Invisible appliance wearing monitoring and adjusting system and method based on computer vision

Abstract

The invention relates to the technical field of invisible orthodontic appliance wearing monitoring and adjusting systems, in particular to an invisible orthodontic appliance wearing monitoring and adjusting system and method based on computer vision, and the system comprises a cloud storage module which is used for storing initial oral data and re-consultation oral data of a patient; the cloud processing module is used for analyzing the oral cavity picture uploaded by the patient and executing analysis software; the client module is used for managing an electronic medical record of a patient, providing treatment information, obtaining an analysis result and generating personalized feedback; according to the invention, a computer vision algorithm is combined with oral medical knowledge, so that not only can the position change of teeth be accurately identified and tracked, but also the judgment ability of the device can be continuously optimized through a deep learning algorithm. Meanwhile, the cloud computing architecture adopted by the invention enables the system to process mass data, and powerful data support is provided for formulating a personalized treatment scheme.

Description

Technical Field

[0001] The present invention relates to the technical field of invisible brace appliance wearing monitoring and adjustment systems, and in particular to an invisible brace appliance wearing monitoring and adjustment system and method based on computer vision. Background Art

[0002] As people pay more and more attention to oral health and aesthetics, invisible braces have been widely used in recent years due to their beautiful and comfortable features. Traditional invisible braces treatment methods mainly rely on doctors' experience and regular follow-up visits. However, this method has many shortcomings, which seriously restricts the improvement of treatment effects and patient experience.

[0003] First, it is difficult to achieve accurate treatment monitoring with traditional methods. Doctors usually need to evaluate the movement of teeth through naked eye observation or simple measuring tools. This method is highly subjective and difficult to capture subtle changes. Secondly, fixed follow-up cycles often fail to promptly detect and solve problems that arise during treatment, resulting in poor treatment effects or prolonged treatment time. Furthermore, frequent face-to-face follow-up visits not only increase the patient's time cost, but also put tremendous pressure on medical resources.

[0004] In addition, existing technologies also face the problem of difficulty in formulating personalized treatment plans. Each patient's oral conditions and treatment needs are unique, and it is difficult to formulate the best treatment plan for each patient based solely on the doctor's experience. At the same time, patients also lack intuitive and timely feedback and find it difficult to understand their own treatment progress, which to some extent affects treatment compliance.

[0005] In view of the above problems, there is an urgent need for an invisible braces wearing monitoring and adjustment system that can achieve accurate monitoring, timely adjustment, and personalized treatment. The present invention is an innovative solution proposed to meet this demand. Summary of the invention

[0006] The computer vision-based invisible braces wearing monitoring and adjustment system proposed in this invention aims to solve a series of problems existing in traditional methods, such as inaccurate monitoring, untimely adjustment, and low personalization. By organically combining advanced computer vision, cloud computing, and Internet of Things technologies, the present invention realizes intelligent management of the entire invisible braces treatment process.

[0007] The present invention proposes a computer vision-based invisible brace appliance wearing monitoring and adjustment system, comprising:

[0008] Cloud storage module, used to store the patient's initial oral data and follow-up oral data;

[0009] A cloud processing module that analyzes oral photos uploaded by patients and executes analysis software;

[0010] The client module manages patient electronic medical records, provides treatment information, obtains analysis results, and generates personalized feedback;

[0011] The appliance module is used to obtain the oral data of the patient's follow-up visit;

[0012] The cloud processing module includes a computer vision module, which includes an image preprocessing submodule, an image segmentation and feature extraction submodule, a tooth recognition submodule and a tooth movement assessment submodule. The tooth movement assessment submodule calculates the degree of tooth movement using the following formula:

[0013] M=Σ(w i *||P i -P i ′||) / N

[0014] Among them, M is the tooth movement degree score, w i is the weight factor of each tooth, P i is the current tooth position coordinate, P i ′ is the tooth position coordinate at the last follow-up visit, N is the total number of teeth evaluated, and ||·|| represents the Euclidean distance.

[0015] Preferably, it also includes a data upload module, a data download module and a data update module.

