Identification method for electrode wearing position of wireless mobile monitor

Abstract

The invention provides an automatic identification method for a wearing position of a wireless mobile monitor. The automatic identification method comprises the following steps: step 1, taking calibrated prior data as a database; 2, a terahertz imaging system is adopted to obtain a chest skeleton two-dimensional image; step 3, acquiring a chest visible light image through a monocular camera; 4, registering the visible light image and the terahertz image marked with the reference point; and 5, setting a connecting line for the reference points to calculate the positions of the V1-V6 electrodes. The beneficial effects of the invention are that the auxiliary positioning of the electrode problem is realized, and the mobile monitoring equipment can be correctly worn by a user without the guidance of a professional background person.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless monitoring, and in particular to a method for identifying the electrode wearing positions of a wireless mobile monitor. Background Art

[0002] Currently, monitors in hospitals are such that due to the professionalism of the monitors, all wearing devices must be worn by specialized medical staff such as nurses. However, the number of monitors in hospitals is relatively small and cannot meet the needs of the large chronic disease population. With the development of monitor technology and the reduction of costs, there are also various home monitors outside the hospital, enabling users to perform physical sign monitoring through home monitors outside the hospital. However, home monitors also require a certain amount of professional knowledge, and ordinary users do not know how to wear the devices and sensors. Although home monitors are equipped with instruction manuals or video tutorials, it is somewhat difficult for ordinary users to understand. After wearing, they cannot determine whether the wearing is correct. If the wearing is incorrect, it will lead to sensing errors, inaccurate collected data, affect the final result, and cause the monitor to malfunction. Summary of the Invention

[0003] Aiming at the problem that it is difficult to wear accurately when self-wearing the chest leads of current mobile monitors, and to solve the defect that it is difficult for users to wear electrodes or unable to determine whether the electrodes are worn correctly, resulting in abnormal data obtained by mobile monitoring devices and unable to output effective information, it is achieved through the following technical solutions.

[0004] A method for identifying the electrode wearing positions of a wireless mobile monitor includes the following steps: Step 1: Use the pre-calibrated prior data as a database; Step 2: Use a terahertz imaging system to obtain a two-dimensional image of the chest bones; Step 3: Collect a visible light image of the chest through a monocular camera; Step 4: Register the visible light image and the terahertz image with marked reference points; Step 5: Set a connection line for the reference reference points to calculate the positions of the v1-v6 electrodes.

[0005] The beneficial effect of the present invention is: It realizes the problem of assisting in positioning electrodes, and users can correctly wear mobile monitoring devices by themselves without the guidance of professionals with a professional background. Brief Description of the Drawings

[0006] Figure 1 It is a flowchart of the method for identifying the wearing positions of the monitor in the embodiment of the present invention.

[0007] Figure 2 It is a schematic diagram of the electrode positions based on prior data in the embodiment of the present invention.

[0008] Figure 3 It is a schematic diagram for setting reference benchmark points of chest bones in an embodiment of the present invention.

[0009] Figure 4 It is a schematic diagram for setting connection lines of reference benchmark points of chest bones in an embodiment of the present invention.

[0010] Figure 5 It is a schematic diagram for guiding the path and direction of the real electrode patch of chest bones to the correct position in an embodiment of the present invention.

[0011] Figure 6 It is Figure 5 a schematic diagram for guiding the real electrode patch to approach the correct position in

[0012] Figure 7 It is Figure 5 a schematic diagram showing that the real electrode patch in is in the correct position. Detailed implementation manners

[0013] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, the present invention can be implemented in many different ways defined and covered by the claims.

[0014] The embodiments of the present invention are applicable to the wearing of chest leads, especially for wireless mobile monitors.

[0015] A wireless mobile monitor is a mobile and portable physiological sign monitor, which is a dynamic monitor with mobility and wireless continuous monitoring, and is a component of the in-hospital and out-of-hospital medical systems.

[0016] ECG monitoring is a form of wireless mobile monitor, which wears multiple wired or wireless ECG patches to the correct positions on the chest.

[0017] The embodiments of the present invention take the chest six-lead ECG patch as an example.

[0018] As Figure 1 shown, a method for identifying the electrode wearing positions of a wireless mobile monitor includes the following steps: S1. Using the calibrated prior data as a database; S2. Obtaining a two-dimensional image of chest bones by using a terahertz imaging system; S3. Collecting a visible light image of the chest through a monocular camera; S4. Registering the visible light image and the terahertz image with marked benchmark points; S5. Calculating the positions of v1-v6 electrodes by setting connection lines for reference benchmark points.

