Information processing device, information processing method, and program

Abstract

An information processing device according to an embodiment comprises: a signal splitting unit that splits an input signal of a cardiac measurement waveform at a splitting point between an R-wave peak point and a T-wave peak point; a curved line approximation unit that performs curved line approximation on each signal portion split by the signal splitting unit; and a fitting evaluation unit that evaluates fitting of the curved line approximation performed by the curved line approximation unit. The splitting point at which a result of the evaluation by the fitting evaluation unit satisfies a prescribed condition is set as an approximate J-point.

Description

Information processing apparatus, information processing method, and program

[0001] The present disclosure relates to an information processing apparatus, an information processing method, and a program.

[0002] In an electrocardiogram waveform, which is an example of a cardiac measurement waveform measured for a human heart, it is known that waveforms having specific characteristics such as Q wave, R wave, S wave, and T wave appear. In such an electrocardiogram waveform, there exists a J point which is a junction point between a QRS complex and an ST segment.

[0003] In the cardiac device described in Patent Document 1, a potential map representing inherent cardiac activity is recorded, and changes in the potential map are detected (see Patent Document 1).

[0004] Japanese National Publication of International Patent Application No. 2006-506130

[0005] However, in general, it is difficult to detect the J point in an electrocardiogram, and conventional techniques have insufficient development of apparatuses for detecting the J point in an electrocardiogram. Such a J point may appear not only when processing electrocardiogram waveforms, but also when processing magnetocardiogram waveforms with reference to electrocardiogram waveforms, for example. Alternatively, it may also appear in cases where the J point is detected on a magnetocardiogram and the magnetocardiogram is processed based thereon without referring to an electrocardiogram. Furthermore, there is a possibility that sensors for detecting other cardiac measurement waveforms in which such a J point appears will be developed. The sensor that measures data used for J point detection and the sensor that measures data processed or analyzed based on the information of the detected J point may be of the same type or different types.

[0006] The present disclosure has been made in consideration of such circumstances, and an object thereof is to provide an information processing apparatus, an information processing method, and a program capable of detecting the J point of a cardiac measurement waveform.

[0007] One embodiment is an information processing device comprising: a signal division unit that divides an input signal of a cardiac measurement waveform at a division point between the R-wave peak point and the T-wave peak point; a curve approximation unit that performs curve approximation on each signal portion divided by the signal division unit; and a fitting evaluation unit that evaluates the fitting of the curve approximation performed by the curve approximation unit, wherein the division point is defined as the approximate J point when the result of the evaluation by the fitting evaluation unit satisfies predetermined conditions.

[0008] One embodiment is an information processing method in which a signal splitting unit splits the input signal of a cardiac measurement waveform at a splitting point between the R-wave peak point and the T-wave peak point, a curve approximation unit performs curve approximation on each signal portion split by the signal splitting unit, a fitting evaluation unit evaluates the fitting of the curve approximation performed by the curve approximation unit, and the splitting point when the result of the evaluation by the fitting evaluation unit satisfies predetermined conditions is set as the approximate J point.

[0009] One embodiment is a program for a computer that provides the following functions: a function to divide the input signal of a cardiac measurement waveform at a dividing point between the R-wave peak and the T-wave peak; a function to perform curve approximation on each of the divided signal portions; a function to evaluate the fitting of the curve approximation; and a function to set the dividing point as the approximate J point when the result of the evaluation satisfies predetermined conditions.

[0010] According to this disclosure, an information processing device, an information processing method, and a program can detect the J point of a cardiac measurement waveform.

