Paralyzed patient excrement collecting device and using method

Abstract

The invention belongs to the technical field of robots, and discloses a paralyzed patient excrement collecting device and a using method, the device comprises a sickbed, an excrement collecting groove, a supporting plate, a rectangular frame, a transposition driving mechanism, an excrement collecting box, a water tank and a monitoring mechanism; a transposition driving mechanism is arranged below the sickbed; the transposition driving mechanism is in transmission connection with a rectangular frame; an excrement collecting tank and an electric push rod are arranged on the rectangular frame; a supporting plate is arranged at the upper end of a movable rod of the electric push rod; the lower end of the excrement collecting tank is communicated with an excrement collecting pipe; a water suction pump is fixedly arranged on the water tank; an annular water spraying pipe is fixedly arranged at the upper end of the excrement collecting tank; a monitoring mechanism is arranged on the sickbed. The defecation requirement, the defecation state and the abnormal condition of the patient can be automatically judged in time; the automatic operation function is achieved, and the manual operation steps and time of nursing personnel are reduced; whether a semi-paralyzed patient is controlled by himself or a completely paralyzed patient is assisted by medical care, the operation can be completed on the sickbed.

Description

Technical Field

[0001] The present application relates to the field of robotics, and more specifically, to a feces collection device for paralyzed patients and a method of using the device. Background Art

[0002] For patients who have been bedridden for a long time, defecation in daily life has become a difficult problem. When they need to defecate, they must rely on the assistance of medical staff. Medical staff need to carefully help the patient out of bed first, and then help him to the bathroom step by step. This process is extremely cumbersome, not only consuming a lot of time, but also requiring medical staff to exert great physical strength. Every assistance may make medical staff exhausted and out of breath, especially when facing patients with heavy weight or serious illness, the difficulty is doubled. In order to improve this situation, a series of convenient nursing beds have appeared on the market. These nursing beds are designed with full consideration of the needs of patients, and many of them are equipped with toilets, trying to fundamentally solve the problem of defecation for patients. When patients need to defecate, they can defecate directly on the bed through some simple operations on the nursing bed, without having to be helped to go to the bathroom, which undoubtedly reduces the burden on patients and medical staff to a great extent.

[0003] The document with the prior art publication number CN111166581A provides a feces collection and transportation robot for the nursing bed of the novel coronavirus infection and paralysis patients, including a control box, a walking component is arranged at the bottom of the control box, a collection box with a feces receiving port is arranged at the top of the control box, an automatic cover is arranged at the feces receiving port, an electric heating wire is arranged inside the collection box, a feces discharge port is arranged at the bottom of the collection box, and the control box is used to control the walking of the walking component and the working state of the power supply device. The walking component can automatically walk to the bottom of the nursing bed, receive feces through the feces receiving port, and then sterilize them at high temperature through the electric heating wire, and finally pump them out through the feces discharge port, that is, the feces can be automatically collected and transported without the need for nursing staff to touch and handle, which ensures health and safety while saving labor.

[0004] Although the above-mentioned prior art solutions can achieve relevant beneficial effects through the structure of the prior art, they still have the following defects: 1. The prior art lacks effective automatic monitoring means, and nursing staff need to be on duty at all times to observe the patient's defecation needs, defecation status and abnormal conditions, making it difficult to deal with patient needs in a timely manner. In the preparation and cleaning work before and after the patient's defecation, it mostly relies on manual operation by nursing staff, which is cumbersome and time-consuming, seriously affecting the nursing efficiency. 2. For paralyzed patients with limited mobility, the prior art often requires patients to move their bodies to a special toilet to defecate, which will cause great pain and discomfort to the patients during the process, and cannot provide patients with a comfortable defecation experience. 3. Patients defecate under the prior art, or need to move to a public toilet, or with the assistance of others, lack a relatively private environment, which can easily lead to embarrassment for patients, and there is also a problem of privacy leakage, which damages the dignity of patients. 4. It is difficult for the prior art to accurately evaluate the patient's defecation situation, and it is impossible to analyze it comprehensively from multiple factors, so it is impossible to provide personalized nursing measures for different types of patients such as complete paralysis and semi-paralysis, and it is impossible to meet the diverse needs of patients.

[0005] In view of this, we propose a feces collection device for paralyzed patients and a method of use. Summary of the invention

[0006] 1. Technical problems to be solved

[0007] The purpose of the present application is to provide a feces collection device for paralyzed patients and a method of use, which solves the technical problems raised in the above-mentioned background technology, and realizes the coordinated work of multiple modules of the monitoring mechanism, such as image acquisition, physiological characteristics acquisition, image analysis and comprehensive evaluation modules, which can automatically and timely judge the patient's defecation needs, defecation status and abnormal conditions; with automatic operation function, the device's electric push rod, transposition drive mechanism, electric telescopic rod and other automated components can automatically complete the movement of the support plate, the lifting and lowering of the urine shield and other actions, reducing the steps and time of manual operation of nursing staff; during the defecation process, whether it is a semi-paralyzed patient controlling it by himself or a completely paralyzed patient assisted by medical staff, it can be completed on the bed without moving the body to a special bathroom, reducing the patient's pain and discomfort, and the technical effect is particularly suitable for paralyzed patients with limited mobility.

[0008] 2. Technical solution

[0009] The technical solution of the present application provides a feces collection device for paralyzed patients, including: a hospital bed, a feces collection trough, a support plate, a rectangular frame, a transposition drive mechanism, a movable trolley, a feces collection box, a water tank and a monitoring mechanism.

[0010] The sickbed is provided with a defecation groove; a transposition driving mechanism is fixedly arranged under the sickbed; and a rectangular frame is arranged on the transposition driving mechanism for transmission connection.

[0011] A feces collecting trough and an electric push rod are fixedly arranged on the rectangular frame, and a support plate is fixedly arranged on the upper end of the movable rod of the electric push rod; the shape and size of the support plate are adapted to the defecation trough; when the support plate is stuck in the defecation trough, the upper end of the support plate is flush with the hospital bed to form a complete hospital bed; the transposition drive mechanism can drive the rectangular frame to move and adjust the position, so that the support plate and the feces collecting trough are alternately located under the defecation trough; adjustable armrests are arranged on both sides of the bed to facilitate the patient's grasping and enhance his sense of security and comfort on the bed.

[0012] A feces collection box and a water tank are fixedly arranged above the movable trolley; a plug-in fixing seat is fixedly arranged on the feces collection box, and a feces collection pipe is fixedly arranged at the lower end of the feces collection tank, and the lower end of the feces collection pipe is detachably fixed on the plug-in fixing seat, so that the feces collection pipe is connected to the feces collection box; a one-way valve that can be opened and closed is arranged at the connection between the feces collection pipe and the feces collection tank to prevent feces from flowing back during transportation. A feces discharge pipe is installed at the bottom of the feces collection box, and a valve is arranged on the feces discharge pipe.

[0013] A water pump is fixedly arranged on the water tank, a water supply pipe is fixedly arranged on the output end of the water pump, and the input end of the water pump extends into the bottom of the water tank; an annular water spray pipe is fixedly arranged on the upper end of the feces collecting trough, and the water supply pipe is connected with the annular water spray pipe; a plurality of water spray holes inclined inwardly are arranged on the annular water spray pipe, and water can be sprayed to the inner wall of the feces collecting trough for cleaning through the annular water spray pipe.

[0014] A monitoring device is fixed on the bed to monitor the defecation needs of paralyzed patients and detect abnormal situations in time.

[0015] The bed is equipped with a special controller. The operator can easily switch the position of the support plate and the feces collection tank through the buttons or touch screen on the controller. At the same time, the controller also has an automatic return function. After completing a feces collection operation, the device can be controlled to return to the initial state with one button, that is, the support plate is flush with the bed. The support plate is made of high-strength engineering plastics and covered with a layer of soft silicone pad on the surface to increase the comfort when in contact with the patient's body.

[0016] Through the above technical scheme, the upper end of the support plate is flush with the bed in the initial state to form a complete bed. When the paralyzed patient lies on the bed, the electric push rod is started to drive the support plate to move down and away from the defecation trough when defecation is needed, and then the transposition drive mechanism drives the support plate and the feces collecting trough to move synchronously, the support plate moves to the outside of the defecation trough, and the feces collecting trough moves to the bottom of the defecation trough at the same time; the paralyzed patient can defecate through the defecation trough on the bed; the feces of the patient are received by the feces collecting trough, and the feces discharged by the patient are transported to the feces collecting box through the feces collecting pipe. After the patient finishes defecation, the water in the water tank can be pumped to the annular water spray pipe through the water pump and the water supply pipe, and the inner wall of the feces collecting trough 2 is sprayed with water through the annular water spray pipe for cleaning, so as to keep the inner wall of the feces collecting trough clean. The defecation of the paralyzed patient is monitored by the monitoring mechanism, and abnormal conditions are discovered in time.