[0019] As Figures 2 - 3As shown in the figure, in the first step, use the calibrated prior data as a database. The images in the database are terahertz images of the human chest bones, and multiple reference benchmark points are calibrated in the images. Among them, the first reference benchmark point 10 is located at the middle position of the connection between the left and right collarbones, the second reference benchmark point 20 is located at the connection between the lower edge of the sternal body and the xiphoid process, and the third reference benchmark point 30 is located on the middle line under the left armpit of the body and is on the same horizontal line as the second reference benchmark point.

[0020] In the second step, use a terahertz imaging system to obtain a two-dimensional image of the chest bones.

[0021] The terahertz imaging system includes an optical frequency comb generation module, a phase modulator, an intensity modulator, a radio frequency signal source, a lock-in amplifier, an intermediate frequency electrical signal processing module, an antenna, and a parabolic mirror device. Adjust the wavelength interval between the two lasers in the optical frequency comb module so that it is about 100 GHz. The two laser wavelengths are coupled into a single optical fiber by an optical coupler and input into the phase and intensity modulators. The radio frequency signal source generates an 8 GHz sine signal and enters the phase and intensity modulators. The adjusted optical frequency comb module generates an optical beat frequency signal of 90 - 120 G and is converted into a terahertz signal for emission to form a multi-color light source. The light source is accurately irradiated onto a point on the chest through a parabolic mirror. Then, it is converged by a parabolic mirror onto a terahertz detector and converted into an intermediate frequency electrical signal. After the lock-in amplifier performs lock-in amplification, the signal intensity is read out through the processing module, that is, the corresponding intensity of a point in the chest area. A complete image of the chest bones is obtained by moving the optical frequency comb module to receive different intensity signals.

[0022] The terahertz imaging system is a separate device, and the processed images are stored in a wireless mobile monitor for backup. Or they are stored in a server and downloaded from the server when the wireless mobile monitor is in use.

[0023] In the third step, collect visible light images of the chest using a monocular camera. And use the prior data calibrated in the first step to train a benchmark point positioning model. Take the collected terahertz images as the model input, and the model automatically marks the reference benchmark points on the terahertz images. The reference steps are as follows: Prepare the camera and equipment: Select a monocular camera suitable for collecting visible light images of the chest and ensure that the camera has appropriate lenses and settings. In addition, prepare necessary equipment such as camera brackets and tripods to ensure the stability of shooting and the image quality.

[0024] Determine the shooting position: Select a position suitable for shooting visible light images of the chest. Usually, the shooting position should be on the chest horizontal line to ensure that the entire chest area can be captured. If necessary, adjust the height and angle of the camera to obtain the best shooting effect.

[0025] Set camera parameters: According to the shooting requirements and targets, adjust the parameters of the camera, such as exposure time, aperture size, ISO sensitivity, etc. Ensure that appropriate parameters are set to obtain clear and high-contrast visible light images of the chest.

[0026] Take images: Align the camera with the chest area and press the shutter button to take pictures. You can take multiple images continuously for subsequent selection and use.

[0027] Store and process images: Store the captured images in the wireless mobile monitor for subsequent analysis and processing. Call the image processing software in the wireless mobile monitor to perform operations such as preprocessing, enhancement, and annotation on the images to meet the needs of further analysis and diagnosis.

[0028] Fourth step, register the visible light image and the terahertz image with marked reference points. Obtain the consistency (PC) map of the two images through wavelet transform, and calculate corner points and edge points based on the PC map; use the minimum moment method to detect corner points and the FAST algorithm to perform edge detection on the maximum moment. Construct the MIM (maximum index map) to describe features, use the SIFT method to divide the image into small regions, and calculate the feature vectors of the small regions to form feature descriptors. Perform outlier filtering through nearest neighbor matching to select correct inliers, and calculate the Homography transformation matrix to map the terahertz image to the coordinate system of the visible light image to achieve registration.

[0029] In the embodiment of the present invention, a handheld terahertz imager can be used to align with the human chest to obtain a terahertz image of the chest, and the terahertz image is transmitted to the wireless mobile monitor.

[0030] Fifth step, set the connection lines for the reference reference points to calculate the positions of the v1-v6 electrodes.

[0031] As Figure 4 shown, based on the connection lines to calculate the electrode patch positions, the first reference reference point 10 and the second reference reference point 20 are used as the longitudinal connection line with a length of A, and the second reference reference point 20 and the third reference reference point 30 are used as the transverse connection line with a length of B.

[0032] The V2 electrode is located at a distance of A / 4 from the transverse connection line and a distance of B / 7 from the longitudinal connection line; The V1 electrode is symmetric with the V2 electrode with respect to the longitudinal connection line, at a distance of B / 7 from the longitudinal connection line and a distance of A / 4 from the transverse connection line; the V4 electrode is on the transverse connection line and at a distance of 3B / 7 from the second reference reference point; The V5 electrode is on the transverse connection line and at a distance of 5B / 7 from the second reference reference point; The V3 electrode is in the middle of the V2 electrode and the V4 electrode, at a distance of A / 8 from the transverse connection line and a distance of 2B / 7 from the longitudinal connection line; The V6 electrode is located at the third reference band and is at a distance of A / 8 from the horizontal connection line.

