CN114519380A - Drug administration method and device based on privacy protection and image recognition - Google Patents

Drug administration method and device based on privacy protection and image recognition Download PDF

Info

Publication number
CN114519380A
CN114519380A CN202111664120.8A CN202111664120A CN114519380A CN 114519380 A CN114519380 A CN 114519380A CN 202111664120 A CN202111664120 A CN 202111664120A CN 114519380 A CN114519380 A CN 114519380A
Authority
CN
China
Prior art keywords
oral
drug
patient
drug delivery
oral cavity
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111664120.8A
Other languages
Chinese (zh)
Inventor
熊常春
李海良
王敬贵
李国元
刘昂
吴江川
张富耕
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Jilian Technology Co ltd
Original Assignee
Shenzhen Jilian Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Jilian Technology Co ltd filed Critical Shenzhen Jilian Technology Co ltd
Priority to CN202111664120.8A priority Critical patent/CN114519380A/en
Publication of CN114519380A publication Critical patent/CN114519380A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
    • A61M11/00—Sprayers or atomisers specially adapted for therapeutic purposes
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
    • A61M31/00—Devices for introducing or retaining media, e.g. remedies, in cavities of the body
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/62—Protecting access to data via a platform, e.g. using keys or access control rules
    • G06F21/6218—Protecting access to data via a platform, e.g. using keys or access control rules to a system of files or objects, e.g. local or distributed file system or database
    • G06F21/6245—Protecting personal data, e.g. for financial or medical purposes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/02—Reservations, e.g. for tickets, services or events
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/18—Status alarms
    • G08B21/24—Reminder alarms, e.g. anti-loss alarms
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/10—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
    • G16H40/67—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Public Health (AREA)
  • Business, Economics & Management (AREA)
  • Biomedical Technology (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Medical Informatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Evolutionary Biology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Anesthesiology (AREA)
  • Software Systems (AREA)
  • Hematology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Bioethics (AREA)
  • Veterinary Medicine (AREA)
  • Computer Hardware Design (AREA)
  • Medicinal Chemistry (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Computing Systems (AREA)
  • Databases & Information Systems (AREA)
  • Mathematical Physics (AREA)
  • Computer Security & Cryptography (AREA)
  • Chemical & Material Sciences (AREA)
  • Molecular Biology (AREA)

Abstract

The application provides a drug administration method and device based on privacy protection and image recognition, comprising the following steps: after the patient visits a doctor and purchases the medicine, the doctor orders of the patient input the specified dosage, the medicine name, the oral disease name and the personal diagnosis information into the central processing unit; collecting the oral lesion images of the patients through a hospital, a treatment institution and a drug delivery device at multiple terminals, clustering and carrying out privacy protection according to the number of cases; training an oral positioning model through federal learning for the partition positioning of oral lesions; training a disease analysis model for determining the dosage; initializing the drug delivery device; administering a drug to a patient using the drug delivery device, wherein uploading the image of the oral lesion via a collection comprises; the drug administration device establishes a patient drug administration project, and carries out expiration reminding, expiration locking and device multiplexing; the drug delivery device is a handheld spray drug delivery device which meets the drug delivery requirements of the oral cavity of a human body and comprises a drug storage chamber, an image acquisition and analysis device, a drug delivery probe and an information interconnection device.

