WO2023108331A1 - 自适应实时心电信号质量评估方法 - Google Patents

自适应实时心电信号质量评估方法 Download PDF

Info

Publication number
WO2023108331A1
WO2023108331A1 PCT/CN2021/137405 CN2021137405W WO2023108331A1 WO 2023108331 A1 WO2023108331 A1 WO 2023108331A1 CN 2021137405 W CN2021137405 W CN 2021137405W WO 2023108331 A1 WO2023108331 A1 WO 2023108331A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal quality
heartbeat
ecg
power spectrum
model
Prior art date
Application number
PCT/CN2021/137405
Other languages
English (en)
French (fr)
Inventor
李烨
刘记奎
苗芬
刘增丁
Original Assignee
中国科学院深圳先进技术研究院
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国科学院深圳先进技术研究院 filed Critical 中国科学院深圳先进技术研究院
Priority to PCT/CN2021/137405 priority Critical patent/WO2023108331A1/zh
Publication of WO2023108331A1 publication Critical patent/WO2023108331A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure

Definitions

  • the invention relates to the technical field of wearable monitoring, in particular to an adaptive real-time ECG signal quality evaluation method, device, equipment and storage medium thereof.
  • Wearable monitoring of physiological signals is an important way to achieve early diagnosis of cardiovascular diseases outside the hospital routine.
  • the existing wearable monitoring system has the defect of low model accuracy, which cannot meet the needs of medical-grade continuous physiological monitoring and disease risk prevention and control.
  • One of the important reasons is that the physiological signals monitored by wearable devices are susceptible to noise interference, especially motion artifacts, lead off, and myoelectric interference caused by daily activities. When the noise interference is serious, it is impossible to restore the real signal through the denoising technology. Therefore, in order to improve the accuracy of signal analysis, it is necessary to remove signal fragments with poor quality through signal quality assessment.
  • the waveform parameters of the signal quality are obtained by calculating the area difference under the QRS complex wave between different leads of the ECG signal, and finally the quality evaluation of the multi-lead ECG signal is realized through the statistical method of the histogram and the cumulative histogram; or the existing The signal quality assessment algorithm of multi-lead ECG signal fusion, which realizes the fusion of multi-lead ECG by using the basic idea of local weighted linear prediction.
  • Weighted value estimation based on multi-lead features; or use wavelet transform to decompose ECG signals into different frequency bands, and then calculate time-domain features for each frequency band signal, such as the maximum absolute value amplitude, zero-crossing point, kurtosis and waveform self- Correlation coefficient, and finally classify and identify the signal quality of the merged features through the classifier.
  • the embodiment of the present application provides an adaptive real-time ECG signal quality assessment method, the method includes: using the AR model to extract the features of the power spectrum signal quality; inputting the extracted features into the K-means clustering
  • the K-means clustering algorithm is used to realize the adaptive calculation of the signal quality matching template; the quantitative evaluation of the calculated signal quality is carried out according to the similarity comparison method.
  • the use of the AR model to extract the characteristics of the power spectrum signal quality includes: selecting the order of the AR model between 24 and 28; determining the power of the heartbeat signal through the AR model pair for the ECG m Spectral feature PSD m , where ECG m represents the mth heart beat segment.
  • the clustering algorithm finds the cluster center for the power spectrum feature set PSD m ; selects the preset c heart beats as the similarity matching template.
  • the quantitative evaluation of the calculated signal quality according to the similarity comparison method includes: measuring the similarity between each heartbeat and the template heartbeat by the Pearson correlation coefficient method; A predetermined threshold TR is used to determine whether the quality of the heartbeat signal is acceptable. When the correlation coefficient between each heartbeat segment and the template is greater than the threshold, the quality of the heartbeat signal is considered acceptable, otherwise it is unacceptable.
  • the embodiment of the present application also provides an adaptive real-time ECG signal quality evaluation device, which includes: an extraction unit for extracting features of the power spectrum signal quality using an AR model; a matching unit for Input the extracted features into the K-means clustering algorithm, and realize the adaptive calculation of the signal quality matching template through the K-means clustering algorithm; the evaluation unit is used to quantify the calculated signal quality according to the similarity comparison method Evaluate.
  • the embodiment of the present application also provides a computer device, including a memory, a processor, and a computer program stored in the memory and operable on the processor.
  • the processor executes the program, it implements the The method described in any one of the descriptions of the examples.
  • the embodiment of the present application also provides a computer device, a computer-readable storage medium, on which a computer program is stored, and the computer program is used for: when the computer program is executed by a processor, the computer program according to the present application is implemented.
  • a computer device a computer-readable storage medium, on which a computer program is stored, and the computer program is used for: when the computer program is executed by a processor, the computer program according to the present application is implemented.
