Intelligent electrocardiogram automatic detection and diagnosis device

Abstract

The invention relates to the technical field of electrocardiogram detection, and discloses an intelligent electrocardiogram automatic detection and diagnosis device which comprises a detection bed, a pair of wrist electric chucks, a pair of ankle electric chucks, an AI camera, a tool table, a mechanical arm and an intelligent detection module. According to the application, the detection process does not need manual intervention, and the patient is helped to successfully complete preparation work through voice guidance and visual indication; meanwhile, the position of the electric chuck can be automatically adjusted according to different heights, so that all patients can obtain accurate detection results; before each detection, cotton with alcohol is used for cleaning the skin surface, so that good contact between the probing electrode and the skin is ensured, and the quality of data acquisition is improved; based on an AI camera and an intelligent detection module, a mechanical arm is guided to accurately place a probing electrode at a designated position, and data distortion caused by position deviation is avoided; and through the design of the intelligent detection module, intelligent diagnosis is realized, and the influence of human subjective judgment is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of electrocardiogram detection, and in particular to an intelligent electrocardiogram automatic detection and diagnosis device. Background Art

[0002] Currently, electrocardiogram (ECG) testing, as an important means of evaluating cardiac electrical activity, still has significant technical limitations in its testing process. Traditional ECG testing equipment mainly relies on the operation of medical staff to complete the patient's testing process. Electrodes or electric clamps must be accurately installed on the patient's wrists, ankles, chest, etc., and the patient must be at rest to complete data collection. Although the conventional ECG examination time is short, its testing process has the following key issues:

[0003] (1) Electrode installation requires professional personnel, and patients must expose their body parts and maintain a fixed position, resulting in a time-consuming and cumbersome testing process. The lack of professional personnel, especially in emergency or remote areas, further limits the timeliness of testing. In addition, traditional equipment must be used in fixed medical facilities and cannot meet the real-time monitoring needs of home or mobile scenarios.

[0004] (2) Due to the busy hospital environment, medical staff may not be able to provide high-quality services to each patient in a timely manner. Conventional electrocardiograms in fixed medical facilities only record electrocardiogram signals for tens of seconds, making it difficult to capture occasional arrhythmias or transient cardiac events (such as paroxysmal atrial fibrillation, myocardial ischemia, etc.).

[0005] (3) ECG results need to be manually analyzed by doctors, which is easily influenced by subjective experience. Especially in complex cases, such as the identification of arrhythmia types, the misdiagnosis or missed diagnosis rate is high.

[0006] Therefore, the traditional electrocardiogram detection process has significant shortcomings in terms of operational convenience, detection reliability and diagnostic intelligence. There is an urgent need for an electrocardiogram detection technology that can improve detection efficiency and reduce misdiagnosis rate. Summary of the Invention

[0007] The purpose of this application is to provide an intelligent electrocardiogram automatic detection and diagnosis device to solve the technical problems raised in the above background technology.

[0008] To achieve the above objectives, the present application discloses the following technical solution: an intelligent electrocardiogram automatic detection and diagnosis device, characterized in that it includes:

[0009] 1. Testing bed;

[0010] A pair of wrist electric clamps are installed on the testing bed and can move along the length of the testing bed to clamp and fix the patient's wrists and collect corresponding electrocardiogram signals;

[0011] A pair of ankle electrical clamps, mounted on the testing bed, capable of moving along the length of the testing bed, clamping and fixing the patient's ankles, and collecting corresponding electrocardiogram signals;

[0012] an AI camera, disposed above the examination bed, for capturing a morphological image of the patient on the examination bed;

[0013] A tool table is provided near the testing bed and is provided with a plurality of probing electrodes and a storage box containing sterilized cotton balls;

[0014] a robotic arm, disposed near the inspection bed and the tool table;

[0015] An intelligent detection module is connected to the wrist electric clamp, the ankle electric clamp, the AI camera, the robotic arm, and the probing electrode, and is configured to:

[0016] In response to receiving the morphological image captured by the AI camera, controlling the wrist electric clamp and the ankle electric clamp to move and clamp and fix the patient's wrist and ankle;

[0017] In response to the morphological image captured by the AI camera, after the patient's wrist and ankle are clamped and fixed, control the robotic arm to pick up the disinfectant cotton ball and disinfect the target location on the patient's limb;

[0018] In response to the disinfection result of the target position, controlling the robotic arm to pick up the probing electrode and place it at the target position;

[0019] In response to the placement of the probing electrode, controlling the probing electrode to execute a fixation instruction so that the probing electrode is adsorbed and fixed to the target position;

[0020] In response to feedback signals from the wrist electrical clamp, the ankle electrical clamp and the plurality of probing electrodes, an electrocardiogram of the patient is generated, and a corresponding diagnosis result is generated based on the electrocardiogram.