[0011] This figure shows a schematic configuration example of the information processing device according to the first embodiment. This figure shows a configuration example of the J-point detection processing unit according to the first embodiment. This figure shows an example of a cardiac measurement waveform signal according to the embodiment. This figure shows an example of a J-point in the cardiac measurement waveform signal according to the embodiment. This figure shows another example of a J-point in the cardiac measurement waveform signal according to the embodiment. This figure shows an example of a division point in the cardiac measurement waveform signal according to the embodiment. This figure shows an example of accurate fitting and division points in the cardiac measurement waveform signal according to the embodiment. This figure shows an example of inaccurate fitting and division points in the cardiac measurement waveform signal according to the embodiment. This figure shows an example of the search range in the cardiac measurement waveform signal according to the embodiment. This figure shows an example of an extremum in the cardiac measurement waveform signal according to the embodiment. This figure shows an example of an inflection point in the cardiac measurement waveform signal according to the embodiment. This figure shows an example of an extremum of the second derivative in the cardiac measurement waveform signal according to the embodiment. This figure shows an example of a case where the extremum in the cardiac measurement waveform signal according to the embodiment is close to the approximate J-point. This figure shows an example of a case where the inflection point in the cardiac measurement waveform signal according to the embodiment is close to the approximate J-point. This figure shows an example of a case where the extremum of the second derivative in the cardiac measurement waveform signal according to the embodiment is close to the approximate J-point. This figure shows a configuration example of the J-point detection processing unit of the information processing device according to the second embodiment.

[0012] The embodiments of this disclosure will be described below with reference to the drawings.

[0013] (First Embodiment) [Information Processing Device] Figure 1 is a diagram showing a schematic configuration example of an information processing device 11 according to the first embodiment. The information processing device 11 is configured using, for example, a computer. The information processing device 11 includes an input unit 131, an output unit 132, a communication unit 133, a storage unit 134, a control unit 135, and a J-point detection processing unit 136. The input unit 131 includes an operation unit 151. The output unit 132 includes a display unit 152.

[0014] The input unit 131 receives information from an external source. The input unit 131 has, for example, an operation unit 151 that receives operations performed by a user, and inputs information corresponding to the operations received by the operation unit 151. The operation unit 151 may have, for example, a function to receive operations from a touch panel, or it may have a function to receive operations from physical keys. The operation unit 151 may also have a function to receive operations by voice (for example, the user's voice). The input unit 131 may also be connected to an external device and input information output from that external device. The external device may be, for example, a portable recording medium. The recording medium may also be called, for example, a storage medium.

[0015] The output unit 132 outputs information. The output unit 132 has, for example, a display unit 152, and displays (outputs) information on the screen of the display unit 152. The screen may have touch panel functionality. The output unit 132 may also be connected to an external device and output information to the external device. The external device may be, for example, a portable recording medium. The output unit 132 may also output information in a manner other than display, such as outputting audio.

[0016] In this embodiment, the input unit 131 and the output unit 132 are shown as separate functional units, but the input unit 131 and the output unit 132 may be configured as a common functional unit (input / output unit). This input / output unit may be composed of a touch panel.

[0017] The communication unit 133 has the function of communicating information with an external device. This communication may be wired or wireless. In this embodiment, the communication unit 133 is shown as a separate functional unit from the input unit 131 and the output unit 132, but the receiving function of the communication unit 133 may be considered as being included in the function of the input unit 131, and the transmitting function of the communication unit 133 may be considered as being included in the function of the output unit 132.

[0018] The memory unit 134 stores information. The memory unit 134 may store any information. The information may also be called data.

[0019] The control unit 135 performs various types of control. In this embodiment, the control unit 135 is equipped with a processor such as a CPU (Central Processing Unit), and performs the control defined in a predetermined program (control program) by executing the program using the processor. The program may be stored in, for example, the storage unit 134.

[0020] The J-point detection processing unit 136 performs the process of detecting the J-point. In this embodiment, for the sake of explanation, the control unit 135 and the J-point detection processing unit 136 are shown separately, but for example, the processing performed by the J-point detection processing unit 136 may be considered to be realized by various controls performed by the control unit 135, or the control unit 135 and the J-point detection processing unit 136 may be considered as a single unit. In other words, the control unit 135 may be considered to perform various controls and various processing.

[0021] <Input Signals to the Information Processing Device> In this embodiment, the information processing device 11 receives a predetermined signal (also referred to as an input signal for convenience of explanation) from an external device. The input signal may be stored in, for example, the storage unit 134. The input signal may be input to the information processing device 11 via, for example, the input unit 131, or via the communication unit 133.

[0022] Furthermore, the external device may be, for example, a sensor (measuring instrument) that measures the predetermined signal, or a device that acquires the measurement result (the predetermined signal) from the sensor that measures the predetermined signal. Various types of sensors may be used as the sensor; for example, a sensor that measures electrical signals or a sensor that measures magnetic signals may be used. Here, instead of the word "measurement," words such as "detection," "detection," or "measurement" may be used.

