Patient privacy protection method for intelligent mouth opening trainer

Abstract

The invention discloses a patient privacy protection method for an intelligent mouth opening trainer, and the method comprises the following steps: obtaining training data of a patient from the intelligent mouth opening trainer, and generating a corresponding management background link, namely a hyperlink character string; symmetrically encrypting the original link according to a preset key and an advanced encryption standard (AES) algorithm to obtain an encrypted link character string; a traditional intelligent mouth opening training device records patient data, but due to the fact that a privacy protection mechanism is lacked, the data is prone to being leaked, and potential safety hazards exist, robust watermarking and an information encryption technology are combined, dual protection is provided for patient privacy, information encryption is conducted, an advanced encryption standard (AES) algorithm is adopted to encrypt real information of a patient, and the real information of the patient is encrypted; original information is converted into an encrypted character string which cannot be directly understood, and even if the two-dimensional code is leaked, an unauthorized user cannot acquire real data of a patient.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent mouth opening trainers, and in particular relates to a patient privacy protection method for intelligent mouth opening trainers. Background Art

[0002] Smart healthcare aims to gradually realize informatization by creating a medical information platform to connect patients with doctors, medical institutions with medical equipment, and adopt the most advanced Internet of Things technology. At present, smart healthcare has been applied to many areas of medical services. In the field of maxillofacial diseases, limited mouth opening is a common clinical symptom, which affects patients' eating, speaking, oral hygiene, leads to facial deformities, and even complicates multiple oral diseases.

[0003] Traditional mouth opening training tools include wedge-shaped wooden plugs, duckbill-shaped metal retainers, etc. However, since the patient's training data is their personal privacy and there is a risk of leakage, these mouth opening trainers do not have the function of recording the patient's training data, making it impossible to monitor the patient's prognosis in real time. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the existing defects and provide a patient privacy protection method for an intelligent mouth opening trainer to solve the problem that the traditional mouth opening training tools proposed in the above background technology include wedge-shaped wooden plugs, duckbill-shaped metal retainers, etc., but because the patient's training data belongs to their personal privacy and there is a risk of leakage, these mouth opening trainers do not have the function of recording the patient's training data, and thus cannot monitor the patient's prognosis in real time.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for protecting patient privacy for an intelligent mouth opening trainer, comprising the following steps:

[0006] Step 1: Obtain the patient's training data from the intelligent mouth opening trainer and generate a corresponding management backend link, i.e., a hyperlink string;

[0007] Step 2: According to the preset key and Advanced Encryption Standard (AES) algorithm, the original link is symmetrically encrypted to obtain an encrypted link string;

[0008] Step 3: Input the encrypted link string into the QR code generation module to generate a QR code image containing the encrypted link;

[0009] Step 4: Convert the key into a cyclic complement code and input it into the digital watermark embedding module;

[0010] Step 5: The digital watermark embedding module uses a model based on generative adversarial network (GAN) training to embed the key cyclic complement into the QR code image as an invisible watermark to generate an encrypted QR code image with watermark information;

[0011] Step 6: The authorized party uses the camera of the mobile device to capture the encrypted QR code image with watermark information;

[0012] Step 7: The digital watermark parsing module parses the hidden key cyclic complement code from the captured encrypted QR code image;

[0013] Step 8: Convert the parsed key cyclic complement code into a key;

[0014] Step 9: Use the key and AES algorithm to decrypt the encrypted link and restore the original link string;

[0015] Step 10: Access the original link through a browser or other application to obtain patient training data.

[0016] Preferably, in step 2, the key is generated by an authorized party and kept confidential, that is, a randomly generated string of characters or a predetermined password.

[0017] Preferably, in step 4, the length of the key cyclic complement is adjusted according to the capacity of the QR code and the required security level.

[0018] Preferably, in step five, the model of the digital watermark embedding module adopts a U-Net structure, which can learn encoding from multiple scales to improve the robustness of the digital watermark.

[0019] Preferably, in step five, the digital watermark embedding module splices the key cyclic complement code and the QR code image along the RGB channel dimension, and generates a final QR code image with watermark information through an encoder.

[0020] Preferably, in step six, the mobile device includes a smart phone and a tablet computer.

[0021] Preferably, the model of the digital watermark analysis module in step seven is composed of a multi-layer convolutional neural network and a two-layer multi-layer perceptron (MLP), which can extract the hidden key cyclic complement from the distorted QR code image.

[0022] Preferably, in step seven, the distortion processing includes motion blur, out-of-focus blur, perspective deformation, color adjustment, JPEG compression, illumination change and noise, simulating interference factors encountered during camera shooting.

