Safety Management Identification System and Detection Method for Acupuncture Needles

Abstract

A safety management and identification system for acupuncture needles is provided, which is paired with multiple electronic tags. The multiple electronic tags are attached to the bodies of their respective acupuncture needles. An electronic tag identification device reads the relevant information of the multiple acupuncture needles recorded by the multiple electronic tags. An image capturing device captures the image-related information of the multiple acupuncture needles. A central processing unit, a graphics processing unit, and a computer computing system determine whether the relevant information of the multiple acupuncture needles recorded by the electronic tags is consistent with the image-related information of the multiple acupuncture needles captured by the image capturing device. The relevant information and the judgment result are stored in a database.

Description

Technical Field

[0001] This invention relates to an acupuncture needle safety management and identification system and its detection method, and particularly to an acupuncture needle safety management and identification system and its detection method that uses multiple electronic tags, an electronic tag identification device and a visual identification module to achieve precise positioning, dynamic monitoring and data recording of acupuncture needles. Prior Technology

[0002] Acupuncture is an important treatment method in Traditional Chinese Medicine, which uses acupuncture needles to insert into specific acupoints to achieve the effects of treating diseases and regulating the body. However, acupuncture is a professional treatment method that requires experience and skill. Therefore, doctors are prone to negligence during the acupuncture process, such as forgetting to remove the needles or having a discrepancy between the number of needles inserted and the number removed. These errors can threaten the patient's safety and even lead to more serious medical accidents.

[0003] Meanwhile, acupuncture needles are essential medical equipment used in acupuncture, but the current management of acupuncture needles mainly relies on manual recording by doctors, including the location, number, and removal time of the needles. However, such manual recording is prone to errors due to busy workflows or doctors' negligence, and the efficiency of manual data recording is low, which is not conducive to the data-driven and precise management of acupuncture treatment.

[0004] Therefore, improving the safety of acupuncture procedures, avoiding medical accidents caused by missing needles or incorrect needle counts, reducing the error rate in managing acupuncture needles, and ensuring the accuracy and traceability of data for each treatment are among the urgent problems that the medical and academic communities need to solve. Summary of the Invention

[0005] The purpose of this invention is to provide an acupuncture needle safety management and identification system and its detection method, which uses multiple electronic tags, an electronic tag identification device and a visual identification module to achieve precise positioning, dynamic monitoring and data recording of acupuncture needles.

[0006] To achieve the above objectives, the acupuncture needle safety management and identification system disclosed in the first embodiment of the present invention is used in conjunction with multiple electronic tags, and these multiple electronic tags are affixed to the bodies of their respective multiple acupuncture needles. The acupuncture needle safety management and identification system includes at least: an electronic tag identification device for simultaneously identifying the multiple electronic tags corresponding to each of the multiple acupuncture needles, reading relevant information from the multiple electronic tags, and converting the relevant information into an electronic tag signal for transmission; and an image capturing device for capturing images of the multiple acupuncture needles and transmitting the captured images. The system converts the images of the multiple acupuncture needles into a first image signal and transmits it; a central processing unit receives the electronic tag signal and the first image signal, and transmits them; a computer processing system receives the first image signal, performs image preprocessing on it, and converts it into a second image signal for transmission; a graphics processing unit receives the second image signal, accelerates the detection process of an object frame operation on the second image signal, and generates a third image signal to be sent back to the computer processing system; wherein the computer processing... Upon receiving the third image signal, the system transmits it to the central processing unit and simultaneously performs multiple image matching operations on it. Based on the matching results, it outputs multiple matching result signals. Then, based on the number of matches and average distances derived from each matching result signal, it calculates multiple matching scores. Using these scores and match counts, it outputs multiple valid images. Finally, it averages and weights these valid images to estimate the number and specific location of acupuncture needles on the patient's body. The estimated number of acupuncture needles is then displayed. The number and specific location of the needles on the patient's body are converted into an image comparison result signal and transmitted to the central processing unit. After receiving the image comparison result signal and the third image signal, the central processing unit determines whether the number and specific location of the electronic tag signal corresponding to each of the multiple acupuncture needles on the body are consistent with the number and specific location of the multiple acupuncture needles on the patient's body in the image comparison result signal. It then determines whether the feedback value of the multiple acupuncture needles detected by the object detection of the third image signal is zero, and converts the determination result into a final comparison result signal and presents it.