[0016] Among them, the data upload module is used to collect and upload oral photos of patients for follow-up visits, the data download module is used to obtain treatment data, and the data update module is used to update the treatment data in the client module and the appliance module;

[0017] The cloud storage module includes a storage server, a data interaction server and a data backup server, which are used to store patient data, respond to data access requests and back up data respectively.

[0018] Preferably, the appliance module includes a plurality of pressure sensors and a main control mounted on the surface of the appliance.

[0019] The pressure sensor is used to collect the pressure value of the appliance acting on each tooth, and the main control is used to obtain the information collected by the appliance and upload it to the client module and the cloud processing module;

[0020] The client module includes a personal interface module and a mobile tracking module, wherein the mobile tracking module uses the following formula to calculate the treatment progress:

[0021] P=(M / M exp )*100%

[0022] Among them, P is the percentage of treatment progress, M is the actual degree of tooth movement, and M expThe expected degree of tooth movement.

[0023] The method based on the system comprises the following steps:

[0024] S1. Collect and upload the patient's initial oral data as the initial data set;

[0025] S2. Collect oral photos during the patient's regular follow-up visits to form a follow-up photo collection;

[0026] S3. Upload and analyze the follow-up photo collection to obtain tooth movement information;

[0027] S4. Generate personal wearing monitoring report and appliance adjustment plan;

[0028] S5. Feedback the wearing information report and adjustment plan to doctors and patients;

[0029] Wherein, step S3 includes preprocessing, segmenting, feature extraction and tooth recognition of the follow-up photos, and calculating the degree of tooth movement using the formula in step S3.

[0030] Preferably, the step S1 further includes:

[0031] S11. Separate the facial area where the teeth and the orthodontic appliance are located from the background to form an original image set;

[0032] S12. extracting multiple sub-images based on the original image set, marking them as tooth areas, orthodontic appliance areas, and irrelevant areas;

[0033] S13. Divide the data set into a training set and a test set in a ratio of 8:2, train the model and evaluate it. The model evaluation uses the following formula to calculate the accuracy:

[0034] Accuracy=(TP+TN) / (TP+TN+FP+FN)

[0035] Among them, TP is true positive, TN is true negative, FP is false positive, and FN is false negative.

[0036] Preferably, the preprocessing in step S3 includes:

[0037] S31. Extracting the contours of teeth and orthodontic appliances in the image;

[0038] S32. Correct the patient's oral position and eliminate the inaccurate relative position between teeth;

[0039] S33. Using the trained model to identify the positions of teeth and orthodontic appliances in the image;

[0040] S34. Obtain the position, size and spatial relationship of teeth and appliances;

[0041] S35. Generate a visual image of teeth and orthodontic appliances;

[0042] Among them, step S31 uses the adaptive threshold method to binarize the image, and the formula is as follows:

[0043]

[0044] Where T(x,y) is the threshold of pixel (x,y), μ(x,y) is the average grayscale value of the neighborhood of (x,y), σ(x,y) is the standard deviation of the neighborhood of (x,y), R is the grayscale dynamic range of the image, and k is the weight coefficient.

[0045] Preferably, the step S31 further includes:

[0046] S311. Enhance or reduce noise on the original image of teeth and orthodontic appliances;

[0047] S312. grayscale the tooth image;

[0048] S313. Separating the outlines of teeth and orthodontic appliances from the background using an image segmentation method;

[0049] The noise reduction process in step S311 uses a Gaussian filter, and the formula is as follows:

[0050]

[0051] Among them, G(x,y) is the Gaussian function, (x,y) is the pixel coordinate, and σ is the standard deviation.

[0052] Preferably, the step S3 further includes:

[0053] S36. Compare the appliance image with the initial appliance image to obtain a deviation between the actual appliance position and the ideal appliance position;

[0054] S37. Compare the tooth image with the initial tooth image to obtain the deviation between the actual tooth position and the ideal tooth position;

[0055] S38. Analyze the deviation data to obtain the specific situation of tooth movement;

[0056] The degree of tooth movement is calculated using the following formula:

[0057]

[0058] Where D is the average tooth movement distance, (x i ,y i ,z i ) is the three-dimensional coordinate of the current tooth position, (x i ′,yi ′,z i ′) is the three-dimensional coordinate of the initial tooth position, and N is the total number of teeth.

