TWI863061B - Video assistance system with wound identification, method and computer readable medium thereof - Google Patents

Abstract

The present invention is a video assistance system with wound identification and a method thereof which provide a user to use a video device for remote video consultation or advisory with a doctor after identity verification. During image recognition, an identifiable object is placed in the guide area by a user, and the identifiable object is recognized and corresponding guide lines is generated automatically by a first image recognition module. Then, the user can focus on the wound according to guide lines, and image recognition can be performed and wound measurement data can be displayed by a second image recognition module, and the wound measurement data can be uploaded to a cloud server for storage. In addition, hashing calculation can be used for ensuring the correctness of each file in the cloud server, and can be served as a judgment basis for automatic update. The present invention also provides a computer-readable medium for executing the method of the present invention.

Description

Wound identification video assistance system, method and computer-readable medium thereof

The present invention relates to image recognition and video processing technology, and more particularly to a wound recognition video assistance system, method and computer-readable medium thereof.

Taiwan began to enter an aging society in 1993 and turned into an aging society in 2018. As people age and have chronic diseases, long-term care becomes very important. Long-term care issues include difficult-to-care skin wounds and treatments that are difficult to heal, as well as collisions during treatment and tearing injuries caused by the removal of dressings. In clinical practice, patients often worry about how to get to the hospital for treatment from remote areas, or are afraid of changing wound dressings after returning home, or receiving regular wound care guidance.

The current remote care system is composed of video systems, where doctors or medical staff interact with patients through video, but lack actual observed wound values and analysis, which may lead to misjudgment by doctors or medical staff. For example, in the current wound image recognition mechanism, the overall recognition rate often decreases due to the object to be recognized in the image being skewed or too far away from the camera. In addition, there is also the possibility that the image cannot be recognized smoothly due to the large number of patients, so the image recognition system needs to be updated frequently.

From this, it can be seen that how to provide a technology related to image recognition and video processing, especially how to accurately identify wounds during remote care, to assist doctors or medical staff to identify the wound condition through video, thereby reducing the occurrence of misdiagnosis of the disease, will become a goal that people in this technical field are eager to pursue.

In order to solve the above problems of the prior art, the present invention discloses a wound recognition video assistance system, which includes: an image capture module, used to capture a first wound image from a video acquired by a user's video device; a data storage module, used to store a plurality of guide wire templates; a first image recognition module, used to recognize the first wound image, and when a recognizable object is found in the first wound image, obtain an appropriate object from the data storage module. The guide line template of the first wound image is combined to generate a guide line; the second image recognition module is used to capture the second wound image from the adjusted video with the aid of the guide line to perform recognition and analysis of the second wound image; and the image display module is used to display the recognition result of the second image recognition module, so that after the user confirms the recognition result, the recognition result is uploaded to the cloud server.

In one embodiment, the data storage module further includes storing the first wound image, the second wound image, and temporarily storing the identified measurement values.

In one embodiment, the wound recognition video assistance system further includes a wireless communication module for connecting to the cloud server to upload the recognition results, download new image recognition modules and guide line templates, and communicate with the cloud server.

In one embodiment, the video device includes: a video module for acquiring a video signal of the user's wound; and a display module for displaying the signal value measured by the video module.

In one embodiment, the cloud server includes: a data storage module for storing the physiological measurement values and files of the identification results uploaded by the user; a download module for providing new image recognition modules and guide wire templates for use as module or data updates; a security module for providing various encryption algorithms and hashing algorithms to encrypt and hash various files of the wound identification video assistance system or the cloud server; and a blockchain module for recording the hash code after the hashing operation of the security module.

In addition, the present invention discloses a wound identification video-assisted method, which is executed by a computer device and includes the following steps: enabling an image capture module to capture a first wound image from a video acquired by a user's video device; enabling a first image recognition module to recognize the first wound image, and when a recognizable object is found in the first wound image, obtaining an appropriate object from a storage data module storing a plurality of guide wire templates; The guide line template of the first wound image is combined to generate a guide line; the second image recognition module is enabled to capture the second wound image from the adjusted video with the aid of the guide line by the video device to perform recognition and analysis of the second wound image; and the image display module is enabled to display the recognition result of the second image recognition module, so that after the user confirms the recognition result, the recognition result is uploaded to the cloud server.

