A heart rate determination method, device, and apparatus of a heart-like, and a storage medium

Abstract

The application provides a kind of heart-like heart rate determination method, device, equipment and storage medium. The method comprises: determining the time domain signal of feature pixel points from the heart-like image of continuous multiple frames, and determining the preliminary heart rate value according to the time domain signal of feature pixel points. Then calculate the sliding window using the preliminary heart rate value, and use the sliding window to detect the wave peak of the time domain signal of feature pixel points, determine the heart rate value of the heart-like heart according to the wave peak detection result. In this way, the preliminary heart rate value is determined first, and the sliding window is determined using the preliminary heart rate value to perform wave peak detection, and the preliminary heart rate value is verified according to the positive and negative wave peak detection results, and when the verified preliminary heart rate value is correct, the heart rate of the heart-like heart is determined according to at least one of the wave peak detection result and the preliminary heart rate value, which improves the accuracy of the heart rate determination of the heart-like heart. The determination method is not only suitable for the second generation of heart-like heart, but also suitable for the third generation of heart-like heart.

Description

Technical Field

[0001] This invention relates to the field of heart-like technology, and in particular to a method, apparatus, device, and storage medium for determining the heart rate of a heart-like organ. Background Technology

[0002] The use of artificial hearts to replace or assist the human heart in the treatment of heart diseases is gaining increasing recognition. An artificial heart refers to a human-grown structure that resembles the human heart. In practical applications, the heart rate of the artificial heart needs to be tested to understand its functional status.

[0003] Because the heart-like organ mimics the function of the human heart, but operates in a different environment, conventional heart rate determination methods are unsuitable for determining its rate or current operating state. Therefore, determining the heart rate of the heart-like organ has become a pressing technical problem for those skilled in the art. Summary of the Invention

[0004] In view of this, the present invention provides a method, apparatus, device and storage medium for determining the heart rate of a heart-like organ, with the aim of accurately determining the heart rate value of the heart-like organ.

[0005] In a first aspect, this application provides a method for determining the heart rate of a heart-like organ, the method comprising:

[0006] Based on multiple consecutive frames of heart-like images, determine the temporal signal of feature pixels;

[0007] A preliminary heart rate value is determined based on the temporal signal of the feature pixels;

[0008] Calculate the sliding window based on the initial heart rate value;

[0009] Peak detection is performed on the temporal signal of the feature pixel based on the sliding window to obtain the peak detection result; the peak detection includes forward normal peak detection and reverse normal peak detection.

[0010] The preliminary heart rate value is corroborated by the peak detection result, and when the preliminary heart rate value is correct, the heart rate value of the heart-like organ is determined based on at least one of the peak detection result and the preliminary heart rate value.

[0011] Optionally, determining the temporal signal of feature pixels based on multiple consecutive frames of heart-like images specifically includes:

[0012] The time-domain signal of multiple heart-like pixels in multiple consecutive frames of the heart-like image is converted into a frequency-domain signal;

[0013] Based on the amplitude values ​​corresponding to the frequency domain signals of the multiple heart-like pixels, the pixel corresponding to the frequency domain signal with the largest amplitude value within the normal heart rate range is determined as the feature pixel.

[0014] Optionally, the heart-like pixels include pixels on the heart-like boundary line of the heart-like structure.

[0015] Optionally, before determining the heart rate value of the cardiac-like organ based on at least one of the peak detection result and the preliminary heart rate value when the preliminary heart rate value is correct, the method further includes:

[0016] Based on the peak detection results, a preliminary judgment can be made as to whether the heart rhythm of the cardiac-like organ is regular.

[0017] If the heart rhythm is regular, the preliminary heart rate value is corroborated by the peak detection results.

[0018] Optionally, the step of corroborating the preliminary heart rate value based on the peak detection results specifically includes:

[0019] The first judgment step is to determine whether the difference between the positive normal peak detection result and the negative normal peak detection result is less than a preset threshold; if it is less, then the preliminary heart rate value is determined to be correct.

[0020] Optionally, the step of corroborating the preliminary heart rate value based on the peak detection results specifically includes:

[0021] The second judgment step is to determine whether the difference between the forward normal peak detection result and the reverse normal peak detection result is less than a preset threshold. If it is less, perform enhanced peak detection on the temporal signal of the feature pixel based on the sliding window to obtain the enhanced peak detection result. The enhanced peak detection includes at least one of forward enhanced peak detection and reverse enhanced peak detection.

[0022] The third judgment step is to determine whether the enhanced peak detection result, the positive normal peak detection result, and the reverse normal peak detection result are consistent within a preset range; if they are consistent, the preliminary heart rate value is determined to be correct.

[0023] Optionally, the third determination step further includes: if there is a discrepancy, then determining that the preliminary heart rate value is incorrect;

[0024] The method further includes:

[0025] When the initial heart rate value is incorrect, the heart rate value of the heart-like organ is calculated based on the enhanced peak detection result.

[0026] Optionally, the first or second determination step may further include:

[0027] If it is not less than, then the sliding window is recalculated based on the smaller of the heart rate values ​​represented by the positive normal peak detection result and the reverse normal peak detection result.

[0028] Based on the recalculated sliding window, perform the forward normal peak detection or the reverse normal peak detection corresponding to the smaller value again, and obtain a new forward normal peak detection result or a new reverse normal peak detection result accordingly, and then execute the first judgment step or the second judgment step again accordingly.

[0029] Optionally, the peak detection further includes enhanced peak detection; the enhanced peak detection includes at least one of positive enhanced peak detection and reverse enhanced peak detection;

[0030] The preliminary determination of whether the heart rhythm of the cardiac-like organ is regular based on the peak detection results specifically includes:

[0031] The heart rhythm of the cardiac-like organ is preliminarily determined based on at least one of the positive normal peak detection results, the reverse normal peak detection results, the positive enhanced peak detection results, and the reverse enhanced peak detection results.

[0032] Optionally, the step of preliminarily determining whether the heart rhythm of the cardiac-like organ is regular based on the peak detection results further includes:

[0033] If the heart rhythm is irregular, then multiple new preliminary heart rate values ​​that are greater than the preliminary heart rate value and multiple new preliminary heart rate values ​​that are less than the preliminary heart rate value are set based on the preliminary heart rate value;

[0034] Calculate the corresponding sliding window based on each of the new preliminary heart rate values;

[0035] Peak detection is performed on the temporal signal of the feature pixel based on each sliding window to obtain peak detection results;

[0036] Based on the peak detection results corresponding to each of the multiple new preliminary heart rate values, it is determined whether the heart rhythm of the heart-like organ is truly irregular, and when the heart rhythm of the heart-like organ is truly regular, the heart rate of the heart-like organ is determined based on the peak detection results corresponding to each of the multiple new preliminary heart rate values.