[0007] To achieve the above objectives, the acupuncture needle safety management identification and detection method disclosed in the first embodiment of the present invention is used in conjunction with multiple electronic tags, and these multiple electronic tags are affixed to the bodies of their respective multiple acupuncture needles. The acupuncture needle safety management identification and detection method includes at least the following steps: First step: Using an electronic tag identification device to simultaneously identify the multiple electronic tags corresponding to each of the multiple acupuncture needle bodies, reading the relevant information on the multiple electronic tags, and converting the relevant information into an electronic tag signal and transmitting it to a central processing unit. The first step involves using an image capturing device to capture images of the multiple acupuncture needles, converting these images into a first image signal, and transmitting it to the central processing unit (CPU). The second step involves the CPU transmitting the first image signal to a computer processing system, which then performs image preprocessing on the first image signal and outputs a second image signal to a graphics processing unit (GPU). The fourth step involves the GPU accelerating the object frame detection process of the second image signal and generating a third image signal to be transmitted back to the computer. The computing system, in its fifth step, transmits the third image signal to the central processing unit and simultaneously performs multiple image matching on the third image signal. Based on the results of these multiple image matching, it outputs multiple matching result signals. Then, based on the multiple matching results signals and the multiple matching numbers and average distances obtained from each matching result signal, it calculates multiple matching scores. Finally, it uses these multiple matching scores and the multiple matching numbers to output multiple valid images. The average and weighted values ​​from these multiple valid images are then used to estimate the number of acupuncture needles on a patient's body. The specific location is determined, and the estimated number and specific location of the multiple acupuncture needles on the patient's body are converted into an image comparison result signal and transmitted to the central processing unit; the sixth step: the central processing unit determines whether the number and specific location of the multiple acupuncture needles in the image comparison result signal are consistent with the number and specific location of the corresponding electronic tag signals on the bodies of the multiple acupuncture needles, and then determines whether the feedback value of the multiple acupuncture needle bodies detected by the object detection of the third image signal is zero, and converts the judgment result into a final comparison result signal and presents it.

[0008] To achieve the above objectives, the acupuncture needle safety management and identification system disclosed in the second embodiment of the present invention includes at least: an image capturing device for capturing images of the plurality of acupuncture needles and converting the captured images of the plurality of acupuncture needles into a first image signal for transmission; a central processing unit for receiving the first image signal and transmitting the first image signal; a computer processing system for receiving the first image signal, performing image preprocessing on the first image signal, and converting it into a second image signal for transmission; and a graphics processing unit for receiving the second image signal, accelerating the detection process of an object frame operation of the second image signal, and generating a third image signal to be sent back to the computer processing system; wherein when the computer processing system receives the third image signal, it transmits the third image signal to the central processing unit, and simultaneously performs multiple image matching on the third image signal, and outputs the results based on the results of the multiple image matching. Multiple matching result signals are generated, and multiple matching scores are calculated based on the multiple matching quantities and multiple average distances obtained from each of the multiple matching result signals. Multiple valid images are then output using the multiple matching scores and multiple matching quantities. The average and weighted values ​​of the multiple valid images are then taken to estimate the number and specific location of the acupuncture needles on the patient's body. The estimated number and specific location of the acupuncture needles on the patient's body are then converted into an image comparison result signal and transmitted to the central processing unit. After receiving the image comparison result signal and the third image signal, the central processing unit determines whether the number and specific location of the multiple acupuncture needles on the patient's body in the image comparison result signal are consistent with the number and specific location of the multiple acupuncture needles in the first image signal. It then determines whether the feedback value of the multiple acupuncture needles detected by the object detection of the third image signal is zero, and converts the determination result into a final comparison result signal for presentation.

[0009] To achieve the above objectives, the acupuncture needle safety management identification and detection method disclosed in the second embodiment of the present invention includes at least the following steps: First step: Using an image capturing device, images of the multiple acupuncture needles are captured, and the captured images of the multiple acupuncture needles are converted into a first image signal and transmitted to the central processing unit; Second step: The central processing unit transmits the first image signal to a computer computing system, which then performs image preprocessing on the first image signal and outputs a second image signal to a graphics processing unit; Third step: The graphics processing unit accelerates the object frame detection process of the second image signal and generates a third image signal, which is then transmitted back to the computer computing system; Fourth step: The computer computing system transmits the third image signal to the central processing unit and simultaneously performs multiple image matching on the third image signal, and based on the results of the multiple image matching... The system outputs multiple matching result signals, calculates multiple matching scores based on the number of matches and average distances obtained from each matching result signal, outputs multiple valid images using the multiple matching scores and the multiple matching numbers, and then averages and weights these valid images to estimate the number and specific location of the multiple acupuncture needles on the patient's body. The estimated number and specific location of the multiple acupuncture needles on the patient's body are then converted into an image comparison result signal and transmitted to the central processing unit. Fifth step: The central processing unit determines whether the number and specific location of the multiple acupuncture needles in the image comparison result signal are consistent with the number and specific location of the multiple acupuncture needles in the first image signal. It then determines whether the feedback value of the multiple acupuncture needles detected by the object detection of the third image signal is zero, and converts the determination result into a final comparison result signal for presentation.

[0010] The above description of the content of this invention and the following description of the embodiments are used to demonstrate and explain the principles of this invention, and to provide a further explanation of the scope of the patent application of this invention. Simple Explanation of the Diagram

[0011]

[0012] Figure 1 is a block diagram of the overall safety management and identification system for acupuncture needles according to the first embodiment of the present invention.

[0013] Figure 2 is a flowchart of the steps of the safety management identification and detection method for acupuncture needles according to the first embodiment of the present invention.

[0014] Figure 3 is a block diagram of the overall safety management and identification system for acupuncture needles according to the second embodiment of the present invention.

[0015] Figure 4 is a flowchart of the steps of the safety management identification and detection method for acupuncture needles according to the second embodiment of the present invention. Implementation

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please refer to Figure 1, which is a schematic block diagram of the overall block diagram of the acupuncture needle safety management and identification system according to the first embodiment of the present invention. The acupuncture needle safety management and identification system 1 of the first embodiment of the present invention is used in conjunction with multiple electronic tags 2, and the multiple electronic tags 2 are set on the bodies of their respective multiple acupuncture needles 3. The acupuncture needle safety management and identification system 1 includes at least: an electronic tag identification device 11, an image capturing device 12, a central processing unit (CPU) 13, a graphics processing unit (GPU) 14, a database 15, and a computer computing system 16.