In the above method, the data storage module further includes storing the first wound image, the second wound image and temporarily storing the identified measurement values.

In the above method, the step of obtaining the video from the user's video device further includes: allowing the video module of the video device to obtain the video signal of the user's wound; and allowing the display module of the video device to display the signal value measured by the video module.

In the above method, it further includes: allowing the data storage module of the cloud server to store the physiological measurement values and files of the identification results uploaded by the user.

In the above method, it further includes: allowing the security module of the cloud server to provide various encryption algorithms and hashing algorithms to encrypt and hash various files; and allowing the blockchain module of the cloud server to record the hash code after the hashing operation of the security module.

In addition, the above-mentioned wound identification video assistance method further includes: allowing the download module of the cloud server to provide a new image recognition module and guide line template for use as an update of the module or data.

The present invention further discloses a computer-readable medium for use in a computing device or a computer, which stores instructions for executing the aforementioned wound identification video-assisted method.

In summary, the wound identification video assistance system, method and computer-readable medium of the present invention provide users with the ability to use a video device to conduct a remote video consultation or consultation with a doctor or medical staff. Through the video device, the user places a designated object in the guide area, and the video wound image captured by the image capture module will be recognized by the first image recognition module to automatically generate a corresponding guide line. The user then focuses on the wound according to the guidance of the guide line. After completing the relevant operations, the second image recognition module performs image recognition and displays the recognized wound measurement data. Finally, the data is uploaded to the cloud server for storage through the wireless communication module. In addition, the wound identification video assistance system can automatically download the latest module and guide wire template from the cloud server to update the first image recognition module, the second image recognition module and the guide wire template on the system side. In order to ensure the accuracy of each file in the cloud server, the file hashing calculation is performed through the security module, and the hash code is recorded in the blockchain module in the cloud server to ensure the integrity and accuracy of the files stored in the cloud server. In summary, the present invention allows users to automatically download the latest module and generate wound measurement guide lines through the image recognition module trained by deep learning, so as to ensure that users take images correctly and improve the success rate of image recognition. Finally, the relevant wound measurement data is uploaded to the cloud server for users to query and use later.

1: Wound identification video assistance system

11: Image capture module

12: Data storage module

13: First image recognition module

14: Second image recognition module

15: Image display module

16: Wireless communication module

2: Video device

21: Video module

22: Display module

3: Cloud Server

31: Data storage module

32: Download module

33: Confidentiality module

34: Blockchain module

401-412: Process

501-506: Process

601-603: Process

S301-S304: Steps

Figure 1 is a system architecture diagram of the wound identification video assistance system of the present invention.

Figure 2 is a system architecture diagram of another embodiment of the wound identification video assistance system of the present invention.

Figure 3 is a step diagram of the video-assisted wound identification method of the present invention.

Figure 4 is a flowchart of the image recognition of the present invention.

Figure 5 is a flowchart of the present invention for automatically verifying and updating various files.

Figure 6 is a flowchart of uploading files in the cloud server of the present invention.

The following describes the technical content of the present invention through a specific concrete implementation form. People familiar with this technology can easily understand the advantages and effects of the present invention from the content disclosed in this manual. However, the present invention can also be implemented or applied through other different specific implementation forms.

In order to improve the needs of remote medical treatment and wound identification, the present invention proposes a wound identification video assistance system based on deep learning to improve the recognition rate. It can reduce the recognition error factor caused by human operation and guide the user to correctly place the wound display screen in front of the image capture module. Then, the data is uploaded to the cloud server for storage via the Internet. This can avoid information misjudgment or image recognition omissions caused by human operation. In addition, through real-time video wound measurement, it can assist doctors or medical staff to obtain accurate wound analysis, which can serve as a basis for doctors or medical staff to judge the wound.

FIG1 is a system architecture diagram of the wound identification video assistance system of the present invention. As shown in the figure, the wound identification video assistance system 1 of the present invention includes an image capture module 11, a data storage module 12, a first image recognition module 13, a second image recognition module 14 and an image display module 15.

The image capture module 11 is used to capture the first wound image from the video obtained by the video device 2 of the user or patient. In short, the image capture module 11 is responsible for receiving the wound image from the video device 2 and performing image capture. Specifically, the user or patient can use the video device 2 he owns to shoot the wound, and the image capture module 11 captures the first wound image from the video obtained by the video device 2.

