Stoma nursing guidance system and method based on image recognition

Abstract

The invention discloses a stoma nursing guidance system based on image recognition and a method thereof, and relates to the technical field of stoma nursing, the stoma nursing guidance system comprises a stoma nursing guidance platform, an intelligent guidance module and an operation error correction module, the stoma nursing guidance platform is electrically connected with a guidance model construction module, the output end of the guide model construction module is electrically connected with an intelligent guide module, and the intelligent guide module comprises a stoma image acquisition module, an image calibration module, an image recognition module, a stoma analysis module, a nursing decision module, an operation guide module, a primary evaluation module and an operation error correction module. According to the stoma nursing guidance system and method based on image recognition, the intelligent guidance module can create a 3D model of an abdominal stoma of a patient, 3D teaching operation is achieved, action teaching and step guidance are carried out in the form of combination of animation and voice, the operation error correction module can carry out secondary image collection and carry out error recognition and prompting according to the nursing fitting condition, and the operation efficiency is improved. And nursing operation error correction is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of stoma care, and specifically provides a stoma care guidance system and method based on image recognition. Background Art

[0002] A stoma refers to an artificial opening made on the abdominal wall for the treatment of intestinal diseases, and a section of intestinal tube is pulled out of the opening to output excrement. The main purpose of the stoma is to solve the intestinal obstruction caused by intestinal diseases, facilitate the output of excrement, and relieve the intestinal pressure; stoma care is a special kind of care, including the care of abdominal stoma patients, prevention and treatment of certain stoma complications. The chassis is one of the very important products in stoma care. On the one hand, it can play a role in fixing the stoma bag, and on the other hand, it can also protect the skin and keep the skin around the stoma from being affected by stoma excrement. To achieve a systematic care model for stoma patients, a stoma care guidance system is needed. Existing stomas are inconsistent in type, size, depth, and infection status, making it difficult for novices to perform reasonable and effective care for stomas, and it is also not convenient to correct errors in the nursing operation techniques and the fitting condition of the chassis after care. Summary of the Invention

[0003] Aiming at the deficiencies of the prior art, the present invention provides a stoma care guidance system and method based on image recognition, which solves the problems raised in the above background art.

[0004] To achieve the above objectives, the present invention is realized through the following technical solutions: A stoma care guidance system based on image recognition includes a stoma care guidance platform, an intelligent guidance module, and an operation error correction module. The stoma care guidance platform is electrically connected to a guidance model construction module, and the output end of the guidance model construction module is electrically connected to the intelligent guidance module. The intelligent guidance module includes a stoma image acquisition module, an image calibration module, an image recognition module, a stoma analysis module, a nursing decision module, an operation guidance module, a primary evaluation module, and an operation error correction module. The output end of the stoma image acquisition module is electrically connected to the image calibration module, and the output end of the image calibration module is electrically connected to the image recognition module. The output end of the image recognition module is electrically connected to the stoma analysis module, and the output end of the stoma analysis module is electrically connected to the nursing decision module. The output end of the nursing decision module is electrically connected to the operation guidance module, and the output end of the operation guidance module is electrically connected to the primary evaluation module. And the output end of the primary evaluation module is electrically connected to the operation error correction module. The stoma image acquisition module is electrically connected to an image acquisition terminal, and the image acquisition terminal includes a static image acquisition unit and a dynamic video acquisition unit. And the static image acquisition unit and the dynamic video acquisition unit are respectively used to acquire static stoma images and dynamic hand fitting operation videos.

[0005] Furthermore, the stoma care guidance platform is connected to a dataset construction module, and the dataset construction module is connected to a data privacy protection module. The dataset construction module is used to build its own dataset and construct a medical care knowledge base to make it training data based on clinical photos.

[0006] Furthermore, the guidance model construction module includes an image calling module and an image preprocessing module, and the output end of the image calling module is electrically connected to the image preprocessing module. The image calling module is used to call historical clinical photos in the dataset, and the image preprocessing module is used to perform data correction, denoising, and standardization.

[0007] Furthermore, the guidance model construction module also includes a deep learning modeling module and a sample training module. The output end of the image preprocessing module is electrically connected to the deep learning modeling module, and the output end of the deep learning modeling module is electrically connected to the sample training module. The deep learning modeling module builds a stoma care guidance decision model through machine learning, and the sample training module uses historical clinical data as the dataset for training the model to perform sample training. Then, through the learning and analysis of the training samples, the final guidance decision model is obtained.