[0018] In the first embodiment of the present invention, the electronic tag identification device 11, the image capturing device 12, the graphics processing unit 14, the database 15 and the computer computing system 16 are connected or electrically connected to the central processing unit 13.

[0019] The aforementioned electronic tag identification device 11 may be a radio frequency identification (RFID) device or a near-field communication (NFC) device, etc. However, please note that the listed devices are only for the purpose of understanding this embodiment and are not intended to limit the scope of the invention.

[0020] The aforementioned computer processing system 16 further includes a visual recognition module 161, multiple feature matching units 162, a matching scoring unit 163, a voting mechanism unit 164, and a graphical user interface 165. The visual recognition module 161, the voting mechanism 164, and the graphical user interface 165 are connected or electrically connected to the central processing unit 13. The multiple feature matching units 162 are connected or electrically connected to the visual recognition module 161 and the matching scoring unit 163. The voting mechanism unit 164 is connected or electrically connected to the matching scoring unit 163.

[0021] The aforementioned visual recognition module 161 can be a YOLO model (You Only Look Once), an ELAN model (Efficient Layer Attention Network), a Faster R-CNN model (Region-based Convolutional Neural Network), etc. However, the models listed here are only for the purpose of understanding this embodiment and are not intended to limit the scope of the application.

[0022] It should be specifically noted that the acupuncture needle safety management and identification system 1 of the first embodiment of the present invention utilizes the electronic tag identification device 11 to simultaneously identify the multiple electronic tags 2 corresponding to each of the multiple acupuncture needles 3, and reads the relevant information on the multiple electronic tags 2 (such as the time, number, and specific location of the insertion of the multiple acupuncture needles 3 into the patient's body, or the time, number, or specific location of the removal of the multiple acupuncture needles 3, etc.), and converts the read relevant information into an electronic tag signal St and transmits it to the central processing unit 13. When the central processing unit 13 receives the electronic tag signal St, it transmits the electronic tag signal St to the database 15 to save the relevant information in the electronic tag signal St. The central processing unit 13 also transmits the electronic tag signal St to the graphical user interface 165 in the computer computing system 16, so that the graphical user interface 165 can present the relevant information of the multiple acupuncture needles 3 in a graphical or textual manner. Simultaneously, the acupuncture needle safety management and identification system 1 of this embodiment of the invention also uses the image capturing device 12 to capture images of the multiple acupuncture needles 3, and converts the captured images of the multiple acupuncture needles 3 into a first image signal Si1 and transmits it to the central processing unit 13. When the central processing unit 13 receives the first image signal Si1, it transmits the first image signal Si1 to the visual recognition module 161 in the computer computing system 16. The visual recognition module 161 performs image preprocessing on the first image signal Si1. The image preprocessing first reads out multiple necessary images from the first image signal Si1 (the reading out of multiple necessary images mentioned here can be done through, for example, a region of interest-based detection, a key extraction). (This can be achieved by using methods such as frames or specific regions, etc.). Then, the multiple necessary images are uniformly adjusted into multiple necessary images of a specified size (e.g., 224x224 pixels, etc.). Then, the multiple necessary images of the specified size are subjected to noise suppression through a Gaussian filter or a bilateral filter. Finally, the multiple necessary images with noise suppression are converted into a second image signal Si2 and sent to the graphics processing unit 14. The graphics processing unit 14 will accelerate the detection process of the object frame operation of the second image signal Si2 and generate a third image signal Si3 and send it back to the visual recognition module 161. Then, the visual recognition module 161 will send the third image signal Si3 to the central processing unit 13 and the multiple feature matching units 162.When the third image signal Si3 is used for multiple image matching by the multiple feature matching units 162, multiple matching result signals Sm are output based on the results of the multiple image matching (for example, three matching result signals: a first matching result signal Sm1, a second matching result signal Sm2, and a third matching result signal Sm3). Then, the first matching result signal Sm1, the second matching result signal Sm2, and the third matching result signal Sm3 are used by the matching scoring unit 163 to perform a matching score calculation. The matching score calculation is based on the first matching result signal Sm1. Multiple matching scores (e.g., three matching scores: MS1, MS2, and MS3) are calculated from the multiple matching counts and multiple average distances obtained from the first matching result signal Sm1, the second matching result signal Sm2, and the third matching result signal Sm3, respectively. Then, the voting mechanism unit 164 compares and determines whether the multiple matching scores and the multiple matching counts exceed a preset matching score threshold and a preset minimum matching count threshold. When the multiple matching scores are less than a certain threshold, the system determines the minimum matching count threshold. If the number of matches is less than the preset minimum matching threshold, it is considered an invalid vote; conversely, if the number of matches is greater than the preset matching threshold or the number of matches is greater than the preset minimum matching threshold, it is considered a valid vote. The valid votes obtained from the multiple feature matching units are then used to determine if they exceed a preset vote threshold in the voting mechanism unit. If the valid votes exceed the preset vote threshold, multiple valid images are output. The average and weighted values ​​from these multiple valid images are then used to estimate the current status of the target. The system checks whether the number and specific location of the acupuncture needles 3 on the patient's body still exist (conversely, if the number of valid votes does not exceed the preset vote threshold, the third image signal Si3 is re-matched multiple times until the result is stable or confirmed as invalid). Based on this information, an image comparison result signal Sicr is output and sent to the central processing unit 13. After receiving the image comparison result signal Sicr, the central processing unit 13 sends it to the database 15 to save the relevant content of the image comparison result signal Sicr. The central processing unit 13 can also send the image comparison result signal Sicr to the graphical user interface 165, allowing the graphical user interface 165 to present the estimated information of the multiple acupuncture needles 3 using one or more graphics and one or more text.Finally, the central processing unit 13 determines whether the number and specific location of the electronic tag signal St corresponding to each of the multiple acupuncture needles are consistent with the number and specific location of the multiple acupuncture needles 3 in the image comparison result signal Sicr. If the number and specific location of the electronic tag signal St are consistent with the number and specific location of the multiple acupuncture needles 3 in the image comparison result signal Sicr, it means that the location of the patient's part captured by the image capturing device 12 is correct, and the needle removal operation can begin. When the needle removal operation is to be performed, the multiple acupuncture needles 3 will be detected again through an object detection in the third image signal Si3. The system uses feedback values ​​for judgment. When the feedback value is zero (meaning no acupuncture needle 3 is detected), it means that the acupuncture needles on the patient have been completely removed, and the judgment process can end. Conversely, if the number and specific location of the electronic tag signal St are consistent with the number and specific location of the multiple acupuncture needles 3 in the image comparison result signal Sicr, but the feedback value of the multiple acupuncture needles 3 detected by the object detection of the third image signal Si3 is not zero (meaning that the acupuncture needles 3 are detected), it means that the removal of the multiple acupuncture needles 3 is not complete, and some acupuncture needles 3 may still be on the patient's body. In this case, a warning message will be issued to remind the user. If the number and specific position of the electronic tag signal St corresponding to each of the multiple acupuncture needles 3 are inconsistent with the number and specific position of the multiple acupuncture needles 3 in the image comparison result signal Sicr, it indicates that the position of the patient's body captured by the image capturing device 12 is incorrect. The image capturing device 12 needs to be moved to the correct position. Finally, based on the comparison result, a final comparison result signal Sfcr is output and transmitted to the graphical user interface 165. The comparison result information can be presented using one or more graphics or one or more text.