The data storage module 12 is used to store multiple guide wire templates. In short, in order to allow the wound image to meet the diagnostic requirements when captured, the present invention uses guide wires to allow the user or patient to place the wound in a suitable position, and the data storage module 12 pre-stores various guide wire templates.

The first image recognition module 13 is used to recognize the first wound image, so as to obtain a guide line template suitable for the first wound image from the storage data module 12 to generate a guide line when finding a recognizable object in the first wound image. In short, the first image recognition module 13 is responsible for processing the first image from the image capture module 11. Since the user or patient will adjust the position of the wound for shooting, the position relationship of the wound in the wound image must be confirmed first. By inserting a recognizable object into the video first, the first image recognition module 13 analyzes the video and obtains the position of the recognizable object, and then obtains the position and distance relationship between the lens of the video device 2 and the wound of the user or patient. In this way, a suitable guide line can be provided, that is, the guide line template suitable for the first wound image is obtained from the data storage module 12, and the guide line is generated and displayed on the image display module 15, so that the user or patient can adjust the shooting position of the wound according to the guide line.

The second image recognition module 14 is used to extract the second wound image from the adjusted video with the help of the guide line of the video device 2 to identify and analyze the second wound image. The second image recognition module 14 is responsible for processing and obtaining relevant values from the image. Specifically, the first image recognition module 13 will find a suitable guide line to guide the user or patient to shoot the wound, and the second image recognition module 14 will extract the second wound image from the new video after the video device 2 adjusts the shooting position to perform judgment and analysis, thereby obtaining relevant data of the second wound image.

The image display module 15 is used to display the recognition result of the second image recognition module 14, so that after the user or patient confirms the recognition result, the recognition result is uploaded to the cloud server 3. In short, the image display module 15 is responsible for processing the image captured by the image capture module 11, displaying the guide wire frame, and displaying a screen for the user or patient to confirm the wound image. Therefore, after the second image recognition module 14 obtains the second wound image, the image display module 15 displays the wound image for the user or patient to confirm whether the recognition result of the second image recognition module 14 is correct. If correct, the recognition result of the second wound image (i.e., image and data) is uploaded to the cloud server 3.

In specific implementation, the wound identification video assistance system 1 can be installed in a system host including various computing modules and image capture modules. The system host can be, for example, a smart mobile device, various computers, etc.

In addition, in addition to the guide wire template, the storage data module 12 can also store the first wound image, the second wound image, and the temporarily recognized measurement values. In short, the relevant data in the wound recognition video assistance system 1 can be stored in the storage data module 12. In addition, because the cloud server 3 will provide the wound recognition video assistance system 1 with relevant data and updates, such as the first image recognition module, the second image recognition module, and the guide wire template files, the file data transmitted by the cloud server 3 is also stored in the storage data module 12.

As can be seen from the above, the image obtained by the image capture module 11 of the present invention is detected and recognized by the first image recognition module 13. When the image recognition is completed, the guide line template corresponding to the recognition result will be obtained from the storage data module 12, and the frame of this guide line template will be presented in the image display module 15, so that the effect of dynamic guide line can be achieved. In addition, the second image recognition module 14 of the present invention can simultaneously identify a variety of physiological measurement devices (i.e., video device 2), and based on the deep learning method, the server administrator can upload the latest image recognition module in real time to respond to various wound conditions. Among them, the second image recognition module 14 is mainly responsible for identifying the image in the guide wire. The second image recognition module 14 can analyze the wound in the image, and after the recognition is completed, the value is displayed on the image display module 15. The user or patient can confirm again whether the digital value of the recognition result is correct, and after confirmation, it can be uploaded to the cloud server 3 for storage.

FIG2 is a system architecture diagram of another embodiment of the wound identification video assistance system of the present invention. As shown in the figure, the image capture module 11, the data storage module 12, the first image recognition module 13, the second image recognition module 14 and the image display module 15 are the same as those shown in FIG1 and will not be described in detail here. In this embodiment, the video device 2 and the internal modules of the cloud server 3 will be further described. In addition, the wound identification video assistance system 1 of the present invention further includes a wireless communication module 16.

