WO2022032947A1 - 长程心电数据下快速标注方法及装置 - Google Patents

长程心电数据下快速标注方法及装置 Download PDF

Info

Publication number
WO2022032947A1
WO2022032947A1 PCT/CN2020/135535 CN2020135535W WO2022032947A1 WO 2022032947 A1 WO2022032947 A1 WO 2022032947A1 CN 2020135535 W CN2020135535 W CN 2020135535W WO 2022032947 A1 WO2022032947 A1 WO 2022032947A1
Authority
WO
WIPO (PCT)
Prior art keywords
corrected
ecg data
points
long
interval
Prior art date
Application number
PCT/CN2020/135535
Other languages
English (en)
French (fr)
Inventor
罗琼
罗申
王宏燕
Original Assignee
深圳星康医疗科技有限公司
肇庆星网医疗科技有限公司
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 深圳星康医疗科技有限公司, 肇庆星网医疗科技有限公司 filed Critical 深圳星康医疗科技有限公司
Publication of WO2022032947A1 publication Critical patent/WO2022032947A1/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
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes

Definitions

  • the present invention relates to the technical field of electrocardiographic data processing, and in particular, to a method and device for fast labeling under long-range electrocardiographic data.
  • the present invention provides a method and device for fast labeling under long-range ECG data, so as to solve the related technical problems.
  • a rapid labeling method under long-range ECG data includes:
  • a Cartesian coordinate system is established for the ECG data of adjacent RR intervals in the selected t-RR scattergram.
  • the adjacent RR intervals are the X and Y coordinate systems, respectively;
  • the cardiac pacemaker points that need to be corrected and the cardiac pacemaker points that do not need to be corrected are separated, and the cardiac pacemaker points that need to be corrected are processed into a spring compression map, and noise is marked in batches by labeling the spring compression map.
  • drawing a t-RR scatterplot according to the long-range ECG data includes:
  • establishing a rectangular coordinate system for the ECG data of adjacent RR intervals in the selected t-RR scatterplot including:
  • a rectangular coordinate system is established to form a scatter diagram of cardiac pacemakers. Periods are continuous in time.
  • separating the heartbeat points that need to be corrected and the heartbeat points that do not need to be corrected according to the circled instructions, and processing the heartbeat points that need to be corrected into a spring compression map including:
  • the cardiac pacing points in the index coordinate system are divided into cardiac pacing points that need to be corrected and cardiac pacing points that do not need to be corrected;
  • a second aspect of the embodiments of the present invention provides a method and device for rapid labeling under long-range ECG data, including:
  • a first tool module for receiving long-range ECG data, and drawing a t-RR scatterplot according to the long-range ECG data
  • the second tool module is configured to establish a Cartesian coordinate system for the ECG data of adjacent RR intervals in the selected t-RR scattergram according to one or more selected sections of the t-RR scattergram, a
  • the RR interval and the adjacent RR interval are the X and Y coordinate systems, respectively;
  • the third tool module is used to separate the cardiac pacemaker points that need to be corrected and the cardiac pacemaker points that do not need to be corrected according to the circled instructions, and process the cardiac pacemaker points that need to be corrected into a spring compression map. Spring Compression Map Batch Label Noise.
  • the first tool module includes:
  • the first processing unit is configured to draw a t-RR scatterplot with time as the horizontal axis and RR interval as the vertical axis, wherein the long-range ECG data includes time and RR interval.
  • the second tool module includes:
  • the second processing unit is configured to use the i-th RR interval as the horizontal axis and the i+1-th RR interval as the vertical axis to establish a Cartesian coordinate system to form a scatter diagram of cardiac pacemakers.
  • the RR intervals of the vertical axis are continuous in time.
  • the third tool module includes:
  • the third processing unit is configured to divide the cardiac pacing points in the index coordinate system into cardiac pacing points that need to be corrected and cardiac pacing points that do not need to be corrected according to the area circled by the user on the rectangular coordinate system point;
  • a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, the aforementioned method for quickly labeling long-range ECG data is implemented.
  • the ECG data is visualized, the approximate time period of the ECG data noise is first determined by the t-RR scattergram, and then the cardiac pacing points of each selected time period are drawn correspondingly. Finally, the characteristics of arrhythmia are analyzed, and the normal heartbeat points and ectopic heartbeat points are circled and compressed to facilitate rapid batch labeling.
  • FIG. 1 is a schematic flowchart of a method for fast labeling under long-range ECG data in an embodiment of the present invention
  • FIG. 2 is a RR-t scatter diagram in an embodiment of the present invention
  • FIG. 4 is a Lawrence scatter diagram after being circled in an embodiment of the present invention.
  • FIG. 5 is a compression diagram of a spring in an embodiment of the present invention.
  • FIG. 6 is a schematic block diagram of a device for quickly labeling long-range ECG data according to an embodiment of the present invention.
  • FIG. 1 is a schematic flowchart of an embodiment of a method for quickly labeling long-range ECG data according to an embodiment of the present invention.
  • the rapid labeling method under the long-range ECG data includes the following steps:
  • Step S100 Receive long-range ECG data, and draw a t-RR scatterplot according to the long-range ECG data.
  • the RR interval is the time period between two R waves on the ECG.
  • the normal RR interval time limit should be between 0.6-1.0 seconds, with less than 0.6 seconds indicating tachycardia, and greater than 1.0 seconds indicating bradycardia.
  • the RR intervals when the RR intervals are not equal, it indicates that arrhythmia occurs, and there will be obvious RR intervals on the ECG of atrial fibrillation.
  • the RR interval should be equal to the PP interval, which is the time limit between two P waves, and should also be between 0.6-1.0 seconds.
  • the RR interval reflects the ventricular rhythm
  • the PP interval reflects the atrial rhythm. When the atrial rhythm and ventricular rhythm are not equal, conduction block may occur. In the ECG of third-degree AV block, absolute disparity between PP and RR intervals occurs.
  • the present invention draws long-range ECG data into a t-RR scattergram based on time and RR interval, which is used to assist users in selecting areas with more interference.
  • the horizontal axis of the t-RR scatter plot is the time t, and the unit can be the length of time such as hours and days.
  • the vertical axis of the scatter plot is the RR interval, and the unit is mm. Users can judge where the interference is based on the characteristics of irregular distribution and high dispersion of the scatter plot, and can select the corresponding scatter plot from one or more sections on the t-RR scatter plot.
  • Step S200 according to the selected one or more sections of the t-RR scatter diagram, the ECG data of the adjacent RR intervals in the selected t-RR scatter diagram are established in a rectangular coordinate system, an RR interval and The adjacent RR intervals are the X and Y coordinate systems, respectively.
  • the graph formed by the established Cartesian coordinate system is a Lawrence scatter plot.
  • the main advantage of the Lawrence scatter plot is that it can carry a large amount of sample information on a two-dimensional plane. Atrial, supraventricular, ventricular arrhythmias, conduction block have the ability to distinguish. Using the reverse search method, it is convenient to confirm the arrhythmia and its classification, and also to find the noise interference and the problems of missed and false positives. Since the region selected in step 100 is irregular and with high dispersion, the advantages of the Lawrence scatter plot can be used to circle the regions with regular distribution and relatively concentrated.
  • the x-axis of the abscissa is the current RR interval, which is RR(t); the y-axis of the corresponding ordinate is the next RR interval, which is called RR(t+1).
  • the continuous RR interval generated on the long-range ECG data, according to this forward calibration drawing rule, is the Lawrence scatter plot.
  • two adjacent RR interval values in the time dimension are used as scatter points to locate, so as to distinguish the noise distribution range of ECG data, so as to facilitate the user to circle and determine abnormal data.
  • Step S300 separate the cardiac pacemaker points that need to be corrected and the cardiac pacemaker points that do not need to be corrected according to the circled instructions, and process the cardiac pacemaker points that need to be corrected as a spring compression map, by labeling the spring compression map batches. Label noise.
  • the user can circle the selected area on the Lawrence scatter plot, for example, use a brush to frame selection, paint with a brush, etc. to determine the range.
  • normal cardiac pacemakers are concentrated on the Lawrence scatter plot, which is also a user circle. It is also convenient to circle the selected main areas. After circle selection, the areas that are not circled will be regarded as the set of ectopic cardiac pacing points.
  • the user can also directly circle the ectopic cardiac pacemaker to locate the cardiac pacemaker that needs to be corrected.
  • the determination method of the cardiac pacemaker point (segment) that needs to be corrected includes:
  • the cardiac pacemaker (segment) cannot identify the QRS complex due to the interference of the patient's exercise and life.
  • the data does not collect valid ECG waveform due to lead shedding, and the data is straight line or irregular noise.
  • the cardiac pacemaker that needs to be corrected is expanded to form ECG data for the user to mark.
  • the present invention first determines the approximate time period of the ECG data noise through the RR-t scatter diagram, then draws the corresponding Lawrence scatter diagram for the selected heartbeat points in each time period, and finally can analyze the arrhythmia according to the Lawrence diagram.
  • the normal heart beat points and ectopic heart beat points are circled and compressed to form a spring compression map for users to check and mark abnormal ECG data.
  • ECG data is received, and the long-range ECG data includes time and RR interval.
  • time As the horizontal axis and RR interval as the vertical axis, draw a t-RR scatterplot, as shown in Figure 2.
  • the regions in Figure 2 can be selected by the user, and the black shaded part in Figure 3 is the multi-segment scatter diagram selected by the user.
  • a rectangular coordinate system is established for the ECG data of adjacent RR intervals in the t-RR scattergram.
  • a Cartesian coordinate system is established to form a scatter diagram of cardiac pacemakers.
  • the RR intervals on the vertical axis are continuous in time.
  • i and i+1 are the RR intervals in time.
  • the cardiac pacing points in the index coordinate system are divided into cardiac pacing points that need to be corrected and cardiac pacing points that do not need to be corrected.
  • the user eliminates the interference data according to experience (the method for determining the pacemaker (segment) of the heart that needs to be corrected). Expand the pacemaker points that need to be corrected for batch labeling of noise (abnormalities).
  • the present invention also provides a method and device for fast labeling under long-range ECG data, including:
  • a first tool module 100 configured to receive long-range ECG data, and draw a t-RR scatterplot according to the long-range ECG data;
  • the second tool module 200 is configured to establish a Cartesian coordinate system for the ECG data of adjacent RR intervals in the selected t-RR scattergram according to one or more selected sections of the t-RR scattergram, An RR interval and its adjacent RR interval are the X and Y coordinate systems, respectively;
  • the third tool module 300 is configured to separate the cardiac pacing point that needs to be corrected and the cardiac pacing point that does not need to be corrected according to the circled instruction, and process the cardiac pacing point that needs to be corrected as a spring compression map, The described spring compression map batch labels noise.
  • the first tool module 100 includes:
  • the first processing unit is configured to draw a t-RR scatterplot with time as the horizontal axis and RR interval as the vertical axis, wherein the long-range ECG data includes time and RR interval.
  • the second tool module 200 includes:
  • the second processing unit is configured to use the i-th RR interval as the horizontal axis and the i+1-th RR interval as the vertical axis to establish a Cartesian coordinate system to form a scatter diagram of cardiac pacemakers.
  • the RR intervals of the vertical axis are continuous in time.
  • the third tool module 300 includes:
  • the third processing unit is configured to divide the cardiac pacing points in the index coordinate system into cardiac pacing points that need to be corrected and cardiac pacing points that do not need to be corrected according to the area circled by the user on the rectangular coordinate system point;
  • the present invention also provides a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, the method described in the embodiment shown in FIG. 1 is implemented.
  • the computer-readable storage medium may be various computer-readable storage media that can store program codes, such as a U disk, a removable hard disk, a read-only memory (Read-Only Memory, ROM), a magnetic disk, or an optical disk.
  • the disclosed apparatus and method may be implemented in other manners.
  • the apparatus embodiments described above are merely illustrative.
  • the division of each unit is only a logical function division, and other division methods may be used in actual implementation.
  • multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.
  • the integrated unit if implemented in the form of a software functional unit and sold or used as an independent product, may be stored in a storage medium.
  • the technical solution of the present invention is essentially or a part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to cause a computer device (which may be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Pathology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Psychiatry (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Biophysics (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Physiology (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)
  • Electrotherapy Devices (AREA)

Abstract

本发明公开了一种长程心电数据下快速标注方法及装置，所述方法包括接收长程心电数据，根据所述长程心电数据绘制t-RR散点图；根据选取的一段或多段所述t-RR散点图，将所述选取的t-RR散点图中的相邻RR间期的心电数据建立直角坐标系，一个RR间期及与之相邻的RR间期分别为X、Y坐标系；根据圈选的指令分离出需要修正的心脏起搏点和不需要修正的心脏起搏点，并将需要修正的心脏起搏点处理为弹簧压缩图，通过标注所述弹簧压缩图批量标注噪声。克服了现有可视化工具中噪声标注准确不好的问题，在人工审阅时噪声干扰判图时，能快速准确剔除噪声，提升修图速度。

Description

长程心电数据下快速标注方法及装置 技术领域
本发明涉及心电数据处理技术领域，尤其涉及长程心电数据下快速标注方法及装置。
背景技术
临床上，为检测出隐匿性强、一过性难捉取、不规律性的心律失常疾病，需要采集长时间的心电数据。现有的办法是先通过计算机对数据进行一次处理，给出自动检测算法的结果，再经过人工审阅，给出临床意见。
但因为没有算法能够100％准确，特别是存在复杂异常数据和有干扰的数据的情况下，算法的准确性存在问题，所以人工审核成为必须步骤。如果数据样本量太大，依赖专业人士通过常规化去审阅全部信息是难以实现的。
现有长程动态分析软件存在对长程心电数据噪声检测易漏检，误检导致人工审阅时需花费大量的时间辨别噪声。因此现有技术还有待于进一步发展。
发明内容
针对上述技术问题，本发明提供了一种长程心电数据下快速标注方法及装置，以解决相关技术问题。
本发明实施例的第一方面，提供一种长程心电数据下快速标注方法，所述方法包括：
接收长程心电数据，根据所述长程心电数据绘制t-RR散点图；
根据选取的一段或多段所述t-RR散点图，将所述选取的t-RR散点图中的相邻RR间期的心电数据建立直角坐标系，一个RR间期及与之相邻的RR间期分别为X、Y坐标系；
根据圈选的指令分离出需要修正的心脏起搏点和不需要修正的心脏起搏点，并将需要修正的心脏起搏点处理为弹簧压缩图，通过标注所述弹簧压缩 图批量标注噪声。
可选地，所述根据所述长程心电数据绘制t-RR散点图，包括：
以时间为横轴，以RR间期为纵轴，绘制t-RR散点图，其中所述长程心电数据包括时间、RR间期。
可选地，所述将所述选取的t-RR散点图中的相邻RR间期的心电数据建立直角坐标系，包括：
以第i个RR间期为横轴，以第i+1个RR间期为纵轴，建立直角坐标系，形成心脏起搏点散点图，所述横轴与所述纵轴的RR间期在时间上连续。
可选地，所述根据圈选的指令分离出需要修正的心搏点和不需要修正的心搏点，并将需要修正的心脏起搏点处理为弹簧压缩图，包括：
根据用户在所述直角坐标系上圈选的区域，将所述指标坐标系中的心脏起搏点分为需要修正的心脏起搏点和不需要修正的心脏起搏点；
将需要修正的心脏起搏点做展开处理，以用于噪声批量标注。
本发明实施例的第二方面，提供一种长程心电数据下快速标注方法装置，包括：
第一工具模块，用于接收长程心电数据，根据所述长程心电数据绘制t-RR散点图；
第二工具模块，用于根据选取的一段或多段所述t-RR散点图，将所述选取的t-RR散点图中的相邻RR间期的心电数据建立直角坐标系，一个RR间期及与之相邻的RR间期分别为X、Y坐标系；
第三工具模块，用于根据圈选的指令分离出需要修正的心脏起搏点和不需要修正的心脏起搏点，并将需要修正的心脏起搏点处理为弹簧压缩图，通过标注所述弹簧压缩图批量标注噪声。
可选地，所述第一工具模块包括：
第一处理单元，用于以时间为横轴，以RR间期为纵轴，绘制t-RR散点 图，其中所述长程心电数据包括时间、RR间期。
可选地，所述第二工具模块包括：
第二处理单元，用于以第i个RR间期为横轴，以第i+1个RR间期为纵轴，建立直角坐标系，形成心脏起搏点散点图，所述横轴与所述纵轴的RR间期在时间上连续。
可选地，所述第三工具模块包括：
第三处理单元，用于根据用户在所述直角坐标系上圈选的区域，将所述指标坐标系中的心脏起搏点分为需要修正的心脏起搏点和不需要修正的心脏起搏点；
将需要修正的心脏起搏点做展开处理，以用于噪声批量标注。
本发明实施例的第三方面，提供一种计算机可读介质，其上存储有计算机程序，所述程序被处理器执行时实现前述的长程心电数据下快速标注方法。
本发明实施例提供的技术方案中，将心电数据可视化，先通过t-RR散点图确定该心电数据噪声的大致时间段，再将选择的各个时间段的心脏起搏点画出相应的散点图，最后分析心律失常的特点，将正常心搏点和异位心搏点圈选进行压缩，便于快速批量标注。
附图说明
图1为本发明实施例中一种长程心电数据下快速标注方法的流程示意图；
图2为本发明一实施例中RR-t散点图；
图3为本发明一实施例中劳伦斯散点图；
图4为本发明一实施例中被圈选后的劳伦斯散点图；
图5为本发明一实施例中弹簧压缩图；
图6为本发明实施例中的一种长程心电数据下快速标注装置的模块示意图。
具体实施方式
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
请参阅图1，图1为本发明实施例中一种长程心电数据下快速标注方法的一实施例的流程示意图。所述长程心电数据下快速标注方法，包括以下步骤：
步骤S100:接收长程心电数据，根据所述长程心电数据绘制t-RR散点图。
RR间期是指心电图上两个R波之间的时限。正常的RR间期时限应该在0.6-1.0秒之间，小于0.6秒说明出现了心动过速，大于1.0秒说明出现了心动过缓。另外当RR间期不相等的时候，说明出现了心律不齐，在房颤的心电图上会出现明显的RR间距不等。另外正常情况下，RR间期应该和PP间期相等，PP间期也就是2个P波之间的时限，同样应该在0.6-1.0秒之间。RR间期反映的是心室律，PP间期反映的是心房律，当心房律和心室律不相等的时候，可能出现了传导阻滞。在三度房室传导阻滞的心电图当中，PP间期和RR间期会出现绝对不等的情况。
为更方便用户识别心电数据中的噪声，本发明将长程心电数据绘制成时间与RR间期为数据基础的t-RR散点图，用于辅助用户选择干扰较多的区域。
t-RR散点图的横轴是由时间t，单位可以是小时、天等时间长度，散点图的纵轴是RR间期，单位是mm。用户可根据散点图分布不规整、离散度高等特征判断干扰所在，可以t-RR散点图上一段或多段的选取相对应的散点图。
步骤S200:根据选取的一段或多段所述t-RR散点图，将所述选取的t-RR散点图中的相邻RR间期的心电数据建立直角坐标系，一个RR间期及与之相邻的RR间期分别为X、Y坐标系。
本实施例步骤中，建立的直角坐标系形成的图为劳伦斯散点图，劳伦斯 散点图的主要优势在于：能在一个二维平面上承载海量样本信息，对反复频发的心律失常，包括房性、室上性、室性心律失常、传导阻滞有区分能力。利用反向查找的方法，能方便确认心律失常及其分类，也能发现噪声干扰以及漏报和误报问题。由于步骤100中选取是不规整、离散度高的区域，可利用劳伦斯图散点图的优势，将分布规整、较集中的区域进行圈选。
所述直角坐标系，横坐标x轴是当前的RR间期，为RR(t)；对应纵坐标y轴上是紧接后面的下一个RR间期，称为RR(t+1)。两个相邻的RR间期值共同标定了二维平面上的一个点(x＝RR(t)，y＝RR(t+1))。长程心电数据上生成的连续RR间隔，按照这种顺向标定作图规则，即为劳伦斯散点图。
可以理解为是利用时间维度上的相邻两个RR间期值作为散点定位，以便于区分心电数据的噪声分布范围，从而方便用户对异常数据圈选、确定。
步骤S300:根据圈选的指令分离出需要修正的心脏起搏点和不需要修正的心脏起搏点，并将需要修正的心脏起搏点处理为弹簧压缩图，通过标注所述弹簧压缩图批量标注噪声。
用户可在所述劳伦斯散点图上圈选的区域，例如用画笔框选、用画笔涂抹等方式确定范围，一般正常心脏起搏点在所述劳伦斯散点图上都较为集中，也是用户圈选的主要区域，圈选也比较方便，圈选后将未被圈选的区域作为需要异位心脏起搏点的集合。当然，用户也可以直接圈选异位心脏起搏点，从而定位成需要修正的心脏起搏点。
其中，需要修正的心脏起搏点(段)的判定方法包括：
1、心脏起搏点(段)因患者运动及生活产生干扰从而无法辨认QRS波群。
2、数据因导联脱落未采集到有效心电波形，数据为直线或无规律的噪声。
确定异位心脏起搏点需要修正，则将需要修正的心脏起搏点展开形成心电数据，供用户标注。
本发明先通过RR-t散点图确定该心电数据噪声的大致时间段，再将选择的各个时间段的心搏点画出相应的劳伦斯散点图，最后根据劳伦斯图可分析心律失常的特点，将正常心搏点和异位心搏点圈选进行压缩，形成弹簧压缩图，供用户查阅并标注异常心电数据。
下面结合具体实施例对本发明做进一步地说明：
接收心电数据，所述长程心电数据包括时间、RR间期。以时间为横轴，以RR间期为纵轴，绘制t-RR散点图，如图2所示。图2中的区域可供用户选择，如图3中黑色阴影部分为用户选取的多段散点图。
根据用户在图3中选取的多段所述t-RR散点图，将t-RR散点图中的相邻RR间期的心电数据建立直角坐标系。如图4所示，以第i个RR间期为横轴，以第i+1个RR间期为纵轴，建立直角坐标系，形成心脏起搏点散点图，所述横轴与所述纵轴的RR间期在时间上连续。i与i+1为时间上的RR间期进一。
根据用户在所述直角坐标系上圈选的区域，将所述指标坐标系中的心脏起搏点分为需要修正的心脏起搏点和不需要修正的心脏起搏点。如图5所示，用户根据经验剔除干扰数据(需要修正的心脏起搏点(段)的判定方法)。将需要修正的心脏起搏点做展开处理，以用于噪声(异常)批量标注。
如图6所示，本发明还提供一种长程心电数据下快速标注方法装置，包括：
第一工具模块100，用于接收长程心电数据，根据所述长程心电数据绘制t-RR散点图；
第二工具模块200，用于根据选取的一段或多段所述t-RR散点图，将所述选取的t-RR散点图中的相邻RR间期的心电数据建立直角坐标系，一个RR间期及与之相邻的RR间期分别为X、Y坐标系；
第三工具模块300，用于根据圈选的指令分离出需要修正的心脏起搏点和不需要修正的心脏起搏点，并将需要修正的心脏起搏点处理为弹簧压缩图， 通过标注所述弹簧压缩图批量标注噪声。
所述第一工具模块100包括：
第一处理单元，用于以时间为横轴，以RR间期为纵轴，绘制t-RR散点图，其中所述长程心电数据包括时间、RR间期。
所述第二工具模块200包括：
第二处理单元，用于以第i个RR间期为横轴，以第i+1个RR间期为纵轴，建立直角坐标系，形成心脏起搏点散点图，所述横轴与所述纵轴的RR间期在时间上连续。
所述第三工具模块300包括：
第三处理单元，用于根据用户在所述直角坐标系上圈选的区域，将所述指标坐标系中的心脏起搏点分为需要修正的心脏起搏点和不需要修正的心脏起搏点；
将需要修正的心脏起搏点做展开处理，以用于噪声批量标注。
本发明还提供一种计算机可读介质，其上存储有计算机程序，所述程序被处理器执行时实现如图1所示实施例所述的方法。所述计算机可读存储介质可以是U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、磁碟或者光盘等各种可以存储程序代码的计算机可读存储介质。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、计算机软件或者二者的结合来实现，为了清楚地说明硬件和软件的可互换性，在上述说明中已经按照功能一般性地描述了各示例的组成及步骤。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本发明的范围。
在本发明所提供的几个实施例中，应该理解到，所揭露的装置和方法， 可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的。例如，各个单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。
本发明实施例方法中的步骤可以根据实际需要进行顺序调整、合并和删减。本发明实施例装置中的单元可以根据实际需要进行合并、划分和删减。另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以是两个或两个以上单元集成在一个单元中。
该集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分，或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，终端，或者网络设备等)执行本发明各个实施例所述方法的全部或部分步骤。
以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。

Claims (9)

  1. 一种长程心电数据下快速标注方法，其特征在于，所述方法包括：
    接收长程心电数据，根据所述长程心电数据绘制t-RR散点图；
    根据选取的一段或多段所述t-RR散点图，将所述选取的t-RR散点图中的相邻RR间期的心电数据建立直角坐标系，一个RR间期及与之相邻的RR间期分别为X、Y坐标系；
    根据圈选的指令分离出需要修正的心脏起搏点和不需要修正的心脏起搏点，并将需要修正的心脏起搏点处理为弹簧压缩图，通过标注所述弹簧压缩图批量标注噪声。
  2. 根据权利要求1所述的长程心电数据下快速标注方法，其特征在于，所述根据所述长程心电数据绘制t-RR散点图，包括：
    以时间为横轴，以RR间期为纵轴，绘制t-RR散点图，其中所述长程心电数据包括时间、RR间期。
  3. 根据权利要求1所述的长程心电数据下快速标注方法，其特征在于，所述将所述选取的t-RR散点图中的相邻RR间期的心电数据建立直角坐标系，包括：
    以第i个RR间期为横轴，以第i+1个RR间期为纵轴，建立直角坐标系，形成心脏起搏点散点图，所述横轴与所述纵轴的RR间期在时间上连续。
  4. 根据权利要求1所述的长程心电数据下快速标注方法，其特征在于，所述根据圈选的指令分离出需要修正的心搏点和不需要修正的心搏点，并将需要修正的心脏起搏点处理为弹簧压缩图，包括：
    根据用户在所述直角坐标系上圈选的区域，将所述指标坐标系中的心脏起搏点分为需要修正的心脏起搏点和不需要修正的心脏起搏点；
    将需要修正的心脏起搏点做展开处理，以用于噪声批量标注。
  5. 一种长程心电数据下快速标注方法装置，其特征在于，包括：
    第一工具模块，用于接收长程心电数据，根据所述长程心电数据绘制t-RR 散点图；
    第二工具模块，用于根据选取的一段或多段所述t-RR散点图，将所述选取的t-RR散点图中的相邻RR间期的心电数据建立直角坐标系，一个RR间期及与之相邻的RR间期分别为X、Y坐标系；
    第三工具模块，用于根据圈选的指令分离出需要修正的心脏起搏点和不需要修正的心脏起搏点，并将需要修正的心脏起搏点处理为弹簧压缩图，通过标注所述弹簧压缩图批量标注噪声。
  6. 根据权利要求5所述的长程心电数据下快速标注装置，其特征在于，
    所述第一工具模块包括：
    第一处理单元，用于以时间为横轴，以RR间期为纵轴，绘制t-RR散点图，其中所述长程心电数据包括时间、RR间期。
  7. 根据权利要求5所述的长程心电数据下快速标注装置，其特征在于，
    所述第二工具模块包括：
    第二处理单元，用于以第i个RR间期为横轴，以第i+1个RR间期为纵轴，建立直角坐标系，形成心脏起搏点散点图，所述横轴与所述纵轴的RR间期在时间上连续。
  8. 根据权利要求5所述的长程心电数据下快速标注装置，其特征在于，
    所述第三工具模块包括：
    第三处理单元，用于根据用户在所述直角坐标系上圈选的区域，将所述指标坐标系中的心脏起搏点分为需要修正的心脏起搏点和不需要修正的心脏起搏点；
    将需要修正的心脏起搏点做展开处理，以用于噪声批量标注。
  9. 一种计算机可读介质，其上存储有计算机程序，其特征在于，所述程序被处理器执行时实现如权利要求1-4中任一所述的方法。
PCT/CN2020/135535 2020-08-13 2020-12-11 长程心电数据下快速标注方法及装置 WO2022032947A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010810302.0 2020-08-13
CN202010810302.0A CN112120689A (zh) 2020-08-13 2020-08-13 长程心电数据下快速标注方法及装置

Publications (1)

Publication Number Publication Date
WO2022032947A1 true WO2022032947A1 (zh) 2022-02-17

Family

ID=73851358

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/135535 WO2022032947A1 (zh) 2020-08-13 2020-12-11 长程心电数据下快速标注方法及装置

Country Status (2)

Country Link
CN (1) CN112120689A (zh)
WO (1) WO2022032947A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114668401B (zh) * 2022-03-11 2023-04-18 肇庆星网医疗科技有限公司 一种ai心电训练数据标注方法、装置、电子设备及介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1843293A (zh) * 2006-03-21 2006-10-11 李方洁 一种长程心电数据分析的方法
JP2008067858A (ja) * 2006-09-13 2008-03-27 Terumo Corp 心拍ゆらぎ検出装置およびその情報処理方法
US20090292180A1 (en) * 2006-04-18 2009-11-26 Susan Mirow Method and Apparatus for Analysis of Psychiatric and Physical Conditions
CN108784680A (zh) * 2018-03-19 2018-11-13 武汉海星通技术股份有限公司 基于散点图的心电图智能分析方法及系统
CN108836317A (zh) * 2018-06-27 2018-11-20 广州迪茂信息科技有限公司 动态心电时间散点图逆向分析方法
CN109953754A (zh) * 2019-03-08 2019-07-02 东南大学 一种室性并行心律快速识别的方法以及装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1843293A (zh) * 2006-03-21 2006-10-11 李方洁 一种长程心电数据分析的方法
US20090292180A1 (en) * 2006-04-18 2009-11-26 Susan Mirow Method and Apparatus for Analysis of Psychiatric and Physical Conditions
JP2008067858A (ja) * 2006-09-13 2008-03-27 Terumo Corp 心拍ゆらぎ検出装置およびその情報処理方法
CN108784680A (zh) * 2018-03-19 2018-11-13 武汉海星通技术股份有限公司 基于散点图的心电图智能分析方法及系统
CN108836317A (zh) * 2018-06-27 2018-11-20 广州迪茂信息科技有限公司 动态心电时间散点图逆向分析方法
CN109953754A (zh) * 2019-03-08 2019-07-02 东南大学 一种室性并行心律快速识别的方法以及装置

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LIU, MING: "New Thoughts on Electrocardiogram Analysis: Labelling Electrocardiogram Waveforms According to the Characteristics of Scattered Data Points)", CHINESE JOURNAL OF CARDIAC PACING AND ELECTROPHYSIOLOGY, vol. 27, no. 1, 7 February 2013 (2013-02-07), pages 76 - 78, XP055899755, ISSN: 1007-2659, DOI: 10.7695/zgxzqb201301024 *
XIANG JINTAO , LI FANGJIE , YANG LING: "Timed RR-interval Scatter Plots and Reverse Technology", CHINESE JOURNAL OF CARDIAC PACING AND ELECTROPHYSIOLOGY, vol. 25, no. 5, 30 September 2011 (2011-09-30), pages 445 - 449, XP055899763, ISSN: 1007-2659, DOI: 10.13333/j.cnki.cjcpe.2011.05.011 *

Also Published As

Publication number Publication date
CN112120689A (zh) 2020-12-25

Similar Documents

Publication Publication Date Title
CN110013247B (zh) 一种心电图p波和t波的检测、区分与定位方法
CA2686877C (en) Ecg signal analysis tool
US5967994A (en) Method and system for characterizing the quality of signals indicative of heart function
CN103549948B (zh) 心脏激动时间检测
CN106815570B (zh) 一种基于动态模式识别的心电信号st-t段识别方法
CN112528783B (zh) 心电图心搏数据聚类方法、装置、电子设备和介质
WO2021169296A1 (zh) 心电图数据处理的方法、装置、计算机设备及存储介质
JP2013150857A (ja) 心電図波形の解析を高速に行う方法と装置
CN113876335B (zh) 心电图心搏分类方法、装置、电子设备和介质
CN111358459A (zh) 一种心律失常识别方法、装置、设备和存储介质
EP3641643B1 (en) System for detecting atrial fibrillation
WO2022032947A1 (zh) 长程心电数据下快速标注方法及装置
WO2015169724A1 (en) Detection of pulmonary vein isolation
Simoons et al. Quantitation of exercise electrocardiography.
CN111920399A (zh) 心率变异性的分析方法及装置
US20070010752A1 (en) System and method for analysing the p-wave of an ecg-signal
CN109171697A (zh) 一种基于双重判断标准的可电击复律心律识别方法
CN109893124B (zh) 一种基于多导联聚类分析的疑难心电图主波识别定位方法
Tu et al. A novel method for automatic identification of motion artifact beats in ECG recordings
CN109431488B (zh) 一种心电图动态实时分析数据中典型数据的识别方法
CN116473569A (zh) 一种心电图数据处理方法、系统及存储介质
CN110916649B (zh) 一种长程心电散点图的处理装置、处理方法及检测装置
CN110811597B (zh) 心电图记录方法、心电图记录仪、装置及存储介质
CN111493856B (zh) 心搏分类的修正方法、装置、心电图检查系统及存储介质
WO2021246923A1 (ru) Способ детектирования фокальных эпилептиформных разрядов в длительной записи ээг

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: 20949439

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 04.04.2023)

122 Ep: pct application non-entry in european phase

Ref document number: 20949439

Country of ref document: EP

Kind code of ref document: A1