[0017] As an optional solution of the present invention, an electric telescopic rod is fixedly arranged at the bottom of the feces collection tank, and a urine shield is fixedly arranged on the movable rod of the electric telescopic rod. The urine shield is in the shape of a structure with a wide upper end and a narrow lower end. The urine shield can receive and block the urine of the patient to prevent the urine from spraying onto the bed and dirtying the quilt. A sponge is arranged inside the urine shield to prevent urine from splashing onto the outside of the urine shield.

[0018] As an optional solution of the present invention, the transposition drive mechanism includes a motor, a lead screw and a polished rod.

[0019] A motor is fixedly arranged under the bed; a screw rod and a smooth rod are rotatably arranged under the bed; the screw rod is coaxially fixedly connected with the output end of the motor; the screw rod is threadedly connected with the rectangular frame, and the smooth rod is slidably connected with the rectangular frame.

[0020] Through the above technical solution, the starting motor drives the screw rod to rotate, the screw rod drives the rectangular frame to move, and the rectangular frame drives the feces collecting trough and the supporting plate to move and adjust the position.

[0021] As an optional solution of the present invention, the monitoring mechanism includes:

[0022] Data collection module: collects patient data, such as medical history information, past bowel movement patterns, and physical data, such as heart rate, blood pressure, abdominal pressure changes, etc. The collected data is annotated in a specific format and used as a reference sample for subsequent analysis.

[0023] Image acquisition module: High-definition cameras are installed on the bed and stool collection trough to collect images of the patient's face, limbs and defecation.

[0024] Image preprocessing module: preprocess the collected images, including filtering, denoising, grayscale and normalization.

[0025] Physiological characteristics collection module: real-time collection of patients' physiological characteristics, such as heart rate, blood pressure, muscle tension, etc.

[0026] Feature extraction module: extract features of the preprocessed images, including color, texture and shape. For facial images, extract facial features, such as frowning, pursing lips and other expressions that may be related to the need for defecation. For limb images, extract color, texture and shape features related to movement amplitude, limb extension direction, etc.; for images in feces collection troughs, extract feces features such as color, shape, size, etc.

[0027] Image analysis module: Analyze and identify the image after feature extraction, and promptly identify when the patient needs to defecate based on the patient's facial expressions and body movements; identify when the patient has finished defecation; and promptly detect abnormal defecation.

[0028] Comprehensive evaluation module: combines the analysis results of the image analysis module with the monitoring results of the physiological characteristics acquisition module (such as heart rate, blood pressure, muscle tension, etc.) to evaluate and predict whether the patient is about to defecate or whether the defecation has been completed.

[0029] Alarm module: includes an alarm, which will issue an alarm in time when abnormal defecation is detected.

[0030] PLC control module: connected with the data collection module, image acquisition module, image preprocessing module, feature extraction module, image analysis module and alarm module network. PLC control module is connected with the controller through a wireless signal network.

[0031] The present invention provides a method for using a feces collection device for paralyzed patients, comprising the following steps:

[0032] S1. In the initial state, the upper end of the support plate is flush with the bed, forming a complete bed, and the paralyzed patient lies on the bed.

[0033] S2. The monitoring agency monitors the defecation needs of paralyzed patients.

[0034] S21. The image acquisition module acquires images of the patient's face, limbs, and bowel movements.

[0035] S22. The physiological characteristics acquisition module acquires the patient's physiological characteristics in real time, such as heart rate, blood pressure, muscle tension, etc.

[0036] S23, the image preprocessing module preprocesses the collected image, including filtering, denoising, grayscale and normalization.

[0037] S24, the feature extraction module performs feature extraction on the preprocessed image, and the extracted features include color, texture and shape; for facial images, it extracts expression features, such as frowning, pursing lips and other expressions that may be related to the need for defecation, which correspond to color, texture and shape features; for limb images, it extracts color, texture and shape features related to movement amplitude, limb extension direction, etc.; for images in the feces collection trough, it extracts feces features such as color, shape, size, etc.

[0038] S25. The image analysis module analyzes and identifies the image after feature extraction, and promptly identifies the patient's need to defecate based on the patient's facial expression and body movements; identifies the patient's end of defecation; and promptly detects abnormal defecation.

[0039] S26. The comprehensive evaluation module combines the analysis results of the image analysis module with the monitoring results of the physiological characteristics acquisition module (such as heart rate, blood pressure, muscle tension, etc.) to evaluate and predict whether the patient is about to defecate or whether the defecation has been completed.

[0040] S27. When abnormal defecation is detected, the alarm module promptly issues an alarm (for example, a completely paralyzed patient needs medical assistance to defecate or has abnormal defecation).

[0041] S3. When defecation is required, the electric push rod is started to drive the support plate to move down and out of the defecation trough, and then the transposition drive mechanism drives the support plate and the feces collection trough to move synchronously. The support plate moves to the outside of the defecation trough, and the feces collection trough moves to the bottom of the defecation trough.

[0042] S4. Start the electric telescopic rod to drive the urine shield to rise, and the urine shield is used to receive and block the patient's urine to prevent the urine from spraying onto the bed and dirtying the quilt, so as to keep the surrounding environment clean.

[0043] S5. For semi-paralyzed patients (patients who can take off their pants to defecate by themselves), the opening and closing can be controlled by the controller; for completely paralyzed patients, medical staff are required to assist in taking off the patient's pants so that the patient's buttocks are in the defecation trough for defecation.

[0044] S6. Paralyzed patients can defecate through the defecation trough on the hospital bed; the feces of the patient are received by the feces collection trough, and the feces discharged by the patient are transported to the feces collection box through the feces collection pipe.

[0045] S7. After the patient finishes defecating, the urine shield is automatically controlled to descend and be stored in the feces collection trough. Then the shifting drive mechanism drives the support plate and the feces collection trough to move synchronously. The support plate moves to the bottom of the defecation trough, and the feces collection trough moves to the outside of the defecation trough at the same time. The electric push rod drives the support plate to move up and get stuck in the defecation trough, and the support plate is flush with the upper end of the bed.

[0046] S8. The water in the water tank is pumped to the annular water spray pipe through a water pump and a water supply pipe, and water is sprayed to the inner wall of the feces collecting tank through the annular water spray pipe for cleaning, so as to keep the inner wall of the feces collecting tank clean.

[0047] 3. Beneficial effects

[0048] One or more technical solutions provided in the technical solution of this application have at least the following technical effects or advantages:

[0049] 1. The present invention can automatically and timely judge the patient's defecation needs, defecation status and abnormal conditions through the coordinated work of multiple modules of the monitoring mechanism, such as image acquisition, physiological characteristics acquisition, image analysis and comprehensive evaluation modules. For example, when abnormal defecation is detected, the alarm module automatically issues an alarm, and there is no need for nursing staff to be on duty at all times, which greatly improves the response speed of nursing work and allows nursing staff to arrange work more efficiently and deal with patient needs in a timely manner.

[0050] 2. With automatic operation function, the device's electric push rod, transposition drive mechanism, electric telescopic rod and other automated components can automatically complete actions such as moving the tray and raising and lowering the urine shield, reducing the steps and time of manual operation for nursing staff, and greatly improving nursing efficiency in the preparation and cleaning work before and after defecation.

[0051] 3. It can improve the patient experience. In the initial state, the support plate is flush with the bed, forming a complete bed, providing a comfortable lying surface for the patient. During defecation, whether it is a semi-paralyzed patient who controls it by himself or a completely paralyzed patient who is assisted by medical staff, it can be completed on the bed without moving the body to a special toilet, reducing the patient's pain and discomfort, especially suitable for paralyzed patients with limited mobility.

[0052] 4. It can protect the privacy of patients. The entire defecation process is carried out in a relatively private environment on the bed, avoiding the embarrassment and privacy leakage caused by patients moving to public toilets or defecating with the assistance of others, and protecting the dignity of patients.

[0053] 5. The urine shield can effectively receive and block urine, preventing it from spraying onto the bed and dirtying the quilt, thus avoiding hygiene problems and infection risks caused by urine contamination. At the same time, the design of the feces collection tank and collection box can collect feces in time and reduce the spread of odor and bacteria. After defecation, the inner wall of the feces collection tank is automatically sprayed with water through the water pump and the annular water spray pipe to keep the collection tank clean, further reducing the possibility of bacterial growth and cross infection, and helping to maintain the health of patients and caregivers.