The video device 2 includes a video module 21 for obtaining the video signal of the wound of the user or patient and a display module 22 for displaying the signal value measured by the video module 21. In short, the video device 2 shoots images through its lens, the video module 21 is responsible for receiving the video signal of the video device 2, and the display module 22 is responsible for displaying the signal value measured by the video module 21, such as a liquid crystal display screen that can display data.

In one embodiment, the video device 2 may be a smart mobile device or a computer, and has an image capture function (such as a camera) and an image display function (such as a screen).

Compared to the wound identification video assistance system 1 and the video device 2 which are user-side or patient-side devices, the cloud server 3 is a non-local device and can be connected to the wound identification video assistance system 1 through the network. The cloud server 3 includes a data storage module 31, a download module 32, a confidentiality module 33 and a blockchain module 34.

The data storage module 31 is used to store the physiological measurement values and files of the identification results uploaded by the user or patient. In short, the data storage module 31 is responsible for managing the data uploaded by the user or patient and storing various new versions of files, that is, storing the file data uploaded by the wound identification video assistance system 1, or saving the new version of the file on the cloud server 3.

The download module 32 is used to provide new image recognition modules and guide wire templates for module or data updates. In short, the download module 32 is responsible for providing the wound recognition video assistance system 1 with the latest version of the first image recognition module, the second image recognition module and the guide wire template. In short, if the wound recognition video assistance system 1 finds that its internal image recognition or guide wire template is not the latest version, it can obtain the new version of the image recognition module and guide wire template from the download module 32 in the cloud server 3.

The security module 33 is used to provide various encryption algorithms and hashing algorithms to encrypt and hash various files of the wound identification video assistance system 1 or the cloud server 3. In other words, the security module 33 is responsible for providing confidentiality calculation or hashing operation for files or data of the wound identification video assistance system 1 and the cloud server 3. For example, a hashing operation is performed on a new file to be stored in the cloud server 3 to obtain a hash code of the new file. In addition, a hashing operation can also be performed on the file of the wound identification video assistance system 1 to ensure data security by comparing the hash codes of the previous and the next. The difference in the hash code can also be used to determine whether the version needs to be updated.

The blockchain module 34 is used to record the hash code after the hash operation of the security module 33. In other words, the blockchain module 34 is responsible for recording the hash code of the latest file, that is, storing the hash code calculated by the security module 33 for use in subsequent version comparisons. For example, after the hash calculation of the previous module version, if the hash codes stored on the local end and the cloud server 3 end are different, it means that the file versions on both ends are different.

The wireless communication module 16 is used to connect to the cloud server 3 to upload the recognition result, download the new image recognition module and guide line template, and communicate with the cloud server 3. In short, the wound recognition video assistance system 1 is connected to the cloud server 3 through the wireless communication module 16. The wireless communication module 16 can upload the recognition result of the second image recognition module 14, and can also download the new image recognition module and guide line template, that is, download the new image recognition module and guide line template from the cloud server 3. In summary, the wireless communication module 16 is responsible for downloading the first image recognition module, the second image recognition module, the guide wire template, uploading the user's physiological measurement data, querying the pass code, and using the functions of the security module 33 in the cloud server 3, that is, the wireless communication module 16 is responsible for the communication between the wound recognition video assistance system 1 and the cloud server 3.

In another embodiment, the cloud server 3 can perform relevant hash calculations on the files uploaded by the administrator through the security module 33, and then record the hash code in the blockchain module 34. Because the blockchain in the blockchain module 34 has the characteristics of being unalterable and irreversible, the integrity and accuracy of the files stored in the cloud server 3 can be ensured.

Furthermore, in the automatic downloading process of the present invention, in order to ensure that the current wound identification video assistance system 1 is a legitimate user, a pass code mechanism can be set for verification. This pass code can be used to determine whether the current wound identification video assistance system 1 can download the latest file. After the automatic update download program is started, the wound recognition video assistance system 1 will obtain the hash code of the relevant file from the blockchain module 34 of the cloud server 3. Then, the wound recognition video assistance system 1 uses the security module 33 of the cloud server 3 through the wireless communication module 16 to calculate the hash code of the relevant file on the local side (i.e., the wound recognition video assistance system 1). By using the irreversible characteristics of the hash code, it is possible to immediately detect whether the local file is abnormal or not the latest version. If a related abnormal state is found, the file will be updated to reduce the recognition rate of the wound recognition video assistance system 1, i.e., the recognition error. In addition, the hashing algorithm used by the security module 33 in the cloud server 3 must have collision resistance to prevent local files from being tampered with without being discovered.

