Intelligent pressure monitoring and early warning system for end ileum fistulization tube

Abstract

The invention provides an intelligent pressure monitoring and early warning system for a tail ileum fistulization tube, and relates to the field of intelligent medical treatment, the system comprises a data acquisition module, a communication interface module, a terminal processing module, a user management module and an intelligent early warning module; the pressure sensor is used for collecting intestinal fluid pressure in a tail end ileum fistulization tube, a programmed algorithm is designed for data integration processing, an initial alarm threshold value is set, the threshold value is adjusted in real time through online reinforcement learning, and intelligent pipe fitting blockage monitoring and early warning without manual intervention are achieved. According to the system, the lumen pressure change is accurately monitored in real time, the alarm threshold value is adjusted according to the actual condition of a patient, high safety and adaptability are achieved, and a reference basis is provided for doctors to make clinical decisions.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent medical treatment, and more specifically, relates to an intelligent pressure monitoring and early warning system for a terminal ileal fistula tube. Background Art

[0002] Colorectal cancer (CRC) is one of the most common tumors of the digestive tract. Surgical resection is its conventional treatment. Anastomotic fistula is the most serious complication after surgery, with an incidence of 3%-24% and a related mortality rate of 6%-26%. In order to prevent the occurrence of anastomotic fistula, preventive terminal ileostomy is often performed after surgery to reduce anastomotic pressure and contamination to prevent complications such as anastomotic fistula, pelvic infection, and sepsis. For colorectal cancer, intestinal damage, intestinal obstruction caused by inflammatory bowel disease, intestinal perforation, and intestinal fistula, ostomy treatment is also required. However, traditional ostomy surgery may be accompanied by complications such as dehydration, parastomal hernia, stoma prolapse, parastomal dermatitis, anastomotic stenosis, disuse enteritis, and changes in anal sphincter function. Secondary surgery is required to close the stoma. Re-stomosis surgery also has the risks of incision infection, anastomotic leakage, intestinal adhesion, and intestinal obstruction.

[0003] The existing self-closing protective ileal cannula ostomy solves the problem of secondary surgery and avoids the occurrence of related complications, but it cannot completely divert intestinal fluid. At the same time, the tubes after the stoma are often blocked. Some patients even have tube-blocking intestinal obstruction, with an incidence rate of 4.69%-19.2%. Once it occurs, the pressure in the intestinal cavity gradually increases, and the risk of anastomotic leakage increases. The patient will also experience symptoms such as abdominal distension and abdominal pain. If the tube blockage fails to be recanalized, the patient must fast and receive total parenteral nutrition support, which will delay subsequent treatment, be unfavorable for postoperative recovery, and affect the prognosis.

[0004] The applicant has previously applied for a "pressure-sensing terminal ileostomy tube assembly device" (application date: 2023.7.28, application number: 202322031592.0), which is an adaptive medical device for the innovative double-tube terminal ileostomy, avoiding the secondary reduction problem of traditional ostomy surgery. In addition, compared with ileal intubation ostomy, it can completely block intestinal fluid by sensing the changes in fluid pressure in the intestinal cavity, and better prevent anastomotic leakage. However, its intelligent early warning of pipe blockage needs to be improved. Summary of the invention

[0005] In view of the above problems, based on the innovative double-tube terminal ileostomy and the matching medical device, an intelligent pressure monitoring and early warning system for terminal ileostomy tube is invented, which includes: a data acquisition module, a communication interface module, a terminal processing module, a user management module and an intelligent early warning module; the characteristics are as follows: the data acquisition module collects the intestinal fluid pressure data in the ostomy tube cavity, and converts the pressure signal into an electrical signal and then uploads it to the terminal processing module through the communication interface module; the terminal processing module receives and processes the pressure information and then outputs the real-time pressure value F x To the intelligent early warning module, the doctor inputs the patient information through the user management module and sets the initial threshold F0. The intelligent early warning module adjusts the alarm threshold through online reinforcement learning of the real-time prediction results and compares it with the real-time pressure to make an analysis and decision on the patient's intestinal pressure situation. After judgment, it outputs a warning signal to realize intelligent pressure monitoring and early warning of the terminal ileostomy tube.

[0006] Among them, the terminal processing module, user management module and intelligent warning module are all implemented in the host computer; the host computer includes a smart phone, a tablet computer and a smart bracelet.