[0054] 6. The comprehensive assessment module combines image analysis and physiological characteristics monitoring results to more accurately assess the patient's defecation situation and provide a basis for personalized care. According to the monitoring results, targeted nursing measures can be provided for completely paralyzed and semi-paralyzed patients. For example, completely paralyzed patients need medical staff to assist in defecation, while semi-paralyzed patients can control some operations by themselves, meeting the personalized needs of different patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is an overall schematic diagram of a feces collection device for paralyzed patients disclosed in a preferred embodiment of the present application.

[0056] Figure 2 This is a schematic diagram of the bottom structure of a feces collection device for paralyzed patients disclosed in a preferred embodiment of the present application.

[0057] Figure 3 This is a schematic diagram of a state in which a support plate of a feces collection device for paralyzed patients disclosed in a preferred embodiment of the present application is flush with the upper end of a hospital bed.

[0058] Figure 4 This is a schematic diagram of a state in which a feces collection tank of a feces collection device for paralyzed patients disclosed in a preferred embodiment of the present application is located below a defecation tank.

[0059] Figure numerals: 1. hospital bed; 2. feces collecting trough; 3. urine shield; 4. support plate; 5. rectangular frame; 6. transposition drive mechanism; 7. movable trolley; 8. feces collecting box; 9. water tank; 10. plug-in fixing seat; 11. feces collecting pipe; 12. water pump; 13. water supply pipe; 14. feces discharge pipe; 15. valve; 16. feces discharge trough; 21. annular water spray pipe; 31. electric telescopic rod; 41. electric push rod; 61. motor; 62. screw rod; 63. bare rod. DETAILED DESCRIPTION

[0060] The present application is further described in detail below in conjunction with the accompanying drawings.

[0061] Reference Figure 1 , Figure 2 , Figure 3 and Figure 4 The embodiment of the present application provides a feces collection device for paralyzed patients, including: a bed 1, a feces collection trough 2, a support plate 4, a rectangular frame 5, a transposition drive mechanism 6, a movable trolley 7, a feces collection box 8, a water tank 9 and a monitoring mechanism.

[0062] The bed 1 is provided with a defecation groove 16 ; a transposition drive mechanism 6 is fixedly arranged below the bed 1 ; a rectangular frame 5 is arranged on the transposition drive mechanism 6 in transmission connection.

[0063] The rectangular frame 5 is fixed with a feces collection trough 2 and an electric push rod 41, and a support plate 4 is fixed on the upper end of the movable rod of the electric push rod 41; the shape and size of the support plate 4 are adapted to the defecation trough 16; when the support plate 4 is stuck in the defecation trough 16, the upper end of the support plate 4 is flush with the bed 1, forming a complete bed; the transposition drive mechanism 6 can drive the rectangular frame 5 to move and adjust the position, so that the support plate 4 and the feces collection trough 2 are alternately located below the defecation trough 16; the edge of the defecation trough 16 is rounded to prevent scratching the patient. Adjustable armrests are set on both sides of the bed 1 to facilitate the patient's grip and enhance their sense of security and comfort on the bed.

[0064] The movable trolley 7 is a trolley with brakes; a feces collection box 8 and a water tank 9 are fixedly arranged above the movable trolley 7; the wheels of the movable trolley 7 are silent universal wheels with a double brake system, namely a foot brake and a hand brake, to ensure that it will not move at will during use. The body frame is made of stainless steel to enhance its carrying capacity and durability. A storage drawer is set on the side of the trolley for convenient storage of cleaning supplies, disposable gloves and other items.

[0065] A plug-in fixing seat 10 is fixedly provided on the feces collecting box 8, and a feces collecting pipe 11 is fixedly provided at the lower end of the feces collecting tank 2. The lower end of the feces collecting pipe 11 is detachably fixed on the plug-in fixing seat 10, so that the feces collecting pipe 11 is connected to the feces collecting box 8; a one-way valve that can be opened and closed is provided at the connection between the feces collecting pipe 11 and the feces collecting tank 2 to prevent feces from flowing back during transportation.

[0066] A plurality of removable filter screens are arranged inside the feces collection box 8, which can perform preliminary solid-liquid separation on the feces. A defecation pipe 14 is installed at the bottom of the feces collection box 8, and a valve 15 is arranged on the defecation pipe 14; it is convenient to regularly clean the feces in the collection box. In addition, a sealing cover is arranged on the top of the feces collection box 8 to prevent the odor from emitting. The water tank 9 is made of transparent high-strength plastic material, which is convenient for observing the water level. A water level sensor is installed on the outside of the water tank 9. When the water level is lower than the set value, an alarm can be issued through the controller to remind the staff to add water in time.

[0067] A water pump 12 is fixedly provided on the water tank 9, a water supply pipe 13 is fixedly provided at the output end of the water pump 12, and the input end of the water pump 12 extends into the bottom of the water tank 9; an annular water spray pipe 21 is fixedly provided at the upper end of the feces collecting tank 2, and the water supply pipe 13 is connected to the annular water spray pipe 21; a plurality of water spray holes inclined inwardly are provided on the annular water spray pipe 21, and water can be sprayed toward the inner wall of the feces collecting tank 2 for cleaning through the annular water spray pipe 21.

[0068] A monitoring mechanism is fixedly arranged on the bed 1, which monitors the defecation needs of the paralyzed patient and detects abnormal situations in time.

[0069] The bed 1 is equipped with a special controller. The operator can easily switch the positions of the support plate 4 and the feces collection tank 2 through the buttons or touch screen on the controller. At the same time, the controller also has an automatic return function. After completing a feces collection operation, the device can be controlled to return to the initial state with one key, that is, the support plate 4 is flush with the bed 1. The support plate 4 is made of high-strength engineering plastics and covered with a layer of soft silicone pad on the surface to increase the comfort when in contact with the patient's body.

[0070] In this technical solution, the upper end of the support plate 4 is flush with the bed 1 in the initial state, forming a complete bed. The paralyzed patient lies on the bed 1. When the patient needs to defecate, the electric push rod 41 is started to drive the support plate 4 to move downward and away from the defecation trough 16, and then the transposition drive mechanism 6 drives the support plate 4 and the feces collecting trough 2 to move synchronously, the support plate 4 moves to the outside of the defecation trough 16, and the feces collecting trough 2 moves to the bottom of the defecation trough 16; the paralyzed patient can defecate through the defecation trough 16 on the bed 1; the feces of the patient are received by the feces collecting trough 2, and the feces discharged by the patient are transported to the feces collecting box 8 through the feces collecting pipe 11. After the patient finishes defecation, the water in the water tank 9 can be pumped to the annular water spray pipe 21 through the water pump 12 and the water supply pipe 13, and the inner wall of the feces collecting trough 2 is sprayed with water through the annular water spray pipe 21 for cleaning, so as to keep the inner wall of the feces collecting trough 2 clean. The defecation of the paralyzed patient is monitored by the monitoring mechanism, and abnormal conditions are discovered in time.

[0071] Furthermore, a rotary nozzle is provided at the bottom of the feces collecting tank 2. During cleaning, the annular water spray pipe 21 first performs preliminary flushing of the tank wall, and then the rotary nozzle starts to work, flushing the tank bottom in all directions, ensuring that every corner of the feces collecting tank 2 can be thoroughly cleaned.

[0072] Reference Figure 1 and Figure 2 The bottom of the feces collection tank 2 is fixedly provided with an electric telescopic rod 31, and a urine shield 3 is fixedly provided on the movable rod of the electric telescopic rod 31. The urine shield 3 is in the shape of a structure with a wide upper end and a narrow lower end. The urine of the patient can be received and blocked by the urine shield 3 to prevent the urine from being sprayed onto the bed and dirtying the quilt. A sponge is arranged inside the urine shield 3 to prevent urine from splashing onto the outside of the urine shield 3. Keep the surrounding environment clean. In addition, the sponge layer is detachable, which is convenient for regular cleaning and replacement.

[0073] Reference Figure 2 The transposition driving mechanism 6 includes a motor 61 , a lead screw 62 and a polished rod 63 .

[0074] A motor 61 is fixedly arranged below the bed 1; a screw rod 62 and a light rod 63 are rotatably arranged below the bed 1; the screw rod 62 is coaxially fixedly connected to the output end of the motor 61; the screw rod 62 is threadedly connected to the rectangular frame 5, and the light rod 63 is slidably connected to the rectangular frame 5.