[0023] The multiple matching result signals listed above are illustrated using three matching result signals, but are not limited to these three matching result signals. There could also be two, four, five, ten, or more matching result signals.

[0024] The acupuncture needle safety management and identification system 1 of the first embodiment of the present invention can further connect or electrically connect the database 15 and the graphical user interface 165 to a large language model module 17, so that the large language model module 17 can communicate and explain to the user through the graphical user interface 165 based on the relevant information in the database 15.

[0025] The aforementioned large language model module 17 can be a Chinese Medicine LLM, a Med-BERT (Med-Bidirectional Encoder Representation from Transformers) large language model, a Clinical-BERT (Clinical-Bidirectional Encoder Representations from Transformers) large language model, etc. However, the large language models listed here are only for the purpose of understanding this embodiment and are not intended to limit the scope of the large language models listed.

[0026] The aforementioned multiple feature matching units may be a brute-force matching unit, a brute-force ratio test matching unit, a fast approximate nearest neighbor search matching unit (FLANN Matching), etc. However, the listed feature matching units are only for the purpose of understanding this embodiment and are not intended to limit the scope of the embodiments.

[0027] Please refer to Figure 2, which is a flowchart of the steps of the acupuncture needle safety management identification and detection method according to the first embodiment of the present invention. The acupuncture needle safety management identification and detection method of this embodiment is used in conjunction with multiple electronic tags 2, and these multiple electronic tags 2 are affixed to the bodies of their respective corresponding multiple acupuncture needles 3. The acupuncture needle safety management identification and detection method includes at least the following steps:

[0028] First step S10: Use an electronic tag identification device to simultaneously identify the multiple electronic tags corresponding to each of the multiple acupuncture needles, read the relevant information on the multiple electronic tags, and convert the read relevant information into an electronic tag signal and transmit it to a central processing unit;

[0029] Second step S20: Use an image capturing device to capture images of the multiple acupuncture needles, and convert the captured images of the multiple acupuncture needles into a first image signal and transmit it to the central processing unit;

[0030] Third step S30: The central processing unit receives the first image signal and then transmits the first image signal to a visual recognition module. The visual recognition module performs image preprocessing on the first image signal and outputs a second image signal to a graphics processing unit.

[0031] Fourth step S40: The graphics processing unit accelerates the object frame detection process of the second image signal and generates a third image signal to be sent back to the visual recognition module.

[0032] Step S50: The visual recognition module transmits the third image signal to the central processing unit, and simultaneously performs multiple image matching on the third image signal through multiple feature matching units. Based on the results of the multiple image matching, it outputs multiple matching result signals. Then, it performs a matching score calculation on the multiple matching result signals through a matching scoring unit. The matching score calculation calculates multiple matching scores based on the multiple matching numbers and multiple average distances obtained from the multiple matching result signals. Then, it uses a voting mechanism unit to determine whether the multiple matching numbers and multiple matching scores exceed a preset matching score threshold and a preset minimum matching number threshold in the voting mechanism unit to calculate a valid vote. It then determines whether the valid vote exceeds a preset vote threshold in the voting mechanism unit to output multiple valid images. Finally, it takes the average and weights from the multiple valid images to estimate the number and specific location of the acupuncture needles on the patient's body and other information. Based on this information, it outputs an image comparison result signal and sends it back to the central processing unit.