[0007] Furthermore, the data acquisition module converts the collected pressure data into electrical signals and transmits the electrical signals to the terminal processing module through the communication interface module.

[0008] Furthermore, the terminal processing module includes a signal receiving unit, a signal conversion unit, a data processing unit and a data recording unit.

[0009] Furthermore, the signal receiving unit reads the electrical signal from the data acquisition module through a communication protocol.

[0010] Furthermore, the signal conversion unit decodes and analyzes the electrical signal to obtain a pressure value.

[0011] Furthermore, the data processing unit includes noise removal, normalization processing and data integration. The noise removal uses a suitable noise reduction method to reduce the noise of the data. The normalization processing normalizes or standardizes the data to eliminate the influence of different dimensions and ensure that the data is at the same scale. The data integration integrates the pressure values ​​of the three pressure sensors through program calculation and outputs them as real-time pressure values ​​F. x .

[0012] Furthermore, the data recording unit outputs the real-time pressure value F x Perform time stamping to generate real-time pressure curve C in the form of time series x , stored and uploaded to the patient database; the patient database is used to store the intestinal fluid pressure information of the summary data recording unit for subsequent query of patient historical information.

[0013] Furthermore, the user management module includes an information input unit, a pressure curve display unit, a history information query unit and a comprehensive control unit.

[0014] Furthermore, the information input unit can upload the patient's personal information to the system.

[0015] Furthermore, the personal information includes gender, age, height and body mass index (BMI), smoking history, drinking history, and tumor TNM stage, which are used by physicians for clinical management of patients;

[0016] Furthermore, the information input unit can input the clinical test results of the patient and set them as the initial alarm threshold F0, which is used as the initial threshold value of the reinforcement learning model to achieve initial intestinal fluid pressure warning.

[0017] Furthermore, the state space of the reinforcement learning model includes the current threshold, historical prediction probability, and historical alarm records; the action space includes possible threshold adjustments, such as increasing or decreasing F MAX ; The incentive function is a one-to-one mapping from prediction performance to reward score. The possible prediction results include correct alarm, false alarm and missed alarm. When setting the reward score, it is planned to set the correct alarm as a positive score, and the false alarm and missed alarm as a negative score. At the same time, the penalty for missed alarm is increased relative to the false alarm to reduce the impact of missed alarm on clinical treatment.

[0018] Furthermore, the specific steps of the clinical experiment are as follows: in the double-tube terminal ileostomy surgery, the proximal tube body is placed in the terminal ileum, and the balloon is inflated so that the balloon fits the intestinal wall exactly, the outer opening of the fistula tube is closed, the upper intestinal tube is squeezed to simulate intestinal peristalsis, and the pressure value when the intestinal fluid leaks into the distal ileum through the gap between the balloon and the intestinal wall is observed and recorded.

[0019] Furthermore, the pressure curve display unit displays the real-time pressure curve C in the form of a time series. x It is displayed synchronously on the screen of the host computer to meet the doctor's real-time monitoring of the patient's intestinal pressure.

[0020] Furthermore, the historical information query unit can retrieve the historical pressure curve C in the patient database. t The physician combines the real-time pressure curve C x The patient's disease outcome can be assessed.

[0021] Furthermore, the integrated control unit can edit the software program code in the terminal processing module to realize different application functions.

[0022] Furthermore, the intelligent early warning unit includes a threshold generation unit and an early warning output unit.

[0023] Furthermore, the specific working steps of the threshold generation unit are as follows:

[0024] S1: During initialization, the initial pressure threshold F0 is set according to clinical experiments;

[0025] S2: During online training, the model selects actions based on the current state space and changes the state space;

[0026] S3: If an alarm or missed alarm event occurs, adjust the threshold update strategy and model state space according to the prediction results and corresponding incentives;

[0027] S4: Repeat the above online training steps to adjust the blockage warning threshold according to the actual clinical situation;

[0028] S5: Establish a pressure threshold generation model that can be adjusted online in real time to provide real-time warning of fistula tube blockage.

[0029] When initializing the model, the initial stress threshold F0 is set according to the clinical experimental results, and the historical prediction probability and historical alarm record are set to 0; as shown in formula (I), it is the expected reward Q of selecting action a under state s at time t. The model selects the action with the highest expected reward according to the current state space and changes the stress threshold in the state space.