[0059] Preferably, the image comparison in steps S36 and S37 uses the structural similarity index (SSIM), and the formula is as follows:

[0060]

[0061] Among them, μ x and μ y are the average values ​​of images x and y, respectively, x and σ y are the standard deviations of images x and y, σ x y is the covariance of x and y, and C1 and C2 are constants used to avoid the denominator being zero.

[0062] Preferably, the generation of the appliance adjustment plan in step S4 includes:

[0063] Provide appliance adjustment options when teeth are not moving as expected;

[0064] When the teeth move as expected, provide a treatment plan for the next visit;

[0065] The treatment progress was evaluated using the following formula:

[0066] Progress=(1-(∑|D i -D′ i | / ∑|D′ i |))*100%

[0067] Among them, Progress is the percentage of treatment progress, Di is the actual tooth movement distance, Di' is the expected tooth movement distance, and ∑ represents the sum of all teeth;

[0068] For teeth whose actual movement does not match the expected movement, the appliance adjustment amount is calculated as follows:

[0069] Adjustment = k*(D′ i -D i )

[0070] Where Adjustment is the adjustment amount of the appliance, k is the adjustment coefficient (usually between 0.5 and 1.5), and D′ i is the expected tooth movement distance, D i The actual tooth movement distance.

[0071] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0072] In terms of technical principles, the present invention cleverly combines computer vision algorithms with oral medical knowledge to develop an intelligent system specifically for oral image analysis. This system can not only accurately identify and track the position changes of teeth, but also continuously optimize its own judgment ability through deep learning algorithms. At the same time, the cloud computing architecture adopted by the present invention enables the system to process massive amounts of data, providing strong data support for the formulation of personalized treatment plans.

[0073] In terms of complementary and synergistic effects, the various modules of the present invention form an organic whole. For example, the cooperation of the cloud storage module and the cloud processing module not only ensures the security and accessibility of the data, but also realizes the efficient use of computing resources. The real-time pressure data collected by the appliance module and the image data analyzed by the computer vision module verify each other, greatly improving the accuracy of monitoring. The client module converts complex data analysis results into information that can be understood by both patients and doctors through an intuitive interface, which promotes doctor-patient communication.

[0074] It is particularly worth mentioning that the present invention solves the seemingly contradictory problem of accurate monitoring and reducing the number of follow-up visits in traditional methods. Through remote real-time monitoring and intelligent analysis, the system can continuously track the progress of treatment and promptly identify and solve problems without the need for patients to visit the doctor frequently. This not only improves the efficiency and effectiveness of treatment, but also greatly improves the patient's medical experience.

[0075] In addition, the algorithm design of the present invention fully considers the special needs of orthodontics. For example, a weight factor is introduced in the tooth movement assessment, taking into account the importance and movement difficulty of different teeth, so that the assessment results are more in line with clinical practice. The system can also automatically adjust parameters according to the progress of treatment, realizing dynamic optimization of the treatment plan.

[0076] In general, this invention has brought revolutionary changes to invisible braces treatment through technological innovation and the integration of multidisciplinary knowledge. It not only significantly improves the accuracy and efficiency of treatment, but also provides patients with a more comfortable and convenient treatment experience. At the same time, it also provides doctors with a powerful decision-making support tool and improves the efficiency of medical resource utilization. It can be foreseen that this treatment model based on artificial intelligence and remote monitoring is likely to become the mainstream trend in the field of orthodontics in the future, bringing good news to more patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 It is the overall logic block diagram of the system of the present invention.

[0078] Figure 2 This is a logical block diagram of the cloud storage module of the present invention.

[0079] Figure 3 It is a logical block diagram of the cloud processing module of the present invention.

[0080] Figure 4 It is a logic block diagram of the client module of the present invention.

[0081] Figure 5 It is a logic block diagram of the orthodontic appliance module of the present invention.

[0082] Figure 6 It is a logic block diagram of the system workflow of the present invention.