[0008] Furthermore, the stoma image acquisition module is used to capture and collect the patient's stoma image. The image calibration module is used to identify the belly tilt and correct the angle deviation to improve the accuracy of subsequent image recognition. The image recognition module is used to identify the stoma morphology according to the CNN image recognition model and extract the edge of the stoma area to prepare for subsequent judgment of the chassis fitting situation.

[0009] Furthermore, the stoma analysis module is used to analyze the stoma type and the skin condition around the stoma. The care decision module is used to recommend the chassis model, opening size, softness, and edge design according to the recognition result based on the guidance model. The operation guidance module is used to create a 3D model of the patient's abdominal stoma and implement 3D teaching operations, and conduct action teaching and step guidance in the form of a combination of animation and voice.

[0010] Furthermore, the primary evaluation module includes a nursing action acquisition module, an action detection and judgment module, a nursing image acquisition module, and a nursing detection and judgment module. The nursing action acquisition module and the nursing image acquisition module are both electrically connected to the graphic acquisition terminal. The output end of the nursing action acquisition module is electrically connected to the action detection and judgment module, and the output end of the nursing image acquisition module is electrically connected to the carbon nursing detection and judgment module.

[0011] Further, the primary evaluation module further includes an evaluation and analysis module. The output ends of the motion detection and judgment module and the nursing detection and judgment module are electrically connected to the evaluation and analysis module. The motion detection and judgment module is used to perform motion detection based on a dynamic video, judge whether the hand operation is standard, and judge whether the fitting motion is correct. The nursing detection and judgment module is used to judge the fitting degree of the nursing chassis and analyze whether it is properly pasted based on the secondary graphic acquisition after nursing. The evaluation and analysis module is used to comprehensively analyze and judge whether the patient has correctly completed the steps, such as whether the paste is aligned, whether there are wrinkles, and whether it is pressed tightly.

[0012] Further, the output end of the primary evaluation module is electrically connected to an operation error correction module, and the output end of the operation error correction module is electrically connected to a secondary evaluation module. Moreover, the output end of the secondary evaluation module is electrically connected to a comparison and feedback module. The operation error correction module is used to identify and prompt errors based on the nursing fitting situation on the basis of the primary evaluation module, and give feedback suggestions for operation error correction. The secondary evaluation module is used to perform a secondary evaluation of stoma care. The comparison and feedback module is used to comprehensively compare the secondary evaluation with the primary evaluation, detect the nursing effect of the operator, and ensure the effectiveness of nursing.

[0013] Further, the usage method of the stoma care guidance system based on image recognition includes the following specific steps: Step 1: First, the stoma care guidance platform builds its own dataset through the dataset construction module, constructs a medical nursing knowledge base, and makes it the training data based on clinical photos; Step 2: Then, the guidance model construction module models the stoma care guidance decision model through machine learning, and uses historical clinical data as the dataset for training the model for sample training. Then, through the learning and analysis of the training samples, the final guidance decision model is obtained; Step 3: Then, the intelligent guidance module captures the patient's stoma image through the stoma image acquisition module. Then, the image calibration module identifies the belly tilt and corrects the angle deviation to improve the accuracy of subsequent image recognition. Then, the image recognition module identifies the stoma morphology according to the CNN image recognition model and extracts the edge of the stoma area to prepare for subsequent judgment of the chassis fitting situation; Step 4: Then, the stoma analysis module analyzes the stoma type and the skin condition around the stoma. Then, based on the recognition results on the basis of the guidance model, the nursing decision module recommends the chassis model, opening size, softness, and edge design. Finally, the operation guidance module creates a 3D model of the patient's abdominal stoma to achieve 3D teaching operation, and conducts action teaching and step guidance in the form of a combination of animation and voice; Step 5: After the operator completes the stoma care according to the guidance, the primary evaluation module comprehensively analyzes and judges whether the patient has correctly completed the steps, such as whether the paste is aligned, whether there are wrinkles, and whether it is pressed tightly. Then, based on the primary evaluation module, the operation error correction module identifies and prompts errors according to the nursing fitting situation, and gives feedback suggestions for operation error correction. Then, the secondary evaluation module conducts a secondary evaluation of the stoma care. Finally, the comparison and feedback module conducts a comprehensive comparison of the secondary evaluation and the primary evaluation to detect the operator's nursing effect and ensure the effectiveness of the nursing.