[0023] As another example, the predetermined signal may be stored in advance in the storage unit 134 of the information processing device 11, in which case, for example, the predetermined signal does not have to be a signal input from the external device.

[0024] In this embodiment, the input signal is a cardiac measurement waveform signal (also referred to as a cardiac measurement waveform signal for convenience of explanation). This cardiac measurement waveform signal may be, for example, a waveform signal measured with respect to the heart of a human subject. In this embodiment, the cardiac measurement waveform signal is an output waveform signal obtained from electrophysiological measurements of the heart. This cardiac measurement waveform signal may be, for example, an electrocardiogram waveform signal, or a magnetocardiogram waveform signal. However, the cardiac measurement waveform signal is not limited to these examples.

[0025] In this embodiment, the case in which the cardiac measurement waveform signal is an electrocardiogram waveform signal will be described below, but the cardiac measurement waveform signal may be other signals. In this embodiment, the cardiac measurement waveform signal is treated as a digital signal having a plurality of signal values ​​(signal levels) arranged in time series. The process of converting the cardiac measurement waveform signal from an analog signal to a digital signal (A / D conversion process) may be performed, for example, by the sensor that measured the cardiac measurement waveform signal, or by the information processing device 11, or by an external device that relays the cardiac measurement waveform signal between the sensor and the information processing device 11.

[0026] Furthermore, the multiple signal values ​​arranged in time series that constitute the cardiac measurement waveform signal are the signal values ​​of the sampling result. Note that the multiple signal values ​​arranged in time series that constitute the cardiac measurement waveform signal do not necessarily have to be the sampling result itself, but may be the result of a decimation process performed on the sampling result (for example, a process that extracts signal values ​​at predetermined intervals and removes other signal values). In other words, the input signal may have, for example, the signal information (signal values) of all sampling timings (sampling numbers) obtained by sampling, or it may have the signal information (signal values) of some of the sampling timings (sampling numbers) obtained by sampling, and as a specific example, the result of decimation at predetermined sampling points may be used.

[0027] Each of the multiple signal values ​​arranged in a time series that constitute the cardiac measurement waveform signal is assigned a sampling number (for example, information representing its temporal position). For example, if the input signal is the result of sampling at regular time intervals, the sampling order (for example, the sampling number) may be used instead of time. In this embodiment, the time may be, for example, absolute time (clock time) or relative time (time that defines the temporal positional relationship of the signal waveform). Time may also be called, for example, time of day.

[0028] <J-point detection processing unit> Figure 2 shows an example of the configuration of the J-point detection processing unit 136 according to the first embodiment. Here, the example of the configuration of the functional blocks of the J-point detection processing unit 136 shown in Figure 2 is just one example for explanatory purposes and is not limited to this; various configurations may be used.

[0029] The J-point detection processing unit 136 includes a signal division unit 211, a curve approximation unit 212, a fitting evaluation unit 213, an approximate J-point storage unit 214, a differentiator 231, an extremum calculation unit 232, a differentiator 233, an inflection point calculation unit 234, a differentiator 235, a second-order differential extremum calculation unit 236, a distance calculation unit 237, and a distance comparison unit 238. In this embodiment, the functional unit that performs differential processing using differentiators 231, 233, and 235 is shown as the differential processing unit H1.

[0030] In the information processing device 11, the J-point detection processing unit 136 receives an input signal (in this embodiment, a cardiac measurement waveform signal) as the signal to be detected as a J-point. This input signal may be a signal input to the information processing device 11 from an external device such as a sensor, and may be acquired by the J-point detection processing unit 136 from the storage unit 134. This input signal is input to the signal division unit 211 and the differentiator 231.

[0031] The following describes an example of the processing performed by each functional block when the signal to be processed by the J-point detection processing unit 136 is a certain input signal. In other words, if the signal to be processed by the J-point detection processing unit 136 switches to a different input signal, the following series of processes will be performed, for example, from a reset state. However, when the signal to be processed by the J-point detection processing unit 136 switches to a different input signal, the processing results for previously processed input signals may be referenced and used.

[0032] Furthermore, as the input signal to be processed by the J-point detection processing unit 136, for example, a cardiac measurement waveform signal in which a similar waveform (theoretically the same waveform) is expected to be repeated may be used as a signal of one cycle of the waveform, or a signal obtained by statistically processing (for example, averaging) waveforms of multiple cycles may be used.