In summary, the user or patient can take images through the video device 2 and connect to the wound identification video assistance system 1 for image identification and analysis. The image capture module 11 is used to capture the current image, the first image recognition module 13 is responsible for object recognition of the video device 2, the second image recognition module 14 is responsible for measurement data recognition of physiological measurements, the image display module 15 is responsible for displaying images and user or patient operation screens, and the storage data module 12 is responsible for storing the first image recognition module 13, the second image module 14, the guide wire template and the temporary The wireless communication module 16 is responsible for uploading measurement data, downloading new image recognition modules and templates, and communicating with the security module 33 and blockchain module 34 in the cloud server 3. In addition, in the cloud server 3, the data storage module 31 is responsible for storing physiological measurement values and various files uploaded by users, the download module 32 is responsible for providing the wound recognition video assistance system 1 for updating programs, the security module 33 is responsible for providing various encryption algorithms and hashing algorithms, and the blockchain module 34 is responsible for recording the hash codes of various files.

FIG3 is a step diagram of the wound identification video-assisted method of the present invention. The method described in this embodiment can be executed in a computer device, that is, by building a wound identification video-assisted system to assist in wound identification and data transmission with a cloud server.

In step S301, the image capture module is instructed to capture the first wound image from the video acquired by the video device of the user or patient. This step is to illustrate that the user or patient uses the video device to take a picture of the wound, and the image capture module will capture the first wound image from the video acquired by the video device.

In one embodiment, the video module of the video device is used to obtain the video signal of the wound of the user or patient, and the display module of the video device is used to display the signal value measured by the video module. In short, the video device uses its lens to shoot the image of the wound, and the video module obtains the video signal of the video, and the relevant video value is displayed by the display module.

In step S302, the first image recognition module is instructed to recognize the first wound image, and when a recognizable object is found in the first wound image, a guide line template suitable for the first wound image is obtained from a storage data module storing a plurality of guide line templates to generate a guide line. Specifically, when the first wound image is to be examined, if there is no reference object or scale, it may be difficult to know the exact size of the wound. In this step, an identifiable object, such as a pen or ruler, will be added to the video image to let the first image recognition module provide what kind of guide line template to the patient. That is, when the wound condition is known through the identifiable object, the guide line template suitable for the wound can be obtained from the storage data module. At this time, a guide line (such as a frame line) will be generated on the screen to guide the patient to move the wound to a better shooting position.

In step S303, the second image recognition module is used to extract the second wound image from the adjusted video with the aid of the guide wire in the video device to identify and analyze the second wound image. Since the guide wire is provided in the previous step to allow the user or patient to move the wound to a better position, the second image recognition module is used in this step to extract the second wound image from the adjusted video, and the second wound image is identified and analyzed to obtain relevant data about the second wound image, such as the size of the wound.

In step S304, the image display module displays the recognition result of the second image recognition module, so that after the user or patient confirms the recognition result, the recognition result is uploaded to the cloud server. This step illustrates that the recognition result of the second image recognition module is displayed through the image display module. The user or patient can confirm the display content of the image display module. If the confirmation is correct, the recognition result can be uploaded to the cloud server.

In one embodiment, when the identification result is uploaded to the cloud server, the data storage module of the cloud server stores the identification result uploaded by the user or patient, that is, the physiological measurement value and file, and the doctor or medical staff can obtain relevant data assistance during the consultation or consultation.

In another embodiment, in order to improve the recognition rate, the image recognition module and the guide line template must be updated at any time. In order to make the image recognition module and the guide line template in the wound recognition video assistance system the latest version, the present invention proposes that the wound recognition video assistance system can be automatically updated, that is, the download module of the cloud server provides a new image recognition module and guide line template for use as an update of the module or data of the wound recognition video assistance system.