[0083] Figure 7 This is a logical flowchart of the workflow of the cloud processing module of the present invention. DETAILED DESCRIPTION

[0084] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0085] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0086] Example 1

[0087] See also Figure 1-7 The present invention relates to a computer vision-based invisible brace wear monitoring and adjustment system and method. The system comprises a cloud storage module 1, a cloud processing module 2, a client module 3 and a brace module 4.

[0088] First, the cloud storage module 1 is used to store the patient's initial oral data and follow-up oral data. These data are the basis for the operation of the entire system and provide important reference information for subsequent analysis and adjustment. Preferably, the cloud storage module 1 includes a storage server 11, a data interaction server 12 and a data backup server 13. The storage server 11 is mainly responsible for storing the patient's electronic medical records, oral photos and oral data collected in real time by the appliance module 4. The data interaction server 12 is responsible for responding to medical staff and patients' access requests to the cloud storage module 1 to ensure the timeliness and accessibility of the data. The data backup server 13 ensures the security and reliability of the data by regularly backing up the data in the storage server 11.

[0089] Secondly, the cloud processing module 2 is the core of the system, which is used to analyze the oral photos uploaded by the patient and execute the analysis software. An important feature of the present invention is that the cloud processing module 2 includes a computer vision module 21. This module is further subdivided into a picture preprocessing submodule 211, an image segmentation and feature extraction submodule 212, a tooth recognition submodule 213 and a tooth movement assessment submodule 214. These submodules work together to achieve a comprehensive analysis of oral photos.

[0090] In particular, the tooth movement assessment submodule 214 uses an innovative algorithm to calculate the degree of tooth movement. The algorithm formula is:

[0091] M=∑(w i *||P i -P i ′||) / N

[0092] Where M represents the tooth movement score, w i is the weight factor of each tooth (usually set according to the importance and difficulty of tooth movement, for example, the weight of anterior teeth may be higher than that of molars), Pi is the coordinate of the current tooth position, and P′ i is the coordinate of the tooth position at the last follow-up visit, N is the total number of teeth evaluated, and ||·|| represents the Euclidean distance. This formula takes into account the importance and movement distance of each tooth and can more accurately reflect the overall treatment progress.

[0093] Again, the client module 3 plays an important interface role in the entire system. It is not only used to manage the patient's electronic medical records, but also provides treatment information, obtains analysis results, and generates personalized feedback. In one embodiment of the present invention, the client module 3 includes a personal interface module 31 and a mobile tracking module 32. The personal interface module 31 provides an intuitive data display and interactive interface for patients and doctors, while the mobile tracking module 32 is responsible for monitoring the position of the patient's oral appliance, evaluating the treatment progress, and generating personalized treatment plans and appliance adjustment plans.

[0094] It is worth noting that the mobile tracking module 32 uses a simple but effective formula to calculate the treatment progress:

[0095] P=(M / M exp )*100%

[0096] Here, P represents the percentage of treatment progress, M is the actual degree of tooth movement (calculated by the tooth movement evaluation submodule 214 mentioned above), and M exp This formula can intuitively reflect the relationship between the actual progress of treatment and the expected progress, helping doctors and patients better understand and adjust treatment plans.

[0097] Finally, the appliance module 4 is the part of the whole system that is in direct contact with the patient. It includes multiple pressure sensors 41 installed on the surface of the appliance and a main control 42. The pressure sensor 41 is used to collect the pressure value of the appliance on each tooth, which is crucial for evaluating the wearing condition and effect of the appliance. The main control 42 is responsible for obtaining the information collected by the appliance from the cloud storage module 1 and uploading this information to the client module 3 and the cloud processing module 2, realizing real-time transmission and analysis of data.

[0098] Next, let us elaborate on the method part of the present invention. The method comprises the following steps:

[0099] S1. Collect and upload the patient's initial oral data as the initial data set;

[0100] S2. Collect oral photos during the patient's regular follow-up visits to form a follow-up photo collection;

[0101] S3. Upload and analyze the follow-up photo collection to obtain tooth movement information;

[0102] S4. Generate personal wearing monitoring report and appliance adjustment plan;

[0103] S5. Feedback the wearing information report and adjustment plan to doctors and patients;

[0104] Wherein, step S3 includes preprocessing, segmenting, feature extraction and tooth recognition of the follow-up photos, and calculating the degree of tooth movement using the formula in step S3.