[0037] Optionally, the method further includes:

[0038] Obtain the video of the heart-like structure;

[0039] Based on the video, a series of consecutive original heart-like images are determined, and the original heart-like images are cropped to obtain the heart-like image.

[0040] Optionally, determining the pixel point corresponding to the frequency domain signal with the largest amplitude value within the normal heart rate range as the feature pixel point specifically includes:

[0041] The pixel corresponding to the frequency domain signal with the largest amplitude value within the normal heart rate range is identified as the preliminary feature pixel.

[0042] Determine the amplitude value of the frequency domain signal corresponding to a preset number of pixels on the cardiac-like boundary line within a preset distance of the preliminary feature pixel;

[0043] Compare whether the difference between the amplitude values ​​of the preset number of pixels and the maximum amplitude value is within a preset difference range;

[0044] If so, the preliminary feature pixel is determined as the feature pixel.

[0045] Secondly, this application provides a heart rate determination device for a type of heart, the device comprising:

[0046] The time-domain signal determination module is used to determine the time-domain signal of feature pixels based on multiple consecutive frames of heart-like images;

[0047] The calculation module is used to determine the preliminary heart rate value based on the temporal signal of the feature pixels; and to calculate the sliding window based on the preliminary heart rate value.

[0048] The peak detection module is used to perform peak detection on the time-domain signal of the feature pixel based on the sliding window, and obtain the peak detection result; the peak detection includes forward normal peak detection and reverse normal peak detection;

[0049] A heart rate determination module is used to corroborate the preliminary heart rate value based on the peak detection result, and, when the preliminary heart rate value is correct, determine the heart rate value of the heart-like organ based on at least one of the peak detection result and the preliminary heart rate value.

[0050] Thirdly, this application provides an electronic device, comprising:

[0051] Memory, used to store computer programs;

[0052] A processor for executing the computer program to implement the steps of the heart rate determination method for a heart-like structure as described in any of the first aspects.

[0053] Fourthly, this application provides a readable storage medium storing a computer program that, when executed by a processor, implements the steps of the heart rate determination method for a heart-like structure as described in any of the first aspects.

[0054] This application provides a method, apparatus, device, and storage medium for determining the heart rate of a heart-like organ. When performing the method: First, the temporal signals of feature pixels are determined from multiple consecutive frames of heart-like organ images, and a preliminary heart rate value is determined based on the temporal signals of the feature pixels. Then, a sliding window is calculated using this preliminary heart rate value, and the sliding window is used to perform forward and reverse normal peak detection on the temporal signals of the feature pixels. Thus, a preliminary heart rate value is first determined, and the sliding window is used to perform forward and reverse normal peak detection. Then, the preliminary heart rate value is verified based on the forward and reverse peak detection results. When the preliminary heart rate value is verified to be correct, the heart rate of the heart-like organ is determined based on at least one of the two peak detection results and the preliminary heart rate value, thereby improving the accuracy of heart rate determination for heart-like organs. This determination method is applicable not only to second-generation heart-like organs but also to third-generation heart-like organs. Attached Figure Description

[0055] Figure 1 A flowchart of a heart-like heart rate determination method provided in an embodiment of this application;

[0056] Figure 2A A flowchart illustrating a method for obtaining a heart-like image by cropping a heart-like original image, as provided in this application embodiment;

[0057] Figure 2B This application provides a schematic diagram of obtaining a heart-like edge region according to an embodiment of the present application;

[0058] Figure 2C This is a schematic diagram of video cropping provided in an embodiment of this application;

[0059] Figure 3A A schematic diagram of a series of heart-like images obtained based on cropping and preprocessing, provided as an embodiment of this application;

[0060] Figure 3B A schematic diagram of the image feature pixels of the heart-like structure provided in the embodiments of this application at the boundary line of the heart-like structure;

[0061] Figure 4 Schematic diagrams of the second-generation and third-generation heart-like structures provided for this application;

[0062] Figure 5 This application provides a schematic diagram of a positive normal peak, a reverse normal peak, and a positive enhanced peak in its embodiments;

[0063] Figure 6 This is a schematic diagram of a heart-like heart rate determination device provided in an embodiment of this application. Detailed Implementation

[0064] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0065] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0066] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0067] As mentioned earlier, existing conventional heart rate determination methods cannot accurately determine the heart rate and current state of a heart-like organ. Therefore, this application proposes a heart rate determination method for a heart-like organ. This method aims to determine a preliminary heart rate value using the temporal signal of feature pixels, and then use this preliminary heart rate value to calculate a sliding window to perform forward and reverse peak detection on the temporal signal of the feature pixels, thus verifying the preliminary heart rate value. When the preliminary heart rate value is correct, the heart rate of the second-generation and third-generation heart-like organs is accurately determined based on at least one of the peak detection results and the preliminary heart rate value.

[0068] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0069] See Figure 1 This is a flowchart illustrating a heart-like heart rate determination method provided in an embodiment of this application. The method can be implemented by a heart-like heart rate determination system. The method includes the following steps:

[0070] S101: Determine the temporal signal of the feature pixel based on multiple consecutive frames of heart-like images.

[0071] A series of consecutive heart-like images are formed by the heart-like heart rate determination system converting the acquired heart-like video into a series of consecutive heart-like images.

[0072] In one possible implementation, the heart-like image can be a preprocessed version of the original heart-like image. The original heart-like image is a multi-frame heart-like image directly converted from a video of the heart-like image. The preprocessing method for the original heart-like image involves performing three Gaussian downsampling operations on each frame of the original image for image compression. Experiments have shown that using three-channel and single-channel data has virtually no impact on the results; both can accurately determine the heart rate of the heart-like image. Therefore, to accelerate computation while ensuring the accuracy of heart rate determination, this embodiment uses the image data after the third downsampling as the preprocessed data, and then uses the g-channel of each preprocessed frame as the analysis channel for that frame. Finally, all the processed data is combined (video frame number, image height, image width) to form a continuous multi-frame heart-like image according to this embodiment.

[0073] It is worth noting that, in addition to using the g channel as the analysis channel, the R or B channel can also be used. Those skilled in the art can adjust this as needed.

[0074] In one possible implementation, the heart-like image can be an image obtained by first cropping and then preprocessing the original heart-like image. Specifically, the heart-like heart rate determination system first crops the original heart-like image to obtain cropped heart-like image data, and then preprocesses the cropped heart-like image data to obtain a series of consecutive frames of heart-like images composed of (video frame number, image height, image width) according to the embodiments of this application.