Description

Drug administration method and device based on privacy protection and image recognition
Technical Field
The invention relates to the technical field of medical intelligent equipment, in particular to a drug administration method and device based on privacy protection and image recognition.
Background
When patients spray and administer medicines for oral diseases, the patients often face the conditions of inaccurate administration position, incapability of adjusting the administration amount according to the state of illness, overdue medicines and the like. Image recognition techniques have been practiced in the medical field, and can be adjusted for drug delivery by adding a recognition device to the drug delivery device to identify the location of the lesion and the severity of the condition. Due to individual differences of the oral conditions of patients, the recognition effect of the same illness state may have larger differences, an information sharing and transmitting function can be added to the drug administration device, the recognition function is further trained through federal learning based on oral image samples from various patients, the problem of privacy protection of the patients is involved, and the problem needs to be solved by a technical scheme. Meanwhile, aiming at the problem of overdue of the spray administration medicine, the administration management can be carried out by combining the information transmission function of the administration device with a block chain and an intelligent contract technology, so that overdue of the oral spray medicine and abuse of the medicine are prevented.
Disclosure of Invention
The present invention is directed to solving the above-mentioned problems and to providing a medication administration method based on privacy protection and image recognition.
Another object of the present invention is to provide an oral drug delivery device based on privacy protection and image recognition.
The purpose of the invention can be achieved by adopting the following technical scheme: after the patient visits a doctor and purchases the medicine, the doctor orders of the patient input the specified dosage, the medicine name, the oral disease name and the personal diagnosis information into the central processing unit; the patient oral cavity focus images are collected through the hospital side, the treatment mechanism and the drug delivery device at multiple terminals, clustered and subjected to privacy protection according to the number of cases; training an oral positioning model through federal learning for the partition positioning of oral lesions; training a disease analysis model for determining the dosage; initializing the drug delivery device; the patient uses the drug delivery device to deliver drugs, and the oral lesion image is uploaded through a collection; the medication administration device establishes patient medication administration items, performs expiration reminding, expiration locking and device reuse.
Further optionally, after the patient visits and purchases the medication, the patient orders a prescribed dosage, a medication name, an oral disease name, and personal visit information to enter into the central processing device, including:
the oral disease names, included in the following oral disease categories: caries, oral mucosa diseases, and periodontal diseases. The prescribed dosage of the medical advice refers to the dosage suggested by the doctor during the treatment, and if the doctor does not give the suggested dosage, the prescribed dosage of the medicine instruction book is set. The personal visit information refers to the age, sex, prescription time and other auxiliary information of the patient. The other auxiliary information refers to the past medical history, the cause of illness, and the stage of illness of the patient, which are acquired from the medical record of the patient's visit by a text analysis method, for example: preference is given to spicy foods and sweets, endodontic root canal therapy has been done, etc. The central processing unit is deployed in a hospital or treatment institution information center and can acquire, arrange, store and transmit the prescribed dosage of the medical advice of the patient, the name of the medicine, the name of the oral disease and the personal treatment information data. Further optionally, the collecting, by the hospital, the treatment institution, and the administration device, the oral lesion images of the patient at multiple terminals, clustering, and performing privacy protection according to the number of cases includes: the oral lesion image can be acquired by patient consent of the academies and the treatment institutions, or acquired by the drug administration device. Clustering the oral lesion images based on the oral disease name, the personal encounter information, and the other auxiliary information.
For example: according to the name of the oral disease, a patient suffering from periodontal disease is classified into two groups according to the age of 65 years, which are mainly due to the fact that the oral conditions of the elderly and the youngsters are different.
After clustering is stable, counting the number of cases of each cluster, sequencing the cases in sequence from low to high, performing privacy protection according to sequencing, adding noise to the personal diagnosis information and the other auxiliary information, and distributing more privacy budgets to the clusters with less amount so as to ensure that the clusters with less sample amount can provide more available information during model training. The number of cases refers to the number of images of the oral lesion included in each cluster. The privacy protection means that random noise is used to ensure that the result of inquiring the request to disclose the visible information does not reveal the privacy information of an individual, namely, the individual characteristics are removed on the premise of keeping the statistical characteristics of the database to protect the privacy of the user; the privacy protection technology adopted by the invention is a local differential privacy protection based on a Laplace mechanism, and mainly protects structured information, and comprises the following steps: the greater the privacy budget of the personal visit information and the other auxiliary information, the lower the level of protection of the information, and the more information is available.
Further optionally, the training of the oral localization model through federal learning for the zonal localization of oral lesions comprises: the oral cavity positioning model is used for recognizing the position of a focus in the oral cavity, and the specific training mode comprises the following steps: 1) Partitioning the internal structure of the oral cavity, namely matching the positions of each organ of the oral cavity to a plurality of oral cavity partitions; for example, teeth can be classified according to the clinical dentition, and divided into 4 areas, namely, upper right, upper left, lower right, and the like, the upper right is divided into a first area, namely, the upper right middle incisor is 11 teeth, the upper right second molar is 17 teeth, the upper left is divided into a second area, the upper left middle incisor is 21 teeth, and the like;
2) marking the oral cavity subareas and the oral cavity disease names by the pre-collected oral cavity focus images, and performing machine learning model training;
3) the obtained oral cavity positioning model can judge whether the image contains oral disease focus based on the oral cavity image newly collected by the drug delivery device, and quickly judge the oral cavity subarea where the focus is located; 4) The images of the oral lesions of a patient who is treated by the same doctor are collected and trained uniformly; meanwhile, in order to avoid the diagnosis difference of different doctors, the model trained at multiple terminals belonging to each doctor is uploaded to a federal learning server, parameters are returned after updating, and the model is updated at each terminal.