[0033] As Figures 5 - 7 shown, in the sixth step, use the target detection algorithm to detect whether there are real electrode patches on the chest. When there are real electrode patches, use the nearest neighbor search to match their corresponding correct electrode points. Each group of real chest electrode patches and the correct electrode points are used as a matching pair. Each matching pair generates a path guiding arrow 40 on the registration image. The direction and length of the arrow respectively represent the direction and distance from the real chest electrode to the correct electrode. The longer the arrow distance, the greater the deviation, and the shorter the distance, the closer to the correct electrode position, as Figures 5 - 6 shown. When the length of the guiding arrow becomes 0 and the shape becomes a dot 50, it means that the electrode patch is worn correctly, as Figure 7 shown.

[0034] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for identifying the electrode wearing positions of a wireless mobile monitor, characterized in that, It includes the following steps: Step 1: Using the calibrated prior data as a database, the first reference datum point is located at the middle position of the connection between the left and right collarbones, the second reference datum point is located at the connection between the lower edge of the sternal body and the xiphoid process, and the third reference datum point is located on the middle line under the left armpit of the body and at the same horizontal line as the second reference datum point; Step 2: Obtaining a two-dimensional image of the chest bones using a terahertz imaging system; Step 3: Collecting a visible light image of the chest through a monocular camera; Step 4: Registering the visible light image and the terahertz image with the marked reference datum points; Step 5: Calculating the electrode patch positions based on the connection lines. Let the longitudinal connection line length between the first reference datum point and the second reference datum point be A, and the transverse connection line length between the second reference datum point and the third reference datum point be B. Set the connection lines for the reference datum points to calculate the positions of the V1-V6 electrodes.

2. The electrode wearing position identification method according to claim 1, characterized in that, In step 1, the reference datum points are calibrated in the terahertz image of the human chest bones, where: The first reference datum point is located at the middle position of the connection between the left and right collarbones, the second reference datum point is located at the connection between the lower edge of the sternal body and the xiphoid process, and the third reference datum point is located on the middle line under the left armpit of the body and at the same horizontal line as the second reference datum point.

3. The method for identifying the electrode wearing position according to claim 1, wherein, In step 2, the terahertz imaging system emits a multi-color light source. The light source is accurately irradiated onto a point on the chest through a parabolic mirror, and then converges onto a terahertz detector through the parabolic mirror and is converted into an intermediate frequency electrical signal. After being phase-locked and amplified by a phase-locked amplifier, the signal intensity is read out through a processing module, that is, the corresponding intensity of a point in the chest area. A complete image of the chest bones is obtained by the moving optical frequency comb module receiving different intensity signals.

4. The method for identifying the electrode wearing position according to claim 3, characterized in that, The monocular camera uses the prior data calibrated in the first step to train a reference datum point positioning model, takes the collected terahertz image as the model input, and the model automatically marks the reference datum points on the terahertz image.

5. The method for identifying the electrode wearing position according to claim 1, wherein In step 5, calculating the electrode patch positions based on the connection lines. Let the longitudinal connection line length between the first reference datum point and the second reference datum point be A, and the transverse connection line length between the second reference datum point and the third reference datum point be B; The V2 electrode is located at a distance of A / 4 from the transverse connection line and B / 7 from the longitudinal connection line; The V1 electrode is symmetric with the V2 electrode with respect to the longitudinal connection line, at a distance of B / 7 from the longitudinal connection line and A / 4 from the transverse connection line; The V4 electrode is on the transverse connection line and at a distance of 3B / 7 from the second reference datum point; The V5 electrode is on the transverse connection line and at a distance of 5B / 7 from the second reference datum point; The V3 electrode is in the middle of the V2 electrode and the V4 electrode, at a distance of A / 8 from the transverse connection line and 2B / 7 from the longitudinal connection line; The V6 electrode is located at the third reference band and at a distance of A / 8 from the transverse connection line.

6. The method for identifying the electrode wearing position according to claim 1, characterized in that, It also includes a method for detecting whether the electrode patch is worn correctly, including the following steps: Detecting whether there is a real electrode patch on the chest through an object detection algorithm. When there is a real electrode patch, use the nearest neighbor search to match its corresponding correct electrode point; Each set of real chest electrode patches and the correct electrode points are taken as a matching pair. Each matching pair generates a path guiding arrow on the registration image. The direction and length of the arrow respectively represent the direction and distance from the real chest electrode to the correct electrode. When the length of the guiding arrow becomes 0 and the shape becomes a dot, it means that the electrode patch is worn correctly.

Images