[0021] Preferably, the intelligent electrocardiogram automatic detection and diagnosis device further comprises:

[0022] A glass door is installed above the detection bed and forms a detection space between the door and the detection bed when the door is closed.

[0023] A first mounting plate and a second mounting plate are formed on one side of the detection bed extending upward;

[0024] The AI camera is arranged at the front end of the first mounting plate, and the front end of the first mounting plate is located directly above the center of the detection bed;

[0025] A guide display screen is installed at the front end of the second mounting plate, and the front end of the second mounting plate is located above the position range corresponding to the patient's head on the detection bed; the guide display screen plays the patient operation guide.

[0026] The probing electrode comprises:

[0027] A Y-shaped silicone suction cup, wherein the Y-shaped silicone suction cup is provided with a ventilation cavity, and the ventilation cavity is connected to the compressor through an air tube;

[0028] an electrode holder disposed in the ventilation cavity;

[0029] a conductive head disposed in the electrode holder, the top end of the conductive head being connected to a signal output wire, the signal output wire being connected to the intelligent detection module;

[0030] The Y-shaped silicone suction cup is sleeved on the outside of the electrode seat, and the top side of the Y-shaped silicone suction cup is in contact with the outer wall of the electrode seat; there is a gap between the part of the electrode seat located in the ventilation cavity and the Y-shaped silicone suction cup; the electrode seat is provided with a through limiting cavity and a compression cavity; the top end of the conductive head extends into the limiting cavity, and the bottom end of the conductive head extends into the compression cavity and an electrode sheet is connected to its bottom end; a compression spring is provided in the compression cavity, and the bottom end of the compression spring extends into the inner cavity of the Y-shaped silicone suction cup and is connected to the top of the electrode sheet; a rubber sleeve is provided between the bottom end of the electrode seat and the inner cavity of the Y-shaped silicone suction cup, and the rubber sleeve is used to isolate the compression cavity and the ventilation cavity for gas exchange.

[0031] Preferably, the method of generating an electrocardiogram of the patient in response to feedback signals from the wrist electric clamp, the ankle electric clamp, and the plurality of probing electrodes, and generating a corresponding diagnosis result based on the electrocardiogram, specifically includes:

[0032] D1: receiving feedback signals from a pair of wrist electric clamps, a pair of ankle electric clamps and at least six detection electrodes;

[0033] D2: Mark each feedback signal;

[0034] D3: Calculate the arrhythmia risk probability Risk, which is calculated using the following formula:

[0035]

[0036] Among them, ω i is the weight coefficient of the i-th lead, ΔST i is the ST segment deviation of lead i, RR i is the RR interval of lead i, μRR is the mean RR interval of the reference population, σ RR is the standard deviation of the RR interval of the reference population, β is the HRV sensitivity coefficient, γ is the motion artifact weighting factor, and Motion artifact is the root mean square of the signal difference between leads;

[0037] D4: Generate a diagnosis result based on the result of the arrhythmia risk probability.

[0038] Preferably, the HRV sensitivity coefficient is configured to be dynamically adjusted according to the patient's age, and is specifically defined as:

[0039]

[0040] Preferably, the motion artifact weighting factor is configured to be optimized in real time by a least squares method, and is specifically defined as:

[0041]

[0042] Among them, ECG amplitude is the peak amplitude of the current ECG signal, and ε is a small constant to prevent the denominator from being zero, ε=1e-6.

[0043] Preferably, the weight coefficient of the i-th lead is determined by the following formula:

[0044]

[0045] Among them, Transformer 输出i is the weighted score of the multi-head attention mechanism for the i-th lead, calculated by the self-attention mechanism.