The federated learning is horizontal federated learning, namely all data are distributed to different machines, each machine downloads a model from a server, then the model is trained by utilizing local data, and then parameters which need to be updated are returned to the server; the server aggregates the returned parameters on each machine, updates the model, and feeds back the latest model to each machine. Further optionally, the training of the disease analysis model for dose determination comprises: pre-training a disease analysis model based on image recognition technology, and determining the clustering of oral diseases of the patient based on the oral lesion image newly acquired by the administration device. And determining the specified dosage of the different disease conditions corresponding to the cluster as the dosage of the current administration. The disease stage refers to the severity of the oral disease, including the size and duration of the lesion. The dosage specified by different illness states refers to the dosage of the medicine determined by clinical tests under different illness state severity degrees and physical conditions of patients on the basis of the dosage specified by the medical advice. Further optionally, the initializing the medication delivery device comprises: the initialization means that the central processing unit inputs the prescribed dosage of the medical advice, the name of the medicine, the name of the oral disease, the prescribed dosage of different illness states and the personal clinic information data into the drug administration device; simultaneously embedding the oral cavity positioning model and the disease analysis model into the drug delivery device. Further optionally, the patient administers a drug using the drug delivery device, the step of uploading the image of the oral lesion via the collection comprising: the patient uses the drug delivery device to deliver drugs, and the drug delivery device identifies the position of the focus in the oral cavity through the oral cavity positioning model and the disease condition analysis model and delivers the drugs according to the drug dosage regulated by different disease conditions or the drug dosage regulated by different disease conditions. According to the number of the clustered cases, the central processing device can also carry out the oral lesion image collection, namely, a collection pushing device is sent to the drug administration device of the patient with the clustering condition of less cases so as to expand the sample size; patient's accessible device of dosing uploads oral cavity focus image, supplies the model study after privacy protection, includes: 1) The drug delivery device can collect the oral lesion image and the personal diagnosis information of the patient after the patient agrees; 2) The oral cavity focus image is subjected to image segmentation, and non-oral cavity facial organs are subjected to Gaussian blur processing to prevent the privacy of a patient from being revealed; 3) Removing the identification information when uploading the personal treatment information, for example: name, telephone, date of visit;
and (3) adopting a reward mechanism for the patient who voluntarily uploads the image after collection, namely matching the original doctor by the drug administration device from the date of uploading to obtain a remote appointment consultation for the patient who uploads the image of the oral lesion after collection.
Further optionally, the drug administration device establishes a patient drug administration project, and the performing expiration reminding, expiration locking, and device reusing include:
because oral spray administration often faces the condition of drug expiration, the administration condition can be monitored and the device reuse can be realized through the administration device and the block chain intelligent contract function.
And the intelligent contract can be used for recording the drug name, the drug delivery time and the drug valid period from the drug delivery date of the drug delivery device and establishing the drug delivery item.
The intelligent contract can establish the drug administration items for each drug administration device in a blockchain for management; and meanwhile, time calculation is carried out according to the drug valid period and the drug delivery time, and when the time after the drug delivery device is delivered exceeds the drug valid period, expiration reminding is carried out or expiration locking is carried out on the drug delivery device through the drug delivery device.
The drug expiration date, including the instruction-determined drug expiration date, and the recommended length of use by the order, such as: some drugs may cause pharyngeal discomfort over time, and thus physicians may recommend the duration of use. The expiration lock, i.e. no further dosing, is performed by the dosing device for the drug expiration period as determined by the instructions; the medication delivery device will make the expiration reminder for the drug expiration date recommended by the order, but the patient may still choose to continue using.
After the expiration reminder and the expiration lock, the patient can send an application to the hospital through the drug delivery device, and if the patient needs to continue using the drug or replacing other drugs through diagnosis of a doctor, device cleaning and drug supplementation can be carried out in the hospital, namely, the drug delivery device can be continuously reused.
When other medicines are used, the relevant parameters of the new medicine need to be set again, and a new administration item is established.
The other purpose of the invention can be achieved by adopting the following technical scheme:
a drug administration method and device based on privacy protection and image recognition are characterized in that the device comprises:
the drug delivery device is a handheld spray drug delivery device which meets the drug delivery requirements of the oral cavity of a human body and comprises a drug storage chamber, an image acquisition and analysis device, a drug delivery probe and an information interconnection device. The medicine storage chamber is used for storing medicine liquid. The image acquisition and analysis module can acquire images in the oral cavity at a specified distance and direction in front of the oral cavity, comprises the oral cavity positioning model and the disease analysis model, and can compare the images, divide the oral cavity of the focus and judge the dosage. The administration probe is used for administering the medicine liquid with the administration amount to the focus oral cavity subarea, and specifically comprises: 1) The dosing probe can rotate according to the judgment result of the oral cavity positioning model, a macro camera is arranged at the top of the dosing probe for image acquisition and is transmitted back to the image acquisition and analysis module for image comparison until the orientation of the top of the dosing probe is the same as the direction of the focus oral cavity subarea; 2) After the dosage is determined, the administration probe sucks the liquid medicine of the dosage and carries out administration; the information interconnection module is positioned on the bottle body of the medicine storage chamber and mainly has the following functions:
a. for establishing said medication administration project, performing said due reminder, due lock, whereby the patient issues said device reuse application to the hospital;
b. the system is used for sending an acquisition pushing to a patient, so that the patient uploads the oral lesion image and the personal diagnosis information, and the oral lesion image and the personal diagnosis information are provided for model learning after privacy protection;
c. the system is used for recording the time and the times for uploading the oral cavity focus images and sending appointment push to an original doctor when a patient needs appointment consultation.
The technical scheme provided by the embodiment of the invention has the following beneficial effects: the method and the device provided by the invention can confirm the position of the focus of the patient oral cavity and the serious condition of the disease through image recognition, and carry out accurate administration based on preset administration amounts of different diseases; a federal learning method is adopted, and the clustering result is combined, so that the lacked oral cavity image samples are selectively collected from the target population; meanwhile, the clustering result also influences the privacy protection strength of different samples so as to ensure that the clustering of the number of small cases can present more effective information during model training. Meanwhile, the method and the device adopted by the invention are not only beneficial to expanding the training sample of the oral lesion image, but also beneficial to establishing a reward mechanism for the patient who uploads the image voluntarily, and can establish a drug administration management project, thereby preventing the spray type drugs from being overdue and abused and promoting the device to be reused.