[0023] Preferably, in step nine, the decryption process uses a decryption mode of an AES algorithm to restore the encrypted link string to an original link string.

[0024] Preferably, in step ten, the browser or other application can display detailed information of the patient's training data, including mouth opening, bite force and training time, and support data analysis and chart display functions.

[0025] Compared with the prior art, the present invention provides a patient privacy protection method for an intelligent mouth opening trainer, which has the following beneficial effects:

[0026] 1. Traditional intelligent mouth opening trainers record patient data, but due to the lack of privacy protection mechanism, the data is easily leaked, which poses a security risk. The present invention provides double protection for patient privacy by combining robust watermarking with information encryption technology. Information encryption: The patient's real information is encrypted using the Advanced Encryption Standard (AES) algorithm, and the original information is converted into an encrypted string that cannot be directly understood. Even if the QR code is leaked, unauthorized users cannot obtain the patient's real data;

[0027] Robust watermark: The key required for decryption is embedded in the QR code in the form of an invisible watermark; the watermark information is invisible to the human eye, but is robust to camera shooting. Even if the QR code is blurred, deformed, compressed, etc., the watermark information can still be correctly extracted; after the authorized user (such as a doctor) scans the QR code, the watermark information is first extracted as the key, and then the encrypted data is decrypted to obtain the patient's real data;

[0028] 2. The present invention provides a convenient and efficient way to collect and manage patient data, without the need for cumbersome data transmission, and is convenient for patients and doctors to use; Intelligent mouth opening trainer: The device has a built-in sensor that can record the patient's training data in real time, including mouth opening, bite force and training time; Exclusive QR code: Generate an exclusive QR code for each patient, and print it on the mouth opening trainer used by the patient; Remote medical treatment: The patient only needs to take a photo of the trainer with the QR code with a mobile phone and upload the photo to the doctor; The doctor can scan the QR code to instantly obtain the patient's data on the mobile phone and conduct remote diagnosis and treatment;

[0029] 3. By acquiring patient training data in real time, the present invention allows doctors to have a more comprehensive understanding of the patient's recovery status, thereby formulating more accurate and personalized diagnosis and treatment plans and improving treatment effects; Real-time data monitoring: Doctors can check the patient's training data at any time, understand the patient's recovery progress and training effects, and adjust the treatment plan based on the data; Personalized guidance: According to the different recovery conditions of patients, doctors can provide personalized training guidance and rehabilitation suggestions to help patients recover faster. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0031] Figure 1 A schematic diagram of the use of an intelligent mouth opening trainer used in a patient privacy protection method for an intelligent mouth opening trainer proposed by the present invention;

[0032] Figure 2 A schematic diagram of a user scenario for patient privacy protection involved in a patient privacy protection method for an intelligent mouth opening trainer proposed in the present invention;

[0033] Figure 3 A schematic diagram of an embodiment of a method for protecting patient privacy of an intelligent mouth opening trainer proposed by the present invention;

[0034] Figure 4 This is a schematic diagram of a digital watermark model of a patient privacy protection method for an intelligent mouth opening trainer proposed in the present invention. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] See also Figure 1-Figure 4 The present invention provides a technical solution: a method for protecting patient privacy for an intelligent mouth opening trainer, comprising the following steps:

[0037] Step 1: Obtain the patient's training data from the intelligent mouth opening trainer and generate a corresponding management backend link, i.e., a hyperlink string;

[0038] Step 2: According to the preset key and Advanced Encryption Standard (AES) algorithm, the original link is symmetrically encrypted to obtain an encrypted link string;

[0039] Step 3: Input the encrypted link string into the QR code generation module to generate a QR code image containing the encrypted link;

[0040] Step 4: Convert the key into a cyclic complement code and input it into the digital watermark embedding module;

[0041] Step 5: The digital watermark embedding module uses a model based on generative adversarial network (GAN) training to embed the key cyclic complement into the QR code image as an invisible watermark to generate an encrypted QR code image with watermark information;

[0042] Step 6: The authorized party uses the camera of the mobile device to capture the encrypted QR code image with watermark information;

[0043] Step 7: The digital watermark parsing module parses the hidden key cyclic complement code from the captured encrypted QR code image;

[0044] Step 8: Convert the parsed key cyclic complement code into a key;

[0045] Step 9: Use the key and AES algorithm to decrypt the encrypted link and restore the original link string;

[0046] Step 10: Access the original link through a browser or other application to obtain patient training data.