[0105] The step S1 further comprises: S11. separating the facial area where the teeth and the orthodontic appliance are located from the background to form an original image set;

[0106] S12. extracting multiple sub-images based on the original image set, marking them as tooth areas, orthodontic appliance areas, and irrelevant areas;

[0107] S13. Divide the data set into a training set and a test set in a ratio of 8:2, train the model and evaluate it. The model evaluation uses the following formula to calculate the accuracy:

[0108] Accuracy=(TP+TN) / (TP+TN+FP+FN)

[0109] Among them, TP is true positive, TN is true negative, FP is false positive, and FN is false negative.

[0110] The preprocessing in step S3 includes: S31. extracting the contours of teeth and orthodontic appliances in the image;

[0111] S32. Correct the patient's oral position and eliminate the inaccurate relative position between teeth;

[0112] S33. Using the trained model to identify the positions of teeth and orthodontic appliances in the image;

[0113] S34. Obtain the position, size and spatial relationship of teeth and appliances;

[0114] S35. Generate a visual image of teeth and orthodontic appliances;

[0115] Among them, step S31 uses the adaptive threshold method to binarize the image, and the formula is as follows:

[0116] T(x,y)=μ(x,y)*(1+k*((σ(x,y) / R)-1))

[0117] Among them, T(x,y) is the threshold of pixel (x,y), μ(x,y) is the average gray value of the neighborhood of (x,y), σ(x,y) is the standard deviation of the neighborhood of (x,y), R is the gray dynamic range of the image, and k is the weight coefficient. (Usually 0.2). This method can effectively deal with the problem of uneven illumination in oral photos and improve the accuracy of subsequent analysis.

[0118] The step S31 further includes: S311. performing enhancement or noise reduction processing on the original image of teeth and orthodontic appliances;

[0119] S312. grayscale the tooth image;

[0120] S313. Separating the outlines of teeth and orthodontic appliances from the background using an image segmentation method;

[0121] The noise reduction process in step S311 uses a Gaussian filter, and the formula is as follows:

[0122]

[0123] Where G(x,y) is the Gaussian function, (x,y) is the pixel coordinate, and σ is the standard deviation. Gaussian filtering can effectively remove noise from an image while maintaining the main features of the image.

[0124] After these preprocessing steps, the system performs image segmentation to separate the outlines of teeth and braces from the background. The system then uses the trained model to identify the locations of teeth and braces in the image, obtaining their position, size, and spatial relationship.

[0125] Step S3 further includes: S36. comparing the appliance image with the initial appliance image to obtain a deviation between the actual appliance position and the ideal appliance position;

[0126] S37. Compare the tooth image with the initial tooth image to obtain the deviation between the actual tooth position and the ideal tooth position;

[0127] S38. Analyze the deviation data to obtain the specific situation of tooth movement;

[0128] The degree of tooth movement is calculated using the following formula:

[0129]

[0130] Where D is the average tooth movement distance, (x i ,y i ,z i ) is the three-dimensional coordinate of the current tooth position, (x i ′,y i ′,z i ′) is the three-dimensional coordinate of the initial tooth position, and N is the total number of teeth. SSIM can better reflect the structural similarity of images and is more suitable for evaluating changes in oral images than the traditional mean square error (MSE).

[0131] The generation of the appliance adjustment plan in step S4 includes:

[0132] Provide appliance adjustment options when teeth are not moving as expected;

[0133] When the teeth move as expected, provide a treatment plan for the next visit;

[0134] The treatment progress was evaluated using the following formula:

[0135] Progress=(1-(∑|D i -D i ′| / ∑|D i ′|))*100%

[0136] Among them, Progress is the percentage of treatment progress, Di is the actual tooth movement distance, Di' is the expected tooth movement distance, and Σ represents the sum of all teeth;

[0137] For teeth whose actual movement does not match the expected movement, the appliance adjustment amount is calculated as follows:

[0138] Adjustment = k*(D i ′-D i )

[0139] Where Adjustment is the adjustment amount of the appliance, k is the adjustment coefficient (usually between 0.5 and 1.5), and D i ′ is the expected tooth movement distance, D i The actual tooth movement distance

[0140] After obtaining the tooth movement information, the system will generate a personal wearing monitoring report and an appliance adjustment plan. This step involves a complex data analysis and decision-making process; for teeth whose actual movement does not match the expected movement, the system will calculate the adjustment amount of the appliance:

[0141] Finally, the system will provide the doctor and patient with the wear information report and adjustment plan, which is crucial to ensure the continuity and effectiveness of treatment.