[0075] For a flowchart of the method for obtaining a heart-like image by cropping the original heart-like image, please refer to [link / reference]. Figure 2A As shown. Specifically includes:

[0076] S1011: Convert the first frame of the video to grayscale and detect edge lines to find the boundary lines of the heart-like structure.

[0077] See Figure 2B This is a schematic diagram illustrating how to obtain the boundary line of a heart-like structure according to an embodiment of this application. The first frame of the video image is converted to grayscale to obtain... Figure 2B The grayscale image. Using an edge detection algorithm, the edge lines of the heart-shaped image are obtained, such as... Figure 2B The boundary line of the heart-like structure. Of course, in practical applications, this step can be performed on any frame of the image.

[0078] S1012: Transform the boundary line into a set of points, and find the maximum and minimum values ​​of the points.

[0079] S1013: Based on the maximum and minimum values ​​of the points, customize the edge shape and perform video cropping.

[0080] In this embodiment, the coordinates of the top-left and bottom-right corners are calculated. See also... Figure 2C This is a schematic diagram of video cropping provided in an embodiment of this application. Figure 2C Based on the two gray points in (a), all video frames are cropped according to the rectangular area formed by these two points. The result is as follows: Figure 2C (b) shows a heart-like image.

[0081] The above-described method of cropping the original heart-like image is used because the heart-like shape occupies a very small proportion of the entire original heart-like image. In order to speed up the analysis and save computer resources, the region where the heart-like shape appears can be segmented from the frame image to obtain the heart-like image of this embodiment.

[0082] See Figure 3A This is a schematic diagram of a series of heart-like images obtained based on cropping and preprocessing, provided in an embodiment of this application. Exemplarily, Figure 3A The video has 10 frames per second, a width of W, and a height of H. The heart-like image is an RGB image, and the heart-like structure occupies more than half of the image, which is a significant portion.

[0083] In this step, feature pixels refer to pixels in the heart-like image whose values ​​change dynamically with the heartbeat. In practical applications, the preferred feature pixels are those that best reflect the heartbeat, or pixels located at points of intense beating. For example, Figure 3A Each frame of the heart-like image contains a heart-like region (i.e., the foreground region) and a background region. By selecting any frame of the heart-like image, the heart-like pixels within that region are determined. The position of each pixel remains constant across all frames (in the same coordinate system), but the pixel value may differ across frames due to the heart-like beating. If a pixel beats significantly during a certain time period, its pixel value will vary considerably across multiple frames during that time period. Therefore, the temporal signal of a feature pixel can be understood as the pixel value signal of that feature pixel across multiple consecutive frames over time.

[0084] Optionally, the feature pixels are the pixels in multiple consecutive frames of heart-like images that have the largest amplitude value and whose frequency domain signal is within the correct heart rate range. This method ensures that the frequency of the feature pixels falls within the correct heart rate range of the heart-like image and that the feature pixels reflect a significant heart rate fluctuation, thus improving the accuracy of the preliminary heart rate value determined based on these feature pixels. Specifically, feature pixels can be obtained in the following ways:

[0085] S101-21: Convert the time-domain signal of multiple heart-like pixels in a series of consecutive heart-like images into a frequency-domain signal.

[0086] For a given pixel location in a series of consecutive heart-like images, pixel values ​​are taken along the video frame direction. The pixel values ​​at that location across all frames can form a time-domain signal (or pixel sequence). The changes in the pixel sequence of the heart-like pixels contain information about the frequency of the heartbeat. However, this information is difficult to observe solely from changes in pixel values. Therefore, a Fast Fourier Transform is performed on the pixel sequence of the heart-like pixels to convert the time-domain signal into a frequency-domain signal, resulting in the sum of countless sine and cosine sequences.

[0087] S101-22: Determine the feature pixel based on the amplitude value corresponding to the frequency domain signal of each of the multiple heart-like pixel points.

[0088] To determine the normal heart rate range for the heart-like sensor, experimental data showed that the normal heart rate range is between 6 and 130. First, for each heart-like sensor pixel sequence, a frequency filter between 6 and 130 was performed on the spectrogram, removing all other frequencies (frequency coefficients outside 6-130 were set to 0). For each heart-like sensor pixel sequence within the 6-130 frequency range, the maximum amplitude of each pixel was determined. Finally, the maximum amplitudes of all heart-like sensor pixels were compared, and the pixel corresponding to the maximum amplitude among multiple maximum amplitudes was selected as the feature pixel.

[0089] In one possible implementation, the multiple heart-like pixels in steps S101-21 are pixels within the entire internal region of the heart-like structure (i.e., pixels within the boundary line of the heart-like structure). In this scheme, by simultaneously selecting feature pixels using the methods described above, one can obtain, for example... Figure 3A The gray origin near the center point is used as the feature pixel. This method is more suitable for second-generation heart-like structures. This is mainly because the internal region of the second-generation heart-like structure differs greatly from the background region, and the height difference between the main peak and the secondary peak of the pixel sequence variation is large. Therefore, identifying feature pixels that can better reflect the heartbeat in the internal region of the heart-like structure has higher accuracy.

[0090] In one possible implementation, preferably, the feature pixels are pixels on the boundary line of the heart-like structure. In this scheme, selecting feature pixels using the methods described above simultaneously can yield results such as... Figure 3BFeature pixels on the boundary line of the heart-like structure. Considering the beating pattern of the heart-like structure, the internal beating will propagate to the boundary. Therefore, selecting feature pixels on the boundary line is not only faster than selecting feature pixels throughout the entire heart-like region, but also suitable for both second-generation and third-generation heart-like structures. This is mainly because the internal structure of the third-generation heart-like structure is complex. The pixel values ​​of pixels in the internal region of the heart-like structure will have the same values ​​as the background region. In some cases, the height difference between the main peak and the secondary peak of the pixel sequence will be very small. Therefore, searching for feature pixels in the internal region may lead to errors. Considering the beating pattern of the heart-like structure, the internal beating will propagate to the boundary, so selecting feature pixels on the boundary line is more accurate. Therefore, it is applicable not only to second-generation heart-like structures but also to third-generation heart-like structures. Figure 4 (a) is a schematic diagram of the second-generation artificial heart. For example... Figure 4 (b) is a schematic diagram of the third-generation artificial heart.

[0091] When determining feature pixels, the method provided in this embodiment further includes:

[0092] The pixel corresponding to the frequency domain signal with the largest amplitude value within the normal heart rate range is identified as the preliminary feature pixel.