[0046] Preferably, the diagnosis results include a low-risk arrhythmia result, a medium-risk arrhythmia result, and a high-risk arrhythmia result.

[0047] Preferably, the diagnosis result is determined based on a hierarchical early warning mechanism, which is specifically:

[0048]

[0049] Among them, μ Risk is the mean risk value of the reference population, σ Risk is the standard deviation of the risk value of the reference population.

[0050] The intelligent electrocardiogram automatic detection and diagnosis device of the present application has the following technical effects:

[0051] From the moment the patient enters the testing room to the completion of the entire testing process, no human intervention is required. The device can complete the process independently, helping the patient complete the preparation work smoothly through voice guidance and visual instructions. At the same time, the electric chuck can automatically adjust its position according to different heights to ensure that all patients can obtain accurate test results. Before each test, cotton with alcohol is used to clean the skin surface to ensure good contact between the probing electrode and the skin, thereby improving the quality of data acquisition. Based on the AI camera and intelligent detection module, the robotic arm is guided to place the probing electrode precisely at the specified position, avoiding data distortion caused by position deviation. Through the design of the intelligent detection module, intelligent diagnosis is achieved, reducing the influence of human subjective judgment. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 A schematic structural diagram of an intelligent electrocardiogram automatic detection and diagnosis device without a glass door provided in an embodiment of the present application;

[0054] Figure 2 A schematic structural diagram of an intelligent electrocardiogram automatic detection and diagnosis device with a glass door provided in an embodiment of the present application;

[0055] Figure 3 Schematic diagram of the installation position of the AI camera and guidance display provided in the embodiment of the present application;

[0056] Figure 4 A schematic diagram of the structure of the electrically adjustable clamp provided in an embodiment of the present application;

[0057] Figure 5 A schematic diagram of the structure of the mechanical clamping provided in an embodiment of the present application;

[0058] Figure 6 A three-dimensional schematic diagram of a probing electrode provided in an embodiment of the present application;

[0059] Figure 7 A cross-sectional view of a probing electrode provided in an embodiment of the present application.

[0060] Figure numerals: 1, test bed; 2, wrist electric clamp; 3, ankle electric clamp; 4, AI camera; 5, tool table; 6, exploration electrode; 7, storage box; 8, robotic arm; 9, glass hatch; 10, guidance display; 11, first mounting plate; 12, second mounting plate; 13, rack; 14, moving gear; 15, moving reduction motor; 16, sliding mounting seat; 17, motor mounting plate; 18, L-shaped mounting plate; 19, clamping gear shaft; 20, electric clamp; 2 1. Clamping gear; 22. Driving gear; 23. Driving reduction motor; 24. Driving gear shaft; 25. Rubber pad; 26. Clamping gripper; 27. First connecting rod; 28. Second connecting rod; 29. Connecting slider; 30. Electric cylinder; 31. Base; 32. Trash can; 601. Y-shaped silicone suction cup; 602. Air pipe; 603. Electrode holder; 604. Conductive head; 605. Signal output wire; 606. Electrode sheet; 607. Compression spring; 608. Rubber sleeve. DETAILED DESCRIPTION

[0061] The following is a clear and complete description of the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0062] In this document, the term "comprising" is intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0063] This embodiment provides a Figure 1 The intelligent automatic electrocardiogram detection and diagnosis device shown includes: a detection bed 1, which is generally arranged in a detection room and is used for a patient to lie flat on it for detection;

[0064] A pair of wrist electric clamps 2 are installed on the testing bed 1 and can move along the length direction of the testing bed 1 (i.e., move along the direction of the patient's body to accommodate patients of different heights), clamp the patient's wrist and collect the corresponding ECG signal;

[0065] A pair of ankle electrical clamps 3 are mounted on the testing bed 1 and are capable of moving along the length of the testing bed to clamp and fix the patient's ankles and collect corresponding electrocardiogram signals;

[0066] The AI camera 4 is provided above the testing bed 1 and is used to collect a morphological image of the patient on the testing bed 1, wherein the morphological image specifically includes the patient's body posture when lying flat, and an image of the patient's limbs, which is used to determine the target position of the electrode to be attached;

[0067] A tool table 5 is provided near the detection bed 1 and is provided with a plurality of probing electrodes 6 and a storage box 7 containing sterilized cotton balls;