[0014] The present invention provides a stoma care guidance system and method based on image recognition, which has the following beneficial effects: 1. The stoma care guidance system and method based on image recognition are provided with a guidance model construction module. The dataset construction module is used to build its own dataset and construct a medical care knowledge base to make it training data based on clinical photos. The data privacy protection module is used to protect data privacy and perform encrypted transmission. The image calling module is used to call historical clinical photos in the dataset. The image preprocessing module is used to perform data correction, denoising, and standardization. The deep learning modeling module builds a stoma care guidance decision model through machine learning. The sample training module uses historical clinical data as the dataset for training the model to perform sample training. Then, through the learning and analysis of the training samples, the final guidance decision model is obtained.

[0015] 2. The stoma care guidance system and method based on image recognition are provided with an intelligent guidance module. The stoma image acquisition module is used to capture and collect the patient's stoma image. The image calibration module is used to identify the belly tilt and correct the angle deviation to improve the accuracy of subsequent image recognition. The image recognition module is used to identify the stoma shape according to the CNN image recognition model and extract the edge of the stoma area to prepare for subsequent judgment of the chassis fitting situation. The stoma analysis module is used to analyze the stoma type and the skin condition around the stoma. The nursing decision module is used to recommend the chassis model, opening size, softness, and edge design according to the recognition result based on the guidance model. The operation guidance module is used to create a 3D model of the patient's abdominal stoma to realize 3D teaching operation, and conduct action teaching and step guidance in the form of a combination of animation and voice.

[0016] 3. The ostomy care guidance system and its method based on image recognition are provided with a primary evaluation module. The nursing action acquisition module and the nursing image acquisition module are respectively used to collect static images and dynamic videos. The action detection and judgment module is used to perform action detection based on the dynamic video and judge whether the hand operation is standard and whether the fitting action is correct. The nursing detection and judgment module is used to judge the fitting degree of the nursing chassis based on the secondary graphic acquisition after nursing, analyze whether it is pasted well. The evaluation and analysis module is used to comprehensively analyze and judge whether the patient has correctly completed the steps, such as whether the paste is aligned, whether there are wrinkles, and whether it is pressed tightly.

[0017] 4. The ostomy care guidance system and its method based on image recognition are provided with an operation error correction module and a secondary evaluation module. The operation error correction module is used to perform error recognition and prompt based on the nursing fitting situation on the basis of the primary evaluation module, and give feedback suggestions for operation error correction. The secondary evaluation module is used to perform secondary evaluation of ostomy care, and its process is similar to that of the primary evaluation module. The comparison and feedback module is used to perform comprehensive comparison between the secondary evaluation and the primary evaluation to detect the nursing effect of the operator and ensure the effectiveness of nursing. The learning record module is used to track the learning progress of the operator, construct a learning file, and store historical operation data. The guidance and feedback module is used to analyze the strengths and weaknesses of the operator during the actual operation process and give customized training guidance and prompts according to the actual operation performance of the operator. Description of the Drawings

[0018] Figure 1 It is a schematic diagram of the system framework of the ostomy care guidance system and its method based on image recognition of the present invention; Figure 2 It is a schematic diagram of the intelligent guidance module process of the ostomy care guidance system and its method based on image recognition of the present invention; Figure 3 It is a schematic diagram of the guidance model construction module of the ostomy care guidance system and its method based on image recognition of the present invention; Figure 4 It is a schematic diagram of the ostomy image acquisition module of the ostomy care guidance system and its method based on image recognition of the present invention; Figure 5 It is a schematic diagram of the primary evaluation module of the ostomy care guidance system and its method based on image recognition of the present invention.

