid range, and the preset rigid range is higher than the preset flexible range. The intelligent decision terminal is further configured to send a notification containing complication risk information and catheter position information to a preset terminal when early warning or alarm occurs. The application further relates to a method for positioning a catheter, comprising: step S1 of applying an alternating magnetic field by the magnetic field excitation unit to excite the positioning beacon to emit an ultrasonic positioning signal; step S2 of capturing the ultrasonic positioning signal by the ultrasonic receiving array; step S3 of processing the ultrasonic positioning signal to solve real-time three-dimensional coordinates of a distal end of the catheter body; and step S4 of generating position state information or correction guidance based on deviation of the real-time three-dimensional coordinates from an expected path. The processing of the ultrasonic positioning signal to solve the real-time three-dimensional coordinates of the distal end of the catheter body comprises: ​ 2. The nasojejunal tube system of claim 1, wherein, ​ 3. The nasojejunal feeding tube system of claim 1, wherein, ​ 4. The nasojejunal feeding tube system of claim 1, wherein, ​ 5. The nasojejunal feeding tube system of claim 4, wherein, ​ 6. The nasojejunal feeding tube system of claim 1, wherein, ​ 7. A method of positioning a nasojejunal feeding tube system according to any one of claims 1 to 6, characterized in that, ​ ​ ​ ​ ​ 8. The positioning method according to claim 7, characterized in that, ​ The time difference and / or phase difference of the ultrasonic positioning signal arriving at different array elements in the ultrasonic receiving array is analyzed, and the spatial coordinates of the positioning beacon are determined by a fusion algorithm based on the intensity of the received signal, and the spatial coordinates are taken as the real-time three-dimensional coordinates.

9. A method of safety monitoring of a nasojejunal feeding tube system according to any one of claims 1 to 6, characterized in that, Comprise: Step P1: continuously acquiring fluid pressure monitoring data and pH monitoring data output by the integrated sensing module; Step P2: judging whether a regurgitation event or a blockage event occurs based on real-time changes in the fluid pressure monitoring data and the pH monitoring data; Step P3: if it is judged that a regurgitation event or a blockage event occurs, triggering a corresponding local real-time alarm according to the type of the event judged; Step P4: inputting the continuous real-time three-dimensional coordinates, the fluid pressure monitoring data, the pH monitoring data, and the catheter historical state record into the risk prediction model to obtain a complication risk prediction value in a future period; Step P5: if the complication risk prediction value exceeds a preset warning threshold, triggering a prospective risk warning.

10. The safety monitoring method of claim 9, wherein, In the step P2: The condition for judging that a regurgitation event occurs is that the fluid pressure monitoring data experiences a sudden drop exceeding a pressure threshold within a preset time window, and the pH monitoring data is simultaneously lower than a preset acidity threshold and lasts for a set duration; The condition for judging that a blockage event occurs is that the fluid pressure monitoring data presents a stepwise rise and its gradient exceeds a preset gradient threshold, and the fluid flow rate calculated based on the fluid pressure monitoring data is lower than a preset flow rate threshold.

Images