[0030] Q(s,a)=E[R t+1 +γmax a′ Q(s t+1 ,a′)|s t =s,a t =a] (Ⅰ)

[0031] Where: R t+1 is the reward obtained at t+1, γ is the discount factor that determines the value of future rewards, and s t+1 is the state obtained at t+1, and a′ is the possible action under the new state;

[0032] If an alarm or missed alarm event occurs, the threshold update strategy and model state space will be adjusted according to the prediction result performance and the corresponding incentives. Then the above online training steps will be repeated to adjust the blockage warning threshold according to the actual clinical situation.

[0033] Furthermore, the early warning output unit can realize the blockage prediction of the fistula tube by comparing the real-time threshold with the real-time pressure value. The specific working principle is as follows: the terminal processing module uploads the real-time pressure value to the intelligent early warning module, and the time interval between each upload is 1s. The early warning output unit combines the pressure threshold generated by the threshold generation unit to complete the real-time pressure value F x With real-time pressure threshold F MAXIf the threshold comparison result is F x <F MAX , it indicates that there is no intestinal fluid accumulation or only a small amount of intestinal fluid accumulation in the fistula tube, and no warning signal is generated. If the threshold comparison result is F x ≥F MAX , it indicates that a large amount of intestinal fluid has accumulated in the fistula tube or the fistula tube has become narrow or blocked, generating a warning signal.

[0034] Furthermore, the warning output unit sends the warning signal to the buzzer via Bluetooth. After the buzzer receives the warning signal, the internal driving circuit will provide an alternating current signal to the diaphragm. The diaphragm vibrates with the change of the electrical signal, produces a specific sound, realizes the transformation of the electrical signal into a physical signal, and promptly reminds the patient and the physician to perform clinical treatment.

[0035] Furthermore, after the clinical treatment is completed, the physician may upload the warning performance results to the threshold generation unit to complete the real-time threshold update operation.

[0036] Furthermore, the warning signal has a very strong timeliness, and the upload frequency of the real-time pressure value is 1s / time. The operation time of the intelligent warning module plus the signal transmission delay can be controlled within 2-5s, thereby ensuring that feedback can be obtained in a very short time after the fistula tube is blocked, avoiding irreversible damage to the patient due to poor drainage.

[0037] Furthermore, the clinical treatment includes but is not limited to the following: using a fistula tube adapter brush to rotate and insert it through the external opening to clear the blocked part of the tube body, using a syringe to inject normal saline into the proximal tube, further flushing the tube body to achieve the purpose of complete clearing, flushing and administering medicine to the colon through the distal tube, providing intestinal nutritional support, and reducing contamination at the anastomosis.

[0038] Due to the adoption of the above-mentioned technical scheme, the present invention has the following advantages and positive effects compared with the prior art: the present invention can accurately monitor the internal pressure changes of the fistula tube in real time, and send out a warning signal when the pressure deviates from the normal value, thereby providing a reference basis for doctors to make clinical decisions and preventing the occurrence of tube-blocking intestinal obstruction. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the embodiments or descriptions of the prior art will be briefly introduced below. The drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments that conform to the present application and are used together with the specification to explain the principles of the present application.

[0040] Figure 1 It is the overall system framework diagram of the present invention;

[0041] Figure 2is a system structure diagram of the present invention;

[0042] Figure 3 is the signal conversion roadmap in the present invention;

[0043] Figure 4 It is the reinforcement learning technology roadmap in the present invention;

[0044] Figure 5 Generate a logical route map for the early warning signal in the present invention;

[0045] Figure 6 A schematic diagram of the operation of the double-tube terminal ileostomy provided by the present invention.

[0046] Legend: 1. Patient; 2. Host computer; 3. Buzzer; 4. Physician; 5. Cecum; 6. Appendix; 7. Terminal ileum; 8. Anastomosis; 9. Proximal tube; 10. Distal tube; 11. First valve interface; 12. Second valve interface; 13. Third valve interface; 14. First airbag; 15. Second airbag; 16. Disc-shaped airbag. DETAILED DESCRIPTION

[0047] The following will be combined with the accompanying drawings of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] Please see attached Figure 1-6 , an intelligent pressure monitoring and early warning system for terminal ileal fistula tube, including a data acquisition module, a communication interface module, a terminal processing module, a user management module, and an intelligent early warning module; wherein the terminal processing module, the user management module, and the intelligent early warning module are all implemented in a host computer 2, and the host computer 2 includes a smart phone, a tablet computer, and a smart bracelet.