[0019] In the figure: 1. Stoma care guidance platform; 2. Dataset construction module; 3. Data privacy protection module; 4. Guidance model construction module; 401. Image calling module; 402. Image preprocessing module; 403. Deep learning modeling module; 404. Sample training module; 5. Intelligent guidance module; 6. Learning record module; 7. Guidance feedback module; 8. Stoma image acquisition module; 9. Image calibration module; 10. Image recognition module; 11. Stoma analysis module; 12. Nursing decision-making module; 13. Operation guidance module; 14. First evaluation module; 1401. Nursing action acquisition module; 1402. Action detection and judgment module; 1403. Nursing image acquisition module; 1404. Nursing detection and judgment module; 1405. Evaluation and analysis module; 15. Operation error correction module; 16. Second evaluation module; 17. Comparison and feedback module; 18. Image acquisition terminal; 1801. Static image acquisition unit; 1802. Dynamic video acquisition unit. Detailed implementation manners

[0020] Please refer to Figure 1 and Figure 3 As shown in the figure, the present invention provides a technical solution: A stoma care guidance system based on image recognition, including a stoma care guidance platform 1, an intelligent guidance module 5, and an operation error correction module 15. The stoma care guidance platform 1 is electrically connected to a guidance model construction module 4, and the output end of the guidance model construction module 4 is electrically connected to the intelligent guidance module 5. The stoma care guidance platform 1 is connected to a dataset construction module 2, and the dataset construction module 2 is connected to a data privacy protection module 3. The guidance model construction module 4 includes an image calling module 401 and an image preprocessing module 402, and the output end of the image calling module 401 is electrically connected to the image preprocessing module 402. The guidance model construction module 4 further includes a deep learning modeling module 403 and a sample training module 404. The output end of the image preprocessing module 402 is electrically connected to the deep learning modeling module 403, and the output end of the deep learning modeling module 403 is electrically connected to the sample training module 404; The specific operations are as follows. The dataset construction module 2 is used to build its own dataset and construct a medical care knowledge base to make it training data based on clinical photos. The data privacy protection module 3 is used to protect data privacy and perform encrypted transmission. The image calling module 401 is used to call historical clinical photos in the dataset. The image preprocessing module 402 is used to perform data correction, denoising, and standardization. The deep learning modeling module 403 performs modeling of a stoma care guidance decision model through machine learning. The sample training module 404 uses historical clinical data as the dataset for training the model to perform sample training, and then obtains the final guidance decision model through learning and analysis of the training samples.

[0021] Please refer to Figure 2, the intelligent guidance module 5 includes a stoma image acquisition module 8, an image calibration module 9, an image recognition module 10, a stoma analysis module 11, a nursing decision-making module 12, an operation guidance module 13, a primary evaluation module 14, and an operation error correction module 15. The output end of the stoma image acquisition module 8 is electrically connected to the image calibration module 9, and the output end of the image calibration module 9 is electrically connected to the image recognition module 10. The output end of the image recognition module 10 is electrically connected to the stoma analysis module 11, and the output end of the stoma analysis module 11 is electrically connected to the nursing decision-making module 12. The output end of the nursing decision-making module 12 is electrically connected to the operation guidance module 13, and the output end of the operation guidance module 13 is electrically connected to the primary evaluation module 14. Moreover, the output end of the primary evaluation module 14 is electrically connected to the operation error correction module 15; The specific operations are as follows. The stoma image acquisition module 8 is used to capture and collect the stoma images of the patient. The image calibration module 9 is used to identify the tilt of the belly and correct the angular deviation to improve the accuracy of subsequent image recognition. The image recognition module 10 is used to identify the stoma morphology according to the CNN image recognition model and extract the edge of the stoma area to prepare for subsequent judgment of the chassis fitting situation. The stoma analysis module 11 is used to analyze the stoma type and the skin condition around the stoma. The nursing decision-making module 12 is used to recommend the chassis model, opening size, softness, and edge design based on the recognition results on the basis of the guidance model. The operation guidance module 13 is used to create a 3D model of the patient's abdominal stoma to achieve 3D teaching operations and conduct action teaching and step guidance in the form of a combination of animation and voice.