[0068] A robotic arm 8 is provided close to the testing bed 1 and the tool table 5;

[0069] The intelligent detection module is controlled and connected with the wrist electric clamp 2, the ankle electric clamp 3, the AI camera 4, the robotic arm 8 and the probing electrode 6, and is configured as follows:

[0070] In response to receiving the morphological image captured by the AI camera 4, controlling the wrist electric clamp 2 and the ankle electric clamp 3 to move and clamp and fix the patient's wrist and ankle;

[0071] In response to the morphological image captured by the AI camera 4, after the patient's wrist and ankle are clamped and fixed, the robotic arm 8 is controlled to pick up the disinfection cotton ball and disinfect the target position on the patient's limb; after the disinfection is completed, the used disinfection cotton ball is thrown into the trash can 26;

[0072] In response to the disinfection result of the target position (i.e., after the disinfection is completed), controlling the robotic arm 8 to pick up the probing electrode 6 and place it at the target position;

[0073] In response to the placement of the probing electrode 6 (i.e., after the probing electrode 6 is placed at the target position, at this time, the probing electrode 6 is not adsorbed and fixed to the patient's skin), controlling the probing electrode 6 to execute a fixation instruction so that the probing electrode 6 is adsorbed and fixed to the target position, so as to ensure that the probing electrode 6 can stably detect and avoid the occurrence of situations such as the probing electrode 6 falling off during the detection process;

[0074] In response to the feedback signals from the wrist electric clamp 2 , the ankle electric clamp 3 and the plurality of probing electrodes 6 , an electrocardiogram of the patient is generated, and a corresponding diagnosis result is generated based on the electrocardiogram.

[0075] In this embodiment, if Figure 2 As shown, the intelligent electrocardiogram automatic detection and diagnosis device also includes:

[0076] The glass cabin door 9 is installed above the detection bed 1, and forms a detection space between the glass cabin door 9 and the detection bed after closing. It is feasible that the glass cabin door 9 is arc-shaped, and after the patient lies flat on the detection bed 1, the glass cabin door 9 is closed. After the patient detection is completed, the glass cabin door 9 can be opened by means other than but not limited to voice commands. It is also feasible that the glass cabin door 9 is a segmented cabin door, with one section corresponding to the upper part of the human body and another section corresponding to the lower part of the human body. In a feasible embodiment, the detection room is equipped with a personnel sensing module, which can be any technology in the existing technology for sensing the entry of personnel. The personnel sensing module is communicatively connected to the intelligent detection module. After the detection room senses the entry of personnel, the intelligent detection module responds to the detection signal fed back by the personnel sensing module and controls the automatic opening of the glass cabin door 9.

[0077] In this embodiment, if Figure 3 As shown, one side of the detection bed 1 extends upward to form a first mounting plate 11 and a second mounting plate 12. Corresponding to the shape of the glass cabin door 9, the first mounting plate 11 and the second mounting plate 12 are also arc-shaped structures. The AI camera 4 is arranged at the front end of the first mounting plate 11, and the front end of the first mounting plate 11 is located directly above the center of the detection bed 1 to ensure the image acquisition range of the AI camera 4. It is feasible that the AI camera 4 can be any one of the existing technologies, such as but not limited to a camera structure with 360° rotation. Furthermore, in order to ensure the comprehensiveness and efficiency of the test results, a second AI camera 4 can be set at the position corresponding to the patient's foot and the image processing technology in the existing technology can be used to achieve the combination of different areas to achieve the detection purpose through segmented detection. This text will not elaborate on this. A guidance display screen 10 is installed at the front end of the second mounting plate 12. The front end of the second mounting plate 12 is located above the position range corresponding to the patient's head on the detection bed 1. The guidance display screen 10 is used to play the patient operation guide. In this way, after the patient lies on the detection bed 1, he can follow the simple and easy-to-understand operation guide displayed on the guidance display screen 10, unbutton the shirt based on the image and / or voice prompts and adjust the limb posture / position according to the instructions.