[0075] In this technical solution, the starting motor 61 drives the screw rod 62 to rotate, the screw rod 62 drives the rectangular frame 5 to move, and the rectangular frame 5 drives the feces collecting trough 2 and the supporting plate 4 to move and adjust their positions.

[0076] Furthermore, monitoring agencies include:

[0077] Data collection module: collects patient data, such as medical history information, past bowel movement patterns, and physical data, such as heart rate, blood pressure, and abdominal pressure changes. These data are obtained by connecting to the hospital information system, and connecting to various medical monitoring devices to obtain physical data. The collected data is annotated in a specific format, such as associating bowel movements at different time points with physical data at the corresponding time, as a reference sample for subsequent analysis, providing multi-dimensional data support for accurate judgment of bowel movement-related situations.

[0078] Image acquisition module: High-definition cameras are set on the bed 1 and the feces collection trough 2 to collect images of the patient's face, limbs and defecation.

[0079] Image preprocessing module: preprocess the collected images, including filtering, denoising, grayscale and normalization; remove noise interference in the image through filtering algorithm to improve image clarity; then perform grayscale processing to convert the color image into a grayscale image to simplify subsequent image processing steps; finally, normalize the grayscale image so that the pixel values ​​of different images are in a unified numerical range, which is convenient for subsequent feature extraction and analysis.

[0080] Physiological characteristic acquisition module: real-time acquisition of the patient's physiological characteristics, such as heart rate, blood pressure, muscle tension, etc.; using professional physiological monitoring equipment, such as heart rate monitors that use the principle of photoelectric sensors to detect heart rate by measuring changes in the absorption of light by blood; blood pressure monitors that use the oscillometric method to measure blood pressure by detecting changes in cuff pressure; muscle tension monitors that use surface electromyography sensors to detect electrical activity on the muscle surface to assess muscle tension.

[0081] Feature extraction module: extract features from the preprocessed images, including color, texture and shape; for facial images, extract facial features, such as frowning, pursing lips and other expressions that may be related to defecation needs, corresponding to color, texture and shape features; for limb images, extract color, texture and shape features related to movement amplitude, limb extension direction, etc.; for images in the feces collection tank 2, extract feces color, shape, size and other features. Through these feature extractions, key data is provided for subsequent image analysis.

[0082] Image analysis module: Analyze and identify the image after feature extraction, and promptly identify when the patient needs to defecate based on the patient's facial expressions and body movements; identify when the patient has finished defecation; and promptly detect abnormal defecation.

[0083] Comprehensive evaluation module: combines the analysis results of the image analysis module with the monitoring results of the physiological characteristics acquisition module (such as heart rate, blood pressure, muscle tension, etc.) to evaluate and predict whether the patient is about to defecate or whether the defecation has been completed.

[0084] Alarm module: includes an alarm, which will issue an alarm in time when abnormal defecation is detected.

[0085] PLC control module: connected with the data collection module, image acquisition module, image preprocessing module, feature extraction module, image analysis module and alarm module network. PLC control module is connected with the controller through a wireless signal network.

[0086] Furthermore, the feature extraction module extracts features from the preprocessed image, and the extracted features include color, texture and shape; the steps include:

[0087] 1. Facial image feature extraction.

[0088] 1.1. Color feature extraction: Use the color space conversion algorithm to convert the preprocessed facial image from the common RGB color space to the HSV (hue, saturation, brightness) or Lab (brightness, a*, b*) color space. By analyzing the color distribution and changes of specific facial areas (such as forehead, mouth corners, etc.) under different expressions, representative color feature vectors are extracted, such as the brightness change of the forehead area when frowning, and the saturation change of the mouth corner area when pursing the lips.

[0089] 1.2. Texture feature extraction: The gray level co-occurrence matrix (GLCM) method is used to calculate the gray level co-occurrence relationship of pixels in different directions and distances in the facial image to obtain a texture feature matrix. Texture feature parameters such as energy, contrast, correlation, and entropy are extracted from the matrix to describe the texture differences caused by changes in facial expressions, such as the changes in the texture of forehead wrinkles when frowning.

[0090] 1.3. Shape feature extraction: Use the active shape model (ASM) or active appearance model (AAM) to construct a facial shape model by marking key facial feature points (such as the corners of the eyes, the corners of the mouth, the end points of the eyebrows, etc.). According to the displacement and relative position changes of these feature points when the expression changes, the shape feature parameters are extracted, such as the angle change of the corner of the mouth to reflect the expression.

[0091] 2. Limb image feature extraction:

[0092] 2.1. Color feature extraction: Convert the limb image to a suitable color space and analyze the color changes of specific parts (such as arms and legs) during limb movements. For example, when the body twists, the color contrast between the arms and other parts of the body changes, and the color feature vector is extracted.

[0093] 2.2 Texture feature extraction: The local binary pattern (LBP) algorithm is used to perform texture analysis on the limb image. The LBP algorithm generates a binary pattern image by comparing the grayscale values ​​of the central pixel and the neighboring pixels, and then calculates the texture feature histogram. By analyzing the differences in the texture feature histograms under different limb movements, the texture features related to the movement amplitude and stretching direction are extracted.

[0094] 2.3. Shape feature extraction: Use contour-based shape description methods, such as Fourier descriptors. First, extract the contour of the limb image, then perform Fourier transform on the contour, convert the contour shape information into frequency domain information, and extract shape features that can reflect the amplitude and extension direction of limb movement by analyzing the Fourier coefficients of the low-frequency part.

[0095] 3. Image feature extraction in feces collection tank 2:

[0096] 3.1. Color feature extraction: After converting the image to a suitable color space, analyze the color distribution of the feces area and extract feature parameters such as average color value and color variance to determine whether the feces color is normal.

[0097] 3.2. Shape feature extraction: The edge contour of the feces is extracted through an edge detection algorithm (such as the Canny algorithm), and then the contour is described using a shape descriptor (such as the Hu moment) to extract characteristic parameters that can reflect the shape of the feces, such as circularity, aspect ratio, etc.

[0098] 3.3. Size feature extraction: According to the pixel size of the image and the known image resolution, the pixel area of ​​the feces in the image is calculated, and the actual area size is obtained by conversion, which is used as the size feature.

[0099] Furthermore, the image analysis module analyzes and identifies the image after feature extraction, and timely identifies the patient's need to defecate according to the patient's facial expression and body movements; identifies the patient's end of defecation; and promptly detects abnormal defecation; including the following steps:

[0100] 1. Identify situations where the patient needs to have a bowel movement;

[0101] 1.1. Expression analysis: The extracted facial expression features are input into the trained expression recognition model, which selects a deep learning model based on a convolutional neural network (CNN). The model outputs the expression category (such as frowning, pursing lips, etc.) and its confidence based on the input feature vector. When an expression related to the need for defecation (such as frequent frowning, painful expression, etc.) is detected and the confidence exceeds the set threshold, it is used as one of the bases for judging whether the patient may need to defecate. Perform expression analysis L according to the following formula:

[0102] F → =[∑ n j＝1 (w 1j f j ,),∑ n j＝1 (w 2j f j ,),…,∑ n j＝1 (w nj f j ,);] T Where, F → It is a more discriminative feature representation vector obtained after weighted processing, which will be input into the expression recognition model later. n represents the expression feature vector F → The dimension of indicates the number of extracted expression features and also determines the dimension of the attention weight matrix. j is a loop variable, traversing from 1 to n. w ij is an element in the attention weight matrix (i=1,2,3...n), reflecting the degree of attention of the i-th feature dimension to the j-th feature dimension. j is the extracted facial expression feature vector F → =[f1,f2,...f n The jth element in ] contains rich expression details, such as the range of facial muscle movement, the relative positions of facial key points, etc. T represents the transposition operation, which transposes the weighted sum result originally arranged in rows into a column vector form to meet the requirements of the vector dimension for subsequent model input.

[0103] The classification function is the softmax function: P i =e Zhj / [∑ mj-1 (e Zhj )]; where P i represents the probability of the i-th expression category, Zhj is the vector Zh after fusion of physiological signals → The jth element of , m is the total number of expression categories. → It is obtained by adding the output vector of the fully connected layer and the physiological signal features after weighting by the weight matrix. It contains the comprehensive information of the facial expression features and the physiological signal features after a series of calculations, and plays a key role in determining the probability of the expression category.

[0104] 1.2. Body movement analysis: Select the CNN-LSTM model and input the extracted body movement features into the CNN-LSTM body movement recognition model. The model analyzes the input feature sequence and identifies the body movement type (such as body twisting, leg curling, etc.). When body movements related to the need for defecation are detected and the frequency reaches a certain number of times, combined with the expression analysis results, it is comprehensively judged that the patient needs to defecate.