[0049] In a specific embodiment, based on the original self-closing protective ileal intubation ostomy, an innovative double-tube terminal ileostomy is implemented, and the specific steps during the operation are as follows: the lateral peritoneum is incised along the outer side of the ileocecal region, and the ileocecal region is fully freed so that the ileocecal region is in a relaxed state; a midline incision of about 5 cm long is made below the umbilicus in the lower abdomen, and the skin, subcutaneous tissue, median white line, and peritoneum are incised, and a hole of about 1.5 cm long is made at the intersection 3 cm to the right of the midline incision and about 3 cm below the right side of the umbilicus; a 0.5 cm incision is made in the intestinal wall of the terminal ileum on the mesenteric edge 15.0 cm away from the ileocecal valve, and the front half of the proximal tube 9 is inserted into the proximal side of the terminal ileum 7 through the ileal incision, and air or water is injected through the first valve interface 11 to expand and fill the first airbag 14, thereby blocking the intestinal contents from entering the cecum 5 and preventing the proximal tube from entering the cecum 5. 9 falls off; close the outer opening of the fistula tube, squeeze the upper intestinal tube to simulate intestinal peristalsis, observe and record the pressure value when the intestinal fluid leaks into the distal ileum through the gap between the air bag and the intestinal wall; under the guidance of the guide wire, place the disc-shaped air bag 16 at the front end of the distal tube 10 of the terminal ileostomy tube in the cecum 5, inject air or water through the third valve interface 13 to fill the disc-shaped air bag 16, and then inject air or water through the second valve interface 12 to fill the second air bag 15, and the inflated air bag should not squeeze the ileocecal valve; embed the distal tube 10 and the proximal tube 9 tube body with absorbable thread purse string suture, and fix the parietal peritoneum of the ileum wall with absorbable thread, cut a hole from the right lower abdomen, and lead out and fix it; No. 1 absorbable thread is on the periphery of the peritoneum where the ileum wall is fixed, and after the purse string suture, the two ends of the reserved thread are led out from the skin around the tube; complete the double-tube terminal ileostomy.

[0050] After the operation, the physician 4 collects and organizes the personal information and surgical records of the patient 1, and uploads the personal information of the patient 1 and the initial alarm threshold F0 through the host computer 2. The specific implementation route is as follows: the information input unit in the user management module can upload the personal information of the patient 1 and the initial alarm threshold F0 to the system, and the personal information includes gender, age, height and weight index (BMI), smoking history, drinking history, and tumor TNM stage; the initial alarm threshold F0 is the result of the blockage experiment simulated during the operation, which is used as the initial threshold value of the reinforcement learning in the early warning generation unit.

[0051] After the fistula tube is successfully inserted and the basic function settings of the host computer 2 are completed, the intelligent pressure monitoring and early warning system for the terminal ileal fistula tube officially begins to operate.

[0052] The three MEMS pressure sensors embedded in the wall of the proximal tube 9 monitor the pressure in the lumen in real time and output a voltage signal proportional to the pressure through resistance changes. The communication interface module transmits the electrical signal to the host computer 2 via Bluetooth.

[0053] The terminal processing module in the host computer 2 receives, converts, processes and records the electrical signals, and inputs the processed pressure data to the intelligent early warning module.

[0054] The specific working principle is as follows: the signal receiving unit reads the electrical signal from the data acquisition module through the communication protocol; the signal conversion unit decodes and analyzes the electrical signal to calculate the pressure value; the data processing unit includes noise removal, normalization processing and data integration. The noise removal compares and tests multiple algorithms on the actual data according to the characteristics of the data, the type of noise and the calculation efficiency, and selects the optimal denoising strategy to reduce the noise of the data. The normalization processing normalizes or standardizes the data to eliminate the influence of different dimensions and ensure that the data is on the same scale. The data integration integrates the pressure values ​​of the three pressure sensors through program calculations, and outputs the real-time pressure value Fx for storage and upload to the intelligent early warning module in real time.

[0055] The intelligent early warning module adjusts the real-time threshold to achieve intelligent pressure warning of the terminal ileostomy tube through online reinforcement learning of the real-time prediction results; the reinforcement learning model includes state space, action space, incentive function and strategy, among which the state space includes the current threshold, historical prediction probability and historical alarm record; the action space includes possible threshold adjustment, such as increasing or decreasing F MAX ; The incentive function is a one-to-one mapping from prediction performance to reward score. Possible prediction result performances include correct alarm, false alarm and missed alarm. When setting the reward score, it is proposed to set the correct alarm as a positive score, and the false alarm and missed alarm as negative scores. At the same time, the penalty for missed alarm is increased relative to the false alarm to reduce the impact of missed alarm on clinical treatment.