[0093] Determine the amplitude value of the frequency domain signal of a preset number of pixels within a preset distance of the preliminary feature pixel corresponding to the maximum amplitude value on the boundary line of the heart-like structure;

[0094] Compare whether the difference between the amplitude values ​​of the preset number of pixels and the maximum amplitude value is within a preset difference range;

[0095] If so, the preliminary feature pixel corresponding to the maximum amplitude value is determined as the feature pixel.

[0096] This application discloses a method for verifying feature pixels. First, the frequency domain signal corresponding to the maximum amplitude value within a normal heart rate range is used as a preliminary feature pixel. A predetermined number of other pixels within a predetermined distance of the preliminary feature pixel are compared, and the difference between the amplitude values ​​of the other pixels and the amplitude value of the preliminary feature pixel is calculated. It is then determined whether the difference is within a predetermined range. Although the preliminary feature pixel corresponds to the maximum amplitude value, the amplitude values ​​of its neighboring pixels should not differ significantly from this maximum amplitude value for the preliminary feature pixel to be considered a feature pixel.

[0097] Therefore, it can be understood that the above steps essentially add a noise vibration check when determining feature pixels, avoiding the selection of an incorrect pixel. More specifically, to avoid finding an incorrect initial feature pixel as the point of maximum amplitude, the amplitude of the pixels surrounding the initial feature pixel is also calculated. If the vibration amplitude of the surrounding points is also relatively large, then the initial feature pixel is determined to be the pixel with the maximum amplitude and can be used as a feature pixel.

[0098] S102: Determine the preliminary heart rate value based on the time-domain signal of the feature pixels.

[0099] In this embodiment of the application, after the heart-like heart rate determination system determines the feature pixel, it uses the frequency corresponding to the maximum amplitude value of the feature pixel in the frequency domain signal as the preliminary heart rate value.

[0100] Research analysis shows that in second-generation heart replicas, the differences between the heart content and background are significant, and the height difference between the main and secondary peaks in the pixel sequence changes is large. Changes in pixel values ​​do not affect heart rate detection. However, for third-generation heart replicas, factors such as the selection of feature pixels, and the presence of arrhythmia in the replicas (both second and third generation), can lead to errors in the calculated initial heart rate value. These errors could result in an average of multiple frequencies, a correct heart rate value, or a value artificially inflated due to the influence of secondary peaks.

[0101] Therefore, how to corroborate the preliminary heart rate value and accurately determine the heart rate value of the heart-like organ is the technical problem to be solved below, and the technical problem of how to determine whether the heart-like organ actually has arrhythmia will continue to be solved later.

[0102] S103: Calculate the sliding window based on the initial heart rate value.

[0103] The sliding window is calculated using the initial heart rate value as follows: Assume the initial heart rate is f, and the frame rate is f. ps The sliding window size is fps*(1 / f). It can be understood that the sliding window is a minimum step size.

[0104] S104: Perform peak detection on the time-domain signal of feature pixels based on a sliding window, obtain peak detection results, and use the positive and negative normal peak detection results to corroborate the preliminary heart rate value.

[0105] Peak detection refers to detecting the position of peak points in pixel sequence data and obtaining the detection results. Subsequently, based on the detection results, the frames corresponding to each peak (main peak) can be determined. Based on the time between the frames corresponding to two adjacent main peaks, the frequency can be calculated, and further, the heart rate can be calculated.

[0106] The method for peak detection using a sliding window is as follows: Based on the pixel data of the sliding window (the size of the sliding window), find the maximum value within the data of that sliding window (the maximum value can be found using curve fitting). Then, the sliding window moves to the next position with a certain step size. Determine whether the maximum value in the current sliding window data is greater than the average value of the already processed data. If so, the position of the frame containing the maximum value is the peak position, which is used as the location of the peak.

[0107] (1) Initial assessment of whether the heart rhythm is regular.

[0108] In this embodiment, a sliding window is used to perform peak detection on the time-domain signal of the feature pixel. Specifically, peak detection is performed on the time-domain signal of the pixel sequence in which the feature pixel is located. Based on the peak detection results, two situations can be preliminarily determined: one is a regular heart rhythm, and the other is an irregular heart rhythm.

[0109] The specific judgment method is as follows: Obtain the frame position corresponding to the peak of each sliding window. Based on the difference between these frame positions, calculate whether there are different frequencies in the pixel sequence. If multiple heartbeat frequencies appear, all of these heartbeat frequencies can be counted, and the preliminary judgment result is arrhythmia. Otherwise, the preliminary judgment result is regular heartbeat.

[0110] In one possible implementation, peak detection can be one of the following: positive normal peak detection, reverse normal peak detection, or enhanced peak detection. Enhanced peak detection includes both positive and reverse enhanced peak detection. Based on the peak detection results—that is, the results of positive normal peak detection, reverse normal peak detection, and enhanced peak detection (either positive or reverse enhanced peak detection)—a preliminary assessment of heart rhythm regularity can be made.

[0111] Furthermore, after using peak detection to initially determine the heart rhythm, this application can also determine the variation pattern of the peak amplitude. If the variation pattern of the amplitude is within a preset variation range, the pixel sequence is normal; if the variation pattern of the amplitude is not within the preset variation range, the pixel sequence is initially determined to be abnormal.

[0112] In one possible implementation, to improve the accuracy of the initial judgment, peak detection can be a combination of any two of the following: positive normal peak detection, reverse normal peak detection, positive enhanced peak detection, or reverse enhanced peak detection. If both peak detection methods result in a regular heart rhythm, the heart rhythm is judged to be regular; if both result in an irregular heart rhythm, the heart rhythm is judged to be irregular.

[0113] For second-generation cardiac models with significant differences in the height of the primary and secondary peaks, either forward or reverse normal peak detection is sufficient. However, during the experimental phase, it was found that in third-generation cardiac models where the primary and secondary peaks are relatively close, using only forward and reverse peak detection will also detect the secondary peak as a peak. In this case, the peak detection method for any two detection results should include at least one of forward or reverse enhanced peak detection.

[0114] When the peak detection includes the detection of the enhanced peak, the result of the enhanced peak detection is used as the preliminary judgment result of whether the heart rhythm is regular.

[0115] In one possible implementation, to further improve the accuracy of the initial assessment of heart rhythm regularity, peak detection can be any combination of three of the following: positive normal peak detection, reverse normal peak detection, positive reinforced peak detection, or reverse reinforced peak detection. If two of the three detections indicate the presence of multiple different frequencies, a preliminary assessment is made of arrhythmia; if a relatively uniform frequency is obtained, a preliminary assessment is made of regularity. Considering that both positive and reverse peaks may be incorrect, if the reinforced peak detection result is regularity, a preliminary assessment is made of regularity. If the reinforced peak detection result is arrhythmia, a preliminary assessment is made of arrhythmia.