[0105] 2. Identify the end of the patient's defecation: Continuously monitor the image in the stool collection tank 2, and use a method based on time series analysis to track the changes in the position, shape and size of the stool in the image over a period of time (such as 5-10 minutes). When it is detected that no new stool enters the image in the stool collection tank 2 within the set time, and the position of the existing stool does not change significantly, combined with the historical data of the defecation duration and the empirical threshold judgment, it is determined that the patient's defecation has ended.

[0106] 3. Found abnormal defecation:

[0107] 3.1. Abnormal color judgment: The extracted stool color characteristics are compared with the standard characteristic range of normal stool color. If the stool color characteristics are beyond the normal range, such as too dark (may indicate upper gastrointestinal bleeding), too white (may be related to bile duct obstruction) or too green (may be related to indigestion, etc.), it is judged as abnormal color.

[0108] 3.2. Determination of abnormal shape: Compare the extracted stool shape features with the reference model of normal stool shape. If the stool shape is long and thin, granular, or significantly different from the normal shape, it is determined to be abnormal in shape based on clinical experience and medical knowledge.

[0109] 3.3. Size abnormality judgment: Compare the extracted stool size characteristics with the size range corresponding to normal defecation volume. If the stool area or volume is too large or too small, beyond the normal range, and combined with the patient's diet, physical condition and other factors for comprehensive analysis, it is judged as size abnormal.

[0110] Furthermore, the body movement analysis includes the following steps:

[0111] 1. Feature extraction: Use the OpenPose human posture estimation algorithm to extract body movement features from the captured video frames. These features are mainly related to the position of nodes, the angles formed between joints, etc. The features extracted from each frame are sorted into a feature vector, and the feature vectors of all frames are arranged in order to form a feature sequence. The extracted features are normalized so that features of different dimensions are in a similar numerical range.

[0112] 2. Model construction: Use the CNN-LSTM model. CNN is responsible for extracting spatial features from each frame. For the joint point position and angle information of each frame, CNN can extract spatial features such as the relative position relationship between joint points and joint angle changes. LSTM is responsible for processing time series data. The feature sequence extracted by CNN is input into LSTM, which can capture the temporal changes of limb movements, such as the continuity and rhythm of movements. Perform limb movement analysis according to the following formula:

[0113] P → =Softmax{δ t *LSTM[CNN(A t → )+∑ n j＝1 (M t,ij *e j )+W B *B t → ]}.

[0114] Softmax(x) k =e xk / [∑ M i＝1 (e xi )];δ t =1 / [1+e -(t-0.5T总) ]; where P → Represents the probability distribution vector of the final output of the model, P → =[P1,P2,...,P M ], where P k Indicates the probability of the kth limb action type. Softmax is a function used to convert the input vector into a probability distribution vector, indicating the probability of each action type. (x) k In Softmax(x) k In this example, x is the input vector, (x) kRepresents the kth element of the input vector x. e is a natural constant used as the base in exponential operations. M is the total number of body movement types. When calculating the probability distribution vector, it determines the upper limit of the summation and the dimension of the final probability distribution vector. δ t is a temporal dynamic factor, and its calculation method can be designed according to specific needs. Where T is always the total duration of the entire action sequence, and t is the current time step. It is used to adjust the impact of action features at different time points on the final probability distribution. When t is close to the middle of the sequence, δ t Close to 1, indicating that the action feature at this time has a greater impact on the result; when t is close to the beginning or end of the sequence, δ t Small, the impact of motion features is relatively small. LSTM is a long short-term memory network, a special recurrent neural network that can process time series data and capture long-term dependencies in the data. LSTM(...) means that the LSTM network calculates the input in the brackets and outputs a feature vector. CNN is a convolutional neural network, which is mainly used to extract spatial features in images. CNN(A t → ) represents the body movement feature sequence {A t →} Perform convolutional neural network calculations and output the extracted spatial features. t →} is the preprocessed test set data or the real-time collected body movement feature sequence, A t → =[a t1 ,a t2 ,...,a tn ] is the feature vector of the tth frame, and n is the feature dimension. t,ij is an element in the dynamic environmental impact matrix, M t is an mxn matrix, M t,ij It represents the influence coefficient of the jth environmental factor on the probability of the ith action type at time t. j is the value of the jth environmental factor, which may include temperature, humidity, light intensity, etc. b is the individual physiological feature weight matrix, which is an mxl matrix, W bij represents the influence weight of the jth physiological feature on the probability of the i-th action type. t → is the individual physiological feature vector, B t → =[b t1 ,b t2 ,...,b tl ], where b ti Represents the i-th physiological characteristic of an individual at time t, such as heart rate, blood pressure, muscle tension, etc.

[0115] 3. Model training: The collected body movement data is divided into training set, validation set and test set. The training set is used to train the model so that the model can learn the relationship between body movement characteristics and movement types; the validation set is used to adjust the model parameters to avoid overfitting; the test set is used to evaluate the actual performance of the model after training.

[0116] Select loss function and optimizer: Select the cross entropy loss function to measure the gap between the model prediction results and the actual body movement type labels. Select the Adam optimizer as the optimizer, which can automatically adjust the learning rate according to the training situation and speed up the convergence of the model training.

[0117] The training set data is input into the CNN-LSTM model. The model calculates the prediction results through forward propagation, and then calculates the loss value based on the prediction results and the actual action type label. Then, the back-propagation algorithm is used to calculate the degree of influence of the loss value on the model parameters, and finally the optimizer is used to update the model parameters. During training, the performance of the model is tested regularly on the validation set. According to the loss value and prediction accuracy on the validation set, some hyperparameters of the model are adjusted, such as the learning rate and the number of hidden layer neurons, to prevent the model from overfitting.

[0118] 4. Body movement recognition:

[0119] Model prediction: Input the preprocessed test set data or the real-time collected body movement feature sequence into the trained convolutional neural network model CNN-LSTM model. The model outputs a probability distribution vector through forward propagation calculation. Each element in this vector represents the probability of the corresponding body movement type.

[0120] Action type determination: Use the Softmax classifier to process this probability distribution vector to determine the type of body movement. Find the element with the highest probability in the probability distribution vector, and the action type it corresponds to is the body movement type identified by the model.

[0121] 5. Defecation need judgment: Set a fixed time range, and within this time range, count the number of limb movements related to defecation needs (such as body twisting, leg curling, etc.). Set a set of limb movements related to defecation needs in advance. Once the detected limb movement type belongs to this set, add up the number of times it occurs. When the frequency of limb movements related to defecation needs reaches a pre-set threshold, combine it with the results of the previous expression analysis. If the expression analysis determines that the patient has a defecation need, then it is comprehensively determined that the patient needs to defecate.

[0122] Further, identifying the end of a patient's bowel movement includes the following steps:

[0123] 1. Continuously collect images: At set time intervals, continuously collect images in the feces collection tank 2 to form a series of image sequences. Perform simple preprocessing on each frame of the collected image to remove obvious noise, adjust the brightness and contrast of the image, and enhance the clarity of the image to facilitate subsequent feature extraction.

[0124] 2. Extraction of feces features: For each frame of the image, extract the feces’ location, shape, and size. The location is represented by determining the coordinates of the feces in the image; the shape can be described as an approximate geometric shape, such as a circle, an ellipse, or an irregular polygon; the size is measured by measuring the feces’ area or length, width, and other dimensions in the image.

[0125] 3. Time series construction: Arrange the features of feces in images at different times in chronological order to form a time series of position, shape and size.

[0126] 4. Feature change analysis: Compare the features of feces in images at adjacent times to analyze whether the position has moved, the shape has changed, or the size has increased or decreased. For example, compare the coordinates of feces at different times to determine the position change; compare the shape description to determine whether the shape has changed; and compare the area or size data to determine the size change.

[0127] 5. Defecation end judgment:

[0128] 5.1. No new feces entering the judgment: During the set monitoring time, continuously observe the size of feces in the image and the newly appeared suspected feces area. If the area of ​​feces does not increase in multiple consecutive time intervals, and the area of ​​the newly appeared suspected feces area is very small, it can be judged that no new feces has entered.

[0129] 5.2. Judgment of no obvious change in position: Observe the position of the existing feces. If the coordinates of the feces change very little and the shape does not change significantly within multiple consecutive time intervals, it can be considered that the position of the existing feces has no obvious change.

[0130] 5.3. Defecation time judgment: Record the time from detection to the start of defecation. When the set monitoring time is reached and no new feces enter and the position does not change significantly, combine the historical data of defecation duration and the empirical threshold. If the actual defecation time reaches or exceeds the threshold, it is determined that the patient has finished defecation.