[0056] The construction principle of the threshold generation unit is as follows: when initializing the model, the initial pressure threshold F0 will be set according to the clinical experimental results, and the historical prediction probability and historical alarm record will be set to 0; the model selects actions according to the current state space and changes the pressure threshold in the state space; if an alarm or missed event occurs, the threshold update strategy and model state space will be adjusted according to the prediction result performance and corresponding incentives; then the above online training steps are repeated to adjust the blockage warning threshold according to the actual clinical situation; finally, a pressure threshold generation model that can be adjusted online in real time is established to provide real-time warning of fistula tube blockage.

[0057] The specific method of generating a congestion signal is as follows: the terminal processing module uploads the real-time pressure value to the intelligent early warning module, and the time interval between each upload is 1s. The early warning output unit combines the pressure threshold generated by the threshold generation unit to complete the real-time pressure value F x With real-time pressure threshold F MAX If the threshold comparison result is F x <F MAX , it indicates that there is no intestinal fluid accumulation or only a small amount of intestinal fluid accumulation in the fistula tube, and no warning signal is generated. If the threshold comparison result is F x≥F MAX , it indicates that a large amount of intestinal fluid has accumulated in the fistula tube or the fistula tube has become narrow or blocked, generating a warning signal.

[0058] When the intelligent early warning module outputs the early warning signal, the early warning output unit sends the early warning signal to the buzzer 3 via Bluetooth. After the buzzer 3 receives the early warning signal, the internal driving circuit will provide an alternating current signal to the diaphragm. The diaphragm vibrates with the change of the electric signal, generates a specific sound, realizes the transformation of the electric signal into a physical signal, and promptly reminds the patient 1 and the physician 4 to perform clinical treatment.

[0059] After clinical processing, the physician 4 uploads the warning performance results to the threshold generation unit to complete the real-time threshold update operation.

[0060] Furthermore, the warning signal has a very strong timeliness, and the upload frequency of the real-time pressure value is 1s / time. The operation time of the intelligent warning module plus the signal transmission delay can be controlled within 2-5s, thereby ensuring that feedback can be obtained in a very short time after the fistula tube is blocked, avoiding irreversible damage to patient 1 caused by poor drainage.

[0061] Clinical treatment includes but is not limited to the following: using a fistula tube adapter brush to rotate and insert it through the external opening to clear the blocked part of the tube body, using a syringe to inject normal saline into the proximal tube 9, further flushing the tube body to achieve the purpose of complete clearing, flushing and administering medicine to the colon through the distal tube 10, providing intestinal nutritional support, and reducing contamination at the anastomosis.

[0062] In addition, the doctor 4 can edit the software program code in the terminal processing module through the integrated control unit in the host computer 2 according to actual work requirements to meet the needs of other application functions.

[0063] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent pressure monitoring and early warning system for terminal ileostomy tubes, comprising a data acquisition module, a communication interface module, a terminal processing module, a user management module and an intelligent early warning module, characterized in that: The data acquisition module collects the intestinal fluid pressure in the fistula tube cavity detected by the pressure sensor in real time, and converts the pressure signal into an electrical signal and transmits it to the terminal processing module via the communication interface module; The terminal processing module receives and processes the pressure information and outputs the real-time pressure value F x To the intelligent early warning module, the physician inputs the patient information through the user management module and sets the initial threshold F0; The intelligent early warning module adjusts the alarm threshold through online reinforcement learning of real-time prediction results, and makes analysis and decision on the intestinal pressure situation of the patient by comparing it with the real-time pressure. After judgment, it outputs an early warning signal to realize intelligent pressure monitoring and early warning of the terminal ileal fistula tube.

2. According to the intelligent pressure monitoring and early warning system for terminal ileostomy tube according to claim 1, it is characterized in that: The pressure sensor is configured as a micro-electro-mechanical system sensor (MEMS) with a size of 1 mm×2.5 mm×4 mm. A high-precision semiconductor resistance strain gauge is used to form a Wheatstone bridge as a force-to-electricity conversion measurement circuit. When external pressure acts on the sensor, it will cause a slight deformation of the silicon film, thereby causing a change in the resistance value of the strain gauge. These resistance changes will destroy the balance of the bridge, thereby outputting a voltage signal proportional to the pressure. Three MEMS pressure sensors are arranged in a ring and embedded in the wall of the proximal end of the terminal ileal fistula tube to form a 3-channel pressure sensor group for achieving comprehensive and accurate monitoring of intestinal fluid pressure in the fistula tube.