[0047] In the present invention, preferably, in step 2, the key is generated by the authorized party and kept confidential, that is, a randomly generated string of characters or a predetermined password.

[0048] In the present invention, preferably, in step 4, the length of the key cyclic complement is adjusted according to the capacity of the QR code and the required security level.

[0049] In the present invention, preferably, in step five, the model of the digital watermark embedding module adopts a U-Net structure, which can learn encoding from multiple scales and improve the robustness of the digital watermark.

[0050] In the present invention, preferably, in step five, the digital watermark embedding module splices the key cyclic complement code and the QR code image along the RGB channel dimension, and generates a final QR code image with watermark information through an encoder.

[0051] In the present invention, preferably, in step six, the mobile device includes a smart phone and a tablet computer.

[0052] In the present invention, preferably, the model of the digital watermark parsing module in step seven is composed of a multi-layer convolutional neural network and a two-layer multi-layer perceptron (MLP), which can extract the hidden key cyclic complement from the distorted QR code image.

[0053] In the present invention, preferably, in step seven, the distortion processing includes motion blur, out-of-focus blur, perspective deformation, color adjustment, JPEG compression, illumination change and noise, simulating interference factors encountered during camera shooting.

[0054] In the present invention, preferably, in step nine, the decryption process uses the decryption mode of the AES algorithm to restore the encrypted link string to the original link string.

[0055] In the present invention, preferably, in step ten, the browser or other application can display detailed information of the patient's training data, including mouth opening, bite force and training time, and support data analysis and chart display functions.

[0056] Embodiment 1

[0057] A patient privacy protection method for an intelligent mouth opening trainer comprises the following steps:

[0058] Step 1: Obtain the patient's training data from the intelligent mouth opening trainer and generate the corresponding management backend link;

[0059] Step 2: Use the AES algorithm to encrypt the original link and obtain the encrypted link string;

[0060] Step 3: Input the encrypted link string into the QR code generation module to generate a QR code image containing the encrypted link;

[0061] Step 4: Convert the key into a cyclic complement code and input it into the digital watermark embedding module;

[0062] Step 5: The digital watermark embedding module uses a model based on the U-Net structure to embed the key cyclic complement code into the QR code image as an invisible watermark to generate an encrypted QR code image with watermark information;

[0063] Step 6: The authorized party uses the camera of the mobile device to capture the encrypted QR code image with watermark information;

[0064] Step 7: The digital watermark parsing module parses the hidden key cyclic complement code from the captured encrypted QR code image;

[0065] Step 8: Convert the parsed key cyclic complement code into a key;

[0066] Step 9: Use the key and AES algorithm to decrypt the encrypted link and restore the original link string;

[0067] Step 10: Access the original link through a browser or other application to obtain patient training data.

[0068] Digital watermark embedding module based on U-Net structure: By using the U-Net structure, the watermark embedding module of this embodiment can learn encoding from multiple scales to improve the robustness of the watermark. Even after the QR code is distorted, the watermark information can still be correctly extracted, thereby effectively protecting the privacy of patients.

[0069] Embodiment 2

[0070] A patient privacy protection method for an intelligent mouth opening trainer comprises the following steps:

[0071] Step 1: Obtain the patient's training data from the intelligent mouth opening trainer and generate the corresponding management backend link;

[0072] Step 2: Use the AES algorithm to encrypt the original link and obtain the encrypted link string;

[0073] Step 3: Input the encrypted link string into the QR code generation module to generate a QR code image containing the encrypted link;

[0074] Step 4: Convert the key into a cyclic complement code and input it into the digital watermark embedding module;

[0075] Step 5: The digital watermark embedding module uses a model based on a multi-layer perceptron (MLP) to embed the key cyclic complement code into the QR code image as an invisible watermark to generate an encrypted QR code image with watermark information;

[0076] Step 6: The authorized party uses the camera of the mobile device to capture the encrypted QR code image with watermark information;

[0077] Step 7: The digital watermark parsing module parses the hidden key cyclic complement code from the captured encrypted QR code image;

[0078] Step 8: Convert the parsed key cyclic complement code into a key;

[0079] Step 9: Use the key and AES algorithm to decrypt the encrypted link and restore the original link string;

[0080] Step 10: Access the original link through a browser or other application to obtain patient training data.

[0081] Watermark extraction module based on multi-layer perceptron: By using a multi-layer perceptron (MLP) structure, the watermark extraction module of this embodiment can learn complex nonlinear relationships and improve the accuracy of watermark extraction. Even when there is noise or other interference in the QR code image, the hidden key can still be accurately extracted, thereby achieving secure patient data access.