[0033] Sixth step S60: The central processing unit determines whether the number and specific location of the multiple acupuncture needles in the image comparison result signal are consistent with the number and specific location of the corresponding electronic tag signals on the bodies of the multiple acupuncture needles. Then it determines whether the feedback value of the multiple acupuncture needle bodies detected by the object detection of the third image signal is zero, and converts the judgment result into a final comparison result signal.

[0034] Simultaneously, the central processing unit transmits the electronic tag signal, the image comparison result signal, the third image signal, and the final comparison result signal to a database and a graphical user interface, stores the relevant information in the electronic tag signal, the image comparison result signal, the third image signal, and the final comparison result signal in the database, and simultaneously presents the relevant information in the electronic tag signal, the image comparison result signal, the third image signal, and the final comparison result signal on the graphical user interface.

[0035] If the scores of the multiple matches in the aforementioned fifth step S50 are less than the preset matching score threshold or the number of multiple matches is less than the preset minimum matching number threshold, then it is an invalid vote; conversely, if the scores of the multiple matches are greater than the preset matching score threshold or the number of multiple matches is greater than the preset minimum matching number threshold, then it is a valid vote.

[0036] In the aforementioned sixth step S60, when the central processing unit determines that the number and specific location of the electronic tag signals corresponding to each of the multiple acupuncture needles are consistent with the number and specific location of the multiple acupuncture needles on the patient's body in the image comparison result signal, and the feedback value of the multiple acupuncture needles detected by the object detection of the third image signal is zero, it indicates that the patient is safe and the judgment process can end. Conversely, when the central processing unit determines that the number and specific location of the electronic tag signals corresponding to each of the multiple acupuncture needles are consistent with the number and specific location of the multiple acupuncture needles on the patient's body in the image comparison result signal, but the feedback value of the multiple acupuncture needles detected by the object detection of the third image signal is not zero, it indicates that the acupuncture needle removal is incomplete and some acupuncture needles may still be on the patient's body, and a warning message will be issued to remind the user. However, if the central processing unit determines that the number and specific location of the electronic tag signal corresponding to each of the multiple acupuncture needles is inconsistent with the number and specific location of the multiple acupuncture needles on the patient's body in the image comparison result signal, it means that the position of the patient's part captured by the image capturing device is incorrect, and the image capturing device needs to be moved to the correct shooting position. A warning message will be issued to remind the user.

[0037] The acupuncture needle safety management identification and detection method of the present invention can further connect or electrically connect the database and the graphical user interface to a large language model module, so that the large language model module can communicate and explain to the user through the graphical interface based on the real-time relevant information in the database.

[0038] Please refer to Figure 3, which is a block diagram of the overall structure of the acupuncture needle safety management and identification system according to the second embodiment of the present invention. The difference between the second embodiment and the first embodiment is that the second embodiment does not use multiple electronic tags 2, but only utilizes image recognition for relevant judgments. The acupuncture needle safety management and identification system 1 of the second embodiment of the present invention includes at least: an image capturing device 12, a central processing unit (CPU) 13, a graphics processing unit (GPU) 14, a database 15, and a computer computing system 16.

[0039] In the second embodiment of the present invention, the image capturing device 12, the graphics processing unit 14, the database 15 and the computer computing system 16 are connected or electrically connected to the central processing unit 13.

[0040] The aforementioned computer processing system 16 further includes a visual recognition module 161, multiple feature matching units 162, a matching scoring unit 163, a voting mechanism unit 164, and a graphical user interface 165. The visual recognition module 161, the voting mechanism 164, and the graphical user interface 165 are connected or electrically connected to the central processing unit 13. The multiple feature matching units 162 are connected or electrically connected to the visual recognition module 161 and the matching scoring unit 163. The voting mechanism unit 164 is connected or electrically connected to the matching scoring unit 163.