[0022] Please refer to Figure 4 and Figure 5, the output terminal of the primary evaluation module 14 is electrically connected to the operation error correction module 15, the output terminal of the operation error correction module 15 is electrically connected to the secondary evaluation module 16, and the output terminal of the secondary evaluation module 16 is electrically connected to the comparison feedback module 17. The primary evaluation module 14 includes a nursing action acquisition module 1401, an action detection and judgment module 1402, a nursing image acquisition module 1403, and a nursing detection and judgment module 1404. The nursing action acquisition module 1401 and the nursing image acquisition module 1403 are both electrically connected to the graphic acquisition terminal. The output terminal of the nursing action acquisition module 1401 is electrically connected to the action detection and judgment module 1402, and the output terminal of the nursing image acquisition module 1403 is electrically connected to the carbon nursing detection and judgment module 1404. The primary evaluation module 14 further includes an evaluation and analysis module 1405. The output terminals of the action detection and judgment module 1402 and the nursing detection and judgment module 1404 are electrically connected to the evaluation and analysis module 1405. The stoma image acquisition module 8 is electrically connected to the image acquisition terminal 18, and the image acquisition terminal 18 includes a static image acquisition unit 1801 and a dynamic video acquisition unit 1802. The static image acquisition unit 1801 and the dynamic video acquisition unit 1802 are respectively used to acquire static stoma images and dynamic hand fitting operation videos; The specific operations are as follows. The nursing action acquisition module 1401 and the nursing image acquisition module 1403 are respectively used to acquire static images and dynamic videos. The action detection and judgment module 1402 is used to perform action detection based on the dynamic video and judge whether the hand operation is standard and whether the fitting action is correct. The nursing detection and judgment module 1404 is used to judge the nursing chassis fitting degree based on the secondary graphic acquisition after nursing, analyze whether it is properly pasted. The evaluation and analysis module 1405 is used to comprehensively analyze and judge whether the patient has correctly completed the steps, such as whether the paste is aligned, whether there are wrinkles, and whether it is pressed tightly. The operation error correction module 15 is used to, based on the primary evaluation module 14, perform error recognition and prompt according to the nursing fitting situation, and give feedback suggestions for operation error correction. The secondary evaluation module 16 is used to perform secondary evaluation of stoma care, which is similar to the process of the primary evaluation module 14. The comparison feedback module 17 is used to perform comprehensive comparison between the secondary evaluation and the primary evaluation, detect the nursing effect of the operator, and ensure the effectiveness of nursing.

[0023] In summary, please refer to Figures 1 - 5, for the ostomy care guidance system and its method based on image recognition, during use, first, the ostomy care guidance platform 1 builds its own dataset through the dataset construction module 2, constructs a medical care knowledge base to make it training data based on clinical photos, then the image calling module 401 of the guiding model construction module 4 calls historical clinical photos in the dataset, then the image preprocessing module 402 corrects, denoises, and normalizes the data, then the deep learning modeling module 403 builds a decision-making model for ostomy care guidance through machine learning, and finally the sample training module 404 uses the historical clinical data as the dataset for training the model to conduct sample training, and then through learning and analysis of the training samples, obtains the final guiding decision-making model; Then, the intelligent guiding module 5 captures and collects the patient's ostomy image through the ostomy image acquisition module 8. Next, the image calibration module 9 identifies the belly tilt and corrects the angle deviation to improve the accuracy of subsequent image recognition. Then, the image recognition module 10 identifies the ostomy morphology according to the CNN image recognition model and extracts the edge of the ostomy area to prepare for subsequent judgment of the chassis fitting condition. Then, the ostomy analysis module 11 analyzes the ostomy type and the skin condition around the ostomy. Then, based on the guiding model, the nursing decision-making module 12 recommends the chassis model, opening size, softness, and edge design according to the recognition result. Finally, the operation guidance module 13 creates a 3D model of the patient's abdominal ostomy to achieve 3D teaching operations, and conducts action teaching and step guidance in the form of a combination of animation and voice; After the operator completes ostomy care according to the guidance, the primary evaluation module 14 respectively captures static images and dynamic videos through the nursing action acquisition module 1401 and the nursing image acquisition module 1403. Then, the action detection and judgment module 1402 can, based on the dynamic video, conduct action detection and judge whether the hand operation is standard and whether the fitting action is correct. Then, the nursing detection and judgment module 1404, based on the secondary graphic acquisition after nursing, judges the fitting degree of the nursing chassis and analyzes whether it is properly pasted. Then, the evaluation and analysis module 1405 comprehensively analyzes and judges whether the patient has correctly completed the steps, such as whether the paste is aligned, whether there are wrinkles, and whether it is pressed tightly. Then, based on the primary evaluation module 14, the operation error correction module 15 conducts error recognition and prompts according to the nursing fitting situation, and gives feedback suggestions for operation error correction; Then, the secondary evaluation module 16 conducts a secondary evaluation of stoma care, which is similar to the process of the primary evaluation module 14. Finally, the comparison and feedback module 17 conducts a comprehensive comparison between the secondary evaluation and the primary evaluation to detect the care effect of the operator and ensure the effectiveness of the care. During this process, the learning record module 6 can track the learning progress of the operator, construct a learning file, and store historical operation data. Then, the guidance and feedback module 7 can analyze the strengths and weaknesses of the operator during the practical operation process and give customized training guidance and tips according to the operator's practical performance. In this way, the use process of the entire stoma care guidance system and method based on image recognition is completed.