[0142] In general, the present invention provides a comprehensive, accurate, and automated invisible braces wearing monitoring and adjustment system and method. It can not only accurately evaluate the progress of treatment, but also provide personalized adjustment solutions, greatly improving the efficiency and effectiveness of orthodontic treatment. By combining computer vision and advanced data analysis methods, the present invention has brought revolutionary changes to the field of orthodontics, providing patients with a more comfortable and efficient treatment experience.

[0143] In order to verify the superiority of the present invention, we conducted a series of experiments to compare the effects of the embodiments of the present invention with those of the traditional method. The following is a detailed description of the embodiments and comparative examples, as well as the corresponding test results and analysis.

[0144] Example 1: In this example, we selected 100 patients who needed invisible braces treatment, with an age range of 18-45 years old. We used the invisible braces wearing monitoring and adjustment system based on computer vision of the present invention for treatment. According to the above method, the system regularly collects oral photos of patients, analyzes them through the cloud processing module, and generates personalized wearing monitoring reports and braces adjustment plans.

[0145] Comparative Example 1: In the comparative experiment, we also selected 100 patients who needed invisible braces treatment, whose age and treatment needs were similar to those in Example 1. This group of patients used the traditional invisible braces treatment method, that is, they went to the clinic regularly for manual inspection and adjustment.

[0146] We followed up these two groups of patients for 12 months, focusing on the following indicators:

[0147] 1. Treatment accuracy: assessing how well actual tooth movement matches expected movement.

[0148] 2. Treatment time: the time required to achieve the desired correction effect.

[0149] 3. Follow-up frequency: The number of times the patient needs to have a face-to-face follow-up visit.

[0150] 4. Patient satisfaction: the patient's satisfaction with the treatment process and results.

[0151] 5. Doctor’s work efficiency: the time the doctor spends on each case.

[0152] Testing methods and standards:

[0153] 1. Treatment accuracy: By comparing the 3D oral scan models before and after treatment, the deviation between actual tooth movement and expected movement is calculated. A deviation of less than 0.5mm is considered accurate.

[0154] 2. Treatment time: record the time from the start of treatment to achieving the expected correction effect.

[0155] 3. Follow-up frequency: Count the number of face-to-face follow-up visits by patients in the clinic within 12 months.

[0156] 4. Patient satisfaction: A scale of 1-10 was used to investigate the patient's satisfaction through a questionnaire.

[0157] 5. Doctor work efficiency: records the average time doctors spend on each case, including the time for examination, analysis and formulation of adjustment plans.

[0158] The test results are as follows:

[0159] index Embodiment 1 (present invention) Comparative Example 1 (Traditional Method) Treatment accuracy 92％ 78％ Average treatment time 10.5 months 13.2 months Average frequency of follow-up visits 4 times / year 12 times / year Patient satisfaction 8.7 / 10 7.2 / 10 Doctor work efficiency 15 minutes / case 35 minutes / case

[0160] The analysis and explanation are as follows:

[0161] 1. Treatment accuracy: The method of the present invention significantly improves the accuracy of treatment. This is mainly due to the application of computer vision. The system can accurately identify and quantify the movement of teeth, thereby providing more accurate adjustment suggestions. In contrast, traditional methods rely on the doctor's experience and judgment and are easily affected by subjective factors.

[0162] 2. Treatment time: The method of the present invention shortened the treatment time by an average of 2.7 months. This is because the system can detect and correct deviations in a timely manner, avoiding "ineffective" time during the treatment process. Traditional methods may require waiting until the next follow-up visit to discover the problem, resulting in longer treatment time.