[0033] The signal splitting unit 211 splits the input signal by setting splitting points in the input signal. The signal splitting unit 211 does not necessarily have to generate a split signal by dividing the input signal into multiple signals; it can simply split the input signal into multiple signal parts. The points where the splitting points are set in the input signal may be, for example, any of the multiple sampling points, or other points.

[0034] When the signal splitting unit 211 sets a splitting point for the input signal for the first time, it outputs information indicating the position of the splitting point to the curve approximation unit 212. The signal splitting unit 211 also outputs information indicating the input signal to the curve approximation unit 212.

[0035] Furthermore, the signal splitting unit 211 has a function to change the position where the splitting points are set for the input signal. In other words, the signal splitting unit 211 has a function to adjust the position where the splitting points are set for the input signal. The signal splitting unit 211 changes the position where the splitting points are set for the input signal in a predetermined manner, for example, but as another example, it may change the position where the splitting points are set for the input signal based on input information (fitting evaluation results) from the fitting evaluation unit 213.

[0036] Based on the fitting evaluation result, the signal splitting unit 211 determines whether to adopt the currently set splitting point (i.e., the splitting point at which the fitting evaluation result was obtained) or to reset the splitting point to another location. If the signal splitting unit 211 determines to adopt the splitting point, it outputs information to the curve approximation unit 212 indicating this. On the other hand, if the signal splitting unit 211 determines to reset the splitting point to another location, it sets the splitting point to another location (i.e., a location different from the currently set location) and outputs information to the curve approximation unit 212 indicating the location of the splitting point.

[0037] Here, a method for determining whether or not to adopt the currently set division point based on the fitting evaluation result may be used, for example, to determine whether the fitting evaluation result satisfies a predetermined condition of being good. If it is determined that the condition is met, the division point is adopted; on the other hand, if it is determined that the condition is not met, the division point is not adopted. In this case, for example, the fitting evaluation result may be expressed by a predetermined evaluation value (e.g., fitting rate), and the condition may be determined to be met when the evaluation value is the maximum value among the set division points. As the evaluation value, an evaluation value using any method such as the coefficient of determination R² may be used. Note that here, a larger evaluation value is shown to indicate a better result, but the relationship may be reversed.

[0038] In this embodiment, when the signal splitting unit 211, the curve approximation unit 212, and the fitting evaluation unit 213 search for a splitting point where the fitting evaluation result satisfies predetermined conditions, for example, a process may be performed to save (store) information indicating the fitting result for each splitting point set at each point in the storage unit 134 or the like. This process may be performed, for example, by the signal splitting unit 211 or the fitting evaluation unit 213. In this embodiment, the processing by the signal splitting unit 211, the curve approximation unit 212, and the fitting evaluation unit 213 is repeated while sequentially shifting the position of the splitting point, and when a splitting point where the fitting evaluation result satisfies predetermined conditions is found, the approximate J point is notified to the approximate J point storage unit 214 to adopt that splitting point.

[0039] As another example, the fitting evaluation result may be expressed as a predetermined evaluation value (e.g., fitting rate), and the condition may be expressed as a threshold related to the evaluation value. If the evaluation value exceeds the threshold (or is equal to or greater than the threshold), it may be determined that the condition is met. In other words, a configuration may be used in which it is determined that the predetermined condition is met even if the fitting evaluation result is not the maximum value, as long as it is reasonably large. Note that here we have shown a case where a larger evaluation value is better, but the relationship between magnitude may be reversed.

[0040] In this embodiment, when determining whether the fitting evaluation result is good or not, both the fitting evaluation result of the first signal portion and the fitting evaluation result of the second signal portion are considered. For example, the fitting evaluation result of the first signal portion (e.g., evaluation value) and the fitting evaluation result of the second signal portion (e.g., evaluation value) may be obtained separately, and then the sum of these two fitting evaluation results (e.g., evaluation values) may be calculated. This sum may then be used as the overall fitting evaluation result (e.g., evaluation value) to determine whether the overall fitting result is good or not, and this determination result may be used as the final determination result.