[0024] The embodiments of the present invention are given for purposes of illustration and description, and are not exhaustive or limit the invention to the disclosed form. Many modifications and variations are obvious to those of ordinary skill in the art. The embodiments are chosen and described in order to better explain the principles of the invention and its practical application, and to enable those of ordinary skill in the art to understand the invention and design various embodiments suitable for specific purposes with various modifications.

Claims

1. A stoma care guidance system based on image recognition, characterized in that, It includes a stoma care guidance platform (1), an intelligent guidance module (5) and an operation error correction module (15). The stoma care guidance platform (1) is electrically connected to a guidance model construction module (4), and the output end of the guidance model construction module (4) is electrically connected to the intelligent guidance module (5). The intelligent guidance module (5) includes a stoma image acquisition module (8), an image calibration module (9), an image recognition module (10), a stoma analysis module (11), a nursing decision-making module (12), an operation guidance module (13), a primary evaluation module (14) and an operation error correction module (15). The output end of the stoma image acquisition module (8) is electrically connected to the image calibration module (9), and the output end of the image calibration module (9) is electrically connected to the image recognition module (10). The output end of the image recognition module (10) is electrically connected to the stoma analysis module (11), and the output end of the stoma analysis module (11) is electrically connected to the nursing decision-making module (12). The output end of the nursing decision-making module (12) is electrically connected to the operation guidance module (13), and the output end of the operation guidance module (13) is electrically connected to the primary evaluation module (14). Moreover, the output end of the primary evaluation module (14) is electrically connected to the operation error correction module (15). The stoma image acquisition module (8) is electrically connected to an image acquisition terminal (18), and the image acquisition terminal (18) includes a static image acquisition unit (1801) and a dynamic video acquisition unit (1802). The static image acquisition unit (1801) and the dynamic video acquisition unit (1802) are respectively used to acquire static stoma images and dynamic hand fitting operation videos.

2. The ostomy care guidance system based on image recognition according to claim 1, characterized in that The stoma care guidance platform (1) is connected to a dataset construction module (2), and the dataset construction module (2) is connected to a data privacy protection module (3). The dataset construction module (2) is used to build its own dataset and construct a medical care knowledge base to make it training data based on clinical photos.

3. The ostomy care guidance system based on image recognition according to claim 1, wherein, The guidance model construction module (4) includes an image calling module (401) and an image preprocessing module (402). The output end of the image calling module (401) is electrically connected to the image preprocessing module (402). The image calling module (401) is used to call historical clinical photos in the dataset, and the image preprocessing module (402) is used to perform data correction, denoising and standardization.

4. The ostomy care guidance system based on image recognition according to claim 3, wherein The guidance model construction module (4) also includes a deep learning modeling module (403) and a sample training module (404). The output end of the image preprocessing module (402) is electrically connected to the deep learning modeling module (403), and the output end of the deep learning modeling module (403) is electrically connected to the sample training module (404). The deep learning modeling module (403) builds a stoma care guidance decision model by means of machine learning. The sample training module (404) uses historical clinical data as the dataset for training the model to perform sample training, and then obtains the final guidance decision model through learning and analysis of the training samples.

5. The ostomy care guidance system based on image recognition according to claim 1, characterized in that, The stoma image acquisition module (8) is used to capture the stoma images of the patient. The image calibration module (9) is used to identify the abdominal tilt and correct the angle deviation, so as to improve the accuracy of subsequent image recognition. The image recognition module (10) is used to identify the stoma morphology according to the CNN image recognition model and extract the edge of the stoma area, so as to prepare for subsequent judgment of the chassis fitting situation.

6. The ostomy care guidance system based on image recognition according to claim 1, characterized in that, The stoma analysis module (11) is used to analyze the stoma type and the skin condition around the stoma. The nursing decision-making module (12) is used to recommend the chassis model, opening size, softness and edge design according to the recognition result based on the guidance model. The operation guidance module (13) is used to create a 3D model of the patient's abdominal stoma, realize 3D teaching operation, and conduct action teaching and step guidance in the form of combination of animation and voice.