[0163] 3. Frequency of follow-up visits: The method of the present invention greatly reduces the number of times patients need to visit the clinic in person, from 12 times a year to 4 times. This not only saves time and energy for patients, but also reduces the workload of clinics. The remote monitoring function of the system allows most adjustments to be completed through remote guidance.

[0164] 4. Patient satisfaction: The method of the present invention achieved higher patient satisfaction. This may be the result of many factors, including shorter treatment time, fewer follow-up visits, and the fact that patients can understand their treatment progress in real time through the app.

[0165] 5. Doctors’ work efficiency: The method of the present invention significantly improves doctors’ work efficiency, saving an average of 20 minutes per case. This is because the system automatically completes most of the analysis work, and doctors only need to review and confirm the plan generated by the system, which greatly saves time.

[0166] In general, the method of the present invention has shown obvious advantages in all aspects. It not only improves the accuracy and efficiency of treatment, but also improves the patient's treatment experience, while also improving the efficiency of medical resource utilization. These results fully demonstrate the innovation and practicality of the present invention in the field of invisible braces treatment.

[0167] It is particularly noteworthy that the method of the present invention not only improves the accuracy of treatment, but also significantly reduces the frequency of follow-up visits. This seemingly contradictory result precisely reflects the superiority of the system. It proves that remote monitoring based on computer vision can effectively replace some face-to-face follow-up visits without affecting the quality of treatment, but can provide more timely and accurate adjustments.

[0168] In addition, the significant improvement in doctors' work efficiency is also an important highlight. In the current context of tight medical resources, this improvement can enable doctors to serve more patients and help improve the accessibility of high-quality medical resources.

[0169] In summary, the present invention has brought revolutionary changes to invisible braces treatment, which not only optimizes the treatment process and improves the treatment effect, but also brings a better experience for both patients and doctors. This treatment model based on artificial intelligence and remote monitoring is likely to represent the future development direction of oral orthodontic treatment.

[0170] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A computer vision-based invisible braces wearing monitoring and adjustment system, characterized in that: include: Cloud storage module, used to store the patient's initial oral data and follow-up oral data; A cloud processing module that analyzes oral photos uploaded by patients and executes analysis software; The client module manages patient electronic medical records, provides treatment information, obtains analysis results, and generates personalized feedback; The appliance module is used to obtain the oral data of the patient's follow-up visit; The cloud processing module includes a computer vision module, which includes an image preprocessing submodule, an image segmentation and feature extraction submodule, a tooth recognition submodule and a tooth movement assessment submodule. The tooth movement assessment submodule calculates the degree of tooth movement using the following formula: M=∑(w i *||P i -P i ′||) / N Among them, M is the tooth movement degree score, w i is the weight factor of each tooth, P i is the current tooth position coordinate, P i ′ is the tooth position coordinate at the last follow-up visit, N is the total number of teeth evaluated, and ||·|| represents the Euclidean distance.

2. The system according to claim 1, characterized in that It also includes data upload module, data download module and data update module. Among them, the data upload module is used to collect and upload oral photos of patients for follow-up visits, the data download module is used to obtain treatment data, and the data update module is used to update the treatment data in the client module and the appliance module; The cloud storage module includes a storage server, a data interaction server and a data backup server, which are used to store patient data, respond to data access requests and back up data respectively.

3. The system according to claim 1, characterized in that The appliance module includes a plurality of pressure sensors and a main control installed on the surface of the appliance. The pressure sensor is used to collect the pressure value of the appliance acting on each tooth, and the main control is used to obtain the information collected by the appliance and upload it to the client module and the cloud processing module; The client module includes a personal interface module and a mobile tracking module, wherein the mobile tracking module uses the following formula to calculate the treatment progress: P=(M / M exp )*100% Among them, P is the percentage of treatment progress, M is the actual degree of tooth movement, and M exp The expected degree of tooth movement.