  • the self-adaptive real-time ECG signal quality assessment method provided by the present invention solves the influence of certain differences in the ECG between different people and at different times of the same person on the accuracy of signal quality assessment; at the same time, it effectively solves the problem caused by diseases
  • the resulting changes in the heartbeat waveform are misidentified as noise interfering with the signal.
  • FIG. 1 shows a schematic flow chart of an adaptive real-time ECG signal quality assessment method provided by an embodiment of the present application
  • FIG. 2 shows an exemplary structural block diagram of an adaptive real-time ECG signal quality evaluation device 200 according to an embodiment of the present application
  • FIG. 3 shows a schematic structural diagram of a computer system suitable for implementing a terminal device according to an embodiment of the present application
  • Fig. 4 shows a schematic diagram of an ECG signal containing noise interference and heartbeat segmentation results provided by the embodiment of the present application
  • Fig. 5 shows a section of ECG signal including premature ventricular beats and heart beat segmentation results provided by the embodiment of the present application.
  • first and second are used for descriptive purposes only, and cannot be interpreted as indicating or implying relative importance or implicitly specifying the quantity of indicated technical features.
  • the features defined as “first” and “second” may explicitly or implicitly include at least one of these features.
  • “plurality” means at least two, such as two, three, etc., unless otherwise specifically defined.
  • the first feature may be in direct contact with the first feature or the first and second feature may be in direct contact with the second feature through an intermediary. touch.
  • “above”, “above” and “above” the first feature on the second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature.
  • “Below”, “beneath” and “beneath” the first feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is less horizontally than the second feature.
  • FIG. 1 shows a schematic flowchart of an adaptive real-time ECG signal quality assessment method provided by an embodiment of the present application.
  • the method includes:
  • Step 110 using the AR model to extract the characteristics of the power spectrum signal quality
  • Step 120 input the extracted features into the K-means clustering algorithm, and realize the adaptive calculation of the signal quality matching template through the K-means clustering algorithm;
  • Step 130 quantitatively evaluate the calculated signal quality according to the similarity comparison method.
  • the above-mentioned technical solution solves the influence of the difference in the ECG between different people and the same person at different times on the accuracy of signal quality evaluation; at the same time, it effectively solves the problem of heartbeat waveform changes caused by diseases being misunderstood. Problems identified as noise interfering with the signal.
  • using the AR model in this application to extract the characteristics of the power spectrum signal quality includes: selecting the order of the AR model between 24 and 28; The power spectrum feature PSD m , wherein ECG m represents the mth heart beat segment.
  • the similar algorithm calculates the cluster center for the power spectrum feature set PSD m ; selects the preset c heart beats as the similarity matching template.
  • the quantitative evaluation of the calculated signal quality according to the similarity comparison method in the present application includes: measuring the similarity between each heartbeat and the template heartbeat by the Pearson correlation coefficient method; The set threshold TR judges whether the quality of the heartbeat signal is acceptable. When the correlation coefficient between each heartbeat segment and the template is greater than the threshold, the quality of the heartbeat signal is considered acceptable, otherwise it is unacceptable.
  • the technical solution of the present invention mainly requires the following three links: (1) power spectrum signal quality feature extraction: the present invention proposes to use the power spectrum estimated by the AR model as the signal quality feature, and the advantage of this method is high efficiency , the power spectrum curve is smooth. In power spectrum estimation, lower AR model order will reduce the accuracy of power spectrum estimation, while higher node number will produce false peaks. Therefore, selecting the optimal AR model order is particularly important for the accurate estimation of the power spectrum.
  • the invention verifies through experiments that the most accurate power spectrum estimation can be obtained when the order of the AR model is between 24 and 28.
  • the Euclidean distance calculates the true distance between two sample feature vectors, and the number has no exact range
  • the cosine distance calculates the cosine of the angle between the two sample feature vectors, and the value ranges from -1 to 1, -1 means that the direction of the two vectors is opposite, and 1 means that the direction is the same
  • PCC is a measure of the linear correlation between two sample vectors, and the value ranges from 0 to 1. The larger the value, the greater the correlation between the two. Obviously, the Pearson correlation coefficient is more suitable for this problem.
  • the unsupervised beat-to-beat signal quality detection mainly includes the following steps:
  • the present invention uses time-domain waveform features and frequency-domain power spectrum features to judge the beat-to-beat quality of representative ECG signals.
  • the experimental data comes from a single-lead ECG signal collected by a Huawei watch (model: WATCH 3), with a sampling rate of 1000Hz.
  • Figure 4 and Table 1 show the signal quality evaluation results of a section containing noisy heart beats;
  • Figure 5 and Table 2 show the signal quality evaluation results of a section of ventricular premature beats; it can be seen that the power spectrum features estimated based on the AR model It is more sensitive to noise and avoids judging premature ventricular beats as signals with poor signal quality.
  • FIG. 2 shows an exemplary structural block diagram of an adaptive real-time ECG signal quality evaluation device 200 according to an embodiment of the present application.
  • the device includes:
  • An extraction unit 210 configured to extract features of the power spectrum signal quality by using an AR model
  • the matching unit 220 is used to input the extracted features into the K-means clustering algorithm, and realize the adaptive calculation of the signal quality matching template by the K-means clustering algorithm;
  • the evaluation unit 230 is configured to evaluate the quantification of the calculated signal quality according to the similarity comparison method.
  • the units or modules recorded in the device 200 correspond to the steps in the method described with reference to FIG. 1 . Therefore, the operations and features described above for the method are also applicable to the device 200 and the units contained therein, and will not be repeated here.
  • the apparatus 200 may be pre-implemented in the browser of the electronic device or other security applications, and may also be loaded into the browser of the electronic device or its security applications by downloading or other means.
  • the corresponding units in the apparatus 200 may cooperate with the units in the electronic device to implement the solutions of the embodiments of the present application.
  • FIG. 3 shows a schematic structural diagram of a computer system 300 suitable for implementing a terminal device or a server according to an embodiment of the present application.
  • a computer system 300 includes a central processing unit (CPU) 301 that can operate according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage section 308 into a random-access memory (RAM) 303 Instead, various appropriate actions and processes are performed.
  • ROM read-only memory
  • RAM random-access memory
  • various programs and data required for the operation of the system 300 are also stored.
  • the CPU 301, ROM 302, and RAM 303 are connected to each other through a bus 304.
  • An input/output (I/O) interface 305 is also connected to the bus 304 .
  • the following components are connected to the I/O interface 305: an input section 306 including a keyboard, a mouse, etc.; an output section 307 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage section 308 including a hard disk, etc. and a communication section 309 including a network interface card such as a LAN card, a modem, or the like.
  • the communication section 309 performs communication processing via a network such as the Internet.
  • a drive 310 is also connected to the I/O interface 305 as needed.
  • a removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is mounted on the drive 310 as necessary so that a computer program read therefrom is installed into the storage section 308 as necessary.
  • an embodiment of the present disclosure includes an adaptive real-time ECG signal quality assessment method, which includes a computer program tangibly embodied on a machine-readable medium, the computer program including program code for executing the method of FIG. 1 .
  • the computer program may be downloaded and installed from a network via communication portion 309 and/or installed from removable media 311 .
  • each block in a flowchart or block diagram may represent a module, program segment, or part of code that includes one or more logical functions for implementing specified executable instructions.
  • the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations can be implemented by a dedicated hardware-based system that performs the specified functions or operations , or may be implemented by a combination of dedicated hardware and computer instructions.
  • the units or modules involved in the embodiments described in the present application may be implemented by means of software or by means of hardware.
  • the described units or modules may also be set in a processor.
  • a processor includes a first sub-region generating unit, a second sub-region generating unit, and a display region generating unit.
  • the names of these units or modules do not constitute limitations on the units or modules themselves in some cases, for example, the display area generation unit can also be described as "used to generate The cell of the display area of the text".
  • the present application also provides a computer-readable storage medium, which may be the computer-readable storage medium contained in the aforementioned devices in the above-mentioned embodiments; computer-readable storage media stored in the device.
  • the computer-readable storage medium stores one or more programs, and the aforementioned programs are used by one or more processors to execute the text generation method applied to transparent window envelopes described in this application.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Physiology (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Engineering & Computer Science (AREA)
  • Cardiology (AREA)
  • Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)

Abstract

一种自适应实时心电信号质量评估方法、装置、设备及其存储介质，方法包括：利用AR模型对功率谱信号质量的特征进行提取（110）；将提取的特征输入到K-means聚类算法中，通过K-means聚类算法实现信号质量匹配模板的自适应计算（120）；根据相似度比较法对计算后的信号质量的定量进行评估（130）。解决了心电图在不同人之间以及同一个人的不同时期间都存在一定的差异性对信号质量评估精度的影响；同时，有效解决了因疾病引起的心搏波形的改变被误识别为噪声干扰信号的问题。

Description

自适应实时心电信号质量评估方法 技术领域
本发明涉及可穿戴监测技术领域，具体涉及一种自适应实时心电信号质量评估方法、装置、设备及其存储介质。
背景技术
生理信号的可穿戴监测是实现医院外日常心血管疾病早期诊断的重要途径。然而在实际应用中，现有可穿戴监测系统存在模型精度低缺陷，无法满足医疗级的连续生理监测及疾病风险防控的需求。其中一个重要原因是可穿戴设备监测的生理信号容易受到噪声的干扰，尤其是日常活动引起的运动伪影、导联脱落以及肌电干扰。当这些噪声干扰比较严重时，已无法通过去噪技术恢复出真实信号。因此，为了提高信号分析的准确性，需要通过信号质量评估去除质量较差的信号片段。现有技术中通过计算ECG信号不同导联间QRS复合波下面积差异获得信号质量的波形参数，最后通过直方图和累计直方图的统计学方法实现多导联ECG信号的质量评估；或者现有的多导联ECG信号融合的信号质量评估算法，其通过利用局部加权线性预测的基本思想实现多导联ECG的融合，同时为了有效保留ECG信号的质量相关特征，该方法还将模糊推理系统用于多导联特征的加权值估计；或者利用小波变换将ECG信号分解到不同的频带上，然后对每个频带信号计算时域特征，如最大绝对值振幅、过零点、峰度和波形的自相关系 数，最后通过分类器对合并的特征进行信号质量分类识别。
对于在线可穿戴心血管疾病监测系统，由于实时性需求一般需要依赖于单一类型的特征通过无监督方法进行信号质量评估，这种方法具有实现简单、实时性好的优点，但容易将因心脏疾病引起的信号波形变化误识别为信号质量较差的信号。心脏疾病波形在被识别为差质量信号后会被剔除，从而对该心血管事件的发生失去预警作用，增加了事故发生率。
发明内容
鉴于现有技术中的上述缺陷或不足，期望提供一种自适应实时心电信号质量评估方法、装置、设备及其存储介质。
第一方面，本申请实施例提供了一种自适应实时心电信号质量评估方法，该方法包括：利用AR模型对功率谱信号质量的特征进行提取；将提取的特征输入到K-means聚类算法中，通过K-means聚类算法实现信号质量匹配模板的自适应计算；根据相似度比较法对计算后的信号质量的定量进行评估。
在其中一个实施例中，所述利用AR模型对功率谱信号质量的特征进行提取之前，该方法还包括：通过波形检测算法检测ECG的R峰，以R峰为基准点进行心搏分割，以R n表示第n个R峰，以第m个心搏片段表示R n和R n+2之间的ECG信号片段，其中n＝2，4，6，8...，m＝n/2。
在其中一个实施例中，所述利用AR模型对功率谱信号质量的特征进行提取，包括：选择AR模型阶数在24～28之间；对ECG m通过AR模型对确定该心搏信号的功率谱特征PSD m，其中，ECG m表示第m个心搏片段。
在其中一个实施例中，所述将提取的特征输入到K-means聚类算法中，通过K-means聚类算法实现信号质量匹配模板的自适应计算，包括：通过k＝1的K-means聚类算法对功率谱特征集PSD m求聚类中心；选取预设的c个心搏求平均作为相似度匹配模板。
在其中一个实施例中，所述根据相似度比较法对计算后的信号质量的定量进行评估，包括：通过皮尔逊相关系数方法度量每个心搏与模板心搏之间的相似性；根据设定的阈值TR判断所述心搏信号的质量是否被接受，当每个心搏片段与模板的相关系数大于阈值时认为所述心搏的信号质量可以接受，否则为不可接受。
第二方面，本申请实施例还提供了一种自适应实时心电信号质量评估装置，该装置包括：提取单元，用于利用AR模型对功率谱信号质量的特征进行提取；匹配单元，用于将提取的特征输入到K-means聚类算法中，通过K-means聚类算法实现信号质量匹配模板的自适应计算；评估单元，用于根据相似度比较法对计算后的信号质量的定量进行评估。
第三方面，本申请实施例还提供了一种计算机设备，包括存储器、处理器以及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现如本申请实施例描述中任一所述的方法。
第四方面，本申请实施例还提供了一种计算机设备一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序用于：所述计算机程序被处理器执行时实现如本申请实施例描述中任一所述的方法。
本发明的有益效果：
本发明提供的自适应实时心电信号质量评估方法，解决了心电图在不同人之间以及同一个人的不同时期间都存在一定的差异性对信号质量评估精度的影响；同时，有效解决了因疾病引起的心搏波形的改变被误识别为噪声干扰信号的问题。
附图说明
通过阅读参照以下附图所作的对非限制性实施例所作的详细描述，本申请的其它特征、目的和优点将会变得更明显：
图1示出了本申请实施例提供的自适应实时心电信号质量评估方法的流程示意图；
图2示出了根据本申请一个实施例的自适应实时心电信号质量评估装置200的示例性结构框图；
图3示出了适于用来实现本申请实施例的终端设备的计算机系统的结构示意图；
图4示出了本申请实施例提供的一段包含噪声干扰的ECG信号以及心搏分割结果的示意图；
图5示出了本申请实施例提供的一段包含室性早搏的ECG信号以及心搏分割结果。
具体实施方式
为使本发明的上述目的、特征和优点能够更加明显易懂，下面结合附图对本发明的具体实施方式做详细的说明。在下面的描述中阐述了很多具体细节以便于充分理解本发明。但是本发明能够以很多不同于在此描述的其它方式来实施，本领域技术人员可以在不违背本发明内涵的情况下做类似改进，因此本发明不受下面公开的具体实施例的限制。
在本发明的描述中，需要理解的是，术语“中心”、“纵向”、“横向”、“长度”、“宽度”、“厚度”、“上”、“下”、“前”、“后”、“左”、“右”、“竖直”、“水平”、“顶”、“底”、“内”、“外”、“顺时针”、“逆时针”、 “轴向”、“径向”、“周向”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本发明和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本发明的限制。
此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。在本发明的描述中，“多个”的含义是至少两个，例如两个，三个等，除非另有明确具体的限定。
在本发明中，除非另有明确的规定和限定，术语“安装”、“相连”、“连接”、“固定”等术语应做广义理解，例如，可以是固定连接，也可以是可拆卸连接，或成一体；可以是机械连接，也可以是电连接；可以是直接相连，也可以通过中间媒介间接相连，可以是两个元件内部的连通或两个元件的相互作用关系，除非另有明确的限定。对于本领域的普通技术人员而言，可以根据具体情况理解上述术语在本发明中的具体含义。
在本发明中，除非另有明确的规定和限定，第一特征在第二特征“上”或“下”可以是第一和第二特征直接接触，或第一和第二特征通过中间媒介间接接触。而且，第一特征在第二特征“之上”、“上方”和“上面”可是第一特征在第二特征正上方或斜上方，或仅仅表示第一特征水平高度高于第二特征。第一特征在第二特征“之下”、“下方”和“下面”可以是第一特征在第二特征正下方或斜下方，或仅仅表示第一特征水平高度小于第二特征。
需要说明的是，当元件被称为“固定于”或“设置于”另一个元件，它可以直接在另一个元件上或者也可以存在居中的元件。当一个元件被认为是“连接”另一个元件，它可以是直接连接到另一个元件 或者可能同时存在居中元件。本文所使用的术语“垂直的”、“水平的”、“上”、“下”、“左”、“右”以及类似的表述只是为了说明的目的，并不表示是唯一的实施方式。
请参考图1，图1示出了本申请实施例提供的自适应实时心电信号质量评估方法的流程示意图。
如图1所示，该方法包括：
步骤110，利用AR模型对功率谱信号质量的特征进行提取；
步骤120，将提取的特征输入到K-means聚类算法中，通过K-means聚类算法实现信号质量匹配模板的自适应计算；
步骤130，根据相似度比较法对计算后的信号质量的定量进行评估。
采用上述技术方案，解决了心电图在不同人之间以及同一个人的不同时期间都存在一定的差异性对信号质量评估精度的影响；同时，有效解决了因疾病引起的心搏波形的改变被误识别为噪声干扰信号的问题。
在一些实施例中，本申请中的利用AR模型对功率谱信号质量的特征进行提取之前，该方法还包括：通过波形检测算法检测ECG的R峰，以R峰为基准点进行心搏分割，以R n表示第n个R峰，以第m个心搏片段表示R n和R n+2之间的ECG信号片段，其中n＝2，4，6，8...，m＝n/2。
在一些实施例中，本申请中的利用AR模型对功率谱信号质量的特征进行提取，包括：选择AR模型阶数在24～28之间；对ECG m通过AR模型对确定该心搏信号的功率谱特征PSD m，其中，ECG m表示第m个心搏片段。
在一些实施例中，所述将提取的特征输入到K-means聚类算法中， 通过K-means聚类算法实现信号质量匹配模板的自适应计算，包括：通过k＝1的K-means聚类算法对功率谱特征集PSD m求聚类中心；选取预设的c个心搏求平均作为相似度匹配模板。
在一些实施例中，本申请中的根据相似度比较法对计算后的信号质量的定量进行评估，包括：通过皮尔逊相关系数方法度量每个心搏与模板心搏之间的相似性；根据设定的阈值TR判断所述心搏信号的质量是否被接受，当每个心搏片段与模板的相关系数大于阈值时认为所述心搏的信号质量可以接受，否则为不可接受。
综上所述，本发明的技术方案主要需要以下三个环节：(1)功率谱信号质量特征提取：本发明提出使用AR模型估计出的功率谱作为信号质量特征，该方法的优点是效率高，功率谱曲线平滑。在功率谱估计中，较低的AR模型阶数会使功率谱估计的精度降低，而较高的节数又会产生虚假的峰值。因此，选择最优的AR模型阶数对功率谱的准确估计尤为重要。本发明通过实验验证了当AR模型的阶数在24～28之间时能获得最精确的功率谱估计。(2)信号质量匹配模板计算：对于无监督逐搏信号质量检测方法，首先需要用“干净”的ECG心搏周期去确定一个模板，然后再通过相似性度量方法计算检测心搏与模板的相似度，当相似度超过指定阈值时，则认为该心搏周期的信号质量是“可接受的”。由于心电图在不同人之间以及同一个人的不同时期间都存在一定的差异性，因此为避免这种差异性对精度的影响，需要在信号测量时自适应的选取心搏模板。适应模板选择的关键是要尽可能避免质量较差信号的干扰，这就要求去处异常心搏信号。(3)相似性度量：常见的方法有欧几里得距离(Euclidean Distance)、余弦距离(Cosine Distance)和皮尔逊相关系数(Pearson Correlation Coefficient，PCC)。这三种方法中，欧氏距离计算的是两个样本特征向量间的真实距离，数字没有确切的范围；余弦距离计算的是两个样 本特征向量间的夹角余弦，数值范围为-1到1，-1表示两向量的方向相反，1则表示方向相同；PCC是两个样本向量间的线性相关性的度量，数值范围为0到1，数值越大二者的相关性越大。显然，皮尔逊相关系数更适合用于本问题。
对于给定的一段ECG信号，基于无监督的逐搏信号质量检测的主要包括以下步骤：
1)通过波形检测算法检测ECG的R峰，然后以R峰为基准点进行心搏分割，以R n表示第n个R峰，以第m个心搏片段表示R n和R n+2之间的ECG信号片段，其中n＝2，4，6，8...，m＝n/2。为表达方便，同时，使用ECG m表示第m个心搏片段。
2)对ECG m通过AR模型估计该心搏信号的功率谱特征PSD m。
3)通过k＝1的K-means算法对功率谱特征集PSD m求聚类中心，选取最近的c个心搏求平均作为相似度匹配模板。
4)使用皮尔逊相关系数方法度量每个心搏与模板心搏之间的相似性，相似性越高表明质量评分越高，反之则越低。在应用时，根据设定的阈值TR判断该心搏信号的质量是否被接受，当相关系数大于阈值时认为该心搏的信号质量可以接受，否则为不可接受。
本发明分别使用时域波形特征和频域功率谱特征对有代表性意义的ECG信号进行逐搏质量判断。实验数据来自通过华为手表(型号：WATCH 3)采集的单导联ECG信号，采样率为1000Hz。如图4与表1展示了一段包含噪声心搏的信号质量评估结果；图5与表2展示了一段包含室性早搏心搏的信号质量评估结果；可以看出基于AR模型估计的功率谱特征对于噪声更加敏感，且避免了将室性早搏心搏判别为信号质量较差信号。
表1.在图4中每个心搏片段与模板的相关系数
Figure PCTCN2021137405-appb-000001
表2.在图5中每个心搏片段与模板的相关系数
Figure PCTCN2021137405-appb-000002
进一步地，参考图2，图2示出了根据本申请一个实施例的自适应实时心电信号质量评估装置200的示例性结构框图。
如图2所示，该装置包括：
提取单元210，用于利用AR模型对功率谱信号质量的特征进行提取；
匹配单元220，用于将提取的特征输入到K-means聚类算法中，通过K-means聚类算法实现信号质量匹配模板的自适应计算；
评估单元230，用于根据相似度比较法对计算后的信号质量的定量进行评估。
应当理解，装置200中记载的诸单元或模块与参考图1描述的方法中的各个步骤相对应。由此，上文针对方法描述的操作和特征同样适用于装置200及其中包含的单元，在此不再赘述。装置200可以预先实现在电子设备的浏览器或其他安全应用中，也可以通过下载等方式而加载到电子设备的浏览器或其安全应用中。装置200中的相应单元可以与电子设备中的单元相互配合以实现本申请实施例的方案。
下面参考图3，其示出了适于用来实现本申请实施例的终端设备或服务器的计算机系统300的结构示意图。
如图3所示，计算机系统300包括中央处理单元(CPU)301，其可以根据存储在只读存储器(ROM)302中的程序或者从存储部分308加载到随机访问存储器(RAM)303中的程序而执行各种适当的动作和处理。在RAM 303中，还存储有系统300操作所需的各种程序和数据。CPU 301、ROM 302以及RAM 303通过总线304彼此相连。输入/输出(I/O)接口305也连接至总线304。
以下部件连接至I/O接口305：包括键盘、鼠标等的输入部分306；包括诸如阴极射线管(CRT)、液晶显示器(LCD)等以及扬声器等的输出部分307；包括硬盘等的存储部分308；以及包括诸如LAN卡、调制解调器等的网络接口卡的通信部分309。通信部分309经由诸如因特网的网络执行通信处理。驱动器310也根据需要连接至I/O接口305。可拆卸介质311，诸如磁盘、光盘、磁光盘、半导体存储器等等，根据需要安装在驱动器310上，以便于从其上读出的计算机程序根据需要被安装入存储部分308。
特别地，根据本公开的实施例，上文参考图1描述的过程可以被实现为计算机软件程序。例如，本公开的实施例包括一种自适应实时心电信号质量评估方法，其包括有形地包含在机器可读介质上的计算机程序，所述计算机程序包含用于执行图1的方法的程序代码。在这样的实施例中，该计算机程序可以通过通信部分309从网络上被下载和安装，和/或从可拆卸介质311被安装。
附图中的流程图和框图，图示了按照本发明各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分，前述模块、程序段、或代码的一部分包含一个或多个用于 实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个接连地表示的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或操作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
描述于本申请实施例中所涉及到的单元或模块可以通过软件的方式实现，也可以通过硬件的方式来实现。所描述的单元或模块也可以设置在处理器中，例如，可以描述为：一种处理器包括第一子区域生成单元、第二子区域生成单元以及显示区域生成单元。其中，这些单元或模块的名称在某种情况下并不构成对该单元或模块本身的限定，例如，显示区域生成单元还可以被描述为“用于根据第一子区域和第二子区域生成文本的显示区域的单元”。
作为另一方面，本申请还提供了一种计算机可读存储介质，该计算机可读存储介质可以是上述实施例中前述装置中所包含的计算机可读存储介质；也可以是单独存在，未装配入设备中的计算机可读存储介质。计算机可读存储介质存储有一个或者一个以上程序，前述程序被一个或者一个以上的处理器用来执行描述于本申请的应用于透明窗口信封的文本生成方法。
以上描述仅为本申请的较佳实施例以及对所运用技术原理的说明。本领域技术人员应当理解，本申请中所涉及的发明范围，并不限于上述技术特征的特定组合而成的技术方案，同时也应涵盖在不脱离前述发明构思的情况下，由上述技术特征或其等同特征进行任意组合而形成的其它技术方案。例如上述特征与本申请中公开的(但不限于)具有类似功能的技术特征进行互相替换而形成的技术方案。

Claims (10)

  1. 一种自适应实时心电信号质量评估方法，其特征在于，该方法包括：
    利用AR模型对功率谱信号质量的特征进行提取；
    将提取的特征输入到K-means聚类算法中，通过K-means聚类算法实现信号质量匹配模板的自适应计算；
    根据相似度比较法对计算后的信号质量的定量进行评估。
  2. 根据权利要求1所述的自适应实时心电信号质量评估方法，其特征在于，所述利用AR模型对功率谱信号质量的特征进行提取之前，该方法还包括：
    通过波形检测算法检测ECG的R峰，以R峰为基准点进行心搏分割，以R n表示第n个R峰，以第m个心搏片段表示R n和R n+2之间的ECG信号片段，其中n＝2，4，6，8...，m＝n/2。
  3. 根据权利要求2所述的自适应实时心电信号质量评估方法，其特征在于，所述利用AR模型对功率谱信号质量的特征进行提取，包括：
    选择AR模型阶数在24～28之间；
    对ECG m通过AR模型对确定该心搏信号的功率谱特征PSD m，其中，ECG m表示第m个心搏片段。
  4. 根据权利要求3所述的自适应实时心电信号质量评估方法，其特征在于，所述将提取的特征输入到K-means聚类算法中，通过K-means聚类算法实现信号质量匹配模板的自适应计算，包括：
    通过k＝1的K-means聚类算法对功率谱特征集PSD m求聚类中心；
    选取预设的c个心搏求平均作为相似度匹配模板。
  5. 根据权利要求4所述的自适应实时心电信号质量评估方法，其特征在于，所述根据相似度比较法对计算后的信号质量的定量进行评估，包括：
    通过皮尔逊相关系数方法度量每个心搏与模板心搏之间的相似性；
    根据设定的阈值TR判断所述心搏信号的质量是否被接受，当每个心搏片段与模板的相关系数大于阈值时认为所述心搏的信号质量可以接受，否则为不可接受。
  6. 一种自适应实时心电信号质量评估装置，其特征在于，该装置包括：
    提取单元，用于利用AR模型对功率谱信号质量的特征进行提取；
    匹配单元，用于将提取的特征输入到K-means聚类算法中，通过K-means聚类算法实现信号质量匹配模板的自适应计算；
    评估单元，用于根据相似度比较法对计算后的信号质量的定量进行评估。
  7. 根据权利要求所述的自适应实时心电信号质量评估装置，其特征在于，所述利用AR模型对功率谱信号质量的特征进行提取之前，该装置还包括：
    通过波形检测算法检测ECG的R峰，以R峰为基准点进行心搏分割，以R n表示第n个R峰，以第m个心搏片段表示R n和R n+2之间的ECG信号片段，其中n＝2，4，6，8...，m＝n/2。
  8. 根据权利要求7所述的自适应实时心电信号质量评估装置，其特征在于，所述利用AR模型对功率谱信号质量的特征进行提取，包括：
    选择AR模型阶数在24～28之间；
    对ECG m通过AR模型对确定该心搏信号的功率谱特征PSD m，其中，ECG m表示第m个心搏片段。
  9. 一种计算机设备，包括存储器、处理器以及存储在存储器上并可在处理器上运行的计算机程序，其特征在于，所述处理器执行所述程序时实现如权利要求1-5中任一所述的方法。
  10. 一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序用于：
    所述计算机程序被处理器执行时实现如权利要求1-5中任一所述的方法。
PCT/CN2021/137405 2021-12-13 2021-12-13 自适应实时心电信号质量评估方法 WO2023108331A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/137405 WO2023108331A1 (zh) 2021-12-13 2021-12-13 自适应实时心电信号质量评估方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/137405 WO2023108331A1 (zh) 2021-12-13 2021-12-13 自适应实时心电信号质量评估方法

Publications (1)

Publication Number Publication Date
WO2023108331A1 true WO2023108331A1 (zh) 2023-06-22

Family

ID=86775255

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/137405 WO2023108331A1 (zh) 2021-12-13 2021-12-13 自适应实时心电信号质量评估方法

Country Status (1)

Country Link
WO (1) WO2023108331A1 (zh)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160242664A1 (en) * 2013-10-09 2016-08-25 Koninklijke Philips N.V. An apparatus and method for evaluating multichannel ecg signals
CN106456030A (zh) * 2014-06-13 2017-02-22 日东电工株式会社 用于消除生理测量中的伪影的设备和方法
CN108289615A (zh) * 2015-11-24 2018-07-17 皇家飞利浦有限公司 用于量化光电体积描记（ppg）信号质量的方法
WO2018137300A1 (zh) * 2017-01-25 2018-08-02 华为技术有限公司 一种生理信号质量判断方法及装置
CN109117769A (zh) * 2018-07-31 2019-01-01 东南大学 一种针对穿戴式心电采集的实时质量评估反馈方法
CN110123304A (zh) * 2019-01-22 2019-08-16 东南大学 基于多模板匹配和相关系数矩阵的动态心电噪声滤除方法
US20200037961A1 (en) * 2017-01-17 2020-02-06 Mayo Foundation For Medical Education And Research System and method for assessing physiological signal quality
US20210267530A1 (en) * 2018-06-29 2021-09-02 Mybrain Technologies Multiclass classification method for the estimation of eeg signal quality

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160242664A1 (en) * 2013-10-09 2016-08-25 Koninklijke Philips N.V. An apparatus and method for evaluating multichannel ecg signals
CN106456030A (zh) * 2014-06-13 2017-02-22 日东电工株式会社 用于消除生理测量中的伪影的设备和方法
CN108289615A (zh) * 2015-11-24 2018-07-17 皇家飞利浦有限公司 用于量化光电体积描记（ppg）信号质量的方法
US20200037961A1 (en) * 2017-01-17 2020-02-06 Mayo Foundation For Medical Education And Research System and method for assessing physiological signal quality
WO2018137300A1 (zh) * 2017-01-25 2018-08-02 华为技术有限公司 一种生理信号质量判断方法及装置
US20210267530A1 (en) * 2018-06-29 2021-09-02 Mybrain Technologies Multiclass classification method for the estimation of eeg signal quality
CN109117769A (zh) * 2018-07-31 2019-01-01 东南大学 一种针对穿戴式心电采集的实时质量评估反馈方法
CN110123304A (zh) * 2019-01-22 2019-08-16 东南大学 基于多模板匹配和相关系数矩阵的动态心电噪声滤除方法

Similar Documents

Publication Publication Date Title
Sharma et al. Accurate tunable-Q wavelet transform based method for QRS complex detection
Varghees et al. Effective heart sound segmentation and murmur classification using empirical wavelet transform and instantaneous phase for electronic stethoscope
US9042973B2 (en) Apparatus and method for measuring physiological signal quality
Sharma et al. Accurate automated detection of congestive heart failure using eigenvalue decomposition based features extracted from HRV signals
US9295397B2 (en) Method and apparatus for beat-space frequency domain prediction of cardiovascular death after acute coronary event
US20180235493A1 (en) Patient Signal Analysis Based on Vector Analysis
CN114469131A (zh) 自适应实时心电信号质量评估方法
US8684942B2 (en) System for cardiac impairment detection based on signal regularity
US10912479B2 (en) Method for accurately extracting abnormal potential within QRS
CN112971795B (zh) 心电信号质量评估方法
Jaswal et al. QRS detection using wavelet transform
Bruun et al. Automatic atrial fibrillation detection: A novel approach using discrete wavelet transform and heart rate variability
Bhoi et al. QRS Complex Detection and Analysis of Cardiovascular Abnormalities: A Review.
Jarvis et al. Apnea patients characterized by 0.02 Hz peak in the multitaper spectrogram of electrocardiogram signals
EP3203900A1 (en) Weaning readiness indicator, sleeping status recording device, and air providing system applying nonlinear time-frequency analysis
Yu et al. Accurate wavelet thresholding method for ECG signals
So-In et al. Real-time ECG noise reduction with QRS complex detection for mobile health services
WO2023108331A1 (zh) 自适应实时心电信号质量评估方法
Li et al. Crackles detection method based on time-frequency features analysis and SVM
US9020583B2 (en) Patient signal analysis and characterization
Haibing et al. A method of continuous wavelet transform for qrs wave detection in ecg signal
US20230285004A1 (en) Ultrasound data processor
Liu et al. A Rhythm-Specific ECG Signal Quality Assessment Framework for Robust Cardiac Health Monitoring of AI-based Arrhythmia Classifer
WO2018023696A1 (zh) 一种胎儿心电分离方法及装置
EP3967237A1 (en) Ultrasound data processor

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21967474

Country of ref document: EP

Kind code of ref document: A1