[0078] After the patient has finished lying flat, the AI camera 4 detects the result of lying flat. The intelligent detection module determines that the patient's lying posture is correct based on the image analysis technology, and then determines the target position according to the user's morphological image. Then, the wrist electric clamp 2 and the ankle electric clamp 3 are controlled to move and clamp the patient's wrist and ankle. In this embodiment, the mechanical structure of the wrist electric clamp 2 and the ankle electric clamp 3 is the same, and both can be understood as an electrically adjustable electric clamp. The difference is that the installation positions of the two are different. The structure of this electrically adjustable electric clamp is as follows: Figure 4As shown, specifically, this type of electric adjustment clamp includes: a rack 13 with teeth pointing downward, a moving gear 14 meshing with the teeth of the rack 13, a moving reduction motor 15 driving the moving gear 14 to rotate, a sliding mounting seat 16 slidingly mounted on the top of the rack 13, a motor mounting plate 17 fixed on the side of the sliding mounting seat 16, a pair of L-shaped mounting plates 18 fixedly mounted on the top and side of the sliding mounting seat 16, a pair of clamping gear shafts 19 mounted on the pair of L-shaped mounting plates 18 and located between the top side parts of the sliding mounting seat 16, two electric clamps 20 respectively sleeved on the pair of clamping gear shafts 19, a mounting groove opened on the electric clamp 20, a clamping gear 21 installed in the mounting groove, a driving gear 22 meshing with one of the clamping gears 21, a driving reduction motor 23 driving the driving gear 22 to rotate, and a wire connected to the end of the electric clamp 20. The motor base of the mobile reduction motor 15 is fixedly connected to the motor mounting plate 17. The motor mounting plate 17 is perpendicular to the portion of the pair of L-shaped mounting plates 18 located on the side of the sliding mounting seat 16. Two clamping gears 21 are meshed. A driving gear shaft 24 is installed between the portion of the pair of L-shaped mounting plates 18 located on the side of the sliding mounting seat 16. One end of the clamping gear shaft 24 is rotatably connected to one L-shaped mounting plate 18 via a bearing, and the other end passes through the other L-shaped mounting plate 18 and is connected to the driving end of the driving reduction motor 23. The motor base of the driving reduction motor 23 is correspondingly fixed to the other L-shaped mounting plate 18. When the driving reduction motor 23 is output, the driving gear 22 rotates, driving the clamping gear 21 meshed with it to rotate, thereby rotating the other clamping gear 21, thereby driving the two electric clamps 20 to perform the "clamping" or "loosening" action. When the mobile reduction motor 15 is output, the mobile gear 14 rotates and moves along the rack 13 under its meshing with the rack 13 to achieve "forward" or "backward" movement. When the wire connected to the end of the electric clamp 20 is energized, current passes through the patient's hand or foot, and the electrical signal is transmitted to the intelligent detection module for analysis through the guide.

[0079] In this embodiment, if Figure 5 As shown, the robotic arm 8 includes a robotic arm body (which can be any multi-degree-of-freedom robotic arm structure in the prior art) and a mechanical clamp installed at the end of the robotic arm. The mechanical clamp is mainly composed of a rubber pad 25, a clamping gripper 26, a first connecting rod 27, a second connecting rod 28, a connecting slider 29, an electric cylinder 30, and a base 31. The base 31 is installed at the end of the robotic arm body. The end of the electric cylinder 30 is connected to the base 31, and the front end is connected to the connecting slider 29. The clamping gripper 26 is hingedly connected based on the first connecting rod 27 and the second connecting rod 28 as shown in the figure. When working, the electric cylinder 30 pushes the connecting slider 29, which can enable the clamping gripper 26 to grasp and grasp the exploration electrode 6 and the disinfection cotton ball.