[0041] As another example, the fitting evaluation results for the first signal portion and the fitting evaluation results for the second signal portion may be judged separately as to whether they are good or bad, and the combined result of these judgments may be used as the final judgment result. As yet another example, a fitting result considering both the first signal portion and the second signal portion may be obtained, and the result of judging whether that fitting result is good or bad may be used as the final judgment result.

[0042] Here, for example, the signal dividing unit 211 may set dividing points within a predetermined range that further restricts the position for setting dividing points among signal portions (signal portions of the input signal) for which dividing points are set for the input signal (also referred to as a dividing point setting range for convenience of explanation). The dividing point setting range may be set in advance, or may be changed (adjusted) at any timing. For example, any dividing point setting range may be set by designation via a user operation, or automatically set by the information processing apparatus 11. The dividing point setting range may be, for example, a range of temporal positions, a range of level magnitudes, or both of these ranges. Note that such a dividing point setting range is not necessarily required to be used.

[0043] The curve approximation unit 212 performs the following processing based on input information from the signal dividing unit 211 when the signal dividing unit 211 sets a dividing point for an input signal for the first time, or when a currently set dividing point is reset to another position. That is, in these cases, the curve approximation unit 212, with reference to the dividing point set this time by the signal dividing unit 211, performs curve approximation on an input signal portion (also referred to as a first signal portion for convenience of explanation) on the negative side of the dividing point (for example, a portion temporally before the dividing point), and also performs curve approximation on an input signal portion (also referred to as a second signal portion for convenience of explanation) on the positive side of the dividing point (for example, a portion temporally after the dividing point).

[0044] Furthermore, the curve approximation unit 212 outputs information indicating the curve approximation result of the first signal portion and the curve approximation result of the second signal portion to the fitting evaluation unit 213. Here, the information includes, for example, information indicating the input signal, information indicating a curve obtained when the first signal portion is subjected to curve approximation, and information indicating a curve obtained when the second signal portion is subjected to curve approximation. Note that if information indicating the input signal has already been notified from the curve approximation unit 212 to the fitting evaluation unit 213, for example, the curve approximation unit 212 does not need to notify the fitting evaluation unit 213 of the information again.

[0045] On the other hand, the curve approximation unit 212 outputs information indicating the position of the currently set division point to the approximate J-point storage unit 214 if the currently set division point is adopted based on the input information from the signal division unit 211. The curve approximation unit 212 also outputs information indicating the input signal to the approximate J-point storage unit 214.

[0046] Here, there are no particular limitations on the types of curves used for the curve approximation of the first signal portion and the curves used for the curve approximation of the second signal portion; various types may be used. For example, quadratic curves, sigmoid curves, curves of the beta probability distribution, or curves of the log-normal probability distribution may be used, or derived forms of these curves may be used. Furthermore, the types of curves used for the curve approximation of the first signal portion and the curves used for the curve approximation of the second signal portion may be the same or different. Such curve types may be set in advance or may be changed (adjusted) at any time. For example, any curve type may be set by user operation or automatically by the information processing device 11.

[0047] The fitting evaluation unit 213 evaluates the fitting of the curve approximation to the input signal for the first signal portion and for the second signal portion based on the input information from the curve approximation unit 212. Here, for example, the fitting evaluation unit 213 may perform the fitting evaluation for the first signal portion and the fitting evaluation for the second signal portion together. The fitting evaluation unit 213 also outputs information indicating the result of the fitting evaluation to the signal division unit 211.

[0048] Here, in the present embodiment, the case where the process of calculating a comprehensive fitting evaluation result from the fitting evaluation result of the first signal portion and the fitting evaluation result of the second signal portion is performed by the signal dividing section 211 has been described. However, as another example, the process may be performed by the fitting evaluation section 213, and in this case, the comprehensive fitting evaluation result is notified from the fitting evaluation section 213 to the signal dividing section 211.

[0049] The approximate J point storage section 214 stores information indicating the position of an adopted dividing point as information indicating an approximate J point, based on input information from the curve approximation section 212. Said information may be stored (memorized) in the storage section 134, for example. The approximate J point storage section 214 also outputs information indicating the approximate J point to the distance calculation section 237. The approximate J point storage section 214 also outputs information indicating an input signal to the distance calculation section 237.