[0082] Embodiment 3

[0083] A patient privacy protection method for an intelligent mouth opening trainer comprises the following steps:

[0084] Step 1: Obtain the patient's training data from the intelligent mouth opening trainer and generate the corresponding management backend link;

[0085] Step 2: Use the AES algorithm to encrypt the original link and obtain the encrypted link string;

[0086] Step 3: Input the encrypted link string into the QR code generation module to generate a QR code image containing the encrypted link;

[0087] Step 4: Convert the key into a cyclic complement code and input it into the digital watermark embedding module;

[0088] Step 5: The digital watermark embedding module uses a model based on the generative adversarial network (GAN) to embed the key cyclic complement into the QR code image as an invisible watermark to generate an encrypted QR code image with watermark information;

[0089] Step 6: The authorized party uses the camera of the mobile device to capture the encrypted QR code image with watermark information;

[0090] Step 7: The digital watermark parsing module parses the hidden key cyclic complement code from the captured encrypted QR code image;

[0091] Step 8: Convert the parsed key cyclic complement code into a key;

[0092] Step 9: Use the key and AES algorithm to decrypt the encrypted link and restore the original link string;

[0093] Step 10: Access the original link through a browser or other application to obtain patient training data.

[0094] Invisible watermark embedding based on GAN: By using a model based on generative adversarial networks (GAN), the watermark embedding module of this embodiment can generate a QR code image with a watermark. The watermark information is invisible to the human eye, but can effectively resist various attacks such as tampering and copying, thereby ensuring the security and integrity of patient data.

[0095] Embodiment 4

[0096] A patient privacy protection method for an intelligent mouth opening trainer comprises the following steps:

[0097] Step 1: Obtain the patient's training data from the intelligent mouth opening trainer and generate the corresponding management backend link;

[0098] Step 2: Use the AES algorithm to encrypt the original link and obtain the encrypted link string;

[0099] Step 3: Input the encrypted link string into the QR code generation module to generate a QR code image containing the encrypted link;

[0100] Step 4: Convert the key into a cyclic complement code and input it into the digital watermark embedding module;

[0101] Step 5: A JND term is added to the loss function of the digital watermark embedding module to control the distortion within the maximum distortion range that cannot be detected by the human eye, and generate an encrypted QR code image with watermark information;

[0102] Step 6: The authorized party uses the camera of the mobile device to capture the encrypted QR code image with watermark information;

[0103] Step 7: The digital watermark parsing module parses the hidden key cyclic complement code from the captured encrypted QR code image;

[0104] Step 8: Convert the parsed key cyclic complement code into a key;

[0105] Step 9: Use the key and AES algorithm to decrypt the encrypted link and restore the original link string;

[0106] Step 10: Access the original link through a browser or other application to obtain patient training data.

[0107] Distortion control based on JND: By adding the JND term to the loss function, the watermark embedding module of this embodiment can control the image distortion caused by watermark embedding and control the distortion within the maximum distortion range that is imperceptible to the human eye, thereby ensuring the clarity and aesthetics of the QR code image without affecting the robustness of the watermark.

[0108] Embodiment 5

[0109] A patient privacy protection method for an intelligent mouth opening trainer comprises the following steps:

[0110] Step 1: Obtain the patient's training data from the intelligent mouth opening trainer and generate the corresponding management backend link;

[0111] Step 2: Use the AES algorithm to encrypt the original link and obtain the encrypted link string;

[0112] Step 3: Input the encrypted link string into the QR code generation module to generate a QR code image containing the encrypted link;

[0113] Step 4: Convert the key into a cyclic complement code and input it into the digital watermark embedding module;

[0114] Step 5: The digital watermark embedding module embeds the key cyclic complement code into the QR code image as an invisible watermark to generate an encrypted QR code image with watermark information;

[0115] Step 6: The authorized party uses the camera of the mobile device to capture the encrypted QR code image with watermark information;

[0116] Step 7: The captured encrypted QR code image is distorted, including motion blur, out-of-focus blur, perspective deformation, color adjustment, JPEG compression, lighting change and noise;

[0117] Step 8: The digital watermark parsing module parses the hidden key cyclic complement code from the distorted encrypted QR code image;

[0118] Step 9: Convert the parsed key cyclic complement code into a key;

[0119] Step 10: Use the key and AES algorithm to decrypt the encrypted link and restore the original link string;

[0120] Step 11: Access the original link through a browser or other application to obtain patient training data.