[0131] Furthermore, the comprehensive evaluation module combines the analysis results of the image analysis module with the monitoring results of the physiological characteristics acquisition module (such as heart rate, blood pressure, muscle tension and facial expression, etc.) to evaluate and predict whether the patient is about to defecate or whether the defecation has been completed, including the following steps:

[0132] 1. Data collection: Obtain stool feature-related information from the image analysis module, including the location, shape, and size trends of stool, as well as defecation-related action data (such as whether new stool has entered, whether the location of existing stool has changed, etc.) determined based on the image. Collect the patient's heart rate, blood pressure, muscle tension and other real-time monitoring data from the physiological feature acquisition module to ensure that the data timestamp is accurate so that it can be synchronized with the image analysis data.

[0133] 2. Data collation and alignment: The acquired image analysis data and physiological characteristic data are sorted in chronological order to ensure that different types of data at the same time point can accurately correspond to each other, which is convenient for subsequent analysis.

[0134] 3. Feature judgment analysis;

[0135] 3.1、Image feature judgment:

[0136] 3.1.1. Defecation action judgment: Check whether there are body movements related to defecation in the image, such as changes in body posture, abdominal ups and downs, etc. If these movements appear, record the frequency and duration of the movements.

[0137] 3.1.2. Determination of changes in stool characteristics: Analyze whether the position of the stool has moved significantly, whether the shape has changed significantly, and whether the size has increased. If the position of the stool changes frequently, the shape continues to change, or the size continues to increase over a period of time, it may indicate that defecation is in progress.

[0138] 3.2. Physiological characteristics judgment:

[0139] 3.2.1. Heart rate analysis: Observe the changes in heart rate. If the heart rate increases significantly in a short period of time and exceeds the normal fluctuation range, it may be related to factors such as straining during defecation.

[0140] 3.2.2. Blood pressure analysis: Monitor changes in blood pressure. Blood pressure usually rises during bowel movements. If systolic or diastolic blood pressure is significantly higher than the patient's baseline blood pressure, it is used as an indicator of an increased likelihood of bowel movements.

[0141] 3.2.3. Muscle tension analysis: Check the muscle tension data. The muscle tension in the abdomen, pelvic floor and other parts will change during defecation. If the muscle tension in the relevant parts increases significantly and lasts for a period of time, record the change.

[0142] 4. Comprehensive evaluation: Set corresponding weights for different features. The weights can be adjusted according to the actual situation and a large amount of experimental data. Calculate the comprehensive evaluation score based on the judgment results of each feature and the set weights. Set an evaluation score threshold. When the comprehensive evaluation score is greater than or equal to the threshold, it is predicted that the patient may have a bowel movement; when the score is less than the threshold, it is considered that the patient is less likely to have a bowel movement in the near future.

[0143] If within a period of time (such as 5-10 minutes in a row), the image characteristics show that the position, shape, and size of the stool have not changed, and no new stool has entered; at the same time, the physiological characteristics show that the heart rate, facial expression, blood pressure, and muscle tension have returned to normal levels and remain stable, a comprehensive assessment score threshold for the end of defecation is set (such as 0.3, which can be adjusted according to actual conditions). When the comprehensive assessment score is lower than this threshold, it is judged that the patient has finished defecation.

[0144] Perform a comprehensive evaluation according to the following formula: S 综合 =∑ 5 i＝1 (u i *S i )+∑ 4 i＝1 ∑ 5 j＝i+1 (β ij *S i *S j );where S 综 It is a comprehensive evaluation score, which is obtained by comprehensively calculating the judgment results of each feature and the mutual influence between features. It is used to evaluate the possibility of defecation of the patient or to judge whether the defecation is completed. i is the dynamic weight of the ith feature, which is dynamically adjusted based on historical data and all current feature judgment results through Bayesian inference. It reflects the relative importance of the ith feature to the comprehensive evaluation score in the current situation. For example, u1 represents the dynamic weight of the defecation-related limb action feature, and its value changes according to the association between the defecation action and other features in the historical data and the current judgment result. S i is the judgment result of the ith feature (heart rate, blood pressure, muscle tension, facial expression, etc.), and takes a value of 0 or 1. 1 indicates that the feature has a significant change or a related action has occurred, and 0 indicates that there is no significant change or no related action has occurred. Corresponding to the defecation action judgment score, S1 = 1 when there is a defecation-related action, and S1 = 0 when it does not appear; S2 corresponds to the stool feature change score, S2 = 1 when there is a significant change, and S2 = 0 when there is no significant change, and so on. β ij is the feature influence coefficient, which indicates the degree of mutual influence between the i-th feature and the j-th feature, and its value range is [-1,1]. ijWhen β > 0, it indicates that the two features have a positive correlation, that is, the change of one feature will enhance the impact of the other feature on the comprehensive evaluation score; when β ij When <0, it means that the two features have a negative correlation, and the change of one feature will weaken the impact of the other feature.

[0145] 5. Result feedback adjustment: Feedback the evaluation results to medical staff or related systems in a timely manner so that corresponding measures can be taken, such as preparing nursing work in advance or reminding patients to pay attention. Dynamically adjust the evaluation index system and threshold according to the actual defecation situation. If the evaluation results are inconsistent with the actual defecation situation for many times, analyze the reasons, adjust the feature weights or thresholds, and improve the accuracy of the evaluation.

[0146] The present invention provides a method for using a feces collection device for paralyzed patients, comprising the following steps:

[0147] S1, initial state: the upper end of the support plate 4 is flush with the hospital bed 1, forming a complete hospital bed, and the paralyzed patient lies on the hospital bed 1.

[0148] S2. The monitoring agency monitors the defecation needs of paralyzed patients.

[0149] S21. The image acquisition module acquires images of the patient's face, limbs, and bowel movements.

[0150] S22. The physiological characteristics acquisition module acquires the patient's physiological characteristics in real time, such as heart rate, blood pressure, muscle tension, etc.

[0151] S23, the image preprocessing module preprocesses the collected image, including filtering, denoising, grayscale and normalization.

[0152] S24, the feature extraction module performs feature extraction on the preprocessed image, and the extracted features include color, texture and shape; for facial images, it extracts expression features, such as frowning, pursing lips and other expressions that may be related to the need for defecation, which correspond to color, texture and shape features; for limb images, it extracts color, texture and shape features related to movement amplitude, limb extension direction, etc.; for images in the feces collection trough 2, it extracts feces features such as color, shape, size, etc.

[0153] S25. The image analysis module analyzes and identifies the image after feature extraction, and promptly identifies the patient's need to defecate based on the patient's facial expression and body movements; identifies the patient's end of defecation; and promptly detects abnormal defecation.

[0154] S26. The comprehensive evaluation module combines the analysis results of the image analysis module with the monitoring results of the physiological characteristics acquisition module (such as heart rate, blood pressure, muscle tension, etc.) to evaluate and predict whether the patient is about to defecate or whether the defecation has been completed.

[0155] S27. When abnormal defecation is detected, the alarm module promptly issues an alarm (for example, a completely paralyzed patient needs medical assistance to defecate or has abnormal defecation).

[0156] S3. When defecation is needed, the electric push rod 41 is started to drive the support plate 4 to move down and away from the defecation trough 16, and then the transposition drive mechanism 6 drives the support plate 4 and the feces collecting trough 2 to move synchronously. The support plate 4 moves to the outside of the defecation trough 16, and the feces collecting trough 2 moves to the bottom of the defecation trough 16.

[0157] S4, start the electric telescopic rod 31 to drive the urine shield 3 to rise, and the urine shield 3 is used to receive and block the patient's urine to prevent the urine from spraying onto the bed and dirtying the quilt, so as to keep the surrounding environment clean.

[0158] S5. For semi-paralyzed patients (patients who can take off their pants to defecate by themselves), the opening and closing can be controlled by the controller; for completely paralyzed patients, medical staff are required to assist in taking off the patient's pants so that the patient's buttocks are in the defecation groove 16 for defecation.

[0159] S6. The paralyzed patient can defecate on the bed 1 through the defecation trough 16; the feces of the patient are received by the feces collecting trough 2, and the feces discharged by the patient are transported to the feces collecting box 8 through the feces collecting pipe 11.

[0160] S7. After the patient finishes defecating, the urine shield 3 is automatically controlled to descend and be stored in the feces collecting trough 2, and then the shifting driving mechanism 6 drives the support plate 4 and the feces collecting trough 2 to move synchronously, the support plate 4 moves to the bottom of the defecation trough 16, and the feces collecting trough 2 moves to the outside of the defecation trough 16; the electric push rod 41 drives the support plate 4 to move up and get stuck in the defecation trough 16, and the support plate 4 is flush with the upper end of the bed 1.

