WO2010081292A1 - 心电图导联识别方法及装置 - Google Patents

心电图导联识别方法及装置 Download PDF

Info

Publication number
WO2010081292A1
WO2010081292A1 PCT/CN2009/070123 CN2009070123W WO2010081292A1 WO 2010081292 A1 WO2010081292 A1 WO 2010081292A1 CN 2009070123 W CN2009070123 W CN 2009070123W WO 2010081292 A1 WO2010081292 A1 WO 2010081292A1
Authority
WO
WIPO (PCT)
Prior art keywords
lead
wave group
interval
height
time
Prior art date
Application number
PCT/CN2009/070123
Other languages
English (en)
French (fr)
Inventor
周皓隽
李永安
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to CN200980147163.XA priority Critical patent/CN102438514B/zh
Priority to PCT/CN2009/070123 priority patent/WO2010081292A1/zh
Publication of WO2010081292A1 publication Critical patent/WO2010081292A1/zh

Links

Classifications

    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/30Input circuits therefor
    • A61B5/304Switching circuits
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00Measuring for diagnostic purposes; Identification of persons
    • A61B5/24Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/30Input circuits therefor
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61BDIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B2562/00Details of sensors; Constructional details of sensor housings or probes; Accessories for sensors
    • A61B2562/22Arrangements of medical sensors with cables or leads; Connectors or couplings specifically adapted for medical sensors
    • A61B2562/225Connectors or couplings
    • A61B2562/226Connectors or couplings comprising means for identifying the connector, e.g. to prevent incorrect connection to socket

Definitions

  • the present invention relates to electrocardiography techniques, and more particularly to methods and apparatus for identifying detected ECG leads.
  • An electrocardiogram is an important means of monitoring heart disease and reflects the activity of the heart.
  • a typical ECG usually consists of: P waves (representing the left and right atrial activation processes, usually the earliest appearance of each heartbeat cycle, expressed as a relatively low-level wavelet), QRS complex (representing the process of two ventricular activations)
  • the general QRS complex contains three consecutive fluctuations, the first downward wave is called the Q wave, the next narrow vertical wave is called the R wave, and the next downward wave is connected to the R wave.
  • S wave For S wave).
  • the doctor analyzes and diagnoses heart disease based on the characteristics of the height, width, shape, and spacing of each wave group.
  • a static electrocardiogram typically contains a synchronized twelve-lead signal, with each lead corresponding to a fixed combination of electrodes. In fact, different lead signals represent the activity of the heart as viewed from different directions.
  • One method is to first separately identify each lead. Since the recognition results of the same wave group on different leads may be deviated, in order to obtain the exact position of the wave group in the lead, first record the specific time position of the same wave group in multiple different leads. And set a time threshold. Starting from the time position of the earliest wave group, taking a time period of the same length as the time threshold, calculating the number of positions of the same wave group in the time period; starting from the time position of the latest wave group Forward, take one The time period equal to the length of the time threshold is calculated, and the number of positions of the same wave group occurring in the time period is calculated.
  • Another method is to first combine the signals of the plurality of leads into one lead, and then perform the identification of the group on the integrated generated lead.
  • the integration of multiple lead signals uses a weighted superposition method, that is, first calculates a weight for each lead, and then multiplies the signals of the respective leads by the weights corresponding to the respective leads, and then the respective lead signals. The absolute values of the values obtained by multiplying the corresponding weights are added together as a comprehensively generated lead signal.
  • the wave group is identified on the lead after the integration of the plurality of leads.
  • the solution of the invention provides an ECG lead identification method and device, which can effectively utilize the lead with obvious features in the lead to perform priority recognition, thereby ensuring the recognition effect.
  • embodiments of the present invention use the following technical solutions:
  • An ECG lead identification method includes:
  • An ECG lead identification device includes:
  • the selection unit is specifically configured to select a lead with obvious medical characteristics or select a lead with the largest amplitude of the feature
  • the identification unit is specifically configured to identify the wave group on the selected lead.
  • the present invention selects a lead with a medical characteristic from the measured lead or selects a lead with the largest feature amplitude, and identifies the wave group on the selected lead. Since the selected lead is a lead with obvious medical features or a lead with the largest characteristic amplitude, the selected lead signal quality is better, and the selected lead does not undergo any reprocessing to keep the selected lead intact. Graphic form, which is beneficial to the wave group The recognition effect is improved; and the wave group is identified on the selected lead. Since the recognition process is performed on the selected lead with better quality and obvious features, the wave group is further ensured. The recognition effect improves the accuracy of the wave group recognition and is beneficial to improve the accuracy of disease diagnosis.
  • FIG. 1 is a flowchart of an ECG lead identification method according to an embodiment of the present invention.
  • FIG. 2 is a schematic structural diagram of an electrocardiogram lead identification device according to an embodiment of the present invention.
  • FIG. 3 is a flowchart of a method for identifying an ECG lead that selects an obvious lead of a medical feature according to an embodiment of the present invention
  • FIG. 4 is a schematic structural diagram of an electrocardiogram-based identification device for selecting an obvious lead of medical features according to an embodiment of the present invention
  • FIG. 5 is a flowchart of an ECG lead identification method for selecting a lead with the largest feature amplitude according to an embodiment of the present invention
  • FIG. 6 is a schematic structural diagram of an ECG lead identification device for selecting a lead with the largest feature amplitude according to an embodiment of the present invention
  • FIG. 7 is a flowchart of a method for identifying an ECG lead in a combination of selecting an obvious lead of a medical feature and selecting a lead having the largest feature amplitude according to an embodiment of the present invention
  • FIG. 8 is a schematic structural diagram of an electrocardiogram lead device according to an embodiment of the present invention, in which an obvious lead of a medical feature is selected and a lead with the largest feature amplitude is selected;
  • FIG. 9 is a flowchart of a method for determining whether a misidentified wave group exists on a selected lead in the ECG lead identification method corresponding to FIG. 7 according to an embodiment of the present invention.
  • FIG. 10 is a schematic diagram of an electrocardiogram lead according to an embodiment of the present invention
  • FIG. 11 is a schematic structural diagram of an ECG lead identification device corresponding to FIG. 8 for determining whether a misidentified wave group exists on a selected lead according to an embodiment of the present disclosure
  • FIG. 12 is a flowchart of a method for determining whether a leaked identification wave group exists on a selected lead in an ECG lead identification method according to an embodiment of the present invention
  • FIG. 13 is a schematic structural diagram of determining, by using an electrocardiogram lead identification device according to an embodiment of the present invention, whether a leaked identification wave group exists on a selected lead;
  • FIG. 14 is a flowchart of an ECG lead identification method according to an embodiment of the present invention.
  • An embodiment of the present invention provides a method for identifying an electrocardiogram lead. As shown in FIG. 1, the method includes:
  • the wave group can be identified by any one of the differential threshold method, the template method, the wavelet method, and the like.
  • the embodiment of the present invention further provides an electrocardiogram lead identification device. As shown in FIG. 2, the device includes: a selection unit 21 and an identification unit 22.
  • the selecting unit 21 is specifically configured to select a lead with obvious medical features or select a lead with the largest characteristic amplitude in each measured lead.
  • the selected selected lead is identified by a wave group based on any conventionally known techniques such as a differential threshold method, a template method, or a wavelet method. Since the embodiment of the present invention first selects a lead with a medical characteristic from the measured lead or selects a lead with the largest feature amplitude, so that the quality of the selected lead signal is better, and the graphical feature of the selected lead is selected.
  • Maintaining the original graphical form provides a guarantee for the identification of the selected clusters on the lead.
  • the identification of the wave group on the lead is the wave group identification on the selected lead with better quality and obvious features.
  • the position of the wave group on this lead is accurate and the feature is obvious, which is beneficial to the identification of the wave group.
  • the accuracy of the wave group recognition is improved, which is beneficial to improve the accuracy of disease diagnosis.
  • An ECG lead identification method implemented by a medically conspicuous lead as shown in FIG. 3, the method includes:
  • step 301 Determine, according to the lead signal generated by the electrocardiograph's pre-connected electrode combination, whether the II lead exists in the measured lead. If the II lead is present in the measured lead, step 302 is performed; if the II lead is not present in the measured lead, step 303 is performed.
  • the electrocardiogram Since the electrocardiogram is to be applied to the surface of the human body according to various fixed positions, the electrocardiogram is used to record the voltage change between the electrodes, thereby generating a synchronous lead signal corresponding to the corresponding position, but the above-mentioned lead signal is not Containing all of the twelve synchronized lead signals, when the position at which a certain lead is generated is not suitable for the electrode, the lead is defaulted, so there may be no II lead.
  • step 302. Select the II lead from the measured lead as a lead with obvious medical characteristics, and then perform step 306.
  • the medical characteristics of the VI lead are slightly lower than the medical characteristics of the lead II. Therefore, when there is no II lead in the measured lead, the medical characteristics of each lead are compared with the pre-set medical characteristics compared with the VI lead of the second lead, and the measured lead is judged. Whether there is a VI lead. If there is a VI lead in the measured lead, perform step 304; if there is no VI lead in the measured lead, go to step 305.
  • the wave group identification method is used to identify the selected group of the medical features with obvious leads.
  • the time of the wave group identified in the obvious lead of the medical characteristic is determined according to the position of the identified wave group in the lead with obvious medical characteristics. Start and end of time.
  • the time start point is subtracted from the preset time threshold as a time start point of the recognition time interval; and the time end point is added to the preset time threshold as the time end point of the recognition time interval.
  • the identification of the corresponding wave group can be identified by using a differential threshold method, a template method, a wavelet method, or the like.
  • the embodiment of the present invention provides an electrocardiogram lead identification device. As shown in FIG. 4, the device includes: a selection unit 41, an identification unit 42, a determination unit 43, and an interval calculation unit 44.
  • the selecting unit 41 is specifically configured to select a lead with obvious medical features from the measured leads; the identifying unit 42 is specifically configured to use the differential threshold method, the template method, the wavelet method, and the like to identify the selected medical features.
  • the cluster of the more obvious leads is identified; the determining unit 43 is specifically configured to determine the identified waves in the apparent leads of the medical features when performing wave group recognition on the remaining leads other than the obvious lead of the medical features.
  • the time start point and the time end point of the group; the interval calculating unit 44 is specifically configured to calculate a recognition time interval by using the time start point and the time end point of the wave group and a specific time threshold, and the lead in the identification time interval is obvious to the medical feature Remaining leads outside Perform wave group identification.
  • the selection unit 41 includes a first determination module 411 and a first selection module 412.
  • the first determining module 411 is specifically configured to determine whether there is a II lead in the measured lead. If the II lead exists in the measured lead, the first selected module 412 is specifically used to select the II lead as a medicine. The characteristic is more obvious lead; if there is no II lead in the measured lead, it is judged in the first determining module 411 whether there is a VI lead in the measured lead, if the measured lead is in the lead There is a VI lead, and the first selected module 412 is specifically used to select a VI lead for a medical feature to be significantly lead. If the measured lead in the lead is not present, the first selected module 412 is specifically used to select medical features other than the lead lead and the VI lead from the measured lead. Lead.
  • the interval calculation unit 44 includes a start point calculation module 441 and an end point calculation module 442.
  • the starting point calculation module 441 is specifically configured to calculate the starting point of the identification time interval, that is, the time starting point is subtracted from the preset time threshold as the time starting point of the identifying time interval.
  • the endpoint calculation module 442 is specifically configured to calculate the time end point of the recognition time interval, that is, the end of the time plus a preset time threshold as the time end point of the recognition time interval.
  • the most obvious medical features in the lead are the II lead, followed by the VI lead.
  • the medical features are selected.
  • the VI guide When the II lead and the VI lead are not present, the lead in the lead other than the II lead and the VI lead may be referred to.
  • This method of selecting medical features is more obvious than the way of guiding, so that the medical features are more obvious.
  • the selection process of the lead is careful, and the quality of the lead in the measured lead is avoided.
  • the lead of the guide is a clear lead for medical features, so that the selected medical features are more obvious.
  • the quality of the lead is better and the features are more obvious, which ensures that the medical features are more obvious than the recognition of the wave group on the lead, which is conducive to the accurate diagnosis of the disease. .
  • the wave group identification is performed on the remaining leads other than the selected lead according to the medical feature and the recognition result of the wave group on the lead. Because the recognition result of the wave group on the lead with relatively obvious medical characteristics is relatively accurate, the knowledge of the wave group on the remaining leads other than the selected lead is obtained. The quality is guaranteed to further ensure the accuracy of the disease diagnosis.
  • Embodiments of the present invention provide an ECG lead identification method according to a lead with the largest amplitude of the selected feature. As shown in Figure 5, the method includes:
  • the electrodes should be attached to the surface of the human body according to various fixed positions.
  • the electrocardiogram records the voltage change between the electrodes, thereby generating a synchronous lead signal corresponding to the corresponding position.
  • the characteristic amplitudes of the respective measured leads over a predetermined time range are calculated, the predetermined time being the first n seconds of each of the leads including at least one QRS complex, n being at least greater than two.
  • the calculation of the characteristic amplitude of the lead includes the following three methods:
  • the height value of each measured lead within a predetermined time range is calculated, and the maximum height value among the height values is used as the characteristic amplitude.
  • the peak point of each measured lead within a predetermined time range and the height of the peak point; secondly, determine the minimum height of the waveform within a certain time threshold interval of the peak point, the starting point of the specific time threshold interval is The peak point is determined by subtracting a specific time threshold, and the end point is determined by the peak point plus a specific time threshold, wherein the specific time threshold generally does not exceed 100 milliseconds; finally, the height difference between the amplitude of the peak point and the minimum amplitude is calculated, the height is The largest height difference among the differences is taken as the characteristic amplitude.
  • Secondly determine the specific time threshold interval of the valley point, and the end point is added by the valley point
  • the specific time threshold is determined, wherein the specific time threshold generally does not exceed 100 milliseconds; the minimum height of the waveform within the specific time threshold interval of the peak point is determined, and the starting point of the specific time threshold interval is determined by subtracting the specific time threshold from the peak point, and the end point is determined by the peak Point plus a specific time threshold determination, wherein the specific time threshold generally does not exceed 100 milliseconds; finally calculating a first height difference between the height of the valley point and the maximum height and a height between the peak point and the minimum height
  • the second height difference is the maximum height difference between the first height difference and the second height difference as the characteristic amplitude.
  • Identifying the wave group on the lead with the largest feature amplitude by using a differential threshold method, a template method, a wavelet method, and the like.
  • the remaining leads other than the measured lead having the largest amplitude of the feature are measured according to the position of the identified wave group in the lead having the largest characteristic amplitude.
  • the wave group recognition is performed, and the recognition process is the same as the recognition process of 307 to 309 in Fig. 3 of Embodiment 2.
  • the embodiment of the present invention provides an electrocardiogram lead identification device. As shown in FIG. 6, the device includes: a selection unit 61, an identification unit 62, a determination unit 63, and an interval calculation unit 64.
  • the selecting unit 61 is specifically configured to select a lead with the largest feature amplitude from the measured leads according to the graphic features of the lead; the identifying unit 62 is specifically configured to use a differential threshold method, a template method, a wavelet method, and the like The identification method identifies the wave group on the lead with the largest feature amplitude; the determining unit 63 is specifically configured to determine the feature amplitude when performing wave group identification on the remaining leads other than the lead having the largest feature amplitude The time starting point and the time end point of the identified wave group in the largest lead; the interval calculating unit 64 is specifically configured to calculate an identifying time interval by using the time start point and the time end point of the wave group and the specific time threshold, in the identifying time interval The wave group identification is performed on the remaining leads other than the lead having the largest feature amplitude.
  • the selection unit 61 includes a calculation module 611 and a second selection module 612.
  • the calculating module 611 is specifically configured to: when analyzing the graphical features of each measured lead, calculate the characteristic amplitude of each measured lead according to any one of the foregoing three methods for calculating the lead characteristic amplitude .
  • the second selecting module 612 is specifically configured to select a lead corresponding to the maximum feature amplitude from each lead characteristic amplitude, and use the lead as the characteristic amplitude. The largest lead.
  • the interval calculation unit 64 includes a start point calculation module 641 and an end point calculation module 642.
  • the starting point calculation module 641 is specifically configured to calculate a starting point of the identification time interval, that is, the time starting point is subtracted from the preset time threshold as the time starting point of the identifying time interval.
  • End point calculation module 642 is specifically used The time end point of the recognition time interval is calculated, that is, the time end point plus the preset time threshold is used as the time end point of the recognition time interval.
  • the lead characteristic amplitude is first calculated one by one on the measured lead, so that the obtained feature amplitude is relatively accurate, and then the accurate feature amplitude is obtained. Based on the selection of the maximum feature amplitude, the lead corresponding to the maximum feature amplitude is selected as the lead with the largest feature amplitude.
  • the feature of the lead pattern with the largest feature amplitude selected in this way is obvious, which further improves the recognition accuracy of the wave group on the lead, which is beneficial to the diagnosis of the disease.
  • the present invention specifically describes an electrocardiographic lead identification method provided in accordance with the combination of the medical features of the lead and the graphical features of the lead.
  • the method includes the following steps: As shown in FIG. 7, the method includes: since the medical features of the lead are not calculated, and the quality of the lead with medical features is relatively good, the method may first be performed according to the medical characteristics of the lead. The selection of the lead.
  • the electrodes are attached to the surface of the human body according to various fixed positions, and the electrocardiogram is used to record the voltage change between the electrodes, thereby generating a synchronous lead signal corresponding to the corresponding position.
  • the lead signal generated by the pre-connected electrode combination of the electrocardiograph is first determined to determine whether the lead is present in the measured lead. . If there is an II lead in the measured lead, step 702 is performed; if the II lead is not present in the measured lead, step 703 is performed.
  • step 704 is performed; if the VI lead is not present in the measured lead, step 705 is performed.
  • the VI lead with a medical feature that is slightly less than that of the II lead is to select the lead with the largest lead characteristic from the measured lead based on the measured graphical characteristics of the lead.
  • the characteristic amplitudes of the respective measured leads in the predetermined time range are first calculated.
  • the calculation method of the feature amplitude is the same as the calculation method of the 501 feature amplitude in FIG.
  • Identify the wave group of the selected lead by using a differential threshold method, a template method, a wavelet method, and the like.
  • the recognition process is implemented in the same manner as the recognition process of 307 to 309 in FIG.
  • the embodiment of the present invention further provides an electrocardiogram lead identification device. As shown in FIG. 8, the device includes: a selection unit 81, an identification unit 82, a determination unit 83, and an interval calculation unit 84.
  • the selecting unit 81 is specifically configured to select, from the measured leads, a lead with a more obvious medical feature or a lead with the largest feature amplitude as the selected lead; the identifying unit 82 is specifically configured to use a differential threshold method, a template method, When the wave group identification method of the wavelet method identifies the wave group on the selected lead, the time start and the time end point of the identified wave group in the selected lead are determined; the interval calculating unit 84 is specifically used An identification time interval is calculated using a time start and a time end of the wave group and a specific time threshold, and the remaining leads other than the selected lead are subjected to wave group identification within the identified time interval.
  • the selection unit 81 includes a first determination module 811, a first selection module 812, a calculation module 813, and a second selection module 814.
  • the first determining module 811 is specifically configured to determine whether there is a II lead with the most obvious medical feature in the measured lead. If there is an II lead, the first selected module 812 is specifically configured to select the II lead as described. The selected lead; if there is no II lead, the first mode of judgment
  • the block 811 is specifically configured to determine whether there is a VI lead in the measured lead. If there is a VI lead, the first selected module 812 is specifically configured to select the VI lead as the selected lead. When there is neither the II lead nor the VI lead in the measured lead, the lead with the largest feature magnitude is selected from the measured leads as the selected lead.
  • the calculation module 81 3 is specifically configured to calculate a characteristic amplitude of each measured lead; the second selecting module 814 is specifically configured to select a lead with the largest characteristic amplitude from each lead characteristic amplitude, and the feature The lead with the largest amplitude corresponds to the selected lead.
  • the interval calculation unit 84 includes a start point calculation module 841 and an end point calculation module 842.
  • the starting point calculation module 841 is specifically used to calculate the starting point of the identification time interval, that is, the time starting point is subtracted from the preset time threshold as the time starting point of the identifying time interval.
  • the endpoint calculation module 842 is specifically configured to calculate a time end point of the recognition time interval, that is, the end of the time plus a preset time threshold as the time end point of the recognition time interval.
  • the identified clusters on the selected leads are evaluated to see if there are misidentified clusters. If the wave group corresponding to the specific wave group in the selected lead is not recognized from the remaining leads except the selected lead, whether the wave group on the selected lead is wrong Identify the wave group judgment. As shown in Figure 9, the method includes:
  • step 901. Determine whether the group of the selected lead is a misidentification group. If the group is a misidentification group, perform step 902. If the group is not misidentified, perform step 903.
  • step 901 there are two methods for judging the misidentified wave group in step 901, taking the identification of the group QRSi as an example, as shown in FIG. 10:
  • Time interval R i+1 R i+2 is the first time interval; again calculate the time interval ⁇ ⁇ +1 between the wave group QRSi and the next wave group ( ⁇ ! ⁇ and the previous wave group QRS ⁇ +1 ) a second time interval; then determining whether the second time interval is less than a preset interval coefficient ⁇ multiplied by a first time interval, the predetermined interval coefficient ⁇ is an empirical threshold, generally not greater than 1.5; generally ⁇ ⁇ +1 is not
  • the wavelet group SQRi is a normally recognized group when less than K is multiplied or Ri +1 is not less than ⁇ multiplied by R i+1 R i+2 ; if the second time interval is less than the preset interval coefficient multiplied by For a time interval, the specific wave group is a misidentified wave group.
  • the second type first determining the minimum height and the maximum height of the waveform in the specific time threshold interval of the wave group QRSi; secondly calculating the height difference Hi of the maximum height of the wave group QRSi and the minimum height of the specific wave group;
  • the wave group QRSi is forwardly shifted by a height difference ⁇ ⁇ +1 of the wave group QRS i+1 and a backward wave group QRSi ⁇ ; ; then it is judged whether the height difference of the wave group QRSi is smaller than a preset height coefficient P Multiply by ⁇ ⁇ +1 and the smaller height difference, the height coefficient ⁇ is an empirical value, usually not greater than 0.5.
  • the height difference of the wave group is greater than or equal to the preset height coefficient ⁇ multiplied by ⁇ ⁇ +
  • the wave group SQRi is a normally recognized wave group when 1 and the smaller height difference. If the characteristic amplitude of the particular wave group is less than a preset amplitude coefficient multiplied by a smaller feature magnitude, the wave group is a misidentified wave group.
  • the structure of the other wave group as shown in FIG. 11, includes: a first judging unit 11 and a deleting unit 112.
  • the first determining unit i l l is specifically configured to determine that the wave group is a misidentified wave group when a wave group corresponding to the selected specific beam group is not identified from the lead other than the selected lead.
  • the deleting unit 112 is specifically configured to delete the specific wave group deletion from the selected lead when the specific wave group is a misidentified wave group.
  • the first determining unit 11 1 includes an interval calculating module 11 11 , an interval determining module 1112 , an interval determining module 111 3 , a height difference calculating module 1114 , a height difference determining module 1 115 , and a height difference determining module 1116 .
  • the interval calculation module 1111 is specifically configured to first calculate a first time interval between two groups of waves backward of the specific wave group, and then calculate a first wave group of the specific wave group and a wave group of the previous wave group. Two time intervals.
  • the interval determining module 11 12 is specifically configured to determine whether the first time interval is small. 5 ⁇
  • the preset interval coefficient K is an empirical threshold, generally not greater than 1.5.
  • the interval determining module 111 3 determines that the specific wave group is a misidentified wave group when the first time interval is less than the preset interval coefficient and multiplied by the second time interval.
  • the height difference calculation module 1114 is specifically configured to determine a minimum height and a maximum height of a waveform of a specific wave group in a specific time threshold interval, where a starting point of the specific time group is subtracted from the specific time point by the specific time group Threshold determination, the end point of the specific time threshold interval is determined by the specific time group of the specific wave group plus the specific time threshold, wherein the specific time threshold generally does not exceed 100 milliseconds; secondly, calculating the maximum height of the specific wave group and the The height difference of the minimum height of the specific wave group; the height difference between the forward wave group and the backward wave group of the wave group is calculated separately; the height difference determining module 1115 is specifically configured to determine the characteristic amplitude of the specific wave group 5 ⁇ The value is less than 0.5.
  • the height difference determination module 1116 is specifically configured to: when the height difference of the wave group is less than a preset height coefficient multiplied by a smaller feature amplitude of a height difference between the forward wave group and the backward wave group of the wave group, It is determined that the specific wave group is a misidentified wave group.
  • a method for determining whether there is a leak identification group on the selected lead is as shown in FIG. 12, and the method includes:
  • step 1202 determining whether there is a gap between consecutive two groups of waves greater than a preset coefficient multiplied by the average interval, the predetermined coefficient is an empirical threshold, usually greater than 1.5, if there is a continuous wave group between If the interval is greater than the preset coefficient and multiplied by the average interval, step 1203 is performed. If there is no interval between two consecutive groups of waves greater than the preset coefficient multiplied by the average interval, step 1204 is performed.
  • the embodiment further provides a structure for determining whether there is a leaked identification wave group on the selected lead.
  • the structure includes: an interval calculating unit 1 31 and a second determining unit 1 32. .
  • the interval calculation unit 1 31 is specifically configured to calculate an average interval between all consecutive two groups of the selected leads.
  • the second judging unit 1 32 is specifically configured to determine whether there is a continuous interval between two groups of waves greater than a preset coefficient multiplied by the average interval, the preset coefficient is an empirical threshold, usually greater than 1.5.
  • the preset coefficient is an empirical threshold, usually greater than 1.5.
  • the method and structure for determining whether the wave group on the selected lead is a misidentification wave group and a leak recognition wave group may be, but are not limited to, the ECG lead identification process used in Embodiment 1 and Embodiment 3. in.
  • the above embodiment not only selects the lead of the medical feature from the measured lead as the selected lead, but also takes the feature amplitude from the measured lead without the lead with the medical feature being particularly obvious.
  • the largest lead is the selected lead, so that the quality of the selected lead is further ensured, so that the recognition effect of the guided wave group is better.
  • the search is performed. Add to the leak identified wave group Add to the corresponding position of the selected lead.
  • the identification method of the lead makes the identification of the wave group on the lead accurate, and fully ensures the recognition effect of the wave group on the lead, which is beneficial to the diagnosis of the disease.
  • Embodiments of the present invention also provide an ECG lead identification method, as shown in FIG.
  • the various types of clusters with obvious features in the electrocardiogram are not on the same medical feature or the characteristic features of the graphical features. Therefore, in order to make all kinds of wave groups can be preferentially identified on the guides with obvious characteristics, it is necessary to select different guides or select feature amplitudes corresponding to the medical features of the wave groups for different wave groups.
  • the largest lead is used as the selected lead.
  • the various wave groups respectively identify the various wave groups on the corresponding selected leads.
  • the corresponding wave group identification on the selected lead is completed, the corresponding wave group on the measured lead other than the selected lead is identified according to the identification result of the selected lead wave group. Specifically include:
  • the electrodes should be attached to the surface of the human body according to various fixed positions.
  • the electrocardiogram records the voltage change between the electrodes, thereby generating a synchronous lead signal corresponding to the corresponding position.
  • the wave group is determined according to a differential threshold method, a template method, a wavelet method, or the like on the selected lead. Identification.
  • the type of the wave group is identified on the selected lead, according to the position of the wave group that has been identified by the wave group in the selected lead, in addition to the selected lead.
  • the corresponding type of wave group is searched for at the corresponding position of the remaining leads, and the corresponding wave group is identified.
  • the corresponding wave group is searched for at the corresponding position of the other measured non-selected lead, and the corresponding wave group is The corresponding class of waves is identified.
  • the time start and time end of the type of wave identified in the selected lead are determined.
  • a point and time end point and a preset time threshold determine an identified time interval of the identified type of wave group in the selected lead.
  • the other measured leads other than the selected leads are identified corresponding to the wave group.
  • the identification of the corresponding type of wave group can be identified by using a differential threshold method, a template method, a wavelet method, and the like.
  • a clear lead of medical features for another type of to-be-identified wave group or a selection of a lead with the largest feature amplitude is selected. Identification of the wave group. Before selecting the selected lead for another type of to-be-identified wave group, the identified one type of wave group is first excluded from each measured lead, and then for another type of unidentified wave group. A significant lead of the medical feature is selected in each of the measured leads or a selection of the lead having the largest feature magnitude is selected. The selected lead selection process for another type of unidentified wave group and the identification process of the type of wave group are the same as the selection and identification process of the selected lead of the first type of wave group. Repeat this process until all different classes of wave groups are identified.
  • the method for determining whether the wave group on the selected lead has a misidentification wave group and a leak recognition wave group in the process of identifying the lead of the electrocardiogram in the embodiment may be, but is not limited to, specifically used in the embodiment 4
  • the selected lead corresponding to each wave group feature is selected as the selected lead corresponding to each wave group, and thus the selected lead is selected.
  • the characteristics of each type of wave group on the selected lead are obvious and accurate, ensuring the accuracy of the recognition effect of each type of wave group, and the recognition effect of the wave group on the entire measured lead is better, and the disease diagnosis is ensured. The accuracy.
  • the present invention can be implemented by means of software plus a necessary general hardware platform, and of course, can also be through hardware, but in many cases, the former is a better implementation. the way.
  • the technical solution of the embodiments of the present invention may be embodied in the form of a software product in essence or in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for making Mobile devices (which may be mobile phones, personal computers, media players, etc.) perform the methods described in various embodiments of the present invention.
  • the storage medium referred to herein is, for example, R0M/RAM, magnetic disk, optical disk, and the like.

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Physics & Mathematics (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)

Description

心电图导联识别方法及装置 技术领域
本发明涉及心电图技术, 尤其涉及对检测到的心电图导联进行识别的方 法及装置。
背景技术
心电图是用来监测心脏疾病的一个重要手段, 反映了心脏活动的情况。 一张典型的心电图通常包括: P波(代表左右两心房激动的过程, 一般在每个 心跳周期的最早出现, 表现为一个较为低平的小波), QRS波群(代表两个心 室激动的过程, 一般的 QRS波群包含三个连续的波动, 第一个向下的波称为 Q 波, 紧接着的狭高直立的波称为 R波, 与 R波相连的下一个向下的波称为 S 波)。 医生根据各个波群的高度、 宽度、 形态、 相互间的间距等特征来分析、 诊断心脏疾病。 心电检查时, 需要将电极按照固定的位置贴在人体表面, 心 电图机会记录电极间的电压变化情况。 静态心电图通常包含同步的十二导联 的信号, 每个导联对应一种固定的电极组合。 实际上, 不同的导联信号代表 了从不同的方向上观察心脏的活动情况。
人工阅读心电图是一项费时费力的工作, 随计算机技术的发展, 人们开 始对心电自动分析技术进行广泛的研究。 人们研究的重点首先集中在对单个 导联如何识别各种波群, 并提出了各种对单个波群的识别方法; 随后, 人们 开始研究如何在多导联上识别各种波群。
现有技术中有两种在多导联上识别各种波群的方法: 一种方法是, 首先 对每个导联进行单独识别。 由于同一个波群在不同导联上的识别结果可能出 现偏差, 为了能够得到波群在导联中的准确位置, 先要记录下同一个波群在 多个不同导联中的具体时间位置, 并设置一时间阈值。 从最早的波群出现的 时间位置开始向后, 取一个与时间阈值长度相同的时间段, 计算该时间段内 出现该同一个波群的位置数; 从最晚的波群出现的时间位置开始向前, 取一 个与时间阈值长度相同的时间段, 计算该时间段内出现该同一个波群的位置 数。 取一个与时间阈值时间段内出现同一个波群位置数较少的区域, 将该区 域中的最前或最后的波群位置剔除。 重复上述过程, 直到剩余的波群位置都 落在设定的时间阈值范围内, 剩余的波群位置为波群在导联中的准确的位置。 另一种方法是, 首先将多个导联的信号综合成一个导联, 然后在该综合生成 的导联上进行波群的识别。 多个导联信号的综合釆用了加权叠加的方法, 即 首先对每个导联计算一个权值, 然后将各个导联的信号乘以各导联对应的权 值, 再将各个导联信号乘以对应的权值得到的值的绝对值加起来作为综合生 成的导联信号。 该方法是在将多个导联综合后的导联上进行波群的识别。
发明人在发明的过程中发现, 现有技术中对多导联的识别效果差,导致对 疾病的诊断不准确, 甚至不能直接用于疾病的诊断。 发明内容
本发明方案提供了一种心电图导联识别方法及装置, 能够有效的利用导 联中特征比较明显的导联进行优先识别, 保证了识别的效果。 为达到上述目的, 本发明的实施例釆用如下技术方案:
一种心电图导联识别方法, 包括:
选取医学特征明显的导联或者选取特征幅值最大的导联;
识别所选取的导联上的波群。
一种心电图导联识别装置, 包括:
选择单元, 具体用于选取医学特征明显的导联或者选取特征幅值最大的 导联;
识别单元, 具体用于识别所选取的导联上的波群。
本发明从测量到的导联中选取医学特征明显的导联或者选取特征幅值最 大的导联, 并识别所选取的导联上的波群。 由于选取的导联是医学特征明显 的导联或者特征幅值最大的导联, 使得选取的导联信号质量比较好, 且选取 的导联不经过任何的再处理使选取的导联保持原有的图形形态, 有利于波群 识别效果的提高; 并且在针对于选出的导联上进行波群的识别, 由于该识别 过程是在质量比较好、 特征比较明显的选取的导联上进行的, 所以进一步保 证了波群的识别效果, 提高了波群识别的准确率, 有利于提高疾病诊断的准 确性。
附图说明 施例或现有技术描述中所需要使用的附图作简单地介绍, 显而易见地, 下面 描述中的附图仅仅是本发明的一些实施例, 对于本领域普通技术人员来讲, 在不付出创造性劳动性的前提下, 还可以根据这些附图获得其他的附图。
图 1为本发明实施例提供的心电图导联识别方法流程图;
图 2为本发明实施例提供的心电图导联识别装置结构示意图;
图 3 为本发明实施例提供的选取医学特征明显导联的心电图导联的识别 方法流程图;
图 4 为本发明实施例提供的选取医学特征明显导联的心电图联识别装置 结构示意图;
图 5 为本发明实施例提供的选取特征幅值最大的导联的心电图导联识别 方法流程图;
图 6 为本发明实施例提供的选取特征幅值最大的导联的心电图导联识别 装置结构示意图;
图 7 为本发明实施例提供的选取医学特征明显导联和选取特征幅值最大 的导联相结合的心电图导联识别方法流程图;
图 8 为本发明实施例提供的选取医学特征明显导联和选取特征幅值最大 的导联相结合的心电图导联装置结构示意图;
图 9为本发明实施例提供的与图 7对应的心电图导联识别方法中用于判 断选取的导联上是否存在误识别波群的方法流程图;
图 10为本发明实施例提供的心电图导联示意图; 图 1 1为本发明实施例提供的与图 8对应的心电图导联识别装置中用于判 断选取的导联上是否存在误识别波群的结构示意图;
图 12为本发明实施例提供的心电图导联识别方法中用于判断选取的导联 上是否存在漏识别波群的方法流程图;
图 1 3为本发明实施例提供的心电图导联识别装置中用于判断选取的导联 上是否存在漏识别波群的结构示意图;
图 14为本发明实施例提供的心电图导联识别方法流程图。
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, 本领域普通技术人员在没有作 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。
心电检查时, 要将电极按照各种不同的固定位置贴在人体表面, 心电图 机会记录电极间的电压变化情况, 从而产生相应位置对应的同步导联信号。 本发明实施例提供一种心电图导联识别方法, 如图 1所示, 该方法包括:
101、 从测量到的导联中, 选取医学特征明显的导联或者选取特征幅值最 大的导联。
102、在选取的导联上进行波群的识别。波群的识别可以釆用差分阈值法, 模板法, 小波法等中的任意一种波群识别方法。
本发明实施例还提供一种心电图导联识别装置, 如图 2所示, 该装置包 括: 选择单元 21和识别单元 22。
在心电图导联的识别时, 导联的质量的好坏, 直接影响导联的识别效果。 为保证导联的识别效果, 选择单元 21 , 具体用于在各测量到的导联中, 选取 医学特征明显的导联或者选取特征幅值最大的导联。 选出所述选取的导联后, 在识别单元 22内, 根据差分阈值法、 模板法或者小波法等任何现有的公知技 术对选出的选取的导联进行波群的识别。 由于本发明实施例首先从测量到的导联中选取医学特征明显的导联或者 选取特征幅值最大的导联, 使得所述选取的导联信号质量比较好, 使选取的 导联的图形特征保持原有的图形形态, 为选取的导联上波群的识别提供了保 证。 并且导联上波群的识别是在质量比较好、 特征比较明显的选取的导联上 进行的波群识别, 这种导联上的波群位置准确、 特征明显, 有利于波群的识 别, 提高了波群识别的准确率, 有利于提高疾病诊断的准确性。
由前述可知, 在实现心电图导联识别方法时, 从测量到的导联中选取医 学特征明显的导联或者选取特征幅值最大的导联, 本实施例具体描述从测量 到的导联中选取医学特征明显的导联实现的心电图导联识别方法, 如图 3 所 示, 该方法包括:
301、 根据心电图机预先接好的电极组合生成的导联信号, 判断测量到的 导联中是否存在 II导联。 若测量到的导联中存在 II导联, 执行步骤 302 ; 若测 量到的导联中不存在 II导联, 执行步骤 303。
由于心电检查时, 要将电极按照各种不同的固定位置贴在人体表面, 心 电图机会记录电极间的电压变化情况, 从而产生相应位置对应的同步导联信 号, 但上述的导联信号并不包含全部的十二种同步的导联信号, 当产生某一 种导联的位置不适合贴电极时,该种导联就缺省, 因此有可能不存在 I I导联。
302、 从测量到的导联中选取 II导联为医学特征明显的导联, 然后执行步 骤 306。
303、在心电图的导联中除了 II导联医学特征比较明显外, VI导联的医学 特征较 II导联医学特征稍次之。 所以在测量到的导联中不存在 II导联时, 根 据各导联本身的医学特征与预先设置的医学特征较 II导联稍次之 VI导联的进 行比较,判断测量到的导联中是否存在 VI导联。 若测量到的导联中存在 VI导 联, 执行步骤 304 ; 若测量到的导联中不存在 VI导联, 执行步骤 305。
304、 从测量到的导联中选取 VI 导联为医学特征明显的导联, 然后执行 步骤 306。 305、 如果测量到的导联中既不存在医学特征最明显的 II导联, 也不存在 医学特征较 II导联稍次之的 VI导联, 则从选取测量到的导联中选取除了 II导 联和 V 1导联之外的其他医学特征比较明显的导联为医学特征明显的导联。 然 后执行步骤 306。
306、 利用差分阈值法, 模板法, 小波法等波群的识别方法对选取的医学 特征明显的导联的波群进行识别。
307、 当医学特征明显的导联上的波群识别完成后, 根据医学特征明显的 导联中已识别出的波群的位置, 确定在医学特征明显的导联中识别出的波群 的时间起点和时间终点。
308、 根据医学特征明显的导联中识别出的波群的时间起点和时间终点以 及预设的时间阈值, 确定医学特征明显的导联中已识别波群的识别时间区间 , 该识别时间区间的具体确定方法如下:
将所述时间起点减去预设的时间阈值作为识别时间区间的时间起点; 将 所述时间终点加上预设的时间阈值作为识别时间区间的时间终点。 别, 该对应波群的识别可以利用差分阈值法, 模板法, 小波法等波群的识别 方法进行识别。
本发明实施例提供一种心电图导联识别装置, 如图 4所示, 该装置包括: 选择单元 41、 识别单元 42、 确定单元 43、 区间计算单元 44。
选择单元 41具体用于从测量到的导联中选取一个医学特征比较明显的导 联; 识别单元 42具体用于利用差分阈值法, 模板法, 小波法等波群的识别方 法对选取的医学特征比较明显的导联的波群进行识别; 确定单元 43具体用于 在对除医学特征比较明显导联以外的剩余导联进行波群识别时, 确定在医学 特征比较明显导联中已识别的波群的时间起点和时间终点; 区间计算单元 44 具体用于利用该波群的时间起点和时间终点以及特定时间阈值计算一个识别 时间区间, 在所述识别时间区间内对除医学特征明显的导联之外的剩余导联 进行波群识别。
其中, 选择单元 41包括第一判断模块 411和第一选定模块 412。 第一判 断模块 411 具体用于判断测量到的导联中是否存在 II导联, 若测量到的导联 中存在 II导联, 则第一选定模块 412 具体用于选定 II导联为医学特征比较明 显导联; 若测量到的导联中不存在 II导联, 则在在所述第一判断模块 411 中 判断测量到的导联中是否存在 VI导联, 若测量到的导联中存在 VI导联, 则 所述第一选定模块 412具体用于选定 VI导联为医学特征比较明显导联。 若测 量到的导联中 VI导联也不存在, 则所述第一选定模块 412具体用于从测量到 的导联中选取除了 Π导联和 VI导联之外的医学特征比较明显的导联。
其中,所述区间计算单元 44包括起点计算模块 441和终点计算模块 442。 起点计算模块 441 具体用于计算识别时间区间的起点, 即将所述时间起点减 去预设的时间阈值作为识别时间区间的时间起点。 终点计算模块 442 具体用 于计算识别时间区间的时间终点, 即将所述时间终点加上预设的时间阈值作 为识别时间区间的时间终点。
在心电测试时得到导联中医学特征最明显的是 II导联, 其次是 VI导联, 上述实施例选取医学特征比较明显导联选取过程中, 当 II导联不存在时, 可 以参照 VI导联; 或者当 II导联和 VI导联都不存在时, 可以参照导联中除了 II导联和 VI导联之外的其他医学特征比较明显的导联。 这种选取医学特征比 较明显导联的方式, 使医学特征比较明显导联的选取过程周密, 避免了在测 量到的导联中缺省医学特征明显的导联的情况下, 选择了质量不好的导联为 医学特征比较明显导联, 从而使选取的医学特征比较明显导联质量比较好、 特征比较明显, 保证了医学特征比较明显导联上波群的识别效果, 有利于疾 病的准确诊断。
上述实施例中, 根据医学特征比较明显导联上波群的识别结果, 对其他 除选取的导联之外的剩余导联进行波群识别。 由于医学特征比较明显的导联 上波群的识别结果比较准确, 使得除选取的导联之外的剩余导联上波群的识 别质量有了保证, 进一步保证了疾病诊断的准确性。
本发明实施例提供根据选取特征幅值最大的导联实现心电图导联识别方 法。 如图 5所示, 该方法包括:
501、 心电检查时, 要将电极按照各种不同的固定位置贴在人体表面, 心 电图机会记录电极间的电压变化情况, 从而产生相应位置对应的同步导联信 号。 计算预定时间范围内各个测量到的导联的特征幅值, 该预定时间为各个 导联至少包含一个 QRS波群的前 n秒, n至少大于 2。 其中导联的特征幅值的 计算包括以下三种方式:
第一种, 计算各个测量到的导联在预定时间范围内的高度值, 将各高度 值中最大高度值作为特征幅值。
第二种, 首选确定各个测量到的导联在预定时间范围内的波峰点及该波 峰点的高度; 其次确定波峰点的特定时间阈值区间内波形的最小高度, 该特 定时间阈值区间的起点由波峰点减去特定时间阈值确定, 终点由波峰点加上 特定时间阈值确定, 其中特定时间阈值一般不超过 1 00 毫秒; 最后计算所述 波峰点幅度与最小幅度之间的高度差, 将该高度差中最大的高度差作为所述 特征幅值。
第三种, 首先分别确定各个测量到的导联在预定时间范围内的波峰点及 波谷点和波峰点高度以及波谷点高度; 其次确定波谷点的特定时间阈值区间 确定, 终点由波谷点加上特定时间阈值确定, 其中特定时间阈值一般不超过 1 00毫秒; 确定波峰点的特定时间阈值区间内波形的最小高度, 该特定时间阈 值区间的起点由波峰点减去特定时间阈值确定, 终点由波峰点加上特定时间 阈值确定, 其中特定时间阈值一般不超过 1 00 毫秒; 最后分别计算所述波谷 点高度和所述最大高度之间的第一高度差以及波峰点高度与所述最小高度之 间的第二高度差, 将第一高度差和第二高度差中最大高度差作为所述特征幅 值。 502、 根据计算出的各测量到的导联的特征幅值, 选取特征幅值中最大的 特征幅值对应的导联。
503、 利用差分阈值法, 模板法, 小波法等波群的识别方法对选取的特征 幅值最大的导联上的波群进行识别。
特征幅值最大的导联上的波群识别完成后, 根据特征幅值最大的导联中 已识别的波群的位置, 在测量到的除特征幅值最大的导联之外的剩余导联上 进行波群识别, 此识别过程的实现和实施例 2的图 3中 307至 309的该识别 过程相同。
本发明实施例提供一种心电图导联识别装置, 如图 6所示, 该装置包括: 选择单元 61、 识别单元 62、 确定单元 63、 区间计算单元 64。
选择单元 61具体用于根据导联的图形特征, 从测量到的导联中选取一个 特征幅值最大的导联; 识别单元 62 , 具体用于利用差分阈值法, 模板法, 小 波法等波群的识别方法对特征幅值最大的导联上的波群进行识别; 确定单元 63具体用于在对除特征幅值最大的导联以外的剩余导联进行波群识别时, 确 定在特征幅值最大的导联中已识别的波群的时间起点和时间终点; 区间计算 单元 64具体用于利用该波群的时间起点和时间终点以及特定时间阈值计算一 个识别时间区间, 在所述识别时间区间内对除特征幅值最大的导联之外的剩 余导联进行波群识别。
其中选择单元 61 包括计算模块 611和第二选定模块 612。 计算模块 611 具体用于在分析各个测量到的导联的图形特征时, 根据上述的三种导联特征 幅值的计算方法中的任意一种, 计算出各测量到的导联的特征幅值。 测量到 的导联的特征幅值计算完以后, 第二选定模块 612 具体用于从各导联特征幅 值中选出与最大特征幅值对应的导联, 将该导联作为特征幅值最大的导联。
其中,所述区间计算单元 64包括起点计算模块 641和终点计算模块 642。 起点计算模块 641 具体用于计算识别时间区间的起点, 即将所述时间起点减 去预设的时间阈值作为识别时间区间的时间起点。 终点计算模块 642 具体用 于计算识别时间区间的时间终点, 即将所述时间终点加上预设的时间阈值作 为识别时间区间的时间终点。
上述实施例中, 选取特征幅值最大的导联时, 首先在测量到的导联上逐 一进行导联特征幅值的计算, 使获得到的特征幅值比较准确, 然后在准确的 特征幅值基础上进行最大特征幅值的选取, 最后选取与最大特征幅值对应的 导联为特征幅值最大的导联。 这种方式选取的特征幅值最大的导联图形特征 明显, 进一步提高了导联上波群的识别准确率, 有利于疾病的诊断。
本发明实例具体描述根据导联的医学特征和导联的图形特征相结合提供 的心电图导联识别方法。 该方法包括如下步骤: 如图 7所示, 该方法包括: 由于导联的医学特征不用计算, 且医学特征比较明显的导联质量也比较 好, 可以首先根据导联的医学特征, 进行所述导联的选取。
701、 心电检查时, 要将电极按照各种不同的固定位置贴在人体表面, 心 电图机会记录电极间的电压变化情况, 从而产生相应位置对应的同步导联信 号。 从测量到的导联中根据导联的医学特征选取所述选取的导联时, 首先根 据心电图机预先接好的电极组合生成的导联信号, 判断测量到的导联中是否 存在 II导联。 若测量到的导联中存在 II导联, 执行步骤 702 ; 若测量到的导联 中不存在 II导联, 执行步骤 703。
702、 从测量到的导联中选取医学特征明显的 II导联为所述选取的导联, 然后执行步骤 707。
703、在心电图的导联中除了 II导联医学特征比较明显外, VI导联的医学 特征较 II导联医学特征稍次之。 所以在测量到的导联中不存在 II导联时,判断 测量到的导联中是否存在 VI导联。 若测量到的导联中存在 VI导联, 执行步 骤 704 ; 若测量到的导联中不存在 VI导联, 执行步骤 705。
704、 从测量到的导联中选取 VI 导联为所述选取的导联, 然后执行步骤
707。
705、 如果测量到的导联中既不存在医学特征最明显的 II导联, 也不存在 医学特征较 II导联稍次之的 VI导联, 就要根据测量到的导联的图形特征, 从 测量到的导联中选取导联特征幅值最大的导联。 从测量到的导联中选取导联 特征幅值最大的导联时, 首先计算预定时间范围内各个测量到的导联的特征 幅值。 其中特征幅值的计算方法和的图 5中 501特征幅值的计算方法相同。
706根据计算出的各测量到的导联的特征幅值,选取特征幅值中最大的特 征幅值, 将与该最大特征幅值对应的导联作为特征幅值最大的导联。
707、 利用差分阈值法, 模板法, 小波法等波群的识别方法对所述选取的 导联的波群进行识别。
在所述选取的导联上的波群识别完成后, 根据所述选取的导联中已识别 出的波群的位置, 在其他测量到的除选取的导联之外剩余的导联的对应位置 上查找相应的波群, 并对该相应波群进行识别, 此识别过程的实现与图 3 中 307至 309的识别过程相同。
本发明实施例还提供了一种心电图导联识别装置, 如图 8 所示, 该装置 包括: 选择单元 81、 识别单元 82、 确定单元 83、 区间计算单元 84。
选择单元 81具体用于从测量到的导联中选取一个医学特征比较明显的导 联或者特征幅值最大的导联作为选取的导联; 识别单元 82具体用于利用差分 阈值法, 模板法, 小波法等波群的识别方法对所述选取的导联上的波群进行 识别时, 确定在所述选取的导联中已识别的波群的时间起点和时间终点; 区 间计算单元 84具体用于利用该波群的时间起点和时间终点以及特定时间阈值 计算一个识别时间区间, 在所述识别时间区间内对除所述选取的导联之外的 剩余导联进行波群识别。
其中, 选择单元 81 包括第一判断模块 811、 第一选定模块 812、 计算模 块 813和第二选定模块 814。第一判断模块 811具体用于判断测量到的导联中 是否存在医学特征最明显的 II导联, 若存在 II导联, 则第一选定模块 812 具 体用于选定 II导联为所述选取的导联; 若不存在 II导联, 则所述第一判断模 块 811具体用于判断测量到的导联中是否存在 VI导联, 若存在 VI导联, 则 所述第一选定模块 812具体用于选定 VI导联为所述选取的导联。 当测量到的 导联中既不存在 II导联也不存在 VI导联时, 从测量到的导联中选取特征幅值 最大的导联作为所述选取的导联。 计算模块 81 3 具体用于计算各测量到的导 联的特征幅值; 第二选定模块 814 具体用于从各导联特征幅值中选出特征幅 值最大的导联, 将与该特征幅值最大对应的导联作为所述选取的导联。
其中,所述区间计算单元 84包括起点计算模块 841和终点计算模块 842。 起点计算模块 841 具体用于计算识别时间区间的起点, 即将所述时间起点减 去预设的时间阈值作为识别时间区间的时间起点。 终点计算模块 842 具体用 于计算识别时间区间的时间终点, 即将所述时间终点加上预设的时间阈值作 为识别时间区间的时间终点。 选取的导联上已识别的波群进行判定, 查看是否存在误识别的波群。 若从除 所述选取的导联之外的剩余导联中没有识别到与所述选取的导联中特定波群 对应的波群, 则进行所述选取的导联上的波群是否为误识别波群判断。如图 9 所示, 该方法包括:
901、 判断所述选取的导联上的波群是否为误识别波群, 若该波群为误识 别波群, 则执行步骤 902 ; 若不是误识别波群, 则执行步骤 903。
902、 根据上述是滞为误识别波群的判断, 若该特定波群为误识别波群, 此波群将没有使用价值, 则将该特定波群从所述选取的导联中删除。
903、 继续进行与选取的导联上已识别的其他波群对应的除选取的导联以 外的导联上波群的识别。
其中, 针对步骤 901 中误识别波群的判断有以下两种方法,以识别波群 QRSi为例, 如图 10所示:
第一种: 首先计算所述波群 QRSi向后的两个波群 QRS H 和 QRS H 的时间 间隔 或者所述特定波群向前的两个波群 QRSi+1和 QRSi+2之间的时间间隔 Ri+1Ri+2为第一时间间隔; 再次计算所述波群 QRSi向后一个波群(^!^^与向前一 个波群 QRS ί+1之间的时间间隔 υί+1为第二时间间隔; 然后判断所述第二时间 间隔是否小于预设间隔系数 Κ乘以第一时间间隔, 该预设间隔系数 Κ为一经 验阈值,一般不大于 1.5;一般 υί+1不小于 K乘以 或者 Ri— +1不小于 Κ 乘以 Ri+1Ri+2时所述波群 SQRi为正常识别的波群; 如果所述第二时间间隔小于 预设间隔系数乘以第一时间间隔, 所述特定波群为误识别波群。
第二种: 首先确定所述波群 QRSi特定时间阈值区间内波形的最小高度和 最大高度; 其次计算所述波群 QRSi的最大高度与所述特定波群最小高度的高 度差 Hi ; 再次分别计算所述波群 QRSi向前一个波群 QRSi+1的高度差 Ηί+1和向后 一个波群 QRSi— 々高度差 ; 然后判断所述波群 QRSi的高度差 是否小于预 设高度系数 P乘以 Ηί+1和 中较小的高度差, 该高度系数 Ρ为一经验值, 通 常不大于 0. 5 ; 一般所述波群的高度差 大于等于预设高度系数 Ρ乘以 Ηί+1和 中较小的高度差时所述波群 SQRi为正常识别的波群。 如果所述特定波群的 特征幅值小于预设幅度系数乘以较小特征幅值, 则所述波群为误识别波群。 别波群的结构,如图 11所示,该结构包括:第一判断单元 1 11和删除单元 112。
第一判断单元 i l l ,具体用于在从除选取的导联之外的导联中未识别到与 选取的导联特定波群对应的波群时, 确定所述波群为误识别波群。 删除单元 112 , 具体用于当该特定波群为误识别波群时, 从选取的导联中删除该特定波 群删除。
其中, 第一判断单元 11 1包括间隔计算模块 11 11、 间隔判断模块 1112、 间隔判定模块 111 3、 高度差计算模块 1114、 高度差判断模块 1 115和高度差 判定模块 1116。
其中, 间隔计算模块 1111具体用于首先计算该特定波群向后的两个波群 之间的第一时间间隔, 然后再计算该特定波群向后一个波群与向前一个波群 的第二时间间隔。 间隔判断模块 11 12具体用于判断所述第一时间间隔是否小 于预设间隔系数乘以第二时间间隔, 该预设间隔系数 K 为一经验阈值, 一般 不大于 1. 5。 间隔判定模块 111 3用在所述第一时间间隔小于预设间隔系数乘 以第二时间间隔时, 判定所述特定波群为误识别波群。
其中, 高度差计算模块 1114首先具体用于确定特定波群在特定时间阈值 区间内波形的最小高度和最大高度, 该特定时间阈值区间的起点由该特定波 群一定时间点减去所述特定时间阈值确定, 该特定时间阈值区间的终点由该 特定波群一定时间点加上所述特定时间阈值确定, 其中特定时间阈值一般不 超过 100 毫秒; 其次计算所述特定波群的最大高度和所述特定波群最小高度 的高度差; 再次分别计算所述波群向前一个波群和向后一个波群的高度差; 高度差判断模块 1115 , 具体用于判断所述特定波群的特征幅值是否小于预设 幅度系数乘以较小特征幅值, 该高度系数 P为一经验值, 通常不大于 0. 5。 高 度差判定模块 1116 , 具体用于当所述波群的高度差小于预设高度系数乘以所 述波群向前一个波群和向后一个波群的高度差中较小特征幅值时, 判定所述 特定波群为误识别波群。
另外当测量到的波群识别完成以后, 可以对整个所述选取的导联上已识 别的波群进行一个整体的判断, 判断选取的导联上是否存在波群漏识别的情 况。 若选取的导联上存在波群漏识别的情况, 则查找是否存在新的波群, 若 选取的导联上不存漏识别的波群, 则导联的识别结束。 所述选取的导联上是 否存在漏识别波群的判断方法如图 12所示, 该方法包括:
1201、 计算所述选取的导联中任意连续两个波群之间的平均间隔。
1202、 判断是否存在连续两个波群之间的间隔大于预设系数乘以所述平 均间隔, 该预设系数为一经验阈值, 通常大于 1. 5 , 若存在连续两个波群之间 的间隔大于预设系数乘以所述平均间隔, 则执行步骤 1203 , 若不存在连续两 个波群之间的间隔大于预设系数乘以所述平均间隔, 则执行步骤 1204。
1203、 在除所述选取的导联之外的剩余导联上识别该连续两个波群之间 是否存在新的波群, 若存在新的波群, 在所选取导联上识别该连续两个波群 之间是否存在新的波群; 如果所选取导联上该连续两个波群之间存在新的波 群, 则将该新的波群添加到已识别导联的识别结果中。
1204、 结束心电图导联的识别。
本实施例还提供一种具体用于判断所述选取的导联上是否存在漏识别波 群的结构,如图 1 3所示,该结构包括:间隔计算单元 1 31和第二判断单元 1 32。
间隔计算单元 1 31 ,具体用于计算所述选取的导联中所有的连续两个波群 之间的平均间隔。
第二判断单元 1 32 具体用于判断是否存在连续两个波群之间的间隔大于 预设系数乘以所述平均间隔, 该预设系数为一经验阈值, 通常大于 1. 5 , 当存 在连续两个波群之间的间隔大于预设系数乘以所述平均间隔时, 在除所述选 取的导联之外的剩余导联上识别该连续两个波群之间是否存在新的波群, 若 存在新的波群, 在所选取导联上识别该连续两个波群之间是否存在新的波群; 如果所选取导联上该连续两个波群之间存在新的波群, 则将该新的波群添加 到已识别导联的识别结果中。
上述实施例中判断选取的导联上的波群是否为误识别波群和漏识别波群 的方法及结构, 也可以但不局限于用在实施例 1和实施例 3的心电图导联识 别过程中。
上述实施例不仅从测量到的导联中选择医学特征明显的导联为选取的导 联, 还可以在没有医学特征特别明显的导联的情况下, 从测量到的导联中取 特征幅值最大的导联为所述选取的导联, 这样选出的导联的质量得到了更进 一步的保证, 使得导波群的识别效果更佳。 且对除所述选取的导联之外的剩 余导联进行识别的过程中, 还对所述选取的导联上已识别的波群进行是否存 在误识别的判断, 若果存在波群的误识别, 及时将该选取的导联上误识别的 波群删掉, 从而进一步保证了波群的识别效果。 在所有导联中的波群识别完 成后, 对选取的导联上的波群进行整体的检查, 判断是否存在波群漏识别的 情况, 若选取的导联上存在漏识别的波群, 查找到该漏识别的波群, 将其添 加到选取的导联的相应位置。 这种导联的识别方式使得导联上波群的识别准 确无误, 充分保证了导联上波群的识别效果, 有利于疾病的诊断。
本发明实施例还提供心电图导联识别方法, 如图 14所示。
由于心电图中特征明显的各类波群并不在同一条医学特征或图形特征明 显的导联上。 因此要使各类波群都能够在各自特征明显的导联上进行优先识 别, 那就要针对不同的波群选取不同的对应于该类波群的医学特征明显的导 联或者选取特征幅值最大的导联作为所述选取的导联。 上述选取的导联选取 后, 各类波群在各自对应的选取的导联上分别进行各类波群的识别。 上述选 取的导联上对应的波群识别完成后, 根据上述选取的导联上波群的识别结果, 对其它的除选取的导联以外的测量到的导联上的对应波群进行识别。 具体包 括:
1401、 心电检查时, 要将电极按照各种不同的固定位置贴在人体表面, 心电图机会记录电极间的电压变化情况, 从而产生相应位置对应的同步导联 信号。 从各类待识别的波群中选取任意一类待识别的波群, 针对该类波群, 从测量到的导联中选取医学特征明显的导联或者选取特征幅值最大的导联作 为所述选取的导联。
1402、 上述选取的导联从测量到的导联中选出后, 在上述选取的导联上 根据差分阈值法、 模板法、 小波法等任意一种波群识别方法, 对该类波群进 行识别。
1403、 当该类波群在上述选取的导联上识别完成后, 根据所述选取的导 联中该类波群已识别出的波群的位置, 在除所述选取的导联之外的剩余导联 的对应位置上查找相应的该类波群, 并对该相应类波群进行识别。
其中, 要实现根据所述选取的导联中该类波群已识别出的波群的位置, 在其他测量到的非选取的导联的对应位置上查找相应的该类波群, 并对该相 应类波群进行识别。 首先要确定在所述选取的导联中识别出的该类波群的时 间起点和时间终点。 然后根据所述选取的导联中识别出的该类波群的时间起 点和时间终点以及预设的时间阈值, 确定所述选取的导联中已识别的该类波 群的识别时间区间。 最后在上述识别时间区间内, 对除所述选取的导联之外 的其他测量到的导联进行对应该类波群的识别。 该对应类波群的识别可以利 用差分阈值法, 模板法, 小波法等波群的识别方法进行识别。
1404、 当该类波群在所有测量到的导联上识别完成后, 将进行针对另一 类待识别波群的医学特征明显的导联或者选取特征幅值最大的导联的选取以 及该类波群的识别。 在针对另一类待识别波群进行所述选取的导联的选取之 前, 首先将已识别的一类波群在各个测量到的导联中排除掉, 然后再针对另 一类未识别波群在各测量到的导联中进行医学特征明显的导联或者选取特征 幅值最大的导联的选取。 针对另一类未识别波群的所述选取的导联选取过程 以及该类波群的识别过程, 同上述第一类波群的所述选取的导联的选取及识 别过程相同。 重复此过程, 直至将所有不同类的波群识别完成。
本实施例心电图导联的识别过程中判断所述选取的导联上的波群是否存 在误识别波群和漏识别波群的法还可以但不局限于釆用实施例 4 中的具体用 于判断所述选取的导联上的波群是否存在误识别波群和漏识别波群的法。
上述事实例, 选取所述选取的导联时根据各类波群的特征, 选取针对于 各类波群特征明显的导联为各类波群对应的所述选取的导联, 这样选取的所 述选取的导联上每类波群的特征明显、 位置准确, 保证各类波群的识别效果 的准确性, 使整个测量到的导联上的波群的识别效果更佳, 保证了疾病诊断 的准确性。
通过以上实施例的描述, 本领域的技术人员可以清楚地了解到本发明可 借助软件加必需的通用硬件平台的方式来实现, 当然也可以通过硬件, 但艮 多情况下前者是更佳的实施方式。 基于这样的理解, 本发明实施例的技术方 案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出 来, 该软件产品存储在一个存储介质中, 包括若干指令用以使得移动设备(可 以是手机, 个人计算机, 媒体播放器等)执行本发明各个实施例所述的方法。 这里所称的存储介质, 如: R0M/RAM、 磁盘、 光盘等。
以上所述, 仅为本发明的具体实施方式, 但本发明的保护范围并不局限 于此, 任何熟悉本技术领域的技术人员在本发明揭露的技术范围内, 可轻易 想到变化或替换, 都应涵盖在本发明的保护范围之内。 因此, 本发明的保护 范围应所述以权利要求的保护范围为准。

Claims

权利 要 求 书
1、 一种心电图导联识别方法, 其特征在于, 包括:
选取医学特征明显的导联或者选取特征幅值最大的导联;
识别所选取的导联上的波群。
2、 根据权利要求 1所述的心电图导联识别方法, 其特征在于, 选取医学特 征明显的导联包括:
判断测量到的导联中是否存在 II导联;
如果存在 II导联, 则将 II导联作为所述医学特征明显的导联;
如果不存在 II导联, 则判断测量到的导联中是否存在 VI导联;
如果存在 VI导联, 则将 VI导联作为所述医学特征明显的导联。
3、 根据权利要求 1所述的心电图导联识别方法, 其特征在于, 选取特征幅 值最大的导联包括:
计算所有测量到的导联在预定时间范围内的特征幅值;
从测量到的导联中选取特征幅值最大的导联。
4、 根据权利要求 3所述的心电图导联识别方法, 其特征在于, 所述计算所 有测量到的导联在预定时间范围内的特征幅值包括:
确定所述预定时间范围内各波群的高度值, 将高度值中最大高度值作为所 述特征幅值; 或者
确定所述预定时间范围内波峰点以及波峰点的特定时间阈值区间内的波形 的最小高度, 计算所述波峰点高度与最小高度之间的高度差, 将该高度差中最 大值高度差作为所述特征幅值; 或者
确定所述预定时间范围内波峰点和波谷点; 以及确定波谷点的特定时间阈 值区间内波形的最大高度和波峰点的特定时间阈值区间内波形的最小高度, 计 算所述波谷点高度与所述最大高度之间的第一高度差以及所述波峰点高度与所 述最小高度之间的第二高度差, 将第一高度差和第二高度差中最大高度差作为 所述特征幅值。
5、 根据权利要求 1所述的心电图导联识别方法, 其特征在于, 该方法还包 括:
确定所述选取的导联中已识别波群的时间起点和时间终点;
居所述时间起点和时间终点计算识别时间区间;
6、 根据权利要求 5所述的心电图导联识别方法, 其特征在于, 该方法还包 括:
若在剩余导联中没有识别到与选取的导联中已识别的特定波群对应的波 群, 则判断所述选取的导联上已识别的特定波群为误识别波群, 并从选取的导 联中删除该特定波群。
7、 根据权利要求 6所述的心电图导联识别方法, 其特征在于, 所述判断所 述特定波群为误识别波群包括:
计算所述特定波群向后或向前的两个波群之间的第一时间间隔;
计算所述特定波群向后一个波群与向前一个波群的第二时间间隔; 判断所述第二时间间隔是否小于预设间隔系数乘以第一时间间隔; 如果所述第二时间间隔小于预设间隔系数乘以第一时间间隔, 则判定所述 特定波群为误识别波群。
8、 根据权利要求 6所述的心电图导联识别方法, 其特征在于, 所述判断所 述特定波群为误识别波群包括:
计算所述特定波群的在预设时间阈值区间内的高度差、 以及与特定波群的 相邻波群在预设时间阈值区间内的高度差;
判断所述特定波群的高度差是否小于预设高度系数乘以与特定波群的相邻 波群中高度差中较小的高度差;
如果所述特定波群的高度差小于预设高度系数乘以与特定波群的相邻波群 中较小的高度差, 则判定所述特定波群为误识别波群。
9、 根据权利要求 5-8中 任一所述的心电图导联识别方法, 其特征在于, 所述根据所述时间起点和时间 终点计算识别时间区间包括:
将所述时间起点减去预设的时间阈值作为识别时间区间的时间起点; 将所述时间终点加上预设的时间阈值作为识别时间区间的时间终点。
10、 根据权利要求 1 所述的心电图导联识别方法, 其特征在于, 该方法还 包括: 判断是否存在连续两个波群之间的间隔大于预设系数乘以所述平均间隔; 如果存在连续两个波群之间的间隔大于预设系数乘以所述平均间隔, 则在 该连续两个波群之间重新进行波群识别。
11、 根据权利要求 10所述的心电图导联识别方法, 其特征在于, 在该连续 两个波群之间重新进行波群识别包括:
在除所述选取的导联之外的剩余导联上识别该连续两个波群之间是否存在 新的波群;
如果剩余导联上该连续两个波群之间存在新的波群, 在所选取导联上识别 该连续两个波群之间是否存在新的波群;
如果所选取导联上该连续两个波群之间存在新的波群, 则将该新的波群添 加到已识别导联的识别结果中。
12、 一种心电图导联识别装置, 其特征在于, 包括:
选择单元, 具体用于选取医学特征明显的导联或者选取特征幅值最大的导 联;
识别单元, 具体用于识别所选取的导联上的波群。
13、 根据权利要求 12所述的心电图导联识别装置, 其特征在于, 所述选择 单元包括:
第一判断模块, 具体用于判断测量到的导联中是否存在 II导联;
第一选定模块, 具体用于当测量到的导联中存在 II导联时, 将 II导联作为 医学特征明显的导联; 当测量到的导联中不存在 II导联时, 所述第一判断模块具体用于判断测量 到的导联中是否存在 VI导联;
当测量到的导联中存在 VI导联时, 所述第一选定模块具体用于将 VI导联 作为所述医学特征明显的导联;
14、 根据权利要求 12所述的心电图导联识别装置, 其特征在于, 所述选择 单元包括:
计算模块, 具体用于计算所有测量到的导联在预定时间范围内的特征幅值; 第二选定模块, 具体用于选定最大的特征幅值对应的导联。
15、 根据权利要求 14所述的心电图导联识别装置, 其特征在于,
所述计算模块具体用于确定所述预定时间范围内各波群的高度值, 并将所 述预定时间范围内的最大高度值作为所述特征幅值; 或者
具体用于确定预定时间范围内波峰点高度以及波峰点的特定时间阈值区间 内波形的最小高度, 计算所述波峰点高度与最小高度之间的高度差, 将该高度 差中的最大高度差作为所述特征幅值; 或者
具体用于确定预定时间范围内波谷点和波峰点; 确定波谷点的特定时间阈 值区间内波形的最大高度和波峰点的特定时间阈值区间内波形的最小高度, 计 算所述波谷点高度与所述最大高度之间的第一高度差以及所述波峰点高度与所 述最小高度之间的第二高度差, 将第一高度差和第二高度差中最大高度差作为 所述特征幅值。
16、 根据权利要求 12所述的心电图导联识别装置, 其特征在于, 该装置还 包括: 区间计算单元, 具体用于根据所述时间起点和时间终点计算识别时间区间; 所述识别单元还具体用于在所述识别时间区间内对除选取的导联之外的剩 余导联进行波群识别。
17、 根据权利要求 16所述的心电图导联识别装置, 其特征在于, 所述区间 计算单元包括: 间区间的时间起点;
终点计算模块, 具体用于将所述时间终点加上预设的时间阈值作为识别时 间区间的时间终点。
18、 根据权利要求 12所述的心电图导联识别装置, 其特征在于该装置还包 括:
第一判断单元, 具体用于在剩余导联中没有识别到与选取的导联中已识别 的特定波群对应的波群时, 则判断所述选取的导联上已识别的特定波群是否为 误识别波群;
删除单元, 具体用于在所述选取的导联上已识别的特定波群为误识别波群 时, 则将该特定波群从选取的导联中删除。
19、 根据权利要求 18所述的心电图导联识别装置, 其特征在于, 所述第一 判断单元包括:
间隔计算模块, 具体用于计算所述特定波群向后或向前的两个波群之间的 第一时间间隔, 并计算所述特定波群向后一个波群与向前一个波群的第二时间 间隔;
间隔判断模块, 具体用于判断所述第二时间间隔是否小于预设间隔系数乘 以第一时间间隔;
间隔判定模块, 用在所述第二时间间隔小于预设间隔系数乘以第一时间间 隔时, 判定所述特定波群为误识别波群。
20、 根据权利要求 18所述的心电图导联识别装置, 其特征在于, 所述判断 单元包括:
高度差计算模块, 具体用于计算所述特定波群的高度差、 以及与特定波群 的相邻波群的高度差;
高度差判断模块, 具体用于判断所述特定波群的高度差是否小于预设幅值 系数乘以与所述特定波群的相邻波群的高度差中较小的高度差特定波群的相邻 波群;
高度差判定模块, 具体用于在所述特定波群的高度差小于预设幅值系数乘 以与所述特定波群的相邻波群的高度差中较小的高度差时, 判定所述特定波群 为误识别波群。
21、 根据权利要求 12所述的心电图导联识别装置, 其特征在于, 该装置还 包括:
间隔计算单元, 具体用于计算选取的导联中所有的连续两个波群之间的平 均间隔;
第二判断单元, 具体用于判断是否存在连续两个波群之间的间隔大于预设 系数乘以所述平均间隔;
如果存在连续两个波群之间的间隔大于预设系数乘以所述平均间隔, 所述 识别单元具体用于在该连续两个波群之间重新进行波群识别。
22、 根据权利要求 21所述的心电图导联识别装置, 其特征在于, 所述识别
Figure imgf000025_0001
如果剩余导联上该连续两个波群之间存在新的波群, 所述识别单元具体用 于在所选取导联上识别该连续两个波群之间是否存在新的波群; 如果所选取导 联上该连续两个波群之间存在新的波群, 则将该新的波群添加到已识别导联的 识别结果中。
PCT/CN2009/070123 2009-01-13 2009-01-13 心电图导联识别方法及装置 WO2010081292A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN200980147163.XA CN102438514B (zh) 2009-01-13 2009-01-13 心电图导联识别方法及装置
PCT/CN2009/070123 WO2010081292A1 (zh) 2009-01-13 2009-01-13 心电图导联识别方法及装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2009/070123 WO2010081292A1 (zh) 2009-01-13 2009-01-13 心电图导联识别方法及装置

Publications (1)

Publication Number Publication Date
WO2010081292A1 true WO2010081292A1 (zh) 2010-07-22

Family

ID=42339400

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/070123 WO2010081292A1 (zh) 2009-01-13 2009-01-13 心电图导联识别方法及装置

Country Status (2)

Country Link
CN (1) CN102438514B (zh)
WO (1) WO2010081292A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108836313A (zh) * 2018-07-13 2018-11-20 希蓝科技(北京)有限公司 一种智能选取导联的动态心电图分析方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113786200B (zh) * 2020-05-28 2022-10-11 深圳邦健生物医疗设备股份有限公司 心电信号处理方法、装置、设备及可读介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6947789B2 (en) * 2000-11-13 2005-09-20 Innovise Medical, Inc. Method for detecting, sizing and locating old myocardial infarct
EP0853285B1 (en) * 1997-01-09 2007-01-17 Koninklijke Philips Electronics N.V. Method and system for detecting acute myocardial infarction
CN101467879A (zh) * 2007-12-26 2009-07-01 深圳迈瑞生物医疗电子股份有限公司 一种基于多导同步心电信号处理方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0853285B1 (en) * 1997-01-09 2007-01-17 Koninklijke Philips Electronics N.V. Method and system for detecting acute myocardial infarction
US6947789B2 (en) * 2000-11-13 2005-09-20 Innovise Medical, Inc. Method for detecting, sizing and locating old myocardial infarct
CN101467879A (zh) * 2007-12-26 2009-07-01 深圳迈瑞生物医疗电子股份有限公司 一种基于多导同步心电信号处理方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
YU LIANGJUN. ET AL.: "Method of How to Distinguish the Marking Points of 12-lead ECG", MEDICAL EQUIPMENT, vol. 18, no. 12, 2005 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108836313A (zh) * 2018-07-13 2018-11-20 希蓝科技(北京)有限公司 一种智能选取导联的动态心电图分析方法

Also Published As

Publication number Publication date
CN102438514A (zh) 2012-05-02
CN102438514B (zh) 2014-08-20

Similar Documents

Publication Publication Date Title
US11147500B2 (en) Electrocardiogram processing system for delineation and classification
CN108478209B (zh) 心电信息动态监护方法和动态监护系统
CN109171712B (zh) 心房颤动识别方法、装置、设备及计算机可读存储介质
WO2019100565A1 (zh) 基于人工智能自学习的动态心电图分析方法和装置
JP5271718B2 (ja) 複数の時間区分にわたり胎児及び産婦のecgを特定する方法
CN109411042B (zh) 心电信息处理方法和心电工作站
US11147516B2 (en) Methods and systems to quantify and remove asynchronous noise in biophysical signals
US8543194B2 (en) System and method of detecting abnormal movement of a physical object
CN106815570B (zh) 一种基于动态模式识别的心电信号st-t段识别方法
JP2016530019A (ja) 母体腹部のecg記録からの胎児心拍数の抽出
EP2490587A1 (en) Method and system for detecting cardiac arrhythmia
Fotiadou et al. Multi-channel fetal ECG denoising with deep convolutional neural networks
WO2017059569A1 (zh) 起搏信号处理方法、系统和心电监护仪
CN111345816A (zh) 多导联qrs波群检测方法、装置、设备及存储介质
CN115486854B (zh) 一种针对干电极采集的单导联心电图室性早搏识别方法
KR102428259B1 (ko) 수술 중 환자의 저혈압 예측 장치, 방법 및 프로그램
US11672464B2 (en) Electrocardiogram processing system for delineation and classification
CN111528833B (zh) 一种心电信号的快速识别与处理方法及系统
CN110491504B (zh) 一种心音信号医学指标数据的获取方法
WO2010081292A1 (zh) 心电图导联识别方法及装置
CN108836312B (zh) 一种基于人工智能的进行杂波剔除的方法及系统
KR102428260B1 (ko) 비침습 데이터를 활용한 수술 중 환자의 저혈압 예측 장치, 방법 및 프로그램
Vinay et al. An RNN-Bi LSTM based Multi Decision GAN Approach for the Recognition of Cardiovascular Disease (CVD) from Heart Beat Sound: A Feature Optimization Process
CN112438735A (zh) 一种心电图p波检测方法、分析装置以及存储介质
RU2791006C1 (ru) Система и способ автоматизированного анализа и интерпретации электрокардиограммы

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 200980147163.X

Country of ref document: CN

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

Ref document number: 09838079

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09838079

Country of ref document: EP

Kind code of ref document: A1