[0121] Digital watermark parsing based on distortion processing: By performing distortion processing on the captured encrypted QR code image, the watermark parsing module of this embodiment can simulate the interference factors encountered during camera shooting and learn how to extract hidden information from noisy QR codes, thereby improving the robustness of watermark parsing. Even in the case of severe distortion of the QR code image, the hidden key can still be accurately extracted.

[0122] The present invention proposes a patient privacy protection method for an intelligent mouth opening trainer. By combining robust digital watermarking and information encryption technology, the problem of privacy information leakage faced by the intelligent mouth opening trainer when recording patient training data is effectively solved. The method first uses the Advanced Encryption Standard (AES) algorithm to encrypt the patient's real information, and encodes the encrypted information into a QR code. At the same time, the key required for decryption is hidden in the generated QR code in the form of an invisible watermark. After the authorized party (such as a doctor) scans the QR code, the watermark information is first extracted as a key, and then the encrypted data is decrypted, so that the patient information can be correctly accessed. For unauthorized parties, by scanning the code with a mobile phone, only the encrypted information can be obtained, which has no practical value. The present invention can effectively protect patient privacy without affecting the data collection of patients and doctors, ensuring that the patient's private training data will not be easily stolen by an unauthorized party, thereby solving the worries for the real-time management of patient data.

[0123] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for protecting patient privacy for an intelligent mouth opening trainer, characterized in that: The following steps are involved: Step 1: Obtain the patient's training data from the intelligent mouth opening trainer and generate a corresponding management backend link, i.e., a hyperlink string; Step 2: According to the preset key and Advanced Encryption Standard algorithm, the original link is symmetrically encrypted to obtain an encrypted link string; Step 3: Input the encrypted link string into the QR code generation module to generate a QR code image containing the encrypted link; Step 4: Convert the key into a cyclic complement code and input it into the digital watermark embedding module; Step 5: The digital watermark embedding module uses a model based on generative adversarial network training to embed the key cyclic complement into the QR code image as an invisible watermark to generate an encrypted QR code image with watermark information; Step 6: The authorized party uses the camera of the mobile device to capture the encrypted QR code image with watermark information; Step 7: The digital watermark parsing module parses the hidden key cyclic complement code from the captured encrypted QR code image; Step 8: Convert the parsed key cyclic complement code into a key; Step 9: Use the key and AES algorithm to decrypt the encrypted link and restore the original link string; Step 10: Access the original link through a browser or other application to obtain patient training data.

2. A patient privacy protection method for an intelligent mouth opening trainer according to claim 1, characterized in that: In the step 2, the key is generated by the authorized party and kept confidential, that is, a randomly generated string of characters or a predetermined password.

3. A method for protecting patient privacy for an intelligent mouth opening trainer according to claim 1, characterized in that: In the step 4, the length of the key cyclic complement is adjusted according to the capacity of the QR code and the required security level.

4. A method for protecting patient privacy for an intelligent mouth opening trainer according to claim 1, characterized in that: In the step 5, the model of the digital watermark embedding module adopts a U-Net structure, which can learn encoding from multiple scales and improve the robustness of the digital watermark.

5. A method for protecting patient privacy for an intelligent mouth opening trainer according to claim 1, characterized in that: In the step 5, the digital watermark embedding module splices the key cyclic complement code and the QR code image along the RGB channel dimension, and generates a final QR code image with watermark information through an encoder.

6. A method for protecting patient privacy for an intelligent mouth opening trainer according to claim 1, characterized in that: In step six, the mobile devices include smart phones and tablet computers.

7. A method for protecting patient privacy for an intelligent mouth opening trainer according to claim 1, characterized in that: The model of the digital watermark analysis module in step seven is composed of a multi-layer convolutional neural network and a two-layer multi-layer perceptron, and can extract the hidden key cyclic complement from the distorted QR code image.

8. A method for protecting patient privacy for an intelligent mouth opening trainer according to claim 1, characterized in that: In step seven, the distortion processing includes motion blur, out-of-focus blur, perspective deformation, color adjustment, JPEG compression, illumination change and noise, simulating interference factors encountered during camera shooting.

9. A method for protecting patient privacy for an intelligent mouth opening trainer according to claim 1, characterized in that: In step nine, the decryption process uses the decryption mode of the AES algorithm to restore the encrypted link string to the original link string.

10. A method for protecting patient privacy for an intelligent mouth opening trainer according to claim 1, characterized in that: In step ten, the browser or other application can display detailed information of the patient's training data, including mouth opening, bite force and training time, and support data analysis and chart display functions.

Images