In order to ensure the correctness of each file in the cloud server, the present invention proposes to achieve confidentiality through hashing operations, that is, the cloud server's security module provides various encryption algorithms and hashing algorithms to encrypt and hash each file. In addition, the hash code generated after the hashing operation can be recorded in the blockchain of the cloud server's blockchain module.

In one embodiment, the data storage module not only stores the guide line template, but also stores the first wound image, the second wound image, and temporarily stores the recognized measurement values and other data. In addition, the data storage module can also store new image recognition modules and guide line templates and other files provided by the download module from the cloud server.

FIG4 is a flowchart of the image recognition of the present invention. As shown in the figure, in process 401, a video conference or consultation is conducted; in process 402, the image capture screen is opened, that is, the image capture module is used to capture the image; in process 403, the first image recognition is performed, and this process describes the use of the first image module to perform the recognition of the reference object (identifiable object); in process 404, it is determined whether there is an identifiable object, that is, whether there is an identifiable object in the recognition image, if so, enter process 406, if not, enter process 405; in process 405, it is determined whether to recapture, if so, it means to reconfirm whether the identifiable object exists, so return to process 403 to recognize the image again, if not, then end the entire recognition process.

In process 406, a guide line is generated, i.e., wound identification guidance is performed. A guide line is generated to allow the user or patient to move the wound into the guide line. In process 407, it is determined whether to put it into the guidance area. This process is that the wound identification video assistance system confirms whether the video device meets the guidance conditions set by the system. If not, it returns to process 405 and confirms again whether to re-capture. If it meets the guidance conditions, it enters process 408. In process 408, the second image recognition is performed. This process automatically performs the second image recognition and captures physiological measurement data. In short, the second image recognition module is used to identify and analyze the image in the guide line.

In process 409, the wound range is displayed. This process allows the user or patient to view the wound condition in the image through the image display module; in process 410, the wound measurement data is captured. This process measures the wound-related data in the image; in process 411, the user or patient confirms the result. The image display module will display the recognition result operation interface. Through this operation interface, the user or patient can confirm the image recognition result. If the relevant field information is found to be incorrect, it can be corrected immediately; in process 412, the wound measurement data is uploaded. After confirming that the relevant field information is correct, the physiological measurement values can be uploaded to the data storage module in the cloud server for storage, which is convenient for future query and use.

As can be seen from the above, after the image is acquired through the image capture module, the first image recognition module is used to detect the object, and then the recognition result is exchanged with the storage data module of the wound recognition video assistance system to obtain the corresponding guide line template. Finally, the image display module on the wound recognition video assistance system presents the guide frame of the wound. In this way, the guide line for identifying the wound in the current picture can be automatically generated, which can reduce the recognition rate reduction caused by human or environmental factors.

Figure 5 is a flowchart of the automatic verification and update of various files of the present invention. As shown in the figure, in process 501, the system queries the pass code. This process explains that before the user (such as a patient) enters the wound recognition video assistance system, the system will automatically query the user's pass code from the cloud server. This pass code is mainly used to verify the user's identity and the key of its legitimacy; in process 502, it is determined whether the pass code is legal. The verification action of this process is to provide the back-end system platform to manage the user's lease and use status. If the pass code is illegal, it means that the user is currently an illegal user, and the update procedure will not be performed. If the pass code is legal, it will enter process 503.

In process 503, the cloud server security module is used to calculate the system-side file hash code. This process illustrates that the wound identification video assistance system performs a hashing algorithm through the cloud server's security module. In process 504, the hash codes of each file in the blockchain module are compared. This process illustrates that the hash code of the hashing calculation result of process 503 will be compared with the hash codes of each file in the blockchain module in the cloud server. In process 505, the hash code is determined. Whether they are consistent, this process describes the use of hash codes to determine whether the system and cloud data versions are the same. If the hash codes are consistent, it means the versions are the same, and the automatic update process ends. If the hash codes are inconsistent, it means that an abnormality is detected after comparing the files or the version is too old. At this time, the update process will be immediately carried out, that is, entering process 506 to ensure that the files in the local system are the latest and correct versions; process 506, download the latest files from the cloud server.