[0080] In this embodiment, if Figure 6 and 7As shown, the exploration electrode 6 includes: a Y-shaped silicone suction cup 601, which is provided with a ventilation cavity (i.e., a cavity is provided inside), and the ventilation cavity is connected to the compressor through an air tube 602; an electrode seat 603 provided in the ventilation cavity; a conductive head 604 provided in the electrode seat 603, and the top of the conductive head 604 is connected to a signal output wire 605, and the signal output wire 605 is connected to the intelligent detection module. The Y-shaped silicone suction cup sleeve 601 is provided on the outside of the electrode seat 603, and the top side of the Y-shaped silicone suction cup 601 is in contact with the outer wall of the electrode seat 603 to ensure that the inside of the ventilation cavity is sealed. There is a gap between the portion of the electrode seat 603 located in the ventilation cavity and the Y-shaped silicone suction cup 601 to ensure that the air tube 602 can smoothly supply or extract air into the ventilation cavity. The electrode seat 603 is provided with a through limiting cavity and a compression cavity (in the direction of the figure, the limiting cavity is on the top and the compression cavity is on the bottom). The top end of the conductive head 604 extends into the limiting cavity. The limiting cavity is mainly used to limit the movement of the conductive head 604. The bottom end of the conductive head 604 extends into the compression cavity and its bottom end is connected to an electrode sheet 606. A compression spring 607 is provided in the compression chamber, and the compression chamber is mainly used to accommodate the compression spring 607. The bottom end of the compression spring 607 extends into the inner cavity of the Y-shaped silicone suction cup 601 and is connected to the top of the electrode piece 607. A rubber sleeve 608 is provided between the bottom end of the electrode seat 603 and the inner cavity of the Y-shaped silicone suction cup 601. The rubber sleeve 608 is used to isolate the compression chamber from the ventilation chamber for gas exchange. The rubber sleeve 608 wraps and isolates the electrode piece 606 and the compression spring 607, that is, the outer wall of the rubber sleeve 608 and the Y-shaped silicone suction cup 601 form an enclosed space for the entire ventilation chamber, thereby ensuring that the Y-shaped silicone suction cup 601, driven by the compressor, can form a negative pressure in the ventilation chamber when air is extracted. Under the setting of the compression spring 607, it is ensured that the electrode piece 606 is firmly adsorbed at the target position, and the ventilation chamber is relieved of the negative pressure when air is supplied. After the power is turned on, the current flows through the patient's body, and the signal output wire 605 outputs the electrical signal to the intelligent detection module for analysis.

[0081] In this embodiment, the patient's electrocardiogram is generated in response to the feedback signals from the wrist electrical clamp, the ankle electrical clamp, and the plurality of probing electrodes, and a corresponding diagnosis result is generated based on the electrocardiogram, specifically including:

[0082] D1: Receive feedback signals from a pair of wrist electric clamps, a pair of ankle electric clamps and at least six detection electrodes.

[0083] D2: Mark each feedback signal.

[0084] D3: Calculate the arrhythmia risk probability Risk, which is calculated using the following formula:

[0085]

[0086] Among them, ω i is the weight coefficient of the i-th lead, ΔST i is the ST segment deviation of lead i (unit: mV), reflecting myocardial ischemia or damage, RR i is the RR interval of the i-th lead (unit: ms), reflecting the heart rate variability (HRV), μ RR is the mean RR interval of the reference population, obtained through statistics of large-scale healthy population data, σ RR is the standard deviation of the RR interval of the reference population, representing the normal heart rate fluctuation range, β is the HRV sensitivity coefficient, γ is the motion artifact weighting factor, and Motion artifact is the motion artifact intensity, which is calculated by the root mean square of the signal difference between leads; where Dynamically adjust the weight so that the contribution of HRV to risk increases nonlinearly with its value. artifact The introduction of can realize the compensation of motion artifacts, thereby suppressing the signal noise caused by motion in real time and improving the robustness. This design eliminates the dimensional differences in individual RR intervals, unifying them into a standard normal distribution space and preventing the excessive impact of single-lead outliers on overall risk. It also improves the ability to identify arrhythmias (such as premature ventricular contractions and atrial fibrillation) by simultaneously capturing abnormalities in both the ST segment and the RR interval in the time domain.

[0087] D4: Generate a diagnosis result based on the result of the arrhythmia risk probability.

[0088] Furthermore, the HRV sensitivity coefficient is configured to be dynamically adjusted according to the patient's age, and is specifically defined as:

[0089] As known to those skilled in the art, after the age of 40, the HRV of patients decreases with age, and the β value increases to enhance the sensitivity to HRV abnormalities, compensate for physiological differences, and thus make the diagnostic process differentially sensitive to HRV abnormalities in different age groups and reduce the false alarm rate.