[0161] S8. The water in the water tank 9 is pumped to the annular water spray pipe 21 through the water pump 12 and the water supply pipe 13. Water is sprayed to the inner wall of the feces collecting tank 2 through the annular water spray pipe 21 for cleaning, so as to keep the inner wall of the feces collecting tank 2 clean.

[0162] The present invention can automatically and timely judge the patient's defecation needs, defecation status and abnormal conditions through the collaborative work of multiple modules of the monitoring mechanism, such as image acquisition, physiological characteristics acquisition, image analysis and comprehensive evaluation modules. For example, when abnormal defecation is detected, the alarm module automatically issues an alarm, and there is no need for nursing staff to be on duty at all times, which greatly improves the response speed of nursing work, allowing nursing staff to arrange work more efficiently and deal with patient needs in a timely manner. With automatic operation function, the electric push rod, transposition drive mechanism, electric telescopic rod and other automated components of the device can automatically complete actions such as the movement of the support plate and the lifting and lowering of the urine shield, reducing the steps and time of manual operation of the nursing staff, and greatly improving the nursing efficiency in the preparation and cleaning work before and after the patient's defecation. It can improve the patient experience. In the initial state, the support plate is flush with the bed, forming a complete bed, providing a comfortable lying surface for the patient. During the defecation process, whether it is a semi-paralyzed patient who controls it by himself or a completely paralyzed patient who is assisted by medical staff, it can be completed on the bed without moving the body to a special bathroom, reducing the patient's pain and discomfort, and is particularly suitable for paralyzed patients with limited mobility. The privacy of patients can be protected. The entire defecation process is carried out in a relatively private environment on the bed, avoiding the embarrassment and privacy leakage caused by patients moving to public toilets or defecating with the assistance of others, and protecting the dignity of patients. The comprehensive evaluation module combines image analysis and physiological characteristic monitoring results to more accurately evaluate the patient's defecation situation and provide a basis for personalized care. According to the monitoring results, targeted nursing measures can be provided for completely paralyzed and semi-paralyzed patients. For example, completely paralyzed patients need medical staff to assist in defecation, while semi-paralyzed patients can control some operations by themselves, meeting the personalized needs of different patients.

[0163] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for using a feces collection device for paralyzed patients, characterized in that: The following steps are involved: S1, in the initial state, the upper end of the support plate is flush with the bed, and the paralyzed patient lies on the bed; S2. The monitoring agency monitors the defecation needs of paralyzed patients; S3, when defecation is needed, the electric push rod is started to drive the support plate to move down and away from the defecation trough, and the transposition drive mechanism drives the support plate and the feces collection trough to move synchronously, the support plate moves to the outside of the defecation trough, and the feces collection trough moves to the bottom of the defecation trough; S4. The electric telescopic rod drives the urine shield to rise, and the urine shield receives and blocks the patient's urine; S5. Semi-paralyzed patients control the opening and closing of the device by themselves through the controller; for completely paralyzed patients, medical staff assist in taking off the patient's pants so that the patient's buttocks are placed in the defecation trough for defecation; S6, paralyzed patients defecate through the defecation trough; The feces collecting tank receives the feces of the patient, and the feces discharged by the patient are transported to the feces collecting box through the feces collecting pipe; S7. After the patient has finished defecating, the urine shield is automatically controlled to descend into the feces collection tank, and the transposition drive mechanism drives the support plate and the feces collection tank to move, the support plate moves to the bottom of the defecation tank, and the feces collection tank moves to the outside of the defecation tank; the electric push rod drives the support plate to move up and be flush with the upper end of the bed; S8. The water in the water tank is pumped to the annular water spray pipe through a water pump and a water supply pipe, and water is sprayed to the inner wall of the feces collecting tank through the annular water spray pipe for cleaning, so as to keep the inner wall of the feces collecting tank clean.

2. The method for using the feces collection device for paralyzed patients according to claim 1, characterized in that: Step S2 includes the following steps: S21, the image acquisition module acquires images of the patient's face, limbs and defecation; S22, the physiological characteristics collection module collects the patient's physiological characteristics in real time; S23, the image preprocessing module preprocesses the collected image; S24, a feature extraction module extracts features from the preprocessed image, where the extracted features include color, texture, and shape; S25, the image analysis module analyzes and identifies the image after feature extraction, promptly identifies the situation that the patient needs to defecate; identifies the situation that the patient has finished defecation; and promptly detects abnormal defecation; S26, the comprehensive evaluation module combines the analysis result of the image analysis module with the monitoring result of the physiological characteristic acquisition module to evaluate and predict whether the patient is about to defecate or whether the defecation has been completed; S27. When abnormal defecation is detected, the alarm module sounds an alarm.

3. The method for using the feces collection device for paralyzed patients according to claim 2, characterized in that: Step S25 includes the following steps: S251. Identify situations where the patient needs to defecate: S251.1, Expression analysis: The extracted facial expression features are input into the trained convolutional neural network (CNN) expression recognition model. The model outputs the expression category and its confidence level based on the input feature vector. S251.2, Body movement analysis: Select the CNN-LSTM model, input the extracted body movement features into the CNN-LSTM body movement recognition model, the model analyzes the input feature sequence, identifies the body movement type, and when body movements related to the need for defecation are detected and the frequency reaches a certain number of times, combined with the expression analysis results, comprehensively judge that the patient needs to defecate; S252, identifying the patient's end of defecation: continuously monitoring the image in the stool collection tank, tracking the position, shape and size changes of the stool in the image over a period of time, and determining that the patient's defecation has ended when it is detected that no new stool enters the image in the stool collection tank within a set time, and the position of the existing stool has no obvious change, and combining the historical data of the defecation duration and the empirical threshold judgment; S253. Abnormal defecation: S253.

1. Color abnormality judgment: The extracted stool color characteristics are compared with the standard characteristic range of normal stool color. If the stool color characteristics are beyond the normal range, it is judged as color abnormality; S253.2, shape abnormality judgment: compare the extracted stool shape features with the reference model of normal stool shape. If the stool shape is significantly different from the normal shape, it is judged as abnormal shape based on clinical experience and medical knowledge; S253.

3. Judgment of abnormal size: Compare the extracted stool size characteristics with the size range corresponding to the normal bowel movement. If the stool area or volume is too large or too small, beyond the normal range, and combined with the patient's diet and physical condition, a comprehensive analysis is conducted and it is judged to be abnormal in size.

4. The method for using the feces collection device for paralyzed patients according to claim 3, characterized in that: In step S251.1, expression analysis is performed according to the following formula: F → =[∑ n j＝1 (w 1j f j ,),∑ n j＝1 (w 2j f j ,),…,∑ n j＝1 (w nj f j ,);] T Where, F → It is a more discriminative feature representation vector obtained after weighted processing; n represents the expression feature vector F → The dimension of; j is the loop variable, traversing from 1 to n; w ij is an element in the attention weight matrix (i=1,2,3...n), reflecting the degree of attention of the i-th feature dimension to the j-th feature dimension; f j is the extracted facial expression feature vector F → =[f1,f2,...f n ]; T represents the transpose operation; The classification function is the softmax function: P i =e Zhj / [∑ m j-1 (e Zhj )]; where P i represents the probability of the i-th expression category, Zhj is the vector Zh after fusion of physiological signals → The jth element of Zh, m is the total number of expression categories; → It is obtained by adding the output vector of the fully connected layer and the physiological signal features after weighting by the weight matrix, and contains the comprehensive information of the expression features and physiological signal features after a series of calculations.

5. The method for using the feces collection device for paralyzed patients according to claim 3, characterized in that: Step S251.2: Body movement analysis includes the following steps: S251.2.