[0041] The acupuncture needle safety management and identification system 1 of the second embodiment of the present invention utilizes the image capturing device 12 to capture images of multiple acupuncture needles 3, and converts the captured images of the multiple acupuncture needles 3 into a first image signal Si1, which is then transmitted to the central processing unit 13. Upon receiving the first image signal Si1, the central processing unit 13 transmits it to the visual recognition module 161 in the computer processing system 16. The visual recognition module 161 performs image preprocessing on the first image signal Si1. This image preprocessing first extracts multiple necessary images from the first image signal Si1 (the extraction of multiple necessary images can be achieved through, for example, region of interest detection, keyframe extraction, or...). (This can be achieved by methods such as a specific region, etc.) Then, the multiple necessary images are uniformly adjusted to a specified size (e.g., 224x224 pixels, etc.). Then, the multiple necessary images of the specified size are subjected to noise suppression through a Gaussian filter or a bilateral filter. Finally, the multiple necessary images with noise suppression are converted into a second image signal Si2 and sent to the graphics processing unit 14. The graphics processing unit 14 will accelerate the detection process of the object frame operation of the second image signal Si2 and generate a third image signal Si3 and send it back to the visual recognition module 161. Then, the visual recognition module 161 will send the third image signal Si3 to the central processing unit 13 and the multiple feature matching units 162.When the third image signal Si3 performs multiple image matching through the multiple feature matching units 162, and outputs multiple matching result signals Sm based on the results of the multiple image matching (for example, three matching result signals: a first matching result signal Sm1, a second matching result signal Sm2, and a third matching result signal Sm3), then the first matching result signal Sm1, the second matching result signal Sm2, and the third matching result signal Sm3 are used to perform a matching score calculation through the matching scoring unit 163 in the visual recognition module 161. The matching score calculation will then be based on the first matching result signal Sm1, the second matching result signal Sm2, and the third matching result signal Sm3. The matching scores of the multiple feature matching units 162 are calculated using the multiple matching numbers and multiple average distances obtained from the third matching result signal Sm3 (for example, three matching scores: a first matching score MS1, a second matching score MS2, and a third matching score MS3). Then, the voting mechanism unit 164 compares the matching scores and the multiple matching numbers to determine whether they exceed a preset matching score threshold and a preset minimum matching number threshold. If the matching score is less than the preset matching score threshold or the multiple matching numbers are less than the preset minimum matching number threshold, it is an invalid vote; conversely, if the matching score is greater than the preset matching score threshold or the multiple matching numbers are greater than the preset minimum matching number threshold, it is a valid vote. Then, based on the multiple feature matching units... The system uses the number of valid votes obtained to determine whether it exceeds a preset threshold value in the voting mechanism unit. If the number of valid votes exceeds the preset threshold value, multiple valid images are output. The average and weighted values ​​of these multiple valid images are then used to estimate whether the acupuncture needle 3 still exists on the patient's body and other information such as its quantity and specific location. (Conversely, if the number of valid votes does not exceed the preset threshold value, the third image signal Si3 is re-matched multiple times until the result is stable or confirmed as invalid.) Based on this information, an image comparison result signal Sicr is output and sent to the central processing unit 13. When the central processing unit 13 receives the image comparison result signal Sicr, it sends the image comparison result signal Sicr to the database 15 to save the relevant content of the image comparison result signal Sicr. The central processing unit 13 also transmits the image comparison result signal Sicr to the graphical user interface 165, so that the graphical user interface 165 can present the relevant information of the estimated multiple acupuncture needles 3 in the form of one or more graphics and one or more text.Finally, the central processing unit 13 determines whether the number and specific location of the multiple acupuncture needles 3 in the image comparison result signal Sicr are consistent with the number and specific location of the multiple acupuncture needles 3 in the first image signal Si1. If the number and specific location of the multiple acupuncture needles 3 in the image comparison result signal Sicr are consistent with the number and specific location of the multiple acupuncture needles 3 in the first image signal Si1, it means that the location of the patient's part captured by the image capturing device 12 is correct, and the needle removal operation can begin. When the needle removal operation is to be performed, the multiple acupuncture needles 3 detected by the object detector in the third image signal Si3 will be further detected. The system uses the feedback value of the object to determine the needles. If the feedback value is zero (meaning no needle 3 was detected), it indicates that the acupuncture needles 3 have been completely removed from the patient, and the judgment process can end. Conversely, if the number and location of the electronic tag signal St match the number and location of the multiple acupuncture needles 3 in the image comparison result signal Sicr, but the feedback value of the object detection of the multiple acupuncture needles 3 in the third image signal Si3 is not zero (meaning the acupuncture needles 3 were detected), it indicates that the removal of the multiple acupuncture needles 3 is incomplete, and some needles 3 may still be present on the patient's body. In this case, a warning message will be issued to remind the user.

[0042] However, if the number and specific position of the multiple acupuncture needles 3 in the image comparison result signal Sicr are inconsistent with the number and specific position of the multiple acupuncture needles 3 in the first image signal Si1, it means that the position of the patient's body part captured by the image capturing device 12 is incorrect. The image capturing device needs to be moved to the correct position. Finally, based on the comparison result, a final comparison result signal Sfcr is output and transmitted to the graphical user interface 165. The relevant information of the comparison result can be presented in the form of one or more graphics or one or more texts.

[0043] Please refer to Figure 4, which is a flowchart of the steps of the acupuncture needle safety management identification and detection method according to the second embodiment of the present invention. The acupuncture needle safety management identification and detection method of the present invention includes at least the following steps:

[0044] First step S100: Use an image capturing device to capture images of the multiple acupuncture needles, and convert the captured images of the multiple acupuncture needles into a first image signal and transmit it to the central processing unit;

[0045] Second step S200: The central processing unit receives the first image signal and then transmits the first image signal to a visual recognition module. The visual recognition module performs image preprocessing on the first image signal and outputs a second image signal to a graphics processing unit.