[0090] In this embodiment, the motion artifact weighting factor is configured to be optimized in real time by the least squares method, and is specifically defined as:

[0091]

[0092] Among them, ECG amplitudeis the peak amplitude of the current ECG signal (in mV), used to normalize the artifact weight. ε is a small constant that prevents the denominator from reaching zero, ensuring numerical stability (ε = 1e-6). When the ECG signal is strong (such as in the resting state), γ is reduced, thereby reducing the artifact weight. During exercise, the ECG amplitude may decrease, and γ is increased to compensate for the artifact. This effectively reduces the interference of motion artifacts on risk prediction.

[0093] In this embodiment, the weight coefficient of the i-th lead is determined by the following formula:

[0094]

[0095] Among them, Transformer 输出i The multi-head attention mechanism is used to calculate the weighted score of the i-th lead, which is calculated through the self-attention mechanism. The Transformer's multi-head attention mechanism assigns higher weights to key leads (such as ST segment changes in V1-V6), thereby improving sensitivity to abnormal areas.

[0096] In this embodiment, the diagnosis results include a low risk result for arrhythmia, a medium risk result for arrhythmia, and a high risk result for arrhythmia. It is feasible that the diagnosis results are determined based on a graded warning mechanism, and the graded warning mechanism is specifically:

[0097]

[0098] Among them, μ Risk is the mean risk value of the reference population, σ Risk is the standard deviation of the risk value of the reference population.

[0099] To sum up, the intelligent automatic electrocardiogram detection and diagnosis device of this embodiment does not require human intervention from the moment the patient enters the detection room to the completion of the entire detection process. The device can complete the process independently and help the patient complete the preparation work smoothly through voice guidance and visual instructions. At the same time, the electric clamp can automatically adjust its position according to different heights to ensure that all patients can obtain accurate test results. Before each test, cotton with alcohol is used to clean the skin surface to ensure good contact between the probing electrode and the skin, thereby improving the quality of data acquisition. Based on the AI camera and intelligent detection module, the robotic arm is guided to accurately place the probing electrode in the specified position, avoiding data distortion caused by position deviation. Through the design of the intelligent detection module, intelligent diagnosis is realized, and the influence of human subjective judgment is reduced.

[0100] In the embodiments provided herein, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a processor, a controller, a microcontroller, a microprocessor, other electronic units designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the processes of the embodiments can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. Computer-readable storage media include computer storage media and communication media, wherein the communication media include any medium that facilitates the transmission of a computer program from one place to another. The storage medium can be any available medium that a computer can access. The computer-readable storage medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0101] Finally, it should be noted that the above is only a preferred embodiment of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent replacements for some of the technical features therein. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. An intelligent electrocardiogram automatic detection and diagnosis device, characterized in that: include:

1. Testing bed; A pair of wrist electric clamps are installed on the testing bed and can move along the length of the testing bed to clamp and fix the patient's wrists and collect corresponding electrocardiogram signals; A pair of ankle electrical clamps, mounted on the testing bed, capable of moving along the length of the testing bed, clamping and fixing the patient's ankles, and collecting corresponding electrocardiogram signals; an AI camera, disposed above the examination bed, for capturing a morphological image of the patient on the examination bed; A tool table is provided near the testing bed and is provided with a plurality of probing electrodes and a storage box containing sterilized cotton balls; a robotic arm, disposed near the inspection bed and the tool table; An intelligent detection module is connected to the wrist electric clamp, the ankle electric clamp, the AI camera, the robotic arm, and the probing electrode, and is configured to: In response to receiving the morphological image captured by the AI camera, controlling the wrist electric clamp and the ankle electric clamp to move and clamp and fix the patient's wrist and ankle; In response to the morphological image captured by the AI camera, after the patient's wrist and ankle are clamped and fixed, control the robotic arm to pick up the disinfectant cotton ball and disinfect the target location on the patient's limb; In response to the disinfection result of the target position, controlling the robotic arm to pick up the probing electrode and place it at the target position; In response to the placement of the probing electrode, controlling the probing electrode to execute a fixation instruction so that the probing electrode is adsorbed and fixed to the target position; In response to feedback signals from the wrist electrical clamp, the ankle electrical clamp and the plurality of probing electrodes, an electrocardiogram of the patient is generated, and a corresponding diagnosis result is generated based on the electrocardiogram.

2. The intelligent electrocardiogram automatic detection and diagnosis device according to claim 1, characterized in that: The intelligent electrocardiogram automatic detection and diagnosis device also includes: A glass door is installed above the detection bed and forms a detection space between the door and the detection bed when the door is closed.