[0116] In one possible implementation, peak detection can be a combination of four methods: positive normal peak detection, reverse normal peak detection, positive enhanced peak detection, and reverse enhanced peak detection. If the enhanced peak detection result is a regular heart rhythm, the preliminary diagnosis is a regular heart rhythm. If the enhanced peak detection result is an irregular heart rhythm, the preliminary diagnosis is an irregular heart rhythm.

[0117] To help those skilled in the art better understand peak detection, the following section combines... Figure 5 Please provide an explanation.

[0118] Forward normal peak detection refers to detecting the peak position of a forward normal wave, such as... Figure 5 As shown in (a) above. Reverse normal peak detection refers to detecting the peak position of a reverse normal wave, such as... Figure 5As shown in (b) above. Forward enhancement peak detection refers to amplifying the detected peaks within the forward detection peaks based on their numerical values. That is, a magnification factor is applied to each peak; the higher the initial peak, the larger the multiplication factor. For example... Figure 5 As shown in (c) above. Reverse reinforcement wave peak detection refers to amplifying the detected peak positions within the reverse-detection wave peaks based on their numerical value. That is, a magnification factor is applied to each peak position; the higher the initial peak, the larger the factor. Reverse reinforcement waves are similar to forward reinforcement waves and will not be described in detail here.

[0119] Figure 5 This application provides a schematic diagram of a positive normal peak, a reverse normal peak, and a positive reinforced peak. The three peak diagrams represent the Heart Rate Signal Peak Detection (HRS), where the vertical axis represents amplitude and the horizontal axis represents time. For the normal peak, the amplitude ranges from 0 to 0.16, and the time ranges from 0 to 60 seconds. Figure 5 The magnitude of (a) in the middle is related to Figure 5 The amplitudes of (b) are opposite. For reinforced waves, Figure 5 (c) in relation to Figure 5 In (a), the amplitude is multiplied by a large amplification factor. After the peak is reinforced, the secondary peak is suppressed and the main peak is reinforced, which can compensate for the inadequacy of forward and reverse peak detection.

[0120] (2) In the case of a preliminary judgment that the heart rhythm is regular, the results of positive and negative normal peak detection are used to corroborate the preliminary heart rate value.

[0121] When the heart-like organ is a second-generation heart-like organ, the method to corroborate the preliminary heart rate value is as follows:

[0122] First judgment step: Determine whether the difference between the positive normal peak detection result and the reverse normal peak detection result is less than a preset threshold; (a) If it is less than the preset threshold, it means that the difference between the two is small, and the preliminary heart rate value is determined to be correct. (b) If it is not less than the preset threshold, it means that the difference between the two is large, and the sliding window is recalculated based on the smaller value of the heart rate value represented by the positive normal peak detection result and the reverse normal peak detection result.

[0123] The value judged as a large heart rate is incorrect because the error in heart rate monitoring is due to the influence of color values. The difference between the forward and reverse results indicates that there are other secondary peaks with a similar height to the main peak in either the forward or reverse process. Therefore, it is necessary to increase the sliding window. Hence, a smaller heart rate value is used to calculate the sliding window.

[0124] Based on the recalculated sliding window, another forward or reverse normal peak detection corresponding to the smaller value is performed, resulting in a new forward or reverse normal peak detection result. The first judgment step is then executed again. Specifically, if the smaller value is a forward normal peak detection method, then a forward normal peak detection method is performed here; if the smaller value is a reverse normal peak detection method, then a reverse normal peak detection method is performed here.

[0125] In this embodiment, the enhanced peak detection method is not used because the difference between the positive and negative peaks is large. Therefore, the main peak determined by the sliding window corresponding to the heart rate may be incorrect, which would also result in low accuracy when using the enhanced peak method.

[0126] In the embodiments of this application, the detection results can be obtained in various ways. In one possible implementation, the detection results can be determined from the average, median, or most frequently occurring frequency of a set of frequency values. In another possible implementation, the detection results can be a set of frequency values ​​after excluding error frequencies, and the difference between the two sets of frequency values ​​corresponding to the two detection results can be determined by the variance of the two sets of frequency values.

[0127] When the heart-like organ is a third-generation heart-like organ (of course, second-generation heart-like organs are also compatible and applicable), the method to corroborate the preliminary heart rate value is as follows:

[0128] The second judgment step: Determine whether the difference between the forward normal peak detection result and the reverse normal peak detection result is less than a preset threshold. (a) If it is less than the preset threshold, it means that the difference between the two is small. Perform enhanced peak detection on the temporal signal of the feature pixel. This can be either forward enhanced peak detection or reverse enhanced peak detection. Obtain the enhanced peak detection result and execute the third judgment step. (b) If it is not less than the preset threshold, it means that the difference between the two is large. Recalculate the sliding window based on the smaller value of the heart rate value represented by the forward normal peak detection result and the reverse normal peak detection result. The value that is judged to be larger is incorrect because the heart rate monitoring error is due to the influence of the color value. The difference between the forward and reverse results indicates that there are other secondary peaks with a small difference in height from the main peak in either the forward or reverse process. Therefore, it is necessary to increase the sliding window. Thus, a smaller heart rate value is used to calculate the sliding window.

[0129] Based on the recalculated sliding window, another forward or reverse normal peak detection corresponding to the smaller value is performed, resulting in a new forward or reverse normal peak detection result. The second judgment step is then executed again. Specifically, if the smaller value is a forward normal peak detection method, then a forward normal peak detection method is performed here; if the smaller value is a reverse normal peak detection method, then a reverse normal peak detection method is performed here.

[0130] The third judgment step: Determine whether the results of the enhanced heart rate detection, the positive normal heart rate detection, and the reverse normal heart rate detection are consistent within the preset range. If they are consistent, the preliminary heart rate value is determined to be correct. If they are inconsistent, the preliminary heart rate value is determined to be incorrect.

[0131] S105: When the initial heart rate value is correct, determine the heart rate value of the cardiac-like heart based on at least one of the peak detection results and the initial heart rate value.

[0132] When the initial heart rate value is correct, the heart rate determination system for the cardiac-like organ determines the heart rate value based on at least one of the peak detection results and the initial heart rate value.

[0133] Evidence for the correctness of the preliminary heart rate value includes: when the heart rhythm of the second-generation heart-like organ is initially determined to be regular, the preliminary heart rate value is considered correct if the difference between the positive and negative normal peak detection results is less than a preset threshold. The heart rate value of the heart-like organ can be determined based on at least one of the positive and negative normal peak detection results and the preliminary heart rate value. Alternatively, at least two detection results (positive and negative normal peak detection results, and the preliminary heart rate value) can be arbitrarily selected and weighted averaged to calculate the heart rate value of the heart-like organ. Another option is to directly average the positive and negative normal peak detection results and the preliminary heart rate value to calculate the heart rate value of the heart-like organ.