Fig. 1 is a flowchart of a drug administration method based on privacy protection and image recognition according to the present invention. Fig. 2 is a block diagram of an oral administration device based on privacy protection and image recognition according to the present invention.
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention is described in detail below with reference to the accompanying drawings and specific examples. Fig. 1 is a flowchart of a drug administration method based on privacy protection and image recognition according to the present invention. As shown in fig. 1, a medication administration method based on privacy protection and image recognition in this embodiment may specifically include: s1: after the patient visits and purchases the medicine, the doctor orders the prescribed dosage, the medicine name, the oral disease name and the personal visiting information are input into the central processing unit.
The oral disease names, included in the following oral disease categories: caries, oral mucosa diseases, and periodontal diseases. The prescribed dosage of the medical advice refers to the dosage suggested by the doctor during the treatment, and if the doctor does not give the suggested dosage, the prescribed dosage of the medicine instruction book is set. The personal visit information refers to the age, sex, prescription time and other auxiliary information of the patient. The other auxiliary information refers to the past medical history, the cause of illness, and the stage of illness of the patient, which are acquired from the medical record of the patient's visit by a text analysis method, for example: preference is given to spicy foods and sweets, endodontic root canal therapy has been done, etc. The central processing unit is arranged in a hospital or a treatment institution information center and can acquire, arrange, store and transmit the medical advice prescribed dosage, the medicine name, the oral disease name and the personal treatment information data of the patient. S2: through institute side and treatment institution, the device multi-terminal collection patient oral cavity focus image of dosing, the clustering carries out privacy protection according to the case quantity and includes: the oral lesion image can be acquired by the hospital and the treatment institution after patient consent, or acquired by the drug administration device. Clustering the oral lesion images based on the oral disease name, the personal encounter information, and the other auxiliary information.
For example: according to the name of the oral disease, a patient with periodontal disease is classified into two groups when the patient is 65 years old and the patient is 17 years old and has periodontal disease, which is mainly due to the fact that the oral conditions of the elderly and the youngsters are greatly different.
After clustering is stable, counting the number of cases of each cluster, sequencing the cases in sequence from low to high, performing privacy protection according to sequencing, adding noise to the personal diagnosis information and the other auxiliary information, and distributing more privacy budgets to the clusters with less amount so as to ensure that the clusters with less sample amount can provide more available information during model training. The number of cases refers to the number of images of the oral lesion included in each cluster. The privacy protection means that random noise is used to ensure that the result of inquiring the request to disclose the visible information does not reveal the privacy information of an individual, namely, the individual characteristics are removed on the premise of keeping the statistical characteristics of the database to protect the privacy of the user; the privacy protection technology adopted by the invention is a local differential privacy protection based on a Laplace mechanism, and mainly protects structured information, and comprises the following steps: the larger the privacy budget of the personal doctor information and the other auxiliary information is, the lower the protection level of the information is, and the more the available information is.
S3: training an oral positioning model through federal learning, wherein the oral lesion regional positioning model comprises the following steps: the oral cavity positioning model is used for recognizing the position of a focus in the oral cavity, and the specific training mode comprises the following steps: 1) Partitioning the internal structure of the oral cavity, namely matching the positions of each organ of the oral cavity to a plurality of oral cavity partitions; for example, teeth can be classified according to the clinical dentition, and divided into 4 areas, namely, upper right, upper left, lower right, and the like, the upper right is divided into a first area, namely, the upper right middle incisor is 11 teeth, the upper right second molar is 17 teeth, the upper left is divided into a second area, the upper left middle incisor is 21 teeth, and the like;
2) marking the oral cavity subareas and the oral cavity disease names by the pre-collected oral cavity focus images, and performing machine learning model training;
3) the obtained oral cavity positioning model can judge whether the image contains oral disease focus based on the oral cavity image newly collected by the drug delivery device, and quickly judge the oral cavity subarea where the focus is located; 4) The images of the oral lesions of a patient who is treated by the same doctor are collected and trained uniformly; meanwhile, in order to avoid the diagnosis difference of different doctors, the model trained at multiple terminals belonging to each doctor is uploaded to a federal learning server, parameters are returned after updating, and the model is updated at each terminal.
The federated learning is horizontal federated learning, namely all data are distributed to different machines, each machine downloads a model from a server, then the model is trained by utilizing local data, and then parameters which need to be updated are returned to the server; the server aggregates the returned parameters on each machine, updates the model, and feeds back the latest model to each machine. S4: training a disease analysis model for dosing quantity determination comprising: pre-training a disease analysis model based on image recognition technology, and determining the clustering of oral diseases of the patient based on the oral lesion image newly acquired by the administration device. And determining the specified dosage of the different disease conditions corresponding to the cluster as the dosage of the current administration. The disease stage refers to the severity of the oral disease, including the size and duration of the lesion. The dosage specified by different illness states refers to the dosage of the medicine determined by clinical tests under different illness state severity degrees and physical conditions of patients on the basis of the dosage specified by the medical advice. S5: initializing the drug delivery device comprises: the initialization means that the central processing unit inputs the prescribed dosage of the medical advice, the name of the medicine, the name of the oral disease, the prescribed dosage of different illness states and the personal clinic information data into the drug administration device; simultaneously embedding the oral cavity positioning model and the disease analysis model into the drug delivery device. S6: the patient administers the drug using the drug delivery device, and the uploading of the image of the oral lesion via the collection comprises: the patient uses the drug delivery device to deliver drugs, and the drug delivery device identifies the position of the focus in the oral cavity through the oral cavity positioning model and the disease condition analysis model and delivers the drugs based on the different disease condition specified drug dosage or the different disease condition specified drug dosage. According to the number of the clustered cases, the central processing device can also perform the oral cavity focus image collection, namely, a collection pushing is sent to the drug delivery device of the patient with the clustering condition of less number of the cases so as to expand the sample size; patient's accessible device of dosing uploads oral cavity focus image, supplies the model study after privacy protection, includes: 1) The drug delivery device can collect the oral lesion image and the personal diagnosis information of the patient after the patient agrees; 2) The oral cavity focus image is subjected to image segmentation, and non-oral cavity facial organs are subjected to Gaussian blur processing to prevent the privacy of a patient from being revealed; 3) Removing the identification information when uploading the personal treatment information, for example: name, telephone, date of visit;