3. The intelligent electrocardiogram automatic detection and diagnosis device according to claim 1, characterized in that: A first mounting plate and a second mounting plate are formed on one side of the detection bed extending upward; The AI camera is arranged at the front end of the first mounting plate, and the front end of the first mounting plate is located directly above the center of the detection bed; A guide display screen is installed at the front end of the second mounting plate, and the front end of the second mounting plate is located above the position range corresponding to the patient's head on the testing bed; The guidance display screen plays the patient operation guide for the user.

4. The intelligent electrocardiogram automatic detection and diagnosis device according to claim 1, characterized in that: The probing electrode comprises: A Y-shaped silicone suction cup, wherein the Y-shaped silicone suction cup is provided with a ventilation cavity, and the ventilation cavity is connected to the compressor through an air tube; an electrode holder disposed in the ventilation cavity; a conductive head disposed in the electrode holder, the top end of the conductive head being connected to a signal output wire, the signal output wire being connected to the intelligent detection module; The Y-shaped silicone suction cup is sleeved on the outside of the electrode seat, and the top side of the Y-shaped silicone suction cup is in contact with the outer wall of the electrode seat; there is a gap between the part of the electrode seat located in the ventilation cavity and the Y-shaped silicone suction cup; the electrode seat is provided with a through limiting cavity and a compression cavity; the top end of the conductive head extends into the limiting cavity, and the bottom end of the conductive head extends into the compression cavity and an electrode sheet is connected to its bottom end; a compression spring is provided in the compression cavity, and the bottom end of the compression spring extends into the inner cavity of the Y-shaped silicone suction cup and is connected to the top of the electrode sheet; a rubber sleeve is provided between the bottom end of the electrode seat and the inner cavity of the Y-shaped silicone suction cup, and the rubber sleeve is used to isolate the compression cavity and the ventilation cavity for gas exchange.

5. The intelligent electrocardiogram automatic detection and diagnosis device according to claim 1, characterized in that: The generating of the patient's electrocardiogram in response to the feedback signals from the wrist electric clamp, the ankle electric clamp, and the plurality of the probing electrodes, and generating a corresponding diagnosis result based on the electrocardiogram, specifically includes: D1: receiving feedback signals from a pair of wrist electric clamps, a pair of ankle electric clamps and at least six detection electrodes; D2: Mark each feedback signal; D3: Calculate the arrhythmia risk probability Risk, which is calculated using the following formula: Among them, ω i is the weight coefficient of the i-th lead, ΔST i is the ST segment deviation of lead i, RR i is the RR interval of lead i, μ RR is the mean RR interval of the reference population, σ RR is the standard deviation of the RR interval of the reference population, β is the HRV sensitivity coefficient, γ is the motion artifact weighting factor, and Motion artifact is the intensity of motion artifact; D4: Generate a diagnosis result based on the result of the arrhythmia risk probability.

6. The intelligent electrocardiogram automatic detection and diagnosis device according to claim 5, characterized in that: The HRV sensitivity coefficient is configured to dynamically adjust based on the patient's age and is defined as:

7. The intelligent electrocardiogram automatic detection and diagnosis device according to claim 5, characterized in that: The motion artifact weighting factor is configured to be optimized in real time by the least squares method and is specifically defined as: Among them, ECG amplitude is the peak amplitude of the current ECG signal, and ε is a small constant to prevent the denominator from being zero, ε=1e-6.

8. The intelligent electrocardiogram automatic detection and diagnosis device according to claim 5, characterized in that: The weight coefficient of the i-th lead is determined by the following formula: Among them, Transformer 输出i is the weighted score of the multi-head attention mechanism for the i-th lead, calculated by the self-attention mechanism.

9. The intelligent electrocardiogram automatic detection and diagnosis device according to claim 5, characterized in that: The diagnosis results include a low risk result for arrhythmia, a medium risk result for arrhythmia, and a high risk result for arrhythmia.

10. The intelligent electrocardiogram automatic detection and diagnosis device according to claim 9, characterized in that: The diagnosis result is determined based on a hierarchical early warning mechanism, which is specifically: Among them, μ Risk is the mean risk value of the reference population, σ Risk is the standard deviation of the risk value of the reference population.

Images