[0134] When the heart rhythm of the pre-determined cardiac model is regular, for the third-generation cardiac model, if the difference between the positive and negative normal peak detection results is less than a preset threshold, and the enhanced peak detection results are consistent with both the positive and negative normal peak detection results within a preset range, the preliminary heart rate value is confirmed to be correct. The heart rate value of the cardiac model can be calculated based on any one of the enhanced peak detection results, the positive normal peak detection results, the negative normal peak detection results, or the preliminary heart rate value. Alternatively, at least two detection results from the enhanced peak detection results, the positive normal peak detection results, the negative normal peak detection results, and the preliminary heart rate value can be arbitrarily selected and weighted to calculate the heart rate value of the cardiac model. Alternatively, the heart rate value of the cardiac model can be calculated by directly averaging the enhanced peak detection results, the positive normal peak detection results, the negative normal peak detection results, and the preliminary heart rate value.

[0135] In addition, when the heart rate determination system for the cardiac-like heart makes an error in determining the initial heart rate value, that is, when the difference between the positive normal peak detection result and the reverse normal peak detection result is less than a preset threshold, but the enhanced peak detection result and the positive normal peak detection result and the reverse normal peak detection result are inconsistent within a preset range, the heart rate value of the cardiac-like heart can be calculated based on the enhanced peak detection result.

[0136] Analysis of the reason for using enhanced peak detection results to calculate the heart rate value of the cardiac-like organ: Since the difference between the forward and reverse peak detection results is not significant, there are two possibilities: either both forward and reverse peak detection results are incorrect, or both are correct. Therefore, an enhanced peak algorithm is added to judge the forward and reverse results. The principle of enhanced peak is to amplify the peak at the point where it appears, making the peak higher and the trough lower. After peak amplification, if there were secondary peaks in the original data, they have been suppressed, while the main peak is amplified. When the enhanced peak is detected, if the detection result is the same as the previous forward and reverse detection results, it can be inferred that the previous forward and reverse results were correct. If the enhanced peak result is different from the previous forward and reverse results, it proves that the previous forward and reverse results were incorrect.

[0137] In one possible implementation, if the enhancement peak detection in step S104 is a single enhancement wave, for type III hearts with cavities, a single enhancement wave cannot reach the secondary enhancement wave; therefore, the enhancement wave detected here is a multiple enhancement wave. If the enhancement peak detection in step S104 is a multiple enhancement wave, the enhancement in this enhancement peak detection is the same as the enhancement wave in step S104.

[0138] In addition, when a preliminary diagnosis of cardiac-like arrhythmia is made based on peak detection results, the cardiac-like heart rate can be determined using the following methods:

[0139] S1051: Set multiple new initial heart rate values ​​that are greater than the initial heart rate value and multiple new initial heart rate values ​​that are less than the initial heart rate value.

[0140] Example Explanation: Set four new initial heart rate values ​​that are greater than the initial heart rate value: 2 × initial heart rate value, 4 × initial heart rate value, 6 × initial heart rate value, and 8 × initial heart rate value. Set four new initial heart rate values ​​that are less than the initial heart rate value: 1 / 2 × initial heart rate value, 1 / 4 × initial heart rate value, 1 / 6 × initial heart rate value, and 1 / 8 × initial heart rate value.

[0141] It is worth noting that those skilled in the art can set the number and size of new initial heart rate values ​​as needed.

[0142] S1052: Calculate the corresponding sliding window based on each new initial heart rate value.

[0143] The specific calculation method is described in step S103, and will not be repeated here.

[0144] S1053: Perform peak detection on the temporal signal of each feature pixel based on each sliding window to obtain the peak detection result.

[0145] For specific testing details, please refer to step S104, which will not be repeated here.

[0146] S1054: Based on the peak detection results corresponding to multiple new preliminary heart rate values, determine whether the heart rate of the heart-like organ is truly irregular, and when the heart rate of the heart-like organ is truly regular, determine the heart rate of the heart-like organ based on the peak detection results corresponding to multiple new preliminary heart rate values.

[0147] In this embodiment, it is determined whether there are duplicate heart rate values ​​between the first set of peak detection results corresponding to multiple new preliminary heart rate values ​​greater than the preliminary heart rate value and the second set of peak detection results corresponding to multiple new preliminary heart rate values ​​less than the preliminary heart rate value. If duplicate heart rate values ​​are found, the heart rhythm is regular, and the duplicate heart rate values ​​are considered as the heart rate values ​​of the heart-like arrhythmia. If no duplicate heart rate values ​​are found, the heart-like arrhythmia is determined to be real.

[0148] This application provides a method for determining the heart rate of a heart-like sensor. First, the temporal signal of feature pixels is determined from multiple consecutive frames of heart-like sensor images, and a preliminary heart rate value is determined based on this temporal signal. Then, a sliding window is calculated using this preliminary heart rate value, and this sliding window is used to perform forward and reverse normal peak detection on the temporal signal of the feature pixels. The heart rate value of the heart-like sensor is determined based on the peak detection results. Thus, by first determining the preliminary heart rate value, and then using this preliminary heart rate value to determine the sliding window for forward and reverse normal peak detection, the preliminary heart rate value is corroborated based on the forward and reverse peak detection results. When the preliminary heart rate value is corroborated as correct, the heart rate of the heart-like sensor is determined based on at least one of the two peak detection results and the preliminary heart rate value, thereby improving the accuracy of heart rate determination for heart-like sensor images. This method is applicable not only to second-generation heart-like sensor images but also to third-generation heart-like sensor images.

[0149] Furthermore, embodiments of this application also provide a heart-like heart rate determination device. See also Figure 6 The diagram shown is a structural schematic of a heart-like heart rate determination device 600 provided in an embodiment of this application. The device includes:

[0150] The time-domain signal determination module 601 is used to determine the time-domain signal of feature pixels based on multiple consecutive frames of heart-like images;

[0151] The calculation module 602 is used to determine the preliminary heart rate value based on the temporal signal of the feature pixels; and to calculate the sliding window based on the preliminary heart rate value;

[0152] The peak detection value acquisition module 603 is used to perform peak detection on the time-domain signal of feature pixels based on a sliding window and acquire the peak detection result; peak detection includes forward normal peak detection and reverse normal peak detection.

[0153] Heart rate determination module 604 is used to corroborate the preliminary heart rate value based on the peak detection result, and to determine the heart rate value of the cardiac-like heart based on at least one of the peak detection result and the preliminary heart rate value when the preliminary heart rate value is correct.