3. According to the intelligent pressure monitoring and early warning system for terminal ileostomy tubes as described in claim 1, it is characterized by: The communication interface module transmits the collected electrical signals to the terminal processing module via Bluetooth.

4. According to the intelligent pressure monitoring and early warning system for terminal ileostomy tubes as described in claim 1, it is characterized in that: The terminal processing module includes a signal receiving unit, a signal conversion unit, a data processing unit and a data recording unit; The signal receiving unit reads the electrical signal through the communication protocol; The signal conversion unit decodes and analyzes the electrical signal to obtain a pressure value; The data processing unit includes noise removal, normalization processing and data integration. The noise removal uses a suitable noise reduction method to reduce the noise of the data. The normalization processing normalizes or standardizes the data to eliminate the influence of different dimensions and ensure that the data is at the same scale. The data integration integrates the pressure values ​​of the three pressure sensors through program calculation and outputs them as real-time pressure values ​​F x ; The data recording unit outputs the real-time pressure value F x Perform time stamping to generate real-time pressure curve C in the form of time series x , transmitted to the user management module and stored and uploaded to the patient database.

5. According to the intelligent pressure monitoring and early warning system for terminal ileostomy tube according to claim 1, it is characterized in that: The user management module includes an information input unit, a pressure curve display unit, a historical information query unit and a comprehensive control unit; The information input unit uploads the patient's personal information to the system for the physician to manage the patient clinically. The personal information includes gender, age, height and weight index (BMI), smoking history, drinking history, and tumor TNM stage. In addition, the information input unit inputs the patient's clinical test results and sets them as the initial alarm threshold F0, which is used as the initial threshold value of the reinforcement learning model. The pressure curve display unit displays the real-time pressure curve C in the form of a time series. x The data is displayed synchronously on the upper computer screen, so that the physician can view the real-time dynamic changes of the patient's intestinal pressure. The historical information query unit retrieves the historical pressure curve C in the patient database. t , used by doctors to evaluate the patient's condition outcome; The integrated control unit edits the software program codes in the terminal processing module to realize different application functions.

6. According to the intelligent pressure monitoring and early warning system for terminal ileostomy tubes as described in claim 1, it is characterized in that: The intelligent early warning module includes a threshold generation unit and an early warning output unit. The threshold generation unit generates a pressure threshold F MAX The output is sent to the early warning output unit, and the pressure threshold is adjusted online according to the prediction result performance using the reinforcement learning method of the reinforcement learning model.

7. The intelligent pressure monitoring and early warning system for terminal ileostomy tube according to claim 6, characterized in that: The reinforcement learning model includes a state space, an action space and an activation function. The state space includes the current threshold, the historical prediction probability, and the historical alarm record. The action space includes increasing or decreasing F MAX The threshold value is adjusted, the incentive function is a mapping of prediction performance to reward score, and the prediction result performance includes correct alarm, false alarm and missed alarm.

8. The intelligent pressure monitoring and early warning system for terminal ileostomy tube according to claim 6, characterized in that: The specific construction steps of the threshold generation unit are as follows: S1: During initialization, the initial pressure threshold F0 is set according to clinical experiments; S2: During online training, the model selects actions based on the current state space and changes the state space; S3: If an alarm or missed alarm event occurs, adjust the threshold update strategy and model state space according to the prediction results and corresponding incentives; S4: Repeat the above online training steps and adjust the blockage warning threshold according to the actual clinical situation; S5: Establish a pressure threshold generation model that can be adjusted online in real time to provide real-time warning of fistula tube blockage.

9. The intelligent pressure monitoring and early warning system for terminal ileostomy tube according to claim 6, characterized in that: The early warning output unit realizes the blockage prediction of the fistula tube through the real-time threshold value calculated by the threshold value generation unit. x Greater than or equal to the pressure threshold F MAX If a blockage warning signal is generated, an electrical signal will be sent to the buzzer, which will convert the blockage warning signal into a physical signal to remind doctors and patients to perform clinical treatment in a timely manner.

Images