and (3) adopting a reward mechanism for the patient who voluntarily uploads the image after collection, namely matching the original doctor by the drug administration device from the date of uploading to obtain a remote appointment consultation for the patient who uploads the image of the oral lesion after collection.
S7: the medication administration device establishes patient medication administration items, performs expiration reminding, expiration locking and device reuse.
Because oral spray administration is often subjected to the condition of drug expiration, the administration condition can be monitored and the device reuse can be realized by the administration device in combination with the block chain intelligent contract function.
And the intelligent contract can be used for recording the drug name, the drug delivery time and the drug valid period from the drug delivery date of the drug delivery device and establishing the drug delivery item.
The intelligent contract can establish the drug administration items for each drug administration device in a blockchain for management; and meanwhile, time calculation is carried out according to the drug valid period and the drug delivery time, and when the time after the drug delivery device is delivered exceeds the drug valid period, expiration reminding is carried out or expiration locking is carried out on the drug delivery device through the drug delivery device.
The drug expiration date, including the instruction-determined drug expiration date, and the recommended length of use by the order, such as: some drugs may cause pharyngeal discomfort over time, and thus physicians may recommend the duration of use. The expiration lock, i.e. no further dosing, is performed by the dosing device for the drug expiration period as determined by the instructions; the medication delivery device will make the expiration reminder for the drug expiration date recommended by the order, but the patient may still choose to continue using.
After the expiration reminder and the expiration lock, the patient can send an application to the hospital through the drug delivery device, and if the patient needs to continue using the drug or replacing other drugs through diagnosis of a doctor, device cleaning and drug supplementation can be carried out in the hospital, namely, the drug delivery device can be continuously reused.
When other medicines are used, the related parameters of the new medicine need to be set again, and a new administration item is established.
Fig. 2 is a block diagram of an oral administration device based on privacy protection and image recognition according to the present invention. As shown in fig. 2, the oral administration device based on privacy protection and image recognition of the present embodiment may specifically include:
the drug delivery device is a handheld spray drug delivery device which meets the drug delivery requirements of the oral cavity of a human body and comprises a drug storage chamber, an image acquisition and analysis device, a drug delivery probe and an information interconnection device. The medicine storage chamber is used for storing medicine liquid. The image acquisition and analysis module can acquire the internal image of the oral cavity in the specified distance and direction in front of the oral cavity, comprises the oral cavity positioning model and the disease analysis model, and can perform image comparison, partition of the oral cavity of the focus and judgment of the dosage. The administration probe is used for administering the medicine liquid with the administration amount to the focus oral cavity subarea, and specifically comprises: 1) The dosing probe can rotate according to the judgment result of the oral cavity positioning model, a macro camera is arranged at the top of the dosing probe for image acquisition and is transmitted back to the image acquisition and analysis module for image comparison until the orientation of the top of the dosing probe is the same as the direction of the focus oral cavity subarea; 2) After the dosage is determined, the administration probe sucks the liquid medicine of the dosage and carries out administration; the information interconnection module is positioned on the bottle body of the medicine storage chamber and mainly has the following functions:
a. for establishing said medication administration item, performing said due reminder, due lock, whereby the patient issues said device reuse application to the hospital;
b. the system is used for sending an acquisition pushing to a patient, so that the patient uploads the oral lesion image and the personal diagnosis information, and the oral lesion image and the personal diagnosis information are provided for model learning after privacy protection;
c. the system is used for recording the time and the times for uploading the oral cavity focus images and sending appointment push to an original doctor when a patient needs appointment consultation.
Another embodiment of the present invention provides a device for performing federal learning in a drug administration method based on privacy protection and image recognition, which mainly includes: the system comprises an infrastructure module, a cloud network control cnc module and a Dapps application module:
the infrastructure module comprises a virtual host, a virtual storage, a virtual line and TEE trusted hardware, wherein the virtual host, the virtual storage, the virtual line and the TEE trusted hardware are from development cloud manufacturers, enterprise clouds and personal computers.
And the cloud network control cnc module is used for integrating the redundant computing power of all terminal nodes running on the infrastructure module and scheduling the computing power through a cloud network link bottom layer technology, and providing hardware computing support for the Dapps application module.
The infrastructure module provides hardware computing support for the Dapps application module through the cloud network control cnc module.
And the Dapps application module forms a hardware node for bearing operation through an open cloud platform, an enterprise cloud and a personal computer which are included by the infrastructure module, and the hardware node is combined to create a subnet. The subnet carries a software container on which a user uploads an interoperable computing unit containing code, state, privacy computing frameworks and algorithms, including but not limited to Federal Learning (FL) algorithms.
And (4) opening the federal learning service to the internet by combining an infrastructure module, an interconnection chain module, a cloud network control cnc module and a Dapps application module.
The method program of the invention is developed by a DevOps research, development, operation and maintenance integrated development tool.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention. Programs for implementing the information governance of the present invention may be written in computer program code for carrying out operations of the present invention in one or more programming languages, including an object oriented programming language such as Java, python, C + +, or a combination thereof, as well as conventional procedural programming languages, such as the C language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In the embodiments provided in the present invention, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described device embodiments are merely illustrative, and for example, the division of the units is only one logical functional division, and other divisions may be realized in practice. The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment. In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit. The integrated unit implemented in the form of a software functional unit may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute some steps of the methods according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