[0154] Optionally, the time-domain signal determination module 601 is specifically used for:

[0155] Convert the time-domain signal of multiple heart-like pixels in consecutive frames of heart-like images into a frequency-domain signal;

[0156] Based on the amplitude values ​​corresponding to the frequency domain signals of multiple heart-like pixels, the pixel corresponding to the frequency domain signal with the largest amplitude value within the normal heart rate range is identified as the feature pixel. Optionally, heart-like pixels include pixels on the boundary line of the heart-like structure.

[0157] Optionally, the heart rate determination module 604 is also used to: first determine whether the heart rhythm of the heart-like organ is regular based on the peak detection results; if the heart rhythm is regular, then further verify the preliminary heart rate value based on the peak detection results.

[0158] Optionally, the heart rate determination module 604 is specifically used to execute the following steps to corroborate the preliminary heart rate value based on the peak detection results:

[0159] The first judgment step is to determine whether the difference between the positive and negative normal peak detection results is less than a preset threshold; if it is less, the preliminary heart rate value is confirmed to be correct.

[0160] Optional, heart rate determination module 604, specifically used for:

[0161] The second judgment step is to determine whether the difference between the forward normal peak detection result and the reverse normal peak detection result is less than a preset threshold. If it is less, perform enhanced peak detection on the temporal signal of the feature pixel based on the sliding window to obtain the enhanced peak detection result. Enhanced peak detection includes at least one of forward enhanced peak detection and reverse enhanced peak detection.

[0162] The third judgment step: Determine whether the enhanced peak detection result, the positive normal peak detection result, and the reverse normal peak detection result are consistent within the preset range; if they are consistent, then the preliminary heart rate value is confirmed to be correct.

[0163] Optionally, the heart rate determination module 604 is also used to determine that the initial heart rate value is incorrect if there is a discrepancy.

[0164] When the initial heart rate value is incorrect, the heart rate value of the cardiac-like heart is calculated based on the results of the enhanced peak detection.

[0165] Optionally, the first or second judgment step may further include:

[0166] If it is not less than, then the sliding window is recalculated based on the smaller of the heart rate values ​​represented by the positive normal peak detection result and the reverse normal peak detection result.

[0167] Based on the recalculated sliding window, another forward or reverse normal peak detection is performed corresponding to the smaller value, resulting in a new forward or reverse normal peak detection result. The first or second judgment step is then executed again accordingly.

[0168] Optionally, peak detection may also include enhanced peak detection; enhanced peak detection may include at least one of forward enhanced peak detection and reverse enhanced peak detection.

[0169] Based on the peak detection results, a preliminary assessment is made as to whether the heart rhythm of the cardiac-like organ is regular, specifically including:

[0170] The regularity of the heart rhythm in a cardiac-like heart can be preliminarily determined based on at least one of the following: positive normal peak detection results, reverse normal peak detection results, positive enhanced peak detection results, and reverse enhanced peak detection results.

[0171] Optionally, the heart rate determination module 604 is also used for:

[0172] If the heart rhythm is irregular, set multiple new initial heart rate values ​​that are greater than the initial heart rate value and multiple new initial heart rate values ​​that are less than the initial heart rate value.

[0173] Calculate the corresponding sliding window based on each new initial heart rate value;

[0174] Peak detection is performed on the temporal signal of each feature pixel based on each sliding window to obtain the peak detection result.

[0175] Based on the peak detection results corresponding to each of the multiple new preliminary heart rate values, it is determined whether the heart rhythm of the heart-like organ is truly irregular, and when the heart rate of the heart-like organ is truly regular, the heart rate of the heart-like organ is determined based on the peak detection results corresponding to each of the multiple new preliminary heart rate values.

[0176] Optionally, the device 600 also includes a preprocessing module, specifically for acquiring heart-like video;

[0177] Based on the video, a series of consecutive original heart-like images are determined, and the original heart-like images are cropped to obtain heart-like images.

[0178] The time-domain signal determination module 601 is specifically used for:

[0179] The pixel corresponding to the frequency domain signal with the largest amplitude value within the normal heart rate range is identified as the preliminary feature pixel.

[0180] Determine the amplitude value of the frequency domain signal corresponding to a preset number of pixels within a preset distance of the initial feature pixels on the boundary line of the heart-like structure;

[0181] Compare the difference between the amplitude values ​​of a preset number of pixels and the maximum amplitude value to see if it is within a preset difference range;

[0182] If so, determine the preliminary feature pixels as the feature pixels.

[0183] The specific implementation of each module in device 600 is the same as the heart rate determination method for a heart-like device, and will not be described in detail here.

[0184] This application provides a heart rate determination device for a heart-like sensor. A time-domain signal determination module 601 determines the time-domain signal of feature pixels from multiple consecutive frames of heart-like sensor images. A calculation module 602 determines a preliminary heart rate value based on the time-domain signal of the feature pixels. A peak detection value acquisition module 603 calculates a sliding window using the preliminary heart rate value and uses this sliding window to perform forward and reverse normal peak detection on the time-domain signal of the feature pixels. A heart rate determination module 604 determines the heart rate value of the heart-like sensor based on the peak detection results. Thus, a preliminary heart rate value is first determined, and forward and reverse normal peak detection are performed using the preliminary heart rate value and a sliding window. Then, the preliminary heart rate value is verified based on the forward and reverse peak detection results. When the preliminary heart rate value is verified to be correct, the heart rate of the heart-like sensor is determined based on at least one of the two peak detection results and the preliminary heart rate value, improving the accuracy of heart rate determination for heart-like sensors. This determination method is applicable not only to second-generation heart-like sensors but also to third-generation heart-like sensors.

[0185] This application also provides corresponding electronic devices and computer-readable storage media for implementing the heart-like heart rate determination method provided in this application.

[0186] The device includes a memory and a processor. The memory stores instructions or code, and the processor executes the instructions or code to cause the device to perform a heart rate determination method for a heart-like organ as described in any embodiment of this application.

[0187] In practical applications, the computer-readable storage medium can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. For example, a computer-readable storage medium can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0188] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0189] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0190] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0191] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0192] The above description is merely one specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method of heart rate determination for a heart-like object, characterized by, The method comprises: Converting time domain signals of a plurality of cardiac-like pixels in a plurality of continuous cardiac-like images into frequency domain signals; Determining, according to amplitudes of the frequency domain signals of the plurality of cardiac-like pixels, a pixel corresponding to a frequency domain signal with the largest amplitude in a normal heart rate range as a feature pixel; Determining a preliminary heart rate value according to a time domain signal of the feature pixel; the time domain signal of the feature pixel is a pixel value signal of the feature pixel in a plurality of continuous frames over time; the preliminary heart rate value is a frequency corresponding to the largest amplitude of the feature pixel in the frequency domain signal; Calculating a sliding window based on the preliminary heart rate value; Performing peak detection on the time domain signal of the feature pixel based on the sliding window to obtain a peak detection result; the peak detection includes forward normal peak detection and reverse normal peak detection; Using the peak detection result to verify the preliminary heart rate value, and determining a heart rate value of the cardiac-like object according to at least one of the peak detection result and the preliminary heart rate value when the preliminary heart rate value is correct.