1. Feature extraction: Use the OpenPose human posture estimation algorithm to extract body movement features from video frames; organize the features extracted from each frame into a feature vector, and arrange the feature vectors of all frames in order to form a feature sequence; normalize the extracted features; S251.2.2, Model construction: Use CNN-LSTM model; CNN is responsible for extracting spatial features from each frame. For the joint point position and angle information of each frame, CNN extracts the relative position relationship between the joint points and the spatial features of the joint angle change. LSTM is responsible for processing time series data. The feature sequence extracted by CNN is input into LSTM. LSTM captures the temporal change law of limb movements. Perform limb movement analysis according to the following formula: P → =Softmax{δ t *LSTM[CNN(A t → )+∑ n j＝1 (M t,ij *e j )+W B *B t → ]}; Softmax(x) k =e xk / [∑ M i＝1 (e xi )];δ t =1 / [1+e -(t-0.5T总) ]; where P → Represents the probability distribution vector of the final output of the model, P → =[P1,P2,...,P M ], where P k Indicates the probability of the kth body movement type; Softmax is a function that indicates the probability of each movement type; (x) k In Softmax(x) k In this example, x is the input vector, (x) k represents the kth element of the input vector x; e is a natural constant; M is the total number of body movement types; δ t is the time dynamic factor; T is always the total duration of the entire action sequence, and t is the current time step; LSTM is a long short-term memory network, and LSTM(...) means that the LSTM network calculates the input in the brackets and outputs a feature vector; CNN is a convolutional neural network, and CNN(A t → ) represents the body movement feature sequence {A t → } Perform convolutional neural network calculations and output the extracted spatial features. t → } is the preprocessed test set data or the real-time collected body movement feature sequence, A t → =[a t1 ,a t2 ,...,a tn ] is the feature vector of the tth frame, n is the feature dimension, M t,ij is an element in the dynamic environmental impact matrix, M t is an mxn matrix, M t,ij represents the influence coefficient of the jth environmental factor on the probability of the ith action type at time t; e j is the value of the jth environmental factor; W b is the individual physiological feature weight matrix, which is an mxl matrix, W bij represents the influence weight of the jth physiological feature on the probability of the i-th action type; B t → is the individual physiological feature vector, B t → =[b t1 ,b t2 ,...,b tl ], where b ti represents the i-th physiological characteristic of an individual at time t; S251.2.3, Model training: Divide the body movement data into training, validation and test sets; use the cross entropy loss function to measure the gap between the prediction and the true label; train the CNN-LSTM model with the training set data; test the performance on the validation set and adjust the hyperparameters according to the loss value and accuracy; S251.2.4, Body movement recognition: Input the test set data or the real-time collected body movement feature sequence into the trained CNN-LSTM model, output the probability distribution vector through forward propagation, and then process the vector through the Softmax classifier to determine the body movement type; S251.2.

5. Determination of the need for defecation: Set a specific time range and count the number of occurrences of the preset defecation-related limb movements in the time period; when the frequency of the movements reaches the preset threshold and the facial expression analysis determines that the patient has a need for defecation, comprehensively determine that the patient needs to defecate.

6. The method for using the feces collection device for paralyzed patients according to claim 2, characterized in that: Step S252 includes the following steps: S252.

1. Continuously collect images: continuously collect images in the feces collection tank at set time intervals to form a series of image sequences, and pre-process each frame of the collected image; S252.2, feces feature extraction: for each frame of image, extract the feces position, shape and size features; S252.3, Time series construction: Arrange the features of feces in images at different times in time order to form a time series of position, shape and size; S252.

4. Feature change analysis: Compare the features of feces in images at adjacent moments to analyze whether the position has moved, the shape has changed, or the size has increased or decreased; S252.

5. Defecation end judgment: S252.5.

1. Judgment of no new feces entering: If the area of ​​feces does not increase over multiple consecutive time intervals, and the area of ​​the newly appeared suspected feces is very small, it can be judged that no new feces has entered; S252.5.

2. Judgment of no obvious change in position: Observe the position of existing feces. If the coordinates of the feces change very little and the shape does not change significantly over multiple consecutive time intervals, it is considered that the feces position has no obvious change; S252.5.

3. Judge in combination with defecation time: record the start time of defecation, reach the set monitoring time, and meet the requirements of no new feces entering and no obvious change in position. At the same time, refer to the historical data of defecation duration and the empirical threshold. When the actual defecation time reaches or exceeds the threshold, it is determined that the patient has finished defecation.

7. The method for using the feces collection device for paralyzed patients according to claim 2, characterized in that: Step S26 includes the following steps: S26.

1. Data collection: Obtain stool feature-related information from the image analysis module, and collect the patient's heart rate, blood pressure, and muscle tension real-time monitoring data from the physiological feature acquisition module; S26.

2. Data collation and alignment: The acquired image analysis data and physiological characteristic data are collated in chronological order to ensure that different types of data at the same time point can accurately correspond; S26.3, Feature judgment analysis: S26.3.

1. Image feature judgment: Check whether there are any body movements related to defecation in the image, and analyze whether the position of the feces has moved significantly, whether the shape has changed significantly, and whether the size has increased; S26.3.2, Physiological characteristics assessment: Analyze and assess the patient's heart rate, blood pressure, and muscle tension; S26.

4. Comprehensive evaluation: Set corresponding weights for different features, calculate the comprehensive evaluation score, and set an evaluation score threshold. When the comprehensive evaluation score is greater than or equal to the threshold, it is predicted that the patient is about to defecate. If, within a period of time, the image features show that the position, shape, and size of the stool have not changed, and no new stool has entered, and the physiological characteristics have returned to normal levels and remain stable, it is judged that the patient has finished defecation. Perform a comprehensive evaluation according to the following formula: S 综合 =∑ 5 i＝1 (u i *S i )+∑ 4 i＝1 ∑ 5 j＝i+1 (β ij *S i *S j );where S 综 is the comprehensive evaluation score; u i is the dynamic weight of the i-th feature, which is dynamically adjusted based on historical data and all current feature judgment results through Bayesian inference; S i is the judgment result of the i-th feature; β ij is the feature influence coefficient, which indicates the degree of mutual influence between the i-th feature and the j-th feature; S26.

5. Result feedback and adjustment: Provide timely feedback of assessment results to medical staff or relevant systems.

8. The method for using the feces collection device for paralyzed patients according to claim 1, characterized in that: An electric telescopic rod is fixedly arranged at the bottom of the feces collection tank, and a urine shield is fixedly arranged on the movable rod of the electric telescopic rod; a sponge is arranged inside the urine shield; The transposition drive mechanism includes a motor, a screw rod and a polished rod; the motor is fixedly arranged under the bed; the screw rod and the polished rod are rotatably arranged under the bed; the screw rod is coaxially fixedly connected with the output end of the motor; the screw rod is threadedly connected with the rectangular frame, and the polished rod is slidably connected with the rectangular frame.

9. The method for using the feces collection device for paralyzed patients according to claim 1, characterized in that: Monitoring agencies include: Data collection module: collects the patient's condition data and physical data; annotates the data as a reference sample; Image acquisition module: HD cameras are installed on the bed and stool collection trough to collect images of the patient's face, limbs, and defecation; Image preprocessing module: preprocess the collected images; Physiological characteristics collection module: collects the patient's physiological characteristics in real time; Feature extraction module: extract features from preprocessed images; Image analysis module: Analyze and identify the image after feature extraction, identify the situation that the patient needs to defecate, identify the situation that the patient has finished defecation, and promptly detect abnormal defecation; Comprehensive evaluation module: combines the analysis results of the image analysis module with the monitoring results of the physiological characteristics acquisition module to evaluate and predict whether the patient is about to defecate or whether the defecation has been completed; Alarm module: including an alarm, which will promptly issue an alarm when abnormal defecation is detected; PLC control module: connected with the data collection module, image acquisition module, image preprocessing module, feature extraction module, image analysis module and alarm module network.

10. A feces collection device for paralyzed patients, comprising: A sickbed, a feces collection tank, a support plate, a rectangular frame, a transposition drive mechanism, a movable trolley, a feces collection box, a water tank and a monitoring mechanism; the characteristics are: The bed is provided with a defecation slot; a transposition drive mechanism is fixedly arranged below the bed; a rectangular frame is arranged on the transposition drive mechanism for transmission connection; A feces collecting trough and an electric push rod are fixedly arranged on the rectangular frame, and a supporting plate is fixedly arranged on the upper end of the movable rod of the electric push rod; the shape and size of the supporting plate are adapted to the feces trough; and the transposition driving mechanism drives the rectangular frame to move and adjust the position; A feces collection box and a water tank are fixedly arranged above the movable trolley; a plug-in fixing seat is fixedly arranged on the feces collection box, the lower end of the feces collection trough is connected and fixedly arranged with a feces collecting pipe, and the lower end of the feces collecting pipe is detachably fixed on the plug-in fixing seat; a water pump is fixedly arranged on the water tank, a water supply pipe is fixedly arranged at the output end of the water pump, and the input end of the water pump extends into the bottom of the water tank; an annular water spray pipe is fixedly arranged on the upper end of the feces collection trough, and the water supply pipe is connected with the annular water spray pipe; a monitoring mechanism is fixedly arranged on the hospital bed, and the monitoring mechanism monitors the defecation needs of paralyzed patients and discovers abnormal situations in time.

Images