Claims (9)

1. A drug administration method and device based on privacy protection and image recognition are characterized in that the method comprises the following steps:
after the patient visits a doctor and purchases the medicine, the doctor orders of the patient input the specified dosage, the medicine name, the oral disease name and the personal diagnosis information into the central processing device; collecting the oral lesion images of the patients through hospital and treatment institutions and a drug delivery device at multiple terminals, clustering and carrying out privacy protection according to the number of cases; training an oral positioning model through federal learning for the regional positioning of oral lesions; training a disease analysis model for determining the dosage; initializing the drug delivery device; the patient uses the drug delivery device to deliver drugs, and the oral lesion image is uploaded through collection; the drug administration device establishes a patient drug administration project, and carries out expiration reminding, expiration locking and device multiplexing; the drug delivery device is a handheld spray drug delivery device which meets the drug delivery requirements of the oral cavity of a human body and comprises a drug storage chamber, an image acquisition and analysis device, a drug delivery probe and an information interconnection device.
2. The method of claim 1, wherein said patient orders a prescribed dose, medication name, oral disease name, personal visit information entry central processing device after visiting and purchasing a medication, comprising:
the oral disease names, included in the following oral disease categories: caries, oral mucosal diseases, periodontal diseases;
the prescribed dosage of the medical advice refers to the dosage suggested by a doctor during the treatment, and if the doctor does not give the suggested dosage, the prescribed dosage of the medicine specification is set;
the personal visit information refers to the age, sex, prescription time and other auxiliary information of the patient;
the other auxiliary information refers to the information of the past medical history, the disease causes and the disease stages of the patient, which is acquired from the medical record of the patient in the clinic by a text analysis method;
the central processing unit is deployed in a hospital or treatment institution information center and can acquire, arrange, store and transmit the prescribed dosage of the medical advice of the patient, the name of the medicine, the name of the oral disease and the personal treatment information data.
3. The method of claim 1, wherein the multi-terminal collection of the images of the oral lesions of the patient by the hospital and treatment institutions and the drug delivery device, clustering and privacy protection according to the number of cases comprises:
the oral lesion image can be acquired by the hospital and the treatment institution after patient consent, or acquired by the drug administration device; clustering the oral lesion images based on the oral disease name, the personal encounter information, and the other auxiliary information;
after clustering is stable, counting the number of cases of each cluster, sequencing the cases in sequence from low to high, performing privacy protection according to sequencing, adding noise to the personal diagnosis information and the other auxiliary information, and distributing more privacy budgets to the clusters with less amount so as to ensure that the clusters with less sample amount can provide more available information during model training; the number of cases refers to the number of images of the oral lesion included in each cluster; the privacy protection means that random noise is used to ensure that the result of inquiring the request to disclose the visible information does not reveal the privacy information of an individual, namely, the individual characteristics are removed on the premise of keeping the statistical characteristics of the database to protect the privacy of the user; the privacy protection technology adopted by the invention is a Laplace mechanism-based local differential privacy protection, mainly protects structured information, and comprises the following steps: the greater the privacy budget of the personal visit information and the other auxiliary information, the lower the level of protection of the information, and the more information is available.
4. The method of claim 1, wherein the training of the oral localization model by federal learning for oral lesion zoning comprises:
the oral cavity positioning model is used for recognizing the position of a focus in the oral cavity, and the specific training mode comprises the following steps:
1) partitioning the internal structure of the oral cavity, namely matching the positions of each organ of the oral cavity to a plurality of oral cavity partitions;
2) marking the oral cavity subareas and the oral cavity disease names by the pre-collected oral cavity focus images, and performing machine learning model training;
3) the obtained oral cavity positioning model can judge whether the image contains oral disease focus based on the oral cavity image newly collected by the drug delivery device, and quickly judge the oral cavity subarea where the focus is located;
4) the images of the oral lesions of a patient who is treated by the same doctor are collected and trained uniformly; meanwhile, in order to avoid the diagnosis difference of different doctors, the training model is trained at multiple terminals belonging to each doctor, the training parameters are uploaded to the federal learning server, the parameters are returned after updating, and the model is updated at each terminal.
5. The method of claim 1, wherein the training of the disease analysis model for dose determination comprises:
pre-training a disease analysis model based on an image recognition technique, wherein the obtained model can judge the clustering of oral diseases of the patient based on the oral lesion image newly acquired by the drug delivery device;
determining the specified dosage of different disease conditions corresponding to the cluster as the dosage of the current drug administration;
the disease stage refers to the severity of the oral disease, including the size and duration of the lesion;
the dosage specified by different illness states refers to the dosage of the medicine determined by clinical tests under different illness state severity degrees and physical conditions of patients on the basis of the dosage specified by the medical advice.
6. The method of claim 1, wherein the initializing the drug delivery device comprises:
the initialization means that the central processing unit inputs the prescribed dosage of the medical advice, the name of the medicine, the name of the oral disease, the prescribed dosage of different illness states and the personal clinic information data into the drug administration device; simultaneously embedding the oral cavity positioning model and the disease analysis model into the drug delivery device.
7. The method of claim 1, wherein the patient administers the drug using the drug delivery device, and the uploading of the image of the oral lesion via the collection comprises:
the patient uses the drug delivery device to deliver drugs, and the drug delivery device identifies the position of the focus in the oral cavity through the oral cavity positioning model and the disease condition analysis model and delivers the drugs based on the different disease condition specified drug dosage or the different disease condition specified drug dosage;
according to the number of the clustered cases, the central processing device can also carry out the oral lesion image collection, namely, a collection pushing device is sent to the drug administration device of the patient with the clustering condition of less cases so as to expand the sample size; patient's accessible the device of dosing uploads oral cavity focus image, supplies the model study after privacy protection, includes: 1) The drug delivery device can collect the oral lesion image and the personal diagnosis information of the patient after the patient agrees; 2) The oral cavity focus image is subjected to image segmentation, and non-oral cavity facial organs are subjected to Gaussian blur processing to prevent the privacy of a patient from being revealed; 3) Removing the identity identification information when the personal treatment information is uploaded;
a reward mechanism is adopted for the patient who voluntarily uploads the image after collection, namely, the patient who uploads the oral lesion image after collection can be matched with an original doctor through the drug administration device from the uploading date, and a remote appointment consultation is obtained.
8. The method of claim 1, wherein the medication administration device establishes patient medication administration items, expiration reminders, expiration locks, and device reuses, comprising:
because oral spray administration often faces the condition of drug overdue, the administration condition can be monitored by the administration device in combination with the intelligent contract function of a block chain, and the device reuse is realized;
the block chain is preset with an intelligent contract, and the intelligent contract can be used for inputting the drug name, the drug delivery time and the drug valid period from the drug delivery date of the drug delivery device and establishing the drug delivery item;
the intelligent contract can establish the drug administration items for each drug administration device in a blockchain for management; meanwhile, time calculation is carried out according to the drug valid period and the drug delivery time, and when the time after the drug delivery device is delivered exceeds the drug valid period, expiration reminding is carried out or expiration locking is carried out on the drug delivery device through the drug delivery device;
the drug valid period comprises a drug valid period determined by the instruction and a use time recommended by the medical order, and the drug administration device can perform expiration locking on the drug valid period determined by the instruction, namely, the drug administration can not be performed any more; the medication delivery device will make the expiration reminder for the drug expiration date recommended by the order, but the patient may still choose to continue using;
after the expiration reminder and the expiration lock, the patient can send an application to the hospital through the drug administration device, if the patient needs to continue using the drug or replacing other drugs through diagnosis of a doctor, device cleaning and drug supplementation can be carried out in the hospital, namely the drug administration device can be continuously reused;
when other medicines are used, the related parameters of the new medicine need to be set again, and a new administration item is established.
9. The method of claim 1, wherein the drug delivery device is a hand-held spray drug delivery device conforming to the drug delivery requirements of the oral cavity of the human body, and comprises a drug storage chamber, an image acquisition and analysis device, a drug delivery probe and an information interconnection device, and comprises:
the medicine storage chamber is used for storing medicine liquid;
the image acquisition and analysis module can acquire images in the oral cavity at a specified distance and direction in front of the oral cavity, comprises the oral cavity positioning model and the disease analysis model, and can compare the images, divide the oral cavity of the focus and judge the dosage;
the administration probe is used for administering the medicine liquid with the administration amount to the focus oral cavity subarea, and specifically comprises the following components:
1) the dosing probe can rotate according to the judgment result of the oral cavity positioning model, a macro camera is arranged at the top of the dosing probe for image acquisition and is transmitted back to the image acquisition and analysis module for image comparison until the orientation of the top of the dosing probe is the same as the direction of the focus oral cavity subarea;
2) after the dosage is determined, the administration probe sucks the liquid medicine of the dosage and carries out administration;
the information interconnection module is positioned on the bottle body of the medicine storage chamber and mainly has the following functions:
a. for establishing said medication administration project, performing said due reminder, due lock, whereby the patient issues said device reuse application to the hospital;
b. the system is used for sending an acquisition pushing to a patient, so that the patient uploads the oral lesion image and the personal diagnosis information, and the oral lesion image and the personal diagnosis information are provided for model learning after privacy protection;
c. the system is used for recording the time and the times for uploading the oral cavity focus images and sending appointment push to an original doctor when a patient needs appointment consultation.
CN202111664120.8A 2021-12-31 2021-12-31 Drug administration method and device based on privacy protection and image recognition Pending CN114519380A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111664120.8A CN114519380A (en) 2021-12-31 2021-12-31 Drug administration method and device based on privacy protection and image recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111664120.8A CN114519380A (en) 2021-12-31 2021-12-31 Drug administration method and device based on privacy protection and image recognition