7. The ostomy care guidance system based on image recognition according to claim 1, characterized in that, The primary evaluation module (14) includes a nursing action acquisition module (1401), an action detection and judgment module (1402), a nursing image acquisition module (1403) and a nursing detection and judgment module (1404). The nursing action acquisition module (1401) and the nursing image acquisition module (1403) are both electrically connected to the graphic acquisition terminal. The output end of the nursing action acquisition module (1401) is electrically connected to the action detection and judgment module (1402), and the output end of the nursing image acquisition module (1403) is electrically connected to the carbon nursing detection and judgment module (1404).

8. The ostomy care guidance system based on image recognition according to claim 7, wherein The primary evaluation module (14) further includes an evaluation and analysis module (1405). The output ends of the action detection and judgment module (1402) and the nursing detection and judgment module (1404) are electrically connected to the evaluation and analysis module (1405). The action detection and judgment module (1402) is used to perform action detection based on the dynamic video, judge whether the hand operation is standard and whether the fitting action is correct. The nursing detection and judgment module (1404) is used to judge the nursing chassis fitting degree based on the secondary graphic acquisition after nursing, and analyze whether it is pasted well. The evaluation and analysis module (1405) is used to comprehensively analyze and judge whether the patient has correctly completed the steps, such as whether the paste is aligned, whether there are wrinkles, and whether it is pressed tightly.

9. The ostomy care guidance system based on image recognition according to claim 1, wherein, The output end of the primary evaluation module (14) is electrically connected to an operation error correction module (15), and the output end of the operation error correction module (15) is electrically connected to a secondary evaluation module (16). The output end of the secondary evaluation module (16) is electrically connected to a comparison and feedback module (17). The operation error correction module (15) is used to perform error recognition and prompt according to the nursing fitting situation based on the primary evaluation module (14), and give feedback suggestions for operation error correction. The secondary evaluation module (16) is used to conduct secondary evaluation of stoma care. The comparison and feedback module (17) is used to conduct comprehensive comparison between the secondary evaluation and the primary evaluation, detect the nursing effect of the operator, and ensure the effectiveness of nursing.

10. A stoma care guidance system based on image recognition according to any one of claims 1-9, characterized in that, The usage method of the stoma care guidance system based on image recognition includes the following specific steps: Step 1: First, the stoma care guidance platform (1) constructs its own dataset through the dataset construction module (2) to build a medical care knowledge base, making it training data based on clinical photos; Step 2: Next, the guidance model construction module (4) models the stoma care guidance decision model through machine learning, and uses the historical clinical data as the dataset for training the model for sample training. Then, through the learning and analysis of the training samples, the final guidance decision model is obtained; Step 3: Then, the intelligent guidance module (5) captures the patient's stoma image through the stoma image acquisition module (8). Next, the image calibration module (9) identifies the tilt of the belly and corrects the angle deviation to improve the accuracy of subsequent image recognition. Then, the image recognition module (10) identifies the stoma morphology according to the CNN image recognition model and extracts the edge of the stoma area to prepare for the subsequent judgment of the chassis fitting situation; Step 4: Then, the stoma analysis module (11) analyzes the stoma type and the skin condition around the stoma. Then, based on the recognition results, the nursing decision module (12) recommends the chassis model, opening size, softness, and edge design on the basis of the guidance model. Finally, the operation guidance module (13) creates a 3D model of the patient's abdominal stoma to achieve 3D teaching operations, and conducts action teaching and step guidance in the form of a combination of animation and voice; Step 5: After the operator completes the stoma care according to the guidance, the primary evaluation module (14) comprehensively analyzes and judges whether the patient has correctly completed the steps, such as whether the paste is aligned, whether there are wrinkles, and whether it is pressed tightly. Then, based on the primary evaluation module (14), the operation error correction module (15) identifies and prompts errors according to the nursing fitting situation, and gives feedback suggestions for operation error correction. Then, the secondary evaluation module (16) conducts a secondary evaluation of the stoma care. Finally, the comparison feedback module (17) conducts a comprehensive comparison of the secondary evaluation and the primary evaluation to detect the operator's nursing effect and ensure the effectiveness of the nursing.

Images