As can be seen above, users (such as patients) obtain a set of passcodes through the cloud server. This passcode can be used to verify the user's current information and legitimacy. The relevant data obtained through the passcode is provided to the backend system to determine whether the user is a legitimate user and then confirm whether to provide subsequent complete services and related files. In addition, the hash code generated by the hash algorithm provided in the cloud server's security module has the property of being irreversible. This hash code can be used to compare the corresponding files of the local system and the cloud server. The hash algorithm used to generate the hash code must comply with the characteristics of being unmodifiable and highly resistant to collision. Therefore, through this method, users can automatically compare the consistency of the relative files of the local system and the cloud server, reducing the occurrence of abnormalities in the identification system caused by file damage or abnormalities.

FIG6 is a flowchart of uploading files in the cloud server of the present invention, which illustrates the process method of uploading files by the administrator of the cloud server. In process 601, the latest file is uploaded to the data storage module. This process illustrates that when the administrator uploads the latest file to the cloud server, each file is uploaded to the data storage module; in process 602, the hashing algorithm is used to calculate the hash code by the confidentiality module. This process illustrates that the hashing algorithm is used to perform the hashing algorithm on the file; in process 603, the hash code is recorded in the blockchain module. This process illustrates that the hash code calculated by the hashing algorithm will be recorded in the blockchain of the blockchain module for the system to query.

It should be noted that the hashing algorithm used by the confidentiality module is resistant to modification and strong against collision. The optional hashing functions include SHA-2, SHA-3, etc. Using the confidentiality module for hashing and encryption can have the advantages of the wound identification video assistance system being unable to know the content of the hashing algorithm and being able to flexibly manage various keys and algorithms.

As can be seen above, when the administrator uploads a module or file to the cloud server, the module or file will be hashed through the confidentiality module during the upload process. Then, the hash code will be recorded in the blockchain module. The integrity and correctness of various files stored in the cloud server can be ensured by the blockchain's immutability and irreversible sequence.

Each of the aforementioned modules can be software, hardware or firmware; if it is hardware, it can be a processing unit, processor, computer or server with data processing and computing capabilities; if it is software or firmware, it can include instructions that can be executed by the processing unit, processor, computer or server, and can be installed on the same hardware device or distributed on different multiple hardware devices.

In addition, the present invention also discloses a computer-readable medium, which is applied to a computing device or computer having a processor (e.g., CPU, GPU, etc.) and/or a memory, and stores instructions, and the computing device or computer can execute the computer-readable medium through the processor and/or memory to execute the above-mentioned method and each step when executing the computer-readable medium.

In summary, the present invention discloses a wound identification video assistance system, method and computer-readable medium thereof. By using a method of intelligently generating guide lines, the error factor generated by the user when capturing images can be effectively reduced, and doctors or medical staff at a remote end can be assisted in more accurately diagnosing the condition through wound identification and analysis. Therefore, the present invention has the following technical features and advantages.

First, the present invention proposes a method for automatically generating guide lines based on first image recognition. During the image capture process, the first image recognition module is used to detect and identify objects in the image. The first image recognition module is a module that can detect reference objects (identifiable objects) through deep learning training. This module can detect objects in the images presented by the image display module in the wound recognition video assistance system. The detected object can retrieve the corresponding guide line template from the storage data module of the wound recognition video assistance system and insert the template into the existing image display module. Then, the user will completely insert the current wound image into the guide line. When the image display module confirms that the current image meets the guidance conditions, the second image recognition function will be automatically performed.

Second, the present invention proposes a method for the second image recognition module to automatically recognize physiological values based on wounds. The system manager can train multiple wound image values through deep learning to perform image value recognition. The second image recognition module can be used to recognize multiple wound values. Through the values of the reference objects, after calculation, the result values of the second image recognition module will be presented in the form of digital numbers and text on the image display module in the wound recognition video conference assistance system. Because it is real-time video image recognition, caregivers and doctors can get the best auxiliary information for them to make a diagnosis.

Third, the method proposed by the present invention is to encrypt the information of the file and record it in the blockchain through the confidentiality module. The server administrator uploads the latest file to the data storage module of the cloud server. After uploading, the cloud server will immediately use the special hashing algorithm in the confidentiality module to calculate the hash code and record the hash code in the blockchain module's blockchain. Based on the characteristics of the blockchain that cannot be modified and the sequence cannot be reversed, the integrity and non-repudiation of the encrypted information can be guaranteed.