4. A method based on the system according to any one of claims 1 to 3, characterized in that: The following steps are involved: S1. Collect and upload the patient's initial oral data as the initial data set; S2. Collect oral photos during the patient's regular follow-up visits to form a follow-up photo collection; S3. Upload and analyze the follow-up photo collection to obtain tooth movement information; S4. Generate personal wearing monitoring report and appliance adjustment plan; S5. Feedback the wearing information report and adjustment plan to doctors and patients; Wherein, step S3 includes preprocessing, segmenting, feature extraction and tooth recognition of the follow-up photos, and calculating the degree of tooth movement using the formula in step S3.

5. The method according to claim 4, characterized in that The step S1 further comprises: S11. Separate the facial area where the teeth and the orthodontic appliance are located from the background to form an original image set; S12. extracting multiple sub-images based on the original image set, marking them as tooth areas, orthodontic appliance areas, and irrelevant areas; S13. Divide the data set into a training set and a test set in a ratio of 8:2, train the model and evaluate it. The model evaluation uses the following formula to calculate the accuracy: Accuracy=(TP+TN) / (TP+TN+FP+FN) Among them, TP is true positive, TN is true negative, FP is false positive, and FN is false negative.

6. The method according to claim 4, characterized in that The pre-processing in step S3 includes: S31. Extracting the contours of teeth and orthodontic appliances in the image; S32. Correct the patient's oral position and eliminate the inaccurate relative position between teeth; S33. Using the trained model to identify the positions of teeth and orthodontic appliances in the image; S34. Obtain the position, size and spatial relationship of teeth and appliances; S35. Generate a visual image of teeth and orthodontic appliances; Among them, step S31 uses the adaptive threshold method to binarize the image, and the formula is as follows: T(x,y)=μ(x,y)*(1+k*((σ(x,y) / R)-1)) Where T(x,y) is the threshold of pixel (x,y), μ(x,y) is the average grayscale value of the neighborhood of (x,y), σ(x,y) is the standard deviation of the neighborhood of (x,y), R is the grayscale dynamic range of the image, and k is the weight coefficient.

7. The method according to claim 6, characterized in that The step S31 further includes: S311. Enhance or reduce noise on the original image of teeth and orthodontic appliances; S312. grayscale the tooth image; S313. Separating the outlines of teeth and orthodontic appliances from the background using an image segmentation method; The noise reduction process in step S311 uses a Gaussian filter, and the formula is as follows: Among them, G(x,y) is the Gaussian function, (x,y) is the pixel coordinate, and σ is the standard deviation.

8. The method according to claim 4, characterized in that The step S3 further comprises: S36. Compare the appliance image with the initial appliance image to obtain a deviation between the actual appliance position and the ideal appliance position; S37. Compare the tooth image with the initial tooth image to obtain the deviation between the actual tooth position and the ideal tooth position; S38. Analyze the deviation data to obtain the specific situation of tooth movement; The degree of tooth movement is calculated using the following formula: Where D is the average tooth movement distance, (x i ,y i ,z i ) is the three-dimensional coordinate of the current tooth position, (x i ′,y i ′,z i ′) is the three-dimensional coordinate of the initial tooth position, and N is the total number of teeth.

9. The method according to claim 8, characterized in that The image comparison in steps S36 and S37 uses the structural similarity index (SSIM), and the formula is as follows: Among them, μ x and μ y are the average values ​​of images x and y, respectively, x and σ y are the standard deviations of images x and y, σ x y is the covariance of x and y, and C1 and C2 are constants used to avoid the denominator being zero.

10. The method according to claim 4, characterized in that The generation of the appliance adjustment plan in step S4 includes: Provide appliance adjustment options when teeth are not moving as expected; When the teeth move as expected, provide a treatment plan for the next visit; The treatment progress was evaluated using the following formula: Progress=(1-(∑|D i -D i ′| / ∑|D i ′|))*100% Among them, Progress is the percentage of treatment progress, Di is the actual tooth movement distance, Di' is the expected tooth movement distance, and Σ represents the sum of all teeth; For teeth whose actual movement does not match the expected movement, the appliance adjustment amount is calculated as follows: Adjustment=k*(D i ′-D i ) Where Adjustment is the adjustment amount of the appliance, k is the adjustment coefficient (usually between 0.5 and 1.5), and D i ′ is the expected tooth movement distance, D i is the actual tooth movement distance.

Images