[0046] Third step S300: The graphics processing unit accelerates the detection process of the object frame operation of the second image signal and generates a third image signal to be sent back to the visual recognition module;

[0047] Fourth step S400: The visual recognition module transmits the third image signal to the central processing unit, and simultaneously performs multiple image matching on the third image signal through multiple feature matching units. Based on the results of the multiple image matching, it outputs multiple matching result signals. Then, it performs a matching score calculation on the multiple matching result signals through a matching scoring unit. The matching score calculation calculates multiple matching scores based on the multiple matching numbers and multiple average distances obtained from the multiple matching result signals. Then, it uses a voting mechanism unit to determine whether the multiple matching numbers and multiple matching scores exceed a preset matching score threshold and a preset minimum matching number threshold in the voting mechanism unit to calculate a valid vote. It then determines whether the valid vote exceeds a preset vote threshold in the voting mechanism unit to output multiple valid images. Finally, it takes the average and weights from the multiple valid images to estimate the number and specific location of the acupuncture needles on the patient's body and other information. Based on this information, it outputs an image comparison result signal and sends it back to the central processing unit.

[0048] Fifth step S500: The central processing unit determines whether the number and specific location of the multiple acupuncture needles in the image comparison result signal are consistent with the number and specific location of the multiple acupuncture needles in the first image signal, then determines whether the feedback value of the multiple acupuncture needle bodies detected by the object detection of the third image signal is zero, and converts the judgment result into a final comparison result signal.

[0049] Simultaneously, the central processing unit transmits the first image signal, the image comparison result signal, the third image signal, and the final comparison result signal to a database and a graphical user interface, stores the relevant information in the first image signal, the image comparison result signal, the third image signal, and the final comparison result signal in the database, and simultaneously presents the relevant information in the first image signal, the image comparison result signal, the third image signal, and the final comparison result signal to the graphical user interface.

[0050] Although the present invention has been disclosed above with reference to the foregoing embodiments, it is not intended to limit the present invention. Any person skilled in the art may make some modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the scope of patent protection of the present invention shall be determined by the scope of protection of the claims appended to this specification.

[0051]

[0052] 1: Safety Management and Identification System for Acupuncture Needles

[0053] 11: Electronic tag identification device

[0054] 12: Image capturing device

[0055] 13: Central Processing Unit

[0056] 14: Graphics Processing Unit

[0057] 15: Database

[0058] 16: Computer computing system

[0059] 161: Visual Recognition Module

[0060] 162: Feature Matching Unit

[0061] 163: Matching scoring unit

[0062] 164: Voting Mechanism Unit

[0063] 165: Graphical User Interface

[0064] 17: Large Language Model Module

[0065] 2: Electronic tags

[0066] 3: Acupuncture needles

[0067] S10~S60: Steps

[0068] S100~S500: Steps

Claims

1. A safety management and identification system for acupuncture needles, used in conjunction with multiple electronic tags, wherein the multiple electronic tags are affixed to the bodies of their respective corresponding multiple acupuncture needles, wherein the safety management and identification system for acupuncture needles includes at least: An electronic tag identification device is used to simultaneously identify the multiple electronic tags corresponding to each of the multiple acupuncture needles, read the relevant information on the multiple electronic tags, and convert the relevant information into an electronic tag signal for transmission; an image capturing device is used to capture images of the multiple acupuncture needles, convert the captured images into a first image signal for transmission; a central processing unit is used to receive the electronic tag signal and the first image signal, and transmit the electronic tag signal and the first image signal; a computer computing system is used to receive the first image signal, perform image preprocessing on the first image signal, and convert it into a second image signal for transmission; a graphics processing unit is used to receive the second image signal, accelerate the detection process of an object frame operation of the second image signal, and generate a third image signal to be sent back to the computer computing system; When the computer system receives the third image signal, it transmits the third image signal to the central processing unit and simultaneously performs multiple image matching on the third image signal. Based on the results of the multiple image matching, it outputs multiple matching result signals. Then, based on the multiple matching numbers and multiple average distances obtained from the multiple matching result signals, it calculates multiple matching scores. Then, it uses the multiple matching scores and multiple matching numbers to output multiple valid images. Finally, it takes the average and weights from the multiple valid images to estimate the number and specific location of the acupuncture needles on the patient's body. The estimated number and specific location of the acupuncture needles on the patient's body are converted into an image comparison result signal and transmitted to the central processing unit. After receiving the image comparison result signal and the third image signal, the central processing unit will determine whether the number and specific location of the electronic tag signal corresponding to each of the multiple acupuncture needles are consistent with the number and specific location of the multiple acupuncture needles on the patient's body in the image comparison result signal. Then, it will determine whether the feedback value of the multiple acupuncture needles detected by the object detection of the third image signal is zero, and convert the judgment result into a final comparison result signal for presentation.

2. The acupuncture needle safety management and identification system as described in claim 1, wherein the computer computing system comprises at least: a visual recognition module for receiving the first image signal, performing image preprocessing on the first image signal, and converting it into the second image signal for transmission; multiple feature matching units for receiving the second image signal, performing multiple image matching on the second image signal, and outputting multiple matching result signals based on the results of the multiple image matching; and a matching scoring unit for receiving the multiple matching result signals, performing matching score calculation on the multiple matching result signals, wherein the matching score calculation is based on multiple matching quantities and multiple average distances obtained from the multiple matching result signals to calculate multiple matching scores; A voting mechanism unit is used to determine whether the matching score and the number of multiple matches exceed a preset matching score threshold and a preset minimum matching number threshold, and then outputs multiple valid images. The average and weighted values ​​of the multiple valid images are then used to estimate the number and specific location of the multiple acupuncture needles on the patient's body, and the result is converted into an image comparison result signal and transmitted. A graphical user interface is used to receive the image comparison result signal and present the image comparison result signal in one or more graphics or one or more text.