Publications (1)

Publication Number Publication Date
CN114519380A true CN114519380A (en) 2022-05-20

Family

ID=81597566

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111664120.8A Pending CN114519380A (en) 2021-12-31 2021-12-31 Drug administration method and device based on privacy protection and image recognition

Country Status (1)

Country Link
CN (1) CN114519380A (en)

Similar Documents

Publication Publication Date Title
CN106529177B (en) A kind of patient's portrait method and device based on medical big data
WO2020123007A2 (en) Remotely-executed medical diagnosis and therapy including emergency automation
CN112700838A (en) Big data-based medication scheme recommendation method and device and related equipment
Caplehorn et al. Physicians' attitudes and retention of patients in their methadone maintenance programs
US20100211407A1 (en) Method and system for providing a patient therapeutic plan
CN109273098B (en) Medicine curative effect prediction method and device based on intelligent decision
KR20040053269A (en) Prescription fulfillment system and method
JPWO2019244949A1 (en) Biometric information processing methods, biometric information processing devices, and biometric information processing systems
Kruse et al. Healthcare system effects of pay-for-performance for smoking status documentation
CN107526933A (en) A kind of Telemedicine System
TW201913549A (en) Prescription management system and method
US20080319784A1 (en) Online health consultation
WO2023029505A1 (en) Prescription review method and apparatus, electronic device, and storage medium
KR20170022007A (en) System and computer readable recording medium for management health information
CN112085466B (en) Device and method for generating damage assessment result of human injury claim
WO2021253867A1 (en) Framework, method, and system for unmanned smart hospital
US11694249B2 (en) Prepaid bundled health, dental, and veterinary services with virtual payment distribution
CN114519380A (en) Drug administration method and device based on privacy protection and image recognition
CN117059279A (en) Medical service platform, method and device based on virtual digital human body model
Rho et al. Development common data model for adverse drug signal detection based on multi-center emr systems
CN113470769A (en) Medical information integration system and use method thereof
US20160162643A1 (en) Method and apparatus for creating and updating medical records
Langebrake et al. Patient-individual 3D-printing of drugs within a machine-learning-assisted closed-loop medication management–Design and first results of a feasibility study
CN113658659B (en) Medical information processing method, medical information processing device, computer equipment and storage medium
US11830052B2 (en) Prepaid bundled health, dental, and veterinary services with virtual payment distribution

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination