Information processing device, information processing method, and program
The information processing device processes and clusters electrocardiogram data to analyze multiple heartbeats efficiently, addressing the challenge of long-term electrocardiogram analysis and improving diagnostic accuracy by visualizing representative waveforms.
Patent Information
- Application Number
- PCT/JP2024/007831
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-22
- Filing Date
- 2024-03-01
- Publication Date
- 2024-11-28
Smart Images

Figure JP2024007831_28112024_PF_FP_ABST
Abstract
Description
Information processing device, information processing method, and program
[0001] The present invention relates to an information processing device, an information processing method, and a program for processing movement information that electrically indicates the movement of the heart or the like of a living organism.
[0002] Electrocardiographs that are worn on a subject's body and can measure the subject's electrocardiogram over an extended period of time are known (see, for example, Patent Document 1). Also, as methods for statistical processing and correction of information, the individual correction method, the Bazett correction method, and the Fridericia correction method are known (Non-Patent Documents 1 and 2). Also, as a method for processing time-series information such as electrocardiograms, the time-series data clustering method is known (Non-Patent Document 3). Also, as a method for determining the similarity between time-series data, the dynamic time warping (DTW) method is known (Non-Patent Document 4). Also, a method for determining a representative waveform from a signal containing multiple repetitive waveforms, such as the heartbeats in an electrocardiogram, is known (Non-Patent Document 5).
[0003] Utilization of QT measurement indexes and the QT / RR relationship (Kouichi Mizumaki, Heart, 48(4), 2016; Vol. 48 No. 4 (2016) https: / / www.jstage.jst.go.jp / article / shinzo / 48 / 4 / 48_388 / _pdf) Study design and statistical evaluation methods for evaluating QT prolongation (Watashi, Jpn J Biomet Vol. 29, Special Issue, 1 2008; https: / / www.jstage.jst.go.jp / article / jjb / 29 / Special_Issue_1 / 29_Special_Issue_1_S61 / _pdf / -char / ja) Time series data clustering and K-Shape (http: / / www.cybergarage.org / memo / timeseries_clustering-kshape / ) Dynamic Time Domain Wave (DTW) Dynamic time warping (https: / / data-analysis-stats.jp / %E6%A9%9F%E6%A2%B0%E5%AD%A6%E7%BF%92 / dtwdynamic-time-warping%E5%8B%95%E7%9A%84%E6%99%82%E9%96%93%E4%BC%B8%E7%B8%AE%E6%B3%95 / ) QRS wave morphology classification system for Holter ECG using double-layered SOM (Kaneko et al., Journal of Biomedical Engineering, 46(6), 576-586, 2008; https: / / www.jstage.jst.go.jp / article / jsmbe / 46 / 6 / 46_6_576 / _pdf
[0004] An electrocardiogram measured from a single subject over a long period of time, for example, a day, can contain waveforms representing a large number of heartbeats, on the order of 100,000, but analyzing each of these waveforms representing a large number of heartbeats has been difficult even for experts such as doctors.
[0005] The present invention has been made against this background, and its objective is to process movement information indicating multiple repeated movements of a living organism, such as an electrocardiogram obtained by measuring a single human, and to analyze waveforms indicating each of the multiple movements contained in the movement information.
[0006] According to one aspect of the present disclosure, there is provided an information processing device including at least one processor that processes movement information indicating multiple repetitive movements of an organism, wherein the at least one processor is configured to acquire the movement information, divide the acquired movement information into multiple first waveform information indicating waveforms of each of the multiple repetitive movements of the organism, cluster the multiple first waveform information into multiple clusters each including multiple first waveform information, select one or more second waveform information from the first waveform information included in each of the clusters that indicates waveforms representative of waveforms indicated by the first waveform information included in each of the clusters, and execute processing to list the multiple second waveform information indicating waveforms representative of each of the clusters on at least one screen.
[0007] Furthermore, according to one aspect of the present disclosure, there is provided a method for processing movement information indicating multiple repetitive movements of an organism, the method being carried out by the at least one processor executing a predetermined instruction command, the method including the steps of acquiring the movement information, dividing the acquired movement information into a plurality of first waveform information indicating waveforms of each of the multiple repetitive movements of the organism, clustering the plurality of first waveform information into a plurality of clusters, each cluster including a plurality of the first waveform information, selecting, from the first waveform information included in each of the clusters, one or more second waveform information indicating waveforms representative of the waveforms indicated by the first waveform information included in each of the clusters, and displaying, in a list on at least one screen, the plurality of second waveform information indicating waveforms representative of each of the clusters.
[0008] Furthermore, according to one aspect of the present disclosure, there is provided a program that causes an information processing device, which includes at least one processor and processes movement information indicating multiple repeated movements of an organism, to function as a processor configured to execute a process of acquiring the movement information, dividing the acquired movement information into multiple first waveform information indicating waveforms of each of the multiple repeated movements of the organism, clustering the multiple first waveform information into multiple clusters, each cluster including multiple first waveform information, selecting one or more second waveform information indicating waveforms representative of the waveforms indicated by the first waveform information included in each of the clusters, and displaying a list of the multiple second waveform information indicating waveforms representative of each of the clusters on at least one screen.
[0009] According to the present invention, it is possible to process movement information indicating multiple repeated movements of a living organism and analyze waveforms indicating each of the multiple movements included in the movement information.
[0010] It should be noted that the above-described effects are merely exemplary for the sake of convenience of explanation and are not limiting. In addition to or instead of the above-described effects, any effect described in this disclosure or any effect obvious to a person skilled in the art may be achieved.
[0011] 6B 。 FIG. 6C is a diagram illustrating a configuration of an information processing system 1 according to the present disclosure. FIG. 6D is a diagram illustrating a hardware configuration of an electrocardiograph 2 shown in FIG. 1. FIG. 6E is a diagram illustrating a hardware configuration of an information processing device 3 and a terminal device 12 shown in FIG. 1. FIG. 6F is a diagram illustrating an example of a waveform of an electrocardiogram signal indicated by electrocardiogram information. FIG. 6G is a diagram illustrating a waveform indicated by waveform information. FIG. 6H is a diagram illustrating an example of an operation in which a user specifies the end point of a T wave as a measurement target portion of a waveform. FIG. 6I is a diagram illustrating an information table used for information processing in the information processing device 3. FIG. 6I is a diagram illustrating a program module included in an information processing program 34 that performs information processing in the information processing device 3. FIG. 6I is a diagram illustrating an example of a template screen 4. FIG. 6I is a diagram illustrating an example of a template screen 4 including a template 40 characterized by a gradual change in the T wave. FIG. 6I is a diagram illustrating an enlarged view of the T wave of the i-th template 40, which has a particularly gradual change, out of the n templates 40 shown in FIG. 6B (1≦i≦n). FIG. 6I is a diagram illustrating a communication sequence performed between the electrocardiograph 2, the information processing device 3, and the terminal device 12 included in the information processing system 1 shown in FIG. 1. FIG. 6I is a diagram illustrating an example of a template screen 4 including a template 40 annotated with a Δ wave. 9A and 9B are diagrams illustrating an example of a trend graph in which the duration of QTc intervals measured over a long period of time is arranged over time. FIG. 9B is a diagram illustrating a normal heartbeat waveform when no intraventricular conduction disorder occurs in the heart. FIG. 9C is a diagram illustrating an example of an abnormal heartbeat waveform when an intraventricular conduction disorder occurs in the heart. FIG. 9D is a diagram illustrating an example of a trend graph. FIG. 9D is a diagram illustrating an example of a trend graph. FIG. 9D is a diagram illustrating an example of a template screen 4 including a template 40 annotated with "ventricular conduction disorder." FIG. 9E is a diagram illustrating an example of a trend graph. FIG. 9F is a diagram illustrating an example of a template screen 4 including a template 40 annotated with "PVC." FIG. 9G is a diagram illustrating an example of a trend graph. FIG. 9H is a diagram illustrating an example of a template screen 4 including a template 40 annotated with "ST elevation." FIG. 9I is a diagram illustrating an example of a trend graph. FIG. 9I is a diagram illustrating an example of a waveform of the template 40 included in the template 40 shown in FIG. 6A etc. FIG. 9I is a diagram illustrating an example of a heartbeat waveform indicated by waveform information included in cluster information D corresponding to the template 40 selected as shown in FIG. 12A.1 is a diagram illustrating an example of a template screen 4 including a template 40 on which a cardiac waveform of arrhythmia is displayed. FIG. 2 is a diagram illustrating an example of a cardiac waveform of arrhythmia.
[0012] First Embodiment First, a first embodiment of the present disclosure will be described with reference to the accompanying drawings. Note that common components and processes in the drawings are designated by the same reference numerals. For convenience of illustration, the drawings may omit some of the names of components, such as "modules," and may omit lines indicating information transmitted and received between components.
[0013] Furthermore, the information processing device according to the present disclosure can display waveforms of signals corresponding to multiple repetitive movements within the human body, such as electrocardiograms obtained by measuring the human heart, for example, electroencephalograms obtained by measuring the human brain. In other words, this "movement" is not limited to physical movements within the human body, but also includes static movements of the brain and the like.
[0014] Furthermore, the information processing device according to the present disclosure can process signals obtained by measuring various living and non-living things, including not only humans but also mammals, reptiles, amphibians, fish, invertebrates, plants, and natural phenomena, and can process signals indicating their movements. Furthermore, the information processing device according to the present disclosure can process signals obtained by measuring not only a single living thing, but also multiple living things, and can process signals indicating movements within the bodies of these individuals. However, in the following, to make the explanation more specific and easier to understand, a specific example will be given in which the information processing device according to the present disclosure processes an electrocardiogram of a single person (subject) undergoing a health check or treatment for an illness.
[0015] 1. Configuration of Information Processing System First, the configuration of an information processing system 1 according to the present disclosure will be described. Fig. 1 is a diagram illustrating the configuration of the information processing system 1 according to the present disclosure. As shown in Fig. 1, the information processing system 1 has a configuration in which an electrocardiograph 2, an information processing device 3, and a terminal device 12 are connected to each other via a communication network 100 so that they can transmit and receive information to and from each other. However, the electrocardiograph 2 and the information processing device 3, and the information processing device 3 and the terminal device 12 can also be connected via electric wires or cables so that they can transmit information directly without the communication network 100.
[0016] The communication network 100 is a wired communication network or a wireless network capable of data communication, or a combination of a wired communication network and a wireless network. The communication network 100 transmits the electrocardiogram information sent by the electrocardiograph 2 to the information processing device 3. The communication network 100 also transmits information between the information processing device 3 and the terminal device 12 when a user, such as a doctor performing an examination, uses the terminal device 12 to operate the information processing device 3 or to observe and analyze the electrocardiogram information and waveform information processed by the information processing device 3.
[0017] 2. Configuration of Electrocardiograph 2 FIG. 2A is a diagram illustrating an example of the hardware configuration of the electrocardiograph 2 shown in FIG. 1. As shown in FIG. 2A, a communication interface 212 including an electrode interface (electrode IF) 202, a processor 206, a memory 208, and a communication processing circuit 214 is connected via a bus 200 so as to be able to transmit and receive information and data to and from the information processing device 3. The electrocardiograph 2 is, for example, a portable electrocardiograph worn on the subject's body, or a so-called 12-lead electrocardiograph. Using these components, the electrocardiograph 2 measures the subject's heart and transmits the electrocardiogram information in real time to the information processing device 3 via the information processing system 100. Alternatively, the electrocardiograph 2 stores the electrocardiogram information in the memory 208 and transmits the stored electrocardiogram information to the information processing device 3 via the communication network 100 in response to a request from the information processing device 3. Alternatively, the information processing device 3 stores electrocardiogram information obtained over a predetermined period of time, for example, one hour to one day, in memory 208, and outputs the stored electrocardiogram information to the information processing device 3 via a cable (not shown) or an external recording medium 210 that is detachable from the memory 208 and the information processing device 3 and includes a non-volatile memory.
[0018] The electrode interface 202 is connected to n (1≦n) electrodes 204 that are placed on the subject's body surface and detect analog electrical signals (electrocardiogram signals) emitted by the subject's heart. The electrode interface 202 includes an amplifier that amplifies the electrocardiogram signals and an analog-to-digital (A / D) converter that converts the amplified electrocardiogram signals into digital data. The electrode interface 202 amplifies the electrocardiogram signals of the subject and converts them into digital data to generate electrocardiogram information that indicates the electrocardiogram of the subject.
[0019] The processor 206 is composed of at least one processor (CPU) or a combination of at least one CPU and at least one digital signal processor (DSP) specialized for specific processing such as signal processing. The processor 206 functions as a control unit that executes instructions contained in programs stored in the memory 208 and controls other components connected via the bus 200. The programs stored in the memory 208 may include a program including instructions for realizing the functions of an OS (not shown), and an application program including instructions for acquiring an electrocardiogram signal from a subject, generating electrocardiogram information, and transmitting or storing the generated electrocardiogram information.
[0020] The memory 208 is composed of a ROM, a RAM, a nonvolatile memory, a HDD (not shown), an SSD (not shown), etc., and functions as a storage unit. The ROM stores an application program including instructions for using the information provision service according to the present disclosure, and a program including instructions for executing the OS. The RAM is a memory used to write and read data while the processor 206 is executing instructions of the program stored in the ROM. The nonvolatile memory is a memory into which data is written and read by executing a program. For example, if the electrocardiograph 2 is a portable electrocardiograph, electrocardiogram information written to the nonvolatile memory is retained even after the electrocardiograph 2 is turned off. Note that, depending on the specifications of the application program, an external recording medium 210 or the like may be attached to the memory 208, as described above, and the memory 208 may write electrocardiogram information to the external recording medium 210.
[0021] The communication interface 212 functions as a communication unit that transmits, in real time, electrocardiogram information obtained by measurement on the subject to the information processing device 3 connected via the communication network 100, via the communication processing circuit 214 and an antenna (not shown). Depending on the specifications of the application program stored in the memory 208, the communication interface 212 may transmit the electrocardiogram information stored in the memory 208 in response to a request from the information processing device 3, as described above.
[0022] The communication processing circuit 214 performs communication processing for a wideband wireless communication system conforming to the LTE system. Alternatively, the communication processing circuit 214 performs communication processing for a wireless LAN conforming to IEEE 802.11 or a wireless communication system conforming to the Bluetooth (registered trademark) standard. Furthermore, the communication processing circuit 214 can also perform communication processing for wired communication instead of or in addition to wireless communication.
[0023] 3. Configuration of Information Processing Device 3 and Terminal Device 12 Figure 2B is a diagram illustrating an example of the hardware configuration of the information processing device 3 and terminal device 12 shown in Figure 1. The information processing device 3 and terminal device 12 do not need to have all of the components shown in Figure 2B; they may be configured with some components omitted, or other components may be added. Furthermore, the information processing device 3 and terminal device 12 do not need to be configured integrally in a single housing, but may be configured across multiple computers interconnected via the communication network 100. Furthermore, the information processing device 3 and terminal device 12 do not necessarily need to be the same type of device; for example, the information processing device 3 may be a server device, and the terminal device 12 may be a personal computer.
[0024] 2B , the information processing device 3 and the terminal device 12 have a configuration in which an output interface 302, a processor 306, a memory 308, a communication interface 312, and an input interface 316 are connected via a bus 300 so that they can transmit and receive information and data to and from each other. Output devices such as a printer (not shown) and a display 304 are connected to the output interface 302. Input devices such as a mouse 318 and a keyboard 320 are connected to the input interface 316. In other words, the information processing device 3 and the terminal device 12 include components of a general computer.
[0025] The keyboard 320 is, for example, a general keyboard connected to a personal computer or displayed on a smartphone. When the keyboard 320 is a keyboard connected to a personal computer, the keyboard 320 includes hardware key switches used to input alphabets, numbers, and symbols, as well as hardware key switches such as a control key used to control the computer. By simultaneously pressing the control key switch and the key switch for inputting alphabets on the keyboard 320, the user of the information processing system 1 can perform operations on the information processing system 1 using so-called shortcut keys.
[0026] The information processing device 3 processes the subject's electrocardiogram information input from the electrocardiograph 2 using these components. Furthermore, the information processing device 3 can display waveform information, which indicates waveforms corresponding to each of the numerous beats included in the electrocardiogram information, on the display 304 of the information processing device 3 and the terminal device 12, or print it out from a printer for easy observation. In the following description, "beat" and "heartbeat" are essentially synonymous, but "beat" is used for consistency. Using these components, the terminal device 12 functions as an input / output device for the information processing device 3, located remotely from the information processing device 3. In other words, the information processing device 3 transmits and receives information to and from the information processing device 3 so that a user can operate the information processing device 3 from a location remote from the information processing device 3 and observe the electrocardiogram information and waveform information. The information processing device 3 transmits the user's operation details to the information processing device 3, and outputs the electrocardiogram information and waveform information transmitted from the information processing device 3 from an output device, such as the output interface 302, and displays them to the user.
[0027] The output interface 302 is connected to various output devices as described above and functions as an information output unit for outputting information from the connected output devices. As the output interface 302, a known connection format such as a serial port, a parallel port, a USB, or an HDMI (registered trademark) can be adopted depending on the intended use.
[0028] Like the processor 206 of the electrocardiograph 2, the processor 306 is composed of at least one processor (CPU) or a combination of at least one CPU and at least one DSP or GPU specialized for specific processing such as signal processing or image processing. The processor 306 functions as a control unit that executes instructions contained in programs stored in the memory 308 and controls other components connected via the bus 300. In the information processing device 3, the programs stored in the memory 308 may include a program including instructions for implementing OS functions. The programs stored in the memory 308 may also include an application program including instructions for analyzing electrocardiogram information input from the electrocardiograph 2 and displaying waveform information that is included in the electrocardiogram information and indicates waveforms (beat waveforms) corresponding to each of a plurality of beats.
[0029] Furthermore, processor 306 acquires movement information indicating the multiple repeated movements of the organism, divides the acquired movement information into multiple first waveform information indicating the waveforms of each of the multiple repeated movements of the organism, clusters the multiple first waveform information into multiple clusters each including multiple first waveform information, selects one or more second waveform information from the first waveform information included in each cluster, and selects one or more second waveform information indicating waveforms representative of the waveforms indicated by the first waveform information included in each cluster, and performs processing to list the multiple second waveform information indicating waveforms representative of each cluster on at least one screen.
[0030] The memory 308 includes the storage device described above and functions as a storage unit. The memory 308 also stores an application program (described later with reference to FIG. 5 , etc.) for processing the following: acquiring motion information indicating multiple repetitive movements of the living organism; dividing the acquired motion information into multiple first waveform information indicating waveforms of the multiple repetitive movements of the living organism; clustering the multiple first waveform information into multiple clusters, each cluster including multiple first waveform information; selecting, from the first waveform information included in each cluster, one or more second waveform information indicating waveforms representative of the waveforms indicated by the first waveform information included in each cluster; and displaying a list of the multiple second waveform information indicating waveforms representative of each cluster on at least one screen. Similar to the memory 208 of the electrocardiograph 2, the memory 308 can be detachably connected to an external recording medium 210, and the memory 308 can read electrocardiogram information from the external recording medium 210.
[0031] The communication interface 312 performs processing for transmitting and receiving information and data via the communication network 100 in accordance with the wireless communication method described above for the communication interface 212 or a known wired communication method. In the information processing device 3, the communication processing circuit 314, for example, performs processing for receiving electrocardiogram information from the electrocardiograph 2 via the communication network 100. The communication processing circuit 314 also performs processing for transmitting electrocardiogram information and waveform information obtained from the electrocardiogram information to the terminal device 12 via the communication network 100. The communication processing circuit 314 also performs processing for receiving information indicating operations performed by a user on the keyboard 320 or the like from the terminal device 12. In the terminal device 12, the communication processing circuit 314 also performs processing for receiving electrocardiogram information and waveform information from the information processing device 3 via the communication network 100. The communication processing circuit 314 also performs processing for receiving information indicating operations performed by a user on an input device of the terminal device 12 via the communication network 100.
[0032] The input interface 316 accepts operations performed by the user on the input devices described above, and outputs information indicating the accepted operations to the processor 306 via the bus 300.
[0033] 4. Description of Electrocardiogram Information, Waveform Information, etc. Electrocardiogram information and waveform information (first waveform information) will be described below. FIG. 3A is a diagram illustrating an example of the waveform of an electrocardiogram signal indicated by the electrocardiogram information. FIG. 3B is a diagram illustrating an example of a waveform indicated by the waveform information. The waveform indicated by the electrocardiogram information and the waveform information is generally widely known, and does not need to be described with reference to FIGS. 3A and 3B. Electrocardiogram information indicates the electrocardiogram waveform of a subject. The information processing device 3 processes the electrocardiogram waveform and generates waveform information by dividing the electrocardiogram information into information corresponding to each of multiple heartbeats. The electrocardiogram information is displayed as an electrocardiogram waveform on the display 304 or the like in the information processing device 3 and the terminal device 12. The waveform information is also displayed as a beat waveform on the display 304 or the like in the information processing device 3 and the terminal device 12.
[0034] The electrocardiograph 2 generates electrocardiogram information by converting an analog electrocardiogram signal detected from the subject via the electrodes 204 into a digital format using the electrode interface 202. The electrocardiogram information indicates the waveform of the electrocardiogram signal. The electrocardiogram signal includes a beat signal indicating a series of beat waveforms. The information processing device 3 divides the electrocardiogram information into a plurality of pieces of waveform information corresponding to each of the beat signals included in the electrocardiogram signal.
[0035] As shown in FIG. 3B , a single beat waveform included in an electrocardiogram signal includes waveforms called P waves, Q waves, R waves, S waves, T waves, and U waves. Each of the P waves, Q waves, R waves, S waves, T waves, and U waves included in a single beat waveform can be measured for analyzing electrocardiogram information to diagnose the heart. Additionally, the start and end points, maximum and minimum points, etc., of these waveforms included in a single beat waveform can also be measured.
[0036] The time lengths between the onsets, terminations, maximums, and minimums of waveforms contained in one beat waveform may also be measured. The time lengths between the onsets, terminations, maximums, and minimums of waveform components contained in adjacent beat waveforms may also be measured or analyzed. Hereinafter, waveforms and their onsets that are contained in an electrocardiogram signal and that may be measured may also be referred to as measurement portions. Furthermore, the difference in voltage between the maximum and minimum of an ST wave (ST peak value), the difference in voltage between the maximum and minimum of a T wave (T peak value) relative to the difference in voltage between the maximum and minimum of a QRS complex from the onset of a Q wave to the termination of an S wave (peak value of the QRS width), and other such parts may also be measured.
[0037] The time length between the measurement portions included in each pulsatile waveform is, for example, the time length of the QT interval from the start of the Q wave to the end of the T wave, the time length of the QRS width, the time length of the PR interval, the time length from the start of the Q wave to the maximum of the T wave, and the time length from the maximum of the T wave to the end of the T wave. Furthermore, the time length between the measurement portions included in adjacent pulsatile waveforms is, for example, the RR interval, which indicates the time length between the peaks of the R waves included in two adjacent waveforms. Note that in an electrocardiogram waveform, the period from just before the start of a P wave to the time corresponding to the RR interval constitutes one pulsatile waveform. In the explanations of the present disclosure, the multiple pulsations of the heart are described as "repeated" rather than "periodic" because the time corresponding to the RR interval changes depending on the subject's physical condition and exercise state.
[0038] A user can identify any of the points included in the waveform represented by the waveform signal. FIG. 3C illustrates an example of a user identifying the end point of the T wave as the measurement target portion of the waveform. For example, as shown in FIG. 3C , two users can independently observe the waveform represented by the displayed waveform information and identify the end point of the T wave in the displayed waveform. In such a case, if each user assumes a different baseline for the waveform information, the end points of the T wave identified by each user may not match, as indicated by arrows a and b in FIG. 3C . When the identified measurement target portions do not match between multiple users, variations occur in the intervals between measurement target portions, such as the QT interval, obtained by analyzing the waveform information. According to the present disclosure, variations in the identification of measurement target portions and the intervals between measurement target portions can be eliminated or reduced.
[0039] When the electrocardiograph 2 is portable and worn by a subject to generate electrocardiogram information over a long period of time, the electrocardiograph 2 measures electrocardiogram information during the subject's daily life. In such cases, the baseline, which is important in analyzing the electrocardiogram waveform indicated by the electrocardiogram information, changes depending on the subject's movements. Therefore, analyzing the electrocardiogram information assuming a horizontal baseline can result in significant errors in the analysis results. Furthermore, assuming a common baseline for multiple pieces of electrocardiogram information obtained by measuring multiple users can result in significant errors in the analysis results of the electrocardiogram information. According to the present disclosure, when analyzing such electrocardiogram information, errors in the analysis results of the electrocardiogram information due to the assumption of a baseline can be eliminated or reduced. In other words, the present disclosure is not only effective for analyzing electrocardiogram information obtained by short-term measurements on the order of seconds, but also demonstrates significant effectiveness in analyzing electrocardiogram information obtained by long-term measurements, such as several hours to a day.
[0040] 5. Information Regarding Information Processing in Information Processing Device 3 Information regarding information processing in the information processing device 3 will now be described. FIG. 4 is a diagram showing an information table used for information processing in the information processing device 3. This information table is stored in the memory 308 by the processor 306. As shown in FIG. 4, the information table regarding information processing in the information processing device 3 includes multiple subject identification information IDs, each uniquely identifying one of multiple subjects or one of multiple measurements performed on a subject. The information table further includes electrocardiogram information A, waveform information B, time information C, cluster information D, template screen information E, representative waveform information F, measurement target portion information G, analysis result information H, analysis result information I, diagnosis information J, and corrected information K, each associated with one of the multiple subject identification information IDs. These pieces of information associated with the same subject identification information ID correspond to one another. The information table may be created as a single file, or may be generated by dividing it into multiple files.
[0041] The electrocardiogram information A is electrocardiogram information obtained from a subject identified by the associated subject identification information ID or during measurement by the electrocardiograph 2. The waveform information B is waveform information generated from the subject identified by the associated subject identification information ID or from the electrocardiogram information during measurement.
[0042] Time information C is time information indicating the time at which each of the multiple waveform information included in the corresponding waveform information B was generated. In other words, time information C is a timestamp for each of the multiple waveform information included in the corresponding waveform information B. Cluster information D is waveform information included in n (1≦n) clusters obtained by clustering the corresponding waveform information B using time-series data clustering. Template screen information E is screen information indicating the format of a template screen used to display n cardiac waveforms representing each of the n clusters included in the corresponding cluster information D as n templates. Representative waveform information F is n pieces of waveform information indicating each of the n waveforms displayed as templates in the corresponding template screen information E. Note that the representative waveform information F does not have to be waveform information itself indicating one cardiac waveform representing each of the n clusters, but may be, for example, a flag corresponding to each of the multiple waveform information B as described above. For example, when the representative waveform information F is a flag for waveform information B, waveform information corresponding to a flag with a value of 1 is representative waveform information, and waveform information corresponding to a flag with a value of 0 is not representative waveform information. It should be noted that a plurality of pieces of representative waveform information F may be selected from one cluster information D.
[0043] The measurement target portion information G is information indicating the measurement target portion identified in each of the n cardiac waveforms displayed as templates in response to a user operation. The user's operation for each of the n templates is to select one of the n templates and identify the measurement target portion in the selected template. Specifically, the operation of identifying the measurement target portion is, for example, an operation in which the user designates the starting point of the Q wave and the ending point of the T wave in template 40 as points on a coordinate system with the voltage value of the cardiac waveform on the vertical axis and time on the horizontal axis. The annotation information H is information indicating annotations that a user, such as a doctor, can add to each of the n templates. The annotations are, for example, information that, when a user viewing a template recognizes the presence of a delta (δ) wave or a premature extrasystole (PVC), can add a note to the template and display it.
[0044] The analysis result information I is information indicating the results of an analysis process performed on multiple waveform information pieces included in the corresponding waveform information B. The analysis process on the multiple waveform information pieces included in the waveform information B is, for example, a process of displaying the time lengths between the measurement portions of each of the multiple waveform information pieces included in the corresponding waveform information B in a chronologically-series graph format, with the vertical axis representing time and the horizontal axis representing time. The diagnostic information J is information indicating a diagnosis made by a user, such as a doctor, for the heart of a subject identified by the associated subject identification information ID. If the user does not enter this information, the annotation information H and the diagnostic information J will be displayed as a null value (φ), indicating this. In the following description, information such as waveform information will be labeled as waveform information B, as necessary. The corrected information K is information indicating the QT interval duration obtained by correcting the QT interval duration using an individual correction method or the like to eliminate the influence of changes in the heart rate (heart rate), as described in the second embodiment.
[0045] 6. Information Processing Program 34 Executed in Information Processing Device 3 The following describes the information processing program 34 that performs information processing in the information processing device 3. Fig. 5 is a diagram showing program modules included in the information processing program 34 that performs information processing in the information processing device 3. As shown in Fig. 5, the information processing program 34 includes an electrocardiogram information acquisition module 340, a waveform information generation module 342, a clustering processing module 346, a representative waveform selection module 348, a template screen generation module 350, a target portion identification module 352, an annotation generation module 354, and a waveform information analysis module 356.
[0046] These components of the information processing program 34 contain instructions necessary to realize their respective functions and are stored in the memory 308. The instructions contained in each component are read and executed by the processor 306. The information processing program 34, using these components, acquires electrocardiogram information A from the electrocardiograph 2 for multiple subjects or multiple measurements, generates n templates (described below with reference to FIG. 6A , etc.), and displays them on the template screen. Preferably, the n templates are created for each of the multiple subjects. Waveform information contained in electrocardiogram information obtained by measurement of a single subject tends to generate only a few clusters. On the other hand, waveform information contained in electrocardiogram information obtained by measurement of multiple subjects tends to generate many clusters. In other words, because clustering processing can be effectively performed on waveform information obtained by measurement of a single subject, only a small number of clusters of waveform information are generated, and therefore a smaller number of templates are required. Obviously, the fewer the number of templates, the less work the user has to do to identify the measurement target portion within the templates.
[0047] Furthermore, the information processing program 34 identifies the measurement target portion of each of the n displayed templates in response to a user operation and attaches an image such as a cursor to the identified measurement target portion. The information processing program 34 also accepts annotations input for the templates in response to a user operation and displays the annotations attached to the templates. Furthermore, the information processing device 3 performs an analysis process on the waveform information B in response to a user operation and displays the results of the analysis process on the display 304 or the like.
[0048] Each component of the information processing program 34 will be described below. The electrocardiogram information acquisition module 340 acquires all of the electrocardiogram information A required for processing by the information processing program 34 from the electrocardiograph 2 at once via the communication interface 312 and stores it in the memory 308. Hereinafter, a specific example will be given in which the electrocardiogram information A required by the information processing program 34 is electrocardiogram information obtained by measuring one subject (one individual) over a one-day period. Alternatively, the electrocardiogram information acquisition module 340 acquires the electrocardiogram information obtained by measuring the subject from the electrocardiograph 2 in real time or at predetermined intervals, and stores it in the memory 308. When the information processing program 34 has completed acquiring all of the electrocardiogram information required for processing by the information processing program 34, it designates all of the acquired electrocardiogram information stored in the memory 308 as electrocardiogram information A. The electrocardiogram information A stored in the memory 308 is read out by the waveform information generation module 342.
[0049] In response to user operations received via the input interface 316 or the communication interface 312, the waveform information generation module 342 detects the portion of each QRS complex of multiple cardiac waveforms included in the electrocardiogram waveform that exhibits the maximum voltage value (the peak of the QRS complex). The waveform information generation module 342 determines the time length between the peaks of two adjacent QRS complexes as the time length of one cardiac waveform. Furthermore, the waveform information generation module 342 determines the portion of the two Q waveforms in the adjacent cardiac waveforms that is a predetermined time length before the starting point of the preceding Q waveform as the starting point of the preceding cardiac waveform of the two cardiac waveforms. Furthermore, the waveform information generation module 342 determines the portion of the preceding cardiac waveform that is a time length between the peaks of the two adjacent QRS complexes after the starting point of the preceding cardiac waveform as the ending point of the preceding cardiac waveform. In this way, the waveform information generating module 342 repeats the process of extracting one heartbeat signal from the electrocardiogram signal throughout the entire electrocardiogram signal, dividing the electrocardiogram signal into a plurality of consecutive heartbeat signals. Furthermore, the waveform information generating module 342 extracts waveform information corresponding to each of the plurality of heartbeat signals obtained by this division from the electrocardiogram information. Through the processing described above, the waveform information generating module 342 divides the electrocardiogram information into a plurality of pieces of waveform information. The waveform information generating module 342 stores the plurality of pieces of waveform information B generated as described above in the memory 308. The waveform information B stored in the memory 308 is read out by the clustering processing module 346.
[0050] The clustering processing module 346 performs clustering processing on the waveform information B using a time-series data clustering method, such as the K-Shape method, to generate n clusters of multiple waveform information from the waveform information B and store them in the memory 308. The clustering processing module 346 may also perform clustering processing using methods other than the K-Shape method, such as the K-Means method. For example, instead of the K-Means method, the clustering processing module 346 may classify waveform information B of similar shapes into clusters using an algorithm that sets conditions for each shape and performs clustering processing. Furthermore, the clustering processing module 346 may perform supervised or unsupervised machine learning on the cardiac waveform included in the electrocardiogram waveform, and perform clustering processing using an AI-based technique using the results of the machine learning, thereby improving the accuracy of the clustering processing. The cluster information D stored in the memory 308 is read out by the representative waveform selection module 348.
[0051] The representative waveform selection module 348 selects, from each of the n clusters, representative waveform information indicating one cardiac waveform representative of the cardiac waveforms indicated by the plurality of waveform information included in each of these clusters. The representative waveform selection module 348 stores the selected n pieces of representative waveform information in the memory 308 as representative waveform information F. The representative waveform information F stored in the memory 308 is read out by the template screen generation module 350. Note that, in the clustering process, when the clustering processing module 346 determines that the similarity between the representative waveform of a certain cluster and the representative waveforms of the other n-1 clusters is low, it may eliminate this cluster and the waveform information included in this cluster as noise and prevent it from being used in subsequent processing by the information processing program 34. In such a case, the set of all waveform signals included in the n clusters may become a subset of the set of waveform information B.
[0052] The template screen generation module 350 generates template screen information E from the waveform information B and the representative waveform information F. The template screen generation module 350 stores the template screen information E in the memory 308 and outputs it via the output interface 302 or the communication interface 312. The template screen information E output from the template screen generation module 350 is displayed on the display 304 or the like. The template screen information E stored in the memory 308 is read out by the target portion identification module 352.
[0053] FIG. 6A is a diagram illustrating a template screen 4. Note that FIG. 6A illustrates a case where the template screen includes 11 templates (n=11). However, the number n of templates 40 included in each of the multiple template screens 4 does not have to be the same. As shown in FIG. 6A , the template screen 4 includes n templates 40, and each of the n templates 40 displays a heartbeat waveform indicated by representative waveform information F of the waveform information included in the n clusters. The template screen 4 displays the templates 40 in a list. However, the number of template screens 4 is not limited to one and can be two or more. Note that, for example, cursors indicating the onset of the Q wave and the end point of the T wave, and the time length between the onset of the Q wave and the end point of the T wave (QT interval) as measurement target portions are displayed superimposed on the template 40.
[0054] The target portion identification module 352 (FIG. 5) accepts user operations via the communication interface 312 or the input interface 316. In response to the user's operation, the target portion identification module 352 selects one of the templates 40 from the template screen 4, and further identifies the measurement target portion included in the cardiac waveform of the template 40, that is, identifies the coordinate range or coordinate point described above. In this way, by referring to all n templates 40, the user can uniformly identify the measurement target portion for all n clusters.
[0055] The target portion identification module 352 identifies, as the measurement target portion, a coordinate range or coordinate point containing a waveform that matches a waveform near the coordinates of the measurement target portion identified for the representative waveform in each of the n clusters in the cardiac waveform represented by waveform information other than the representative waveform information F of each of the n clusters, or a coordinate range or coordinate point containing a waveform that shows high similarity. The target portion identification module 352 uses a dynamic time warping method to identify the measurement target portion. The dynamic time warping method calculates the distances between all points included in the waveform represented by the representative waveform information F of one cluster in a time series manner and all points included in another waveform information in the same cluster in a time series manner, and finds a time warping path as a set of points that minimizes the distances between these points. Based on the coordinates of the Q wave onset point and the T wave end point identified in the template 40 and the time warping path, the coordinates of the Q wave onset point and the T wave end point in the cardiac waveform representing a waveform other than the representative waveform information F are identified. The target portion identification module 352 stores information indicating all measurement target portions, including the coordinates of the starting point of the Q wave and the coordinates of the ending point of the T wave, included in each of the n clusters identified in this manner, in the memory 308 as measurement target portion information G. The measurement target portion information G stored in the memory 308 is read out by the waveform information analysis module 356.
[0056] The annotation generation module 354 receives, via the output interface 302 or the input interface 316, information indicating annotations input by the user for the cardiac waveforms indicated by the representative waveform information F displayed in each of the n templates 40 in the template screen 4. The annotation generation module 354 stores the information indicating the received annotations in the memory 308 as annotation information H. The annotation information H stored in the memory 308 is read out by the template screen generation module 350. Furthermore, the annotation generation module 354 uses the annotation information assigned to each representative waveform displayed in the template 40 to accept annotations for cardiac waveforms other than those indicated by the representative waveform information F. For example, when the annotation generation module 354 receives annotation information indicating arrhythmia for the representative waveform information F, it also assigns annotation information indicating arrhythmia to other waveform information included in the same cluster. Also, for example, when the annotation generation module 354 receives annotation information indicating that the representative waveform information F is not subject to analysis of the QT interval, it also adds annotation information indicating that the representative waveform information F is not subject to analysis of the QT interval to other waveform information included in the same cluster.
[0057] The template screen generation module 350 displays the read annotation information superimposed on n templates 40. If the user observes the cardiac waveform displayed in the template 40 as a representative waveform and makes a diagnosis of the subject's heart, the user can input the results of the diagnosis via the input device and communication interface 312 or the input interface 316. The annotation generation module 354 stores diagnostic information indicating the results of the diagnosis as diagnostic information J in the memory 308. The diagnostic information J stored in the memory 308 is read by the template screen generation module 350 and displayed superimposed on the template 40.
[0058] The waveform information analysis module 356 performs an analysis process on the waveform information B and the measurement target portion information G. The waveform information analysis module 356, for example, calculates the time lengths between measurement target portions in all waveform signals identified for the template 40 in the template screen 4 and adjacent waveform signals, such as QT intervals and RR intervals, and stores these in the memory 308 as analysis result information I. The waveform information analysis module 356 also arranges, for example, the values of the waveform information B in a chronological graph format, stores this in the memory 308 as analysis result information I, and outputs it via the communication interface 312 or the input interface 316. This analysis process of the measurement target portions of the waveform information B allows the user to know at what time an abnormality, such as a disruption, occurred in the subject's heart rate waveform.
[0059] 7. Template Screen 4 and Identification and Storage of Measurement Targets Using the Template Screen 4 Below, we will explain operations by a user to identify, for example, the starting point of a Q wave and the ending point of a T wave as measurement target portions of waveform information included in electrocardiogram information, and store them in the information processing device 3, in response to a user's operation on the information processing device 3. First, the information processing device 3 starts executing the information processing program 34 ( FIG. 5 ) in response to a user's operation.
[0060] First, the information processing device 3 accepts all electrocardiogram information obtained by the electrocardiograph 2 performing measurements on one subject over the course of one day, stores it as electrocardiogram information A, and generates template screen information E and representative waveform information F. Furthermore, the information processing device 3 uses the generated template screen information E and representative waveform information F to generate the template screen 4 shown in FIG.
[0061] The user performs an operation to select one of the n templates 40 included in the template screen 4. The information processing device 3 highlights the template 40 selected by this operation, for example, by changing the color. Furthermore, the user performs an operation to specify the measurement target portion of the cardiac waveform displayed in the selected template 40. This operation may be performed using the mouse 318 or an operation using directional keys and shortcut keys on the keyboard 320. In response to this operation, the information processing device 3 displays, for example, an image of a cursor as a mark superimposed on the starting point of the Q waveform and the ending point of the T waveform. Note that the mark indicating the starting point of the Q waveform and the ending point of the T waveform specified by the user is not limited to a cursor and may be, for example, a circle or a flag.
[0062] When the user performs this operation on all n templates 40 included in the template screen 4, the information processing device 3 identifies the starting point of the Q waveform and the ending point of the T waveform for each of the n templates 40, and displays a cursor superimposed on each of these identified measurement target portions. Furthermore, when the user performs an operation to store the measurement target portions identified in all n templates 40, the information processing device 3 stores information indicating the identified measurement target portions in response to this operation.
[0063] 6B is a diagram illustrating an example of a template screen 4 including a template 40 characterized by a gradual change in the T wave. FIG. 6C is a diagram illustrating an enlarged view of the T wave of the i-th template 40 (1≦i≦n) among the n templates 40 shown in FIG. 6B , which has a particularly gradual change. The user performs operations on the information processing device 3 to identify the onset of the Q wave and the termination of the T wave in each of the n templates 40 included in the template screen 4 shown in FIG. 6B and to store these measurement target portions. In response to these user operations, the information processing device 3 identifies and stores the onset of the Q wave and the termination of the T wave in each of the n templates 40.
[0064] However, as shown in FIG. 6C , the T wave of the i-th template 40 may change gradually, making it difficult for the user to identify the end point of the T wave. In such a case, the user may be unsure of which of the positions indicated by arrows a and b the T wave end point of the template 40 should be. In this case, for example, the user may postpone identifying the end point of the T wave of the i-th template 40, identify the end points of the T waves of templates 40 other than the i-th template 40, and then identify the end points of the T waves of templates 40 other than the i-th template 40 again. In this way, the user can identify the end point of the T wave of the i-th template 40 while referring to the end points of the T waves of templates 40 other than the i-th template 40 that have already been identified. As a result of such identification, the user can identify, among the positions indicated by arrows a and b included in the i-th template 40, for example, the position indicated by arrow a, which indicates a position close to the end point of the T wave included in another template 40, as the end point of the T wave included in the i-th template 40, and store this in the information processing device 3.
[0065] The template 40 for identifying the onset of the Q wave and the termination of the T wave for the i-th template 40 may be displayed separately from the template screen 4 displaying a list of n templates 40. This allows the template 40 displaying the cardiac waveform for which the onset of the Q wave and the termination of the T wave are to be identified to be enlarged, enabling the onset of the Q wave and the termination of the T wave to be accurately identified. Furthermore, such a template 40 may display cardiac waveforms before and after the displayed cardiac waveform, as well as cardiac waveforms spanning approximately 10 to 30 seconds including the displayed cardiac waveform. This allows the user to easily confirm the position of the baseline connecting the onset of the P wave of one cardiac waveform to the onset of the P wave of the next cardiac waveform, and its long-term fluctuations. This allows the onset of the Q wave and the termination of the T wave to be identified more accurately.
[0066] 8. Communication Sequence in Information Processing System 1 The following describes a communication sequence between the electrocardiograph 2, information processing device 3, and terminal device 12 of the information processing system 1 in the first embodiment. Figure 7 is a diagram showing a communication sequence performed between the electrocardiograph 2, information processing device 3, and terminal device 12 included in the information processing system 1 shown in Figure 1. Note that, although the information processing device 3 in the information processing system 1 can operate without the terminal device 12, the following description will be limited to a specific example in which a user operates the information processing device 3 using the terminal device 12 to perform operations on the template screen 4. As described above, the necessary electrocardiogram information can be transmitted from the electrocardiograph 2 to the information processing device 3 in real time or in multiple batches. However, the following description will be limited to a specific example in which all of the necessary electrocardiogram information is transmitted from the electrocardiograph 2 at once.
[0067] When the user operates the terminal device 12 to perform an operation to instruct the information processing device 3 to start the information processing program 34 and begin processing, the OS read from the memory 308 and executed by the processor 306 of the terminal device 12 accepts these operations via the input interface 316. Furthermore, the OS of the terminal device 12 transmits information indicating the accepted operations to the information processing device 3 via the communication interface 312 and the communication network 100 (S100).
[0068] The OS, which has been read from memory 308 and is being executed by processor 306 of information processing device 3, receives information indicating an operation to instruct startup of information processing program 34 and start processing via communication network 100 and communication interface 312. Furthermore, the OS of information processing device 3 starts information processing program 34 and starts processing. In other words, the OS of information processing device 3 causes processor 306 to read each component of information processing program 34 ( FIG. 5 ) from memory 308 and start execution (S102).
[0069] The electrocardiogram information acquisition module 340 requests the electrocardiograph 2 to transmit electrocardiogram information via the communication interface 312 and the communication network 100 (S104). The OS, which is read from the memory 208 and executed by the processor 206 of the electrocardiograph 2, receives this request via the communication network 100 and the communication interface 212. Furthermore, the OS of the electrocardiograph 2 reads all electrocardiogram information required for processing from the memory 308 and transmits it to the information processing device 3 via the communication interface 312 and the communication network 100 (S106). The electrocardiogram information acquisition module 340 receives the electrocardiogram information via the communication network 100 and the communication interface 312 and stores it in the memory 308 as electrocardiogram information A (S108).
[0070] In response to a user's operation, the waveform information generation module 342 processes the electrocardiogram information A to generate waveform information B and stores it in the memory 308 (S110). The clustering processing module 346 performs clustering on the waveform information B using a shape-based clustering method such as the K-Shape method, generates n clusters from the waveform information B, each cluster including multiple pieces of waveform information, and stores these in the memory 308 (S112). The clustering processing module 346 selects a cardiac waveform representative of each of the n clusters from the cardiac waveforms indicated by the waveform information included in each of the n clusters, and designates the waveform information indicating the selected cardiac waveform as representative waveform information for that cluster. The clustering processing module 346 stores the n pieces of representative waveform information selected for each of the n clusters as representative waveform information F in the memory 308 (S114).
[0071] The template screen generation module 350 generates a template screen 4 (such as FIG. 6A ) from the template screen information E and the representative waveform information F, and stores the information of the template screen 4 in the memory 308 (S116). Furthermore, the template screen generation module 350 transmits the information of the template screen 4 to the terminal device 12 via the communication interface 312 and the communication network 100 (S118). The OS of the terminal device 12 receives the information of the template screen 4 from the information processing device 3 via the communication network 100 and the communication interface 312. Furthermore, the OS of the terminal device 12 displays the received information of the template screen 4 as the template screen 4 on the display 304 or the like via the output interface 302 (S120).
[0072] When the user performs an operation to identify a measurement target portion for each of the n templates 40 included in the template screen 4 displayed on the terminal device 12, the OS of the terminal device 12 accepts the user's operation via the input interface 316 (S122). The OS of the terminal device 12 transmits information indicating the accepted user's operation, that is, information indicating the operation to identify a measurement target portion for the template 40 included in the template screen 4, to the information processing device 3 via the communication interface 312 and the communication network 100. The OS of the information processing device 3 receives the information indicating the user's operation transmitted from the terminal device 12 via the communication network 100 and the communication interface 312 (S124).
[0073] The target portion identification module 352 identifies the measurement target portion for each of the n templates 40 in the template screen 4 based on information indicating the user's operation transmitted from the terminal device 12. Furthermore, the target portion identification module 352 generates measurement target portion information G indicating the identified measurement target portion and stores it in the memory 308. Furthermore, once the measurement target portion has been identified for the waveform indicated by the representative waveform information F, the target portion identification module 352 identifies the measurement target portion for all other waveform information B included in the cluster represented by the representative waveform information F for which the measurement target portion has been identified (S126). The user performs an operation to input annotations for one or more templates 40 included in the template screen 4 displayed on the terminal device 12. Furthermore, the user performs an operation to cause the information processing device 3 to perform an analysis process using the cluster information D and the measurement target portion information G. The OS of the terminal device 12 accepts these operations via the input interface 316 (S128).
[0074] The OS of the terminal device 12 transmits information indicating an operation for inputting the accepted annotation and an operation for performing analysis processing to the information processing device 3 via the communication interface 312 and the communication network 100. The OS of the information processing device 3 receives information indicating these operations from the terminal device 12 via the communication network 100 and the communication interface 312 (S130). Note that the template screen generation module 350 of the information processing device 3 receives information indicating these operations from the terminal device 12 via the communication network 100 and the communication interface 312. Each time the template screen generation module 350 receives these operations, it generates a template screen 4 including the annotation indicated by the received operation, transmits it to the information processing device 3, and displays it on the display 304 via the output interface 302 (not shown).
[0075] The target portion identification module 352 identifies the measurement target portion of each of the n templates 40 based on the information received from the terminal device 12, generates measurement target portion information G, and stores it in the memory 308. The waveform information analysis module 356 processes the cluster information D based on the information received from the terminal device 12, performs analysis processing such as determining the duration of the QT interval in the cardiac waveform indicated by the waveform information contained in each of the n clusters, and generates analysis result information I and stores it in the memory 308 (S132). The target portion identification module 352 and the waveform information analysis module 356 transmit the generated information to the terminal device 12 via the communication interface 312 and the communication network 100. The OS of the terminal device 12 receives this information from the information processing device 3 via the communication network 100 and the communication interface 312 (S134). The OS of the terminal device 12 displays a template screen 4 including a template 40 with a cursor image superimposed on the measurement target portion indicated by the received information. The OS of the terminal device 12 also displays the template screen 4 including the template 40 on which the received analysis results such as the duration of the QT interval are superimposed (S136).
[0076] [Second Embodiment] Next, a second embodiment of the present disclosure will be described with reference to the accompanying drawings. Waveform changes in an electrocardiogram that indicate paroxysmal or sudden beats are unlikely to appear in waveforms included in measurements of about 5 to 10 seconds using a 12-lead electrocardiograph, and are therefore very difficult to detect in such short-term electrocardiogram measurements. On the other hand, if measurements are performed using a 12-lead electrocardiograph for, for example, about 24 hours and all waveforms of beats included in the resulting electrocardiogram are observed, waveform changes that indicate paroxysmal or sudden beats can be detected.
[0077] However, since a 24-hour electrocardiogram includes approximately 100,000 heartbeats, it is extremely difficult for a user to observe all of the waveforms indicating heartbeats included in an electrocardiogram obtained by measurement over such a long period of time. The second embodiment of the present disclosure is configured to automatically detect and visualize waveform changes indicating paroxysmal or sudden heartbeats from waveform information B indicating heartbeats included in electrocardiogram information A obtained by measurement over a long period of time.
[0078] In addition, paroxysmal or sudden changes in the measurement portion itself in the pulsation waveform, changes in the time length between measurement portions, and changes in the shape indicating the pulsation suggest conditions such as WPW syndrome, Brugada syndrome, intraventricular conduction disorders, premature contractions, and myocardial infarction. The second embodiment of the present disclosure is configured to display the template 40 with annotations indicating changes in the pulsation waveform. The second embodiment of the present disclosure is also configured to perform analysis processing, such as arranging the QT interval duration indicated by multiple pieces of analysis result information I, and one or more pieces of information on the cluster to which the pulsation belongs and annotation information (hereinafter referred to as "QT interval duration, etc."), over time to generate a trend graph. The trend graph allows the time at which a change in the pulsation waveform occurred to be visualized.
[0079] A correction screen for the user to correct annotations may be displayed over the template 40. This correction screen may display heartbeat waveforms before and after the heartbeat waveform displayed in the template 40, or a heartbeat waveform of approximately 10 to 30 seconds including the heartbeat waveform. Displaying such heartbeat waveforms allows the user to more accurately determine the presence or absence of WPW syndrome, Brugada syndrome, intraventricular conduction disorder, premature ventricular contractions, myocardial infarction, and the like. The user can also confirm annotations such as WPW syndrome, Brugada syndrome, intraventricular conduction disorder, premature ventricular contractions, and myocardial infarction, which are assigned to each cluster, for each heartbeat waveform. Furthermore, the user can confirm changes in the heartbeat waveform that occur spontaneously and analyze the presence or absence of disease.
[0080] The processes for visualizing delta waves and excluding clusters described below are realized by modifying the processes of the information processing program 34 ( FIG. 5 ) as follows. That is, the clustering processing module 346 is enabled to perform a process of deleting one or more clusters included in the cluster information D in response to a user operation. The clustering processing module 346 is also enabled to control the clustering processing module 346 and the representative waveform selection module 348 to generate clusters and generate the cluster information D. The waveform information analysis module 356 is enabled to perform a process of generating a trend graph. The template screen generation module 350 is enabled to perform a process of surrounding a peculiar portion that appears in the trend graph with a dotted line or the like. The communication sequence ( FIG. 7 ) described in the first embodiment of the present disclosure is also modified so that the information processing device 3 transmits a trend graph or the like to the terminal device 12, and the information processing device 3 receives information indicating an operation to specify a peculiar portion from the terminal device 12.
[0081] 1. Visualization of Delta Waves First, a process for visualizing that the electrocardiogram waveform indicated by electrocardiogram information A and waveform information B contains delta waves (sometimes referred to as "δ waves"). Delta waves suggest the presence of WPW syndrome in the heart. This process adds an annotation to template 40 in template screen 4 indicating that the heartbeat waveform contains delta waves. Furthermore, a trend graph is generated, visualizing the time when delta waves become included in the heartbeat waveform. FIG. 8A is a diagram illustrating an example of template screen 4 including template 40 annotated with delta waves. FIG. 8B is a diagram illustrating a trend graph showing the time lengths of QTc intervals measured over a long period of time.
[0082] First, the user observes the beat waveforms displayed in n templates 40 included in the template screen 4 displayed on the display 304 of the terminal device 12 and discovers, for example, that the beat waveforms displayed in five of the n templates 40 contain delta waves (n=12). The user performs an operation on the template screen 4 to select five templates 40 displaying beat waveforms containing delta waves using an input device such as the mouse 318 of the terminal device 12. The user then performs an operation to display an annotation reading "delta wave" on the selected template 40. The terminal device 12 transmits information indicating these operations to the information processing device 3 via the communication network 100, and the information processing device 3 receives the information indicating these operations.
[0083] The information processing device 3 is running an information processing program 34 that generated the template screen 4 displayed on the terminal device 12. The annotation generation module 354 of the information processing device 3 ( FIG. 5 ) receives information indicating these operations and generates annotation information H for adding the annotation "Δ waves" to the five selected templates 40. Furthermore, the template screen generation module 350 uses the generated annotation information H to generate information indicating a template screen 4 including the template 40 annotated with "Δ waves," as shown in FIG. 8A , and transmits this information to the terminal device 12. The terminal device 12 receives the information indicating this template screen 4 from the information processing device 3 and updates the template screen 4 displayed on the display 304.
[0084] The waveform information analysis module 356 processes the waveform information B, time information C, cluster information D, and analysis result information I. As a result of this processing, the waveform information analysis module 356 generates a trend graph, as shown in FIG. 8B , with the length of the QTc interval on the vertical axis and time on the horizontal axis. Note that, for example, the length of the QTc interval indicated by the vertical axis of the trend graph is in ms (milliseconds), and the time indicated by the horizontal axis is in minutes. In the trend graph, the length of the QT interval of the cardiac waveform indicated by each waveform signal included in the cluster information D is arranged over time. The waveform information analysis module 356 transmits information indicating the generated trend graph to the terminal device 12. The terminal device 12 receives this information from the information processing device 3 and displays it on the display 304.
[0085] By referring to the trend graph, the user can know the times at which beat waveforms containing Δ waves and with long QT intervals occurred throughout the entire period during which waveform information B was generated, as shown in Fig. 8B. Furthermore, the template screen generation module 350 can be configured to generate a screen in response to a user operation, in which heartbeat waveforms containing Δ waves and with long QT intervals are highlighted by being surrounded by a dotted line in the trend graph.
[0086] As shown in FIGS. 8A and 8B , the QTc interval duration displayed on the trend graph is obtained by correcting the QT interval duration using an individual correction method to eliminate the influence of changes in heart rate, and is stored in the memory 308 as corrected information K. When the QT interval is corrected using the Bazett correction method, the corrected QT interval may be referred to as a QTcB interval. When the QT interval is corrected using the Fridericia correction method, the corrected QT interval may be referred to as a QTcF interval. This also applies to the following trend graphs. The information processing program 34 may be configured to apply one of multiple colors to the QTc interval durations obtained from each of the n clusters on the trend graph. This allows the user to easily determine which of the n clusters' waveform information indicates a heartbeat waveform containing a change or abnormality.
[0087] To generate a trend graph of the QTc interval length, the information processing device 3 performs the above-described correction on the QT interval length obtained from, for example, one day's worth of cluster information D. The information processing device 3 may be configured to correct the QT interval length in real time to generate information indicating the QTc interval length. Alternatively, the information processing device 3 may be configured to store all of the QT interval times for one day and then correct them to generate information indicating the QTc interval length.
[0088] Information indicating the QTc interval corrected in this manner is stored in the memory 308 as corrected information K ( FIG. 4 ). The template 40 and the trend graph to which the template is attached may be displayed in the same template screen 4, or may be displayed in two separate windows on the screen of the display 304. The information processing device 3 may be configured to further display, as an annotation, diagnostic information I indicating a cardiac disease suggested by the inclusion of a "delta wave" in the heartbeat waveform, such as "WPW syndrome." Furthermore, the information processing device 3 may be configured to further analyze the trend graph and automatically determine that the heartbeat waveform includes a delta wave.
[0089] 2. Visualization of Intraventricular Conduction Disorder Next, a process for visualizing an electrocardiogram waveform that indicates that an intraventricular conduction disorder is occurring in the heart will be described. Details of the processes of the terminal device 12 and the information processing device 3 described in the section on visualization of delta waves will be omitted below to avoid duplication. Details of the trend graph, correction of the QT interval duration, and modified examples of the information processing device 3 described in the section on visualization of delta waves will also be omitted below to avoid duplication. FIG. 9A illustrates a heartbeat waveform when no intraventricular conduction disorder is occurring in the heart. FIG. 9B illustrates an abnormal heartbeat waveform when an intraventricular conduction disorder is occurring in the heart. When an intraventricular conduction disorder is occurring in the heart, as can be seen by comparing FIGS. 9A and 9B , the duration of the QRS complex (QRS width) included in the heartbeat waveform becomes longer than a predetermined threshold value when the duration of the RR interval of the heartbeat waveform is constant.
[0090] 9C is a diagram illustrating an example of a template screen 4 including a template 40 annotated with "ventricular conduction disorder." FIG. 9D is a diagram illustrating an example of a trend graph. First, the user observes the beat waveforms displayed in n templates 40 (n=12) included in the template screen 4 displayed on the display 304 of the terminal device 12 as shown in FIG. 9C . As a result of this observation, the user may discover, for example, that the duration of the QRS complex included in the beat waveform displayed in one of the n templates 40 is longer than a predetermined threshold, suggesting that intraventricular conduction disorder is occurring paroxysmal in the heart.
[0091] The user performs an operation on the template screen 4 to select one template 40 that suggests the occurrence of an intracardiac conduction disorder in the heart. Furthermore, the user performs an operation to display an annotation of "ventricular conduction disorder / greater than or equal to QRS width threshold" on the selected template 40. Based on information indicating these operations, the information processing device 3 generates a template screen 4 that includes one template 40 annotated with "ventricular conduction disorder / greater than or equal to QRS width threshold," as shown in FIG. 9C . The terminal device 12 displays the generated template screen 4.
[0092] The information processing device 3 processes the waveform information B, time information C, and the like to generate a trend graph as shown in FIG. 9D . By referencing the trend graph, the user can determine, from the cluster information D corresponding to the selected template 40, the times at which cardiac waveforms suggesting the occurrence of ventricular conduction disorder occurred throughout the entire period during which the waveform information B was generated, as shown in FIG. 9D . The user may further manipulate the trend graph to enclose with a dotted line portions where the QT interval duration is particularly long, identifying these portions as anomalous. By enclosing these portions with a dotted line, anomalous portions where cardiac waveforms suggesting the occurrence of ventricular conduction disorder occurred can be highlighted. Measuring the QT interval duration does not allow the user to determine that the QRS width is long. Although not directly illustrated in FIG. 9D , the description here assumes that the user who added the annotation measured the QRS width and determined that the QRS width duration is equal to or greater than a threshold.
[0093] 3. PVC Visualization Next, a process for visualizing whether an electrocardiogram waveform may indicate a premature contraction will be described. Note that whether a premature contraction is occurring in the subject's heart requires the user to compare a specific heartbeat waveform with the heartbeat waveforms before and after it. Therefore, the visualization process described here is performed solely for the purpose of screening for premature contractions. FIG. 10A is a diagram illustrating an example of a template screen 4 including a template 40 annotated with PVC. FIG. 10B is a diagram illustrating an example of a trend graph. Note that there are multiple types of premature contractions, such as PVCs and PACs. The process described here is performed when there is a possibility that a PVC is occurring in the subject's heart.
[0094] First, as shown in Fig. 10A, the user observes the beat waveforms displayed in n templates 40 (n = 12) included in the template screen 4 displayed on the display 304 of the terminal device 12. As a result of this observation, the user may discover that the beat waveforms displayed in three of the n templates 40 indicate the possibility of the occurrence of a premature contraction.
[0095] The user performs an operation on the template screen 4 to select three templates 40 whose pulse waveforms indicate premature contractions. Furthermore, the user performs an operation to display the annotation "PVC" on the selected templates 40. Based on information indicating these operations, the information processing device 3 generates a template screen 4 including templates 40 annotated with "PVC," as shown in FIG. 10A . The terminal device 12 displays the generated template screen 4.
[0096] The information processing device 3 processes the waveform information B, the time information C, and the like to generate the trend graph shown in FIG. 10B . By referring to the trend graph, the user can know the times at which the heartbeat waveform exhibited a premature contraction throughout the entire period for which the waveform information B was generated. As shown in FIG. 10B , when the heartbeat waveform exhibits a premature contraction, for example, three peculiar portions where the QT interval is particularly long appear on the trend graph. The heartbeat waveforms in these three peculiar portions indicate the possibility of a premature contraction. Furthermore, the user may highlight these three peculiar portions by surrounding them with dotted lines. Furthermore, a flag may be set on the template 40 displaying the heartbeat waveform annotated with PVC.
[0097] 4. Visualization of ST Elevation Next, a process for visualizing that an electrocardiogram waveform indicates ST elevation will be described. Fig. 11A is a diagram illustrating an example of a template screen 4 including a template 40 annotated with "ST elevation." Fig. 11B is a diagram illustrating an example of a trend graph.
[0098] 11A, the user observes the beat waveforms displayed in n templates 40 (n=12) included in the template screen 4 displayed on the display 304 of the terminal device 12. As a result of this observation, the user discovers that, for example, the beat waveforms displayed in six of the n templates 40 exhibit ST elevation.
[0099] The user performs an operation on the template screen 4 to select six templates 40 whose beat waveforms indicate ST elevation. Furthermore, the user performs an operation to display the annotation "ST elevation" on the selected template 40. Based on information indicating these operations, the information processing device 3 generates a template screen 4 including templates 40 annotated with "ST elevation," as shown in FIG. 11A . The terminal device 12 displays the generated template screen 4.
[0100] The information processing device 3 processes the waveform information B, the time information C, and the like to generate a trend graph as shown in FIG. 11B . By referring to the trend graph, the user can identify the times at which the heart rate waveform exhibited ST elevation throughout the entire period during which the waveform information B was generated. As shown in FIG. 11B , when the heart rate waveform exhibits a premature contraction, for example, two peculiar portions a and b, in which the duration of the QT interval is disrupted, appear on the trend graph. As shown in FIG. 11A , the heart rate waveforms in these two peculiar portions exhibit ST elevation, while the heart rate waveforms in the remaining portions are normal and do not include ST elevation. Furthermore, the user may highlight these two peculiar portions by surrounding them with dotted lines. It is envisioned that waveform information included in a cluster annotated with ST elevation on the trend graph may be displayed in a different color or with a flag.
[0101] 5. Deleting and Adding Clusters The following describes the process of deleting one or more of the n templates 40 included in the template screen 4, or adding one or more templates 40 to the template screen 4. Fig. 12A is a diagram illustrating an example of a waveform of a template 40 included in the template 40 shown in Fig. 6A etc. Fig. 12B is a diagram illustrating an example of a cardiac waveform indicated by waveform information included in cluster information D corresponding to the template 40 selected as shown in Fig. 12A.
[0102] For example, consider a case where the template 40 selected by the user from the template screen 4 contains a lot of noise after the end point of the T wave, as shown in Fig. 12A, and there are only three pieces of waveform information corresponding to the template 40, as shown in Fig. 12B. In this way, if the heartbeat waveform displayed in the template 40 is noisy and the cluster information D corresponding to the template 40 contains only a small amount of waveform information, the cluster corresponding to this template 40 is not appropriate for processing in the information processing device 3. This is because it is difficult to identify the measurement target portion of the heartbeat waveform indicated by the waveform information included in such a cluster, and there is a high possibility that it does not contain any abnormalities in the heartbeat waveform.
[0103] Therefore, it is desirable that waveform information of a cluster that corresponds to a noisy template 40 and contains only a small amount of waveform information be deleted from the cluster information D. Conversely, if the template screen 4 contains only a small number of templates 40, it is desirable that waveform information of a new cluster be added to the cluster information D.
[0104] Therefore, as described above, if the clustering processing module 346 of the information processing program 34 can perform processing to delete one or more clusters included in the cluster information D in response to a user operation, the waveform information of the above-mentioned clusters can be deleted from the cluster information D. Furthermore, if the clustering processing module 346 can control the clustering processing module 346 and the representative waveform selection module 348 in response to a user operation to generate clusters, it will be possible to add waveform information of a cluster that corresponds to a template 40 with little noise and includes a lot of waveform information to the cluster information D, instead of waveform information of a cluster that corresponds to a template 40 with a lot of noise and includes only little waveform information.
[0105] 6. Others The visualization of delta waves and other components included in the heartbeat waveform has been described above. However, in addition to the above, the information processing device 3 of the information processing system 1 can also visualize waveforms indicated by various heartbeat waveforms and various diseases suggested by the heartbeat waveforms. Fig. 13A is a diagram illustrating an example of a template screen 4 including a template 40 displaying an arrhythmia heartbeat waveform. Fig. 13B is a diagram illustrating an arrhythmia heartbeat waveform.
[0106] 13A , the user observes n templates 40 included in the template screen 4 and finds a cardiac waveform of arrhythmia (n=12). The user performs an operation to annotate such a template 40 as "arrhythmia / not to be analyzed" on the information processing device 3, and the discovered template 40 can be visualized with the annotation "arrhythmia / not to be analyzed."
[0107] The processes and procedures described herein can be realized not only in the manner explicitly described in the embodiments, but also by computer-executed software, computer hardware, or a combination thereof. Specifically, the processes and procedures described herein can be realized by implementing logic corresponding to the processes in media such as integrated circuits, volatile memory, non-volatile memory, magnetic disks, and optical storage. Furthermore, the processes and procedures described herein can be implemented as computer programs and executed by various computers, including terminal devices and server devices.
[0108] Although processes and procedures described herein are described as being performed by a single device, software, component, or module, such processes or procedures may be performed by multiple devices, multiple software, multiple components, and / or multiple modules. Furthermore, although various information described herein is described as being stored in a single memory or storage unit, such information may be stored in multiple memories within a single device or multiple memories distributed across multiple devices. Furthermore, software and hardware elements described herein may be realized by integrating them into fewer components or by decomposing them into more components.
[0109] REFERENCE SIGNS LIST 1 Information processing system 100 Communication network 12 Terminal device 2 Electrocardiograph 3 Information processing device 200, 300 Bus 202 Electrode interface 204 Electrode 206, 306 Processor 208, 308 Memory 210 External recording medium 212, 312 Communication interface 214, 314 Communication processing circuit 302 Output interface 304 Display 316 Input interface 318 Mouse 320 Keyboard 34 Information processing program 340 Electrocardiogram information acquisition module 342 Waveform information generation module 344 Correction module 346 Clustering processing module 348 Representative waveform selection module 350 Template screen generation module 352 Target portion identification module 354 Annotation generation module 356 Waveform information analysis module
Claims
1. An information processing device including at least one processor for processing movement information indicating multiple repeated movements of an organism, wherein the at least one processor is configured to execute the following processes: acquire the movement information; divide the acquired movement information into multiple first waveform information indicating waveforms of each of the multiple repeated movements of the organism; cluster the multiple first waveform information into multiple clusters, each cluster including multiple pieces of the first waveform information; select one or more second waveform information from the first waveform information included in each of the clusters, the second waveform information indicating a waveform representative of the waveform indicated by the first waveform information included in each of the clusters; and display a list of the multiple second waveform information indicating waveforms representative of each of the clusters on at least one screen.
2. The information processing device of claim 1, wherein each of the first waveform information includes at least one third waveform information used to analyze multiple repetitive movements of the organism, and in response to a user's operation on the third waveform information included in the displayed second waveform information, a mark indicating at least one of the third waveform information and a point included in the third waveform information is displayed superimposed on the waveform indicated by the displayed first waveform information.
3. The information processing device of claim 2, wherein the repeated multiple movements of the living organism are heart beats of the living organism, the movement information is electrocardiogram information indicating the heart beats of the living organism, each of the first waveform information indicates each of the multiple heart beats of the living organism, and the third waveform information indicates at least one of the P waves, Q waves, R waves, S waves, T waves and U waves contained in each of the first waveform information, and at least any of the starting points, ending points and coordinates in the electrocardiogram of the P waves, the Q waves, the R waves, the S waves, the T waves and the U waves.
4. An information processing device as described in claim 2, which identifies points contained in the third waveform information in response to a user's operation on the third waveform information contained in the displayed second waveform information, and identifies points contained in the third waveform information of other second waveform information included in the same cluster based on the identified points contained in the third waveform information.
5. The information processing device according to claim 3, wherein the living organism is a human, and the electrocardiogram information is obtained from the human individual via a portable electrocardiograph or a 12-lead electrocardiograph.
6. The information processing device according to claim 3, wherein the acquisition of the electrocardiogram information is repeated for a predetermined period of time.
7. The information processing device according to claim 6, wherein the predetermined time is between 10 and 30 seconds.
8. An information processing device as described in claim 2, wherein the mark indicating at least one of the third waveform information is a cursor, and information indicating the displayed second waveform information and the third waveform information, and the cursor displayed superimposed on each of the displayed second waveform information and the third waveform information is stored.
9. An information processing device as described in claim 8, further displaying information indicating the time intervals between the multiple cursors superimposed on each of the displayed second waveform information and the displayed third waveform information, respectively, superimposed on each of the displayed second waveform information and the displayed third waveform information.
10. The information processing device according to claim 2, further comprising: a keyboard having a shortcut key; and wherein the user's operation is performed on the shortcut key of the keyboard.
11. An information processing method in an information processing device including at least one processor, which processes movement information indicating multiple repetitive movements of an organism, the information processing method comprising the steps of: acquiring the movement information; dividing the acquired movement information into a plurality of first waveform information indicating waveforms of each of the multiple repetitive movements of the organism; clustering the plurality of first waveform information into a plurality of clusters, each cluster including a plurality of the first waveform information; selecting, from the first waveform information included in each of the clusters, one or more second waveform information indicating a waveform representative of the waveform indicated by the first waveform information included in each of the clusters; and displaying, in a list on at least one screen, a plurality of the second waveform information indicating waveforms representative of each of the clusters.
12. A program that causes an information processing device, including at least one processor, to process movement information indicating multiple repetitive movements of an organism to function as a processor configured to execute the following processes: acquire the movement information; divide the acquired movement information into a plurality of first waveform information indicating the waveforms of each of the multiple repetitive movements of the organism; cluster the plurality of first waveform information into a plurality of clusters, each cluster including a plurality of the first waveform information; select, from the first waveform information included in each of the clusters, one or more second waveform information indicating a waveform representative of the waveform indicated by the first waveform information included in each of the clusters; and display, in a list on at least one screen, the plurality of second waveform information indicating the waveforms representative of each of the clusters.
Citation Information
Patent Citations
Organic signal measuring device
JP1993161612A
Method and apparatus for interactive annotation and measurement of time series data to perform central intensive analysis and review
JP2004261583A
Waveform display apparatus, waveform display method, program, and recording medium
JP2018110725A
Electrocardiographic analysis apparatus and control method of the same
JP2023068219A
Cited By
Intelligent monitoring system and method for nursing of critical patient
CN119235279A