3. The acupuncture needle safety management and identification system as described in claim 2 may further connect or electrically connect a database and the graphical user interface to a large language model module, so that the large language model module can communicate and explain to a user through the graphical user interface based on the relevant information in the database.

4. The acupuncture needle safety management and identification system as described in claim 1, wherein the image preprocessing includes at least the following steps: first, reading multiple necessary images from the first image; then, uniformly adjusting the multiple necessary images into multiple necessary images of a specified size; then, performing noise suppression on the multiple necessary images of the specified size through a Gaussian filter or a bilateral filter; and finally, converting the noise-suppressed multiple necessary images into a second image signal and transmitting it out.

5. The acupuncture needle safety management and identification system as described in claim 2, wherein the electronic tag signal, the image comparison result signal, the third image signal and the final comparison result signal can all be presented through the graphical user interface in the form of one or more graphics or one or more text.

6. A method for safety management, identification, and detection of acupuncture needles, which is used in conjunction with multiple electronic tags, wherein the multiple electronic tags are affixed to the bodies of their respective acupuncture needles, and the method includes at least the following steps: First step: Using an electronic tag identification device to simultaneously identify the multiple electronic tags corresponding to the bodies of the multiple acupuncture needles, reading relevant information from the multiple electronic tags, and converting the relevant information into an electronic tag signal and transmitting it to a central processing unit; Second step: Using an image capturing device to capture images of the multiple acupuncture needles, converting the captured images into a first image signal and transmitting it to the central processing unit; Third step: The central processing unit transmits the image signal to a computer processing system, which then performs image preprocessing on the first image signal and outputs a second image signal to a graphics processing unit. Fourth step: The graphics processing unit accelerates the object frame detection process of the second image signal and generates a third image signal to be sent back to the computer processing system; Fifth step: The computer processing system transmits the third image signal to the central processing unit and simultaneously performs multiple image matching on the third image signal, and outputs multiple matching result signals based on the results of the multiple image matching, and calculates multiple matching scores based on the multiple matching numbers and multiple average distances obtained from the multiple matching result signals, and outputs multiple valid images using the multiple matching scores and multiple matching numbers, and then takes the average and weight from the multiple valid images to estimate the number and specific location of the multiple acupuncture needles on the patient's body, and converts the estimated number and specific location of the multiple acupuncture needles on the patient's body into an image comparison result signal and sends it to the central processing unit; Step 6: The central processing unit determines whether the number and specific location of the multiple acupuncture needles in the image comparison result signal are consistent with the number and specific location of the corresponding electronic tag signals on the body of each of the multiple acupuncture needles. Then, it determines whether the feedback value of the multiple acupuncture needles detected by the object detection of the third image signal is zero, and converts the judgment result into a final comparison result signal for presentation.

7. The method for safety management, identification, and detection of acupuncture needles as described in claim 6, wherein the computer computing system further comprises: a visual recognition module for receiving the first image signal, performing image preprocessing on the first image signal, and converting it into the second image signal for transmission; multiple feature matching units for receiving the second image signal, performing multiple image matching on the second image signal, and outputting multiple matching result signals based on the results of the multiple image matching; a matching scoring unit for receiving the multiple matching result signals, performing a matching score calculation on the multiple matching result signals, wherein the matching score calculation is based on multiple matching quantities and multiple average distances obtained from the multiple matching result signals to calculate multiple matching scores; and a voting mechanism unit for determining whether the matching score and the multiple matching quantities exceed a preset matching score threshold and a preset minimum matching quantity threshold, thereby outputting multiple valid images, and then averaging and weighting the multiple valid images to estimate the number and specific location of the multiple acupuncture needles on the patient's body.

8. The acupuncture needle safety management identification and detection method as described in claim 7, wherein if the matching score is less than the preset matching score threshold or the number of multiple matches is less than the preset minimum matching number threshold, it is an invalid vote; conversely, if the matching score is greater than the preset matching score threshold or the number of multiple matches is greater than the preset minimum matching number threshold, it is a valid vote.

9. The method for safety management and identification detection of acupuncture needles as described in claim 6, wherein the central processing unit determines that the number and specific location of the electronic tag signals corresponding to each of the multiple acupuncture needles are consistent with the number and specific location of the multiple acupuncture needles on the patient's body in the image comparison result signal, and the feedback value of the multiple acupuncture needles detected by the object detection in the third image signal is zero, indicating that the acupuncture needles on the patient have been removed and the judgment process can end; conversely, if the number and specific location of the electronic tag signals are consistent with the number and specific location of the multiple acupuncture needles in the image comparison result signal, but the feedback value of the multiple acupuncture needles detected by the object detection in the third image signal is not zero, indicating that the acupuncture needles on the patient have not been removed, a warning message is issued to remind a user.

10. The safety management identification and detection method for acupuncture needles as described in claim 6, wherein the central processing unit determines that the number and specific location of the electronic tag signal corresponding to each of the multiple acupuncture needles on the body are inconsistent with the number and specific location of the multiple acupuncture needles on the patient's body in the image comparison result signal, which indicates that the position of the patient's part captured by the image capturing device is incorrect and the image capturing device needs to be moved to the correct position.

Images