Fourth, the present invention proposes an automatic image module update method. The wound recognition video assistance system queries the cloud server for the current user's passcode. After the passcode is detected, the automatic update process begins. This method uses the cloud server's security module to perform a hashing algorithm on the local first image recognition module, second image recognition module, and guide wire template files in the wound recognition video assistance system to obtain a hash code, and then compares it with the hash codes of each file recorded on the blockchain. If the comparison results are inconsistent, an automatic update process will be performed to replace the relevant files in the wound recognition video assistance system.

The above detailed description is a specific description of a feasible embodiment of the present invention, but the embodiment is not intended to limit the patent scope of the present invention. Any equivalent implementation or modification that does not deviate from the technical spirit of the present invention should be included in the patent scope of the present invention.

1: Wound identification video assistance system

11: Image capture module

12: Data storage module

13: First image recognition module

14: Second image recognition module

15: Image display module

2: Video device

3: Cloud Server

Claims (12)

A wound identification video assistance system includes: an image capture module, used to capture a first wound image from a video acquired by a user's video device, and insert a reference object or a scale into the video as an identifiable object; a data storage module, used to store a plurality of guide line templates; a first image recognition module, used to recognize the first wound image, and when the identifiable object is found in the first wound image and the position of the identifiable object is obtained, the position and distance between the video device and the wound are obtained. The first image recognition module is used to capture the second wound image from the adjusted video with the aid of the guide line, so as to identify and analyze the second wound image; and the image display module is used to display the recognition result of the second image recognition module, so that after the user confirms the recognition result, the recognition result is uploaded to the cloud server. The wound identification video assistance system as described in claim 1, wherein the data storage module further includes storing the first wound image, the second wound image and temporarily storing the identification measurement value. The wound identification video assistance system as described in claim 1 further includes a wireless communication module for connecting to the cloud server to upload the identification results, download new image recognition modules and guide wire templates, and communicate with the cloud server. The wound identification video assistance system as described in claim 1, wherein the video device comprises: a video module for acquiring the video signal of the user's wound; and a display module for displaying the signal value measured by the video module. The wound identification video assistance system as described in claim 1, wherein the cloud server includes: a data storage module for storing the physiological measurement values and files of the identification results uploaded by the user; a download module for providing new image recognition modules and guide wire templates for use as module or data updates; a security module for providing various encryption algorithms and hashing algorithms to encrypt and hash various files of the wound identification video assistance system or the cloud server; and a blockchain module for recording the hash code after the hashing operation of the security module. A wound recognition video assistance method includes the following steps: an image capture module is used to capture a first wound image from a video acquired by a user's video device, and a reference object or a scale is inserted into the video as an identifiable object; a first image recognition module is used to recognize the first wound image, and when the identifiable object is found in the first wound image and the position of the identifiable object is obtained, the position and distance relationship between the video device and the wound and the size of the wound are obtained. , obtaining a guide line template suitable for the first wound image from a storage data module storing multiple guide line templates to generate a guide line; allowing the second image recognition module to extract the second wound image from the adjusted video with the assistance of the guide line to perform recognition and analysis of the second wound image; and allowing the image display module to display the recognition result of the second image recognition module, so that after the user confirms the recognition result, the recognition result is uploaded to the cloud server. The video-assisted wound identification method as described in claim 6, wherein the data storage module further includes storing the first wound image, the second wound image, and temporarily storing the measurement value of the identification. The wound identification video-assisted method as described in claim 6, wherein the step of obtaining a video from a user's video device further includes: allowing a video module of the video device to obtain a video signal of the user's wound; and allowing a display module of the video device to display the signal value measured by the video module. The wound identification video-assisted method as described in claim 6 further includes: allowing the data storage module of the cloud server to store the physiological measurement values and files of the identification results uploaded by the user. The wound identification video assistance method as described in claim 6 further includes: allowing the security module of the cloud server to provide various encryption algorithms and hashing algorithms to encrypt and hash various files; and allowing the blockchain module of the cloud server to record the hash code after the hashing operation of the security module. The wound identification video assistance method as described in claim 6 further includes: allowing the download module of the cloud server to provide a new image recognition module and guide line template for use as an update of the module or data. A computer-readable medium, used in a computing device or a computer, stores instructions for executing a video-assisted wound identification method as described in any one of claims 6 to 11.

Images