2. The method of claim 1, wherein, The cardiac-like pixels include pixels on a cardiac-like boundary line of the cardiac-like object.

3. The method of claim 1, wherein, Before the step of using the peak detection result to verify the preliminary heart rate value, the method further comprises: Preliminarily determining whether the cardiac-like object has a regular rhythm according to the peak detection result. If the cardiac-like object has a regular rhythm, then using the peak detection result to verify the preliminary heart rate value.

4. The method of claim 3, wherein, The step of using the peak detection result to verify the preliminary heart rate value specifically comprises: A first determining step: determining whether a difference between the forward normal peak detection result and the reverse normal peak detection result is less than a preset threshold value; if yes, then determining that the preliminary heart rate value is correct.

5. The method of claim 3, wherein, The step of using the peak detection result to verify the preliminary heart rate value specifically comprises: A second determining step: determining whether a difference between the forward normal peak detection result and the reverse normal peak detection result is less than a preset threshold value; if yes, then performing enhanced peak detection on the time domain signal of the feature pixel based on the sliding window to obtain an enhanced peak detection result; the enhanced peak detection includes at least one of forward enhanced peak detection and reverse enhanced peak detection; A third determining step: determining whether the enhanced peak detection result, the forward normal peak detection result, and the reverse normal peak detection result are consistent within a preset range; if yes, then determining that the preliminary heart rate value is correct.

6. The method of claim 5, wherein, In the third determining step, if no, then determining that the preliminary heart rate value is incorrect. The method further comprises: When the preliminary heart rate value is incorrect, calculating a heart rate value of the cardiac-like object according to the enhanced peak detection result.

7. The method according to claim 4 or 5, characterized in that, The first determining step or the second determining step specifically comprises: If no, then recalculating the sliding window according to a smaller one of heart rate values represented by the forward normal peak detection result and the reverse normal peak detection result; Based on the re-computed sliding window, the forward normal wave peak detection or the reverse normal wave peak detection corresponding to the smaller value is performed again, to obtain a new forward normal wave peak detection result or a new reverse normal wave peak detection result, and the first judging step or the second judging step is performed again.

8. The method of claim 3, wherein, The wave peak detection further comprises a strengthened wave peak detection, and the strengthened wave peak detection comprises at least one of a forward strengthened wave peak detection and a reverse strengthened wave peak detection. The preliminary judgment of whether the heart rhythm of the heart-like object is regular according to the wave peak detection result specifically comprises: The preliminary judgment of whether the heart rhythm of the heart-like object is regular according to at least one of the forward normal wave peak detection result, the reverse normal wave peak detection result, the forward strengthened wave peak detection result and the reverse strengthened wave peak detection result.

9. The method of claim 3, wherein, The preliminary judgment of whether the heart rhythm of the heart-like object is regular according to the wave peak detection result further specifically comprises: If the heart rhythm is irregular, a plurality of new preliminary heart rate values greater than the preliminary heart rate value and a plurality of new preliminary heart rate values less than the preliminary heart rate value are set based on the preliminary heart rate value; A corresponding sliding window is calculated based on each new preliminary heart rate value; Wave peak detection is performed on the time domain signal of the feature pixel point based on each sliding window to obtain a wave peak detection result; Whether the heart rhythm of the heart-like object is truly irregular is judged according to the wave peak detection result corresponding to each new preliminary heart rate value, and the heart rate of the heart-like object is determined according to the wave peak detection result corresponding to each new preliminary heart rate value when the heart rhythm of the heart-like object is truly regular.

10. The method of claim 1, wherein, The method further comprises: A video of the heart-like object is acquired; Based on the video, a plurality of continuous heart-like original images are determined, and the heart-like original images are cropped to obtain the heart-like image.

11. The method of claim 1, wherein, The pixel point corresponding to the frequency domain signal with the maximum amplitude value in the normal heart rate range is determined as the feature pixel point, specifically comprising: The pixel point corresponding to the frequency domain signal with the maximum amplitude value in the normal heart rate range is determined as the preliminary feature pixel point; The amplitude values of the frequency domain signals of a preset number of pixel points within a preset distance of the preliminary feature pixel point on the boundary line of the heart-like object are determined; Whether the difference between the amplitude values of the preset number of pixel points and the maximum amplitude value is within a preset difference range is compared; If yes, the preliminary feature pixel point is determined as the feature pixel point.

12. A heart-like heart rate determination apparatus, characterized by The device comprises: A time domain signal determination module is configured to convert the time domain signals of a plurality of heart-like object pixel points in a plurality of continuous heart-like object images into frequency domain signals, and determine the pixel point corresponding to the frequency domain signal with the maximum amplitude value in the normal heart rate range as the feature pixel point according to the amplitude values corresponding to the frequency domain signals of the plurality of heart-like object pixel points. The computing module is configured to determine a preliminary heart rate value according to a time domain signal of the feature pixel point, and calculate a sliding window based on the preliminary heart rate value; the wave peak detection module is configured to perform wave peak detection on the time domain signal of the feature pixel point based on the sliding window to obtain a wave peak detection result; the wave peak detection includes forward normal wave peak detection and reverse normal wave peak detection; the time domain signal of the feature pixel point is a pixel value signal of the feature pixel point on a plurality of continuous frames as time elapses; and the preliminary heart rate value is a frequency corresponding to a maximum amplitude value in a frequency domain signal of the feature pixel point. The heart rate determination module is configured to corroborate the preliminary heart rate value based on the wave peak detection result, and determine a heart rate value of the heart-like object according to at least one of the wave peak detection result and the preliminary heart rate value when the preliminary heart rate value is correct.

13. An electronic device, comprising: The method comprises the following steps: A memory is configured to store a computer program; A processor is configured to implement the steps of the heart rate determination method of the heart-like object according to any one of claims 1 to 11 when the computer program is executed.

14. A readable storage medium, characterized by, The computer program is stored on the readable storage medium, and the computer program is configured to implement the steps of the heart rate determination method of the heart-like object according to any one of claims 1 to 11 when the computer program is executed by the processor.

Images