[0050] The differentiator 231 performs differentiation processing (i.e., first-order differentiation processing) on an input signal. The differentiator 231 also outputs information indicating a result of said differentiation processing (a result of first-order differentiation on the input signal) to the extreme value calculation section 232 and the differentiator 233.

[0051] The extreme value calculation section 232 calculates an extreme value of the input signal based on a differentiation result of the input signal, based on input information from the differentiator 231. Accordingly, an extremum point that is the point of said extreme value is obtained. The extreme value calculation section 232 also outputs information indicating a calculation result of the extreme value (for example, the position of the extreme value in the input signal) to the distance calculation section 237.

[0052] The differentiator 233 performs differentiation processing (i.e., second-order differentiation processing) on the signal of the differentiation result of the input signal, based on input information from the differentiator 231. The differentiator 233 also outputs information indicating a result of said differentiation processing (a result of second-order differentiation on the input signal) to the inflection point calculation section 234 and the differentiator 235.

[0053] The inflection point calculation unit 234 calculates the inflection point of the input signal based on the second derivative result of the input signal, using the input information from the differentiator 233. The inflection point calculation unit 234 also outputs information indicating the calculation result of the inflection point (for example, the position of the inflection point in the input signal) to the distance calculation unit 237.

[0054] The differentiator 235 performs differentiation (i.e., third-order differentiation) on the signal obtained by taking the second derivative of the input signal, based on the input information from the differentiator 233. The differentiator 235 also outputs information indicating the result of this differentiation (the result of the third derivative with respect to the input signal) to the second-order derivative extreme value calculation unit 236.

[0055] The second differential extremum calculation unit 236 calculates the extremum of the second derivative based on the third derivative result of the input signal, using input information from the differentiator 235. This determines the extreme point of the second derivative. The second differential extremum calculation unit 236 also outputs information indicating the calculation result of the second differential extremum (for example, the position of the extremum in the input signal) to the distance calculation unit 237.

[0056] Here, in each of the differentiators 231, 233, and 235, when performing differential processing on the signal to be differentiated, the number of points (difference points) from one point (temporal position) to the other point (temporal position) for which the difference in signal value (signal level) is taken may be, for example, one (i.e., the one point and the other point are adjacent), or two or more predetermined numbers (i.e., the one point and the other point are not adjacent). The difference points of each of the differentiators 231, 233, and 235 may be set in advance, or they may be changed (adjusted) at any time. For example, the difference points of any number (size) may be set by user operation or automatically by the information processing device 11.

[0057] The number of difference points used in differentiator 231, differentiator 233, and differentiator 235 may all be set to the same value, or two may be set to the same value and the other one to a different value, or all may be set to different values. Generally, it is thought that a smaller number of difference points allows for finer differentiation, while a larger number of difference points can reduce the effects of noise.

[0058] The distance calculation unit 237 calculates the distance between the approximate J point and the extremum point from the extremum calculation unit 232, the distance between the approximate J point and the inflection point from the inflection point calculation unit 234, and the distance between the approximate J point and the second differential extremum point from the second differential extremum calculation unit 236, based on the input information from the approximate J point storage unit 214, the input information from the extremum value calculation unit 232, the input information from the inflection point calculation unit 234, and the input information from the second differential extremum calculation unit 236. The distance calculation unit 237 also outputs information indicating these distances to the distance comparison unit 238.

[0059] The distance comparison unit 238 compares distances based on the input information from the distance calculation unit 237 and determines the position of point J in the input signal. The distance comparison unit 238 also outputs information indicating the determined position of point J.

[0060] In this embodiment, the point closest in distance to the approximate point J (which may be a point that coincides with the approximate point J) is determined as the position of point J. The distance between the two points may be calculated using any method; for example, the Euclidean distance may be used. Such a distance calculation method may be pre-set or changed (adjusted) at any time. For example, an arbitrary calculation method may be set by user operation or automatically by the information processing device 11.

[0061] Here, when the distance calculation unit 237 and the distance comparison unit 238 determine the position of point J, the candidate points for determining point J based on the distance to the approximate point J (in this embodiment, extreme point, inflection point, second differential extreme point) may be limited to points that fall within a predetermined range (for convenience of explanation, also called the search range). As the search range, for example, a search range that limits the range of the temporal position in the input signal may be used, or a search range that limits the range of the waveform level in the input signal may be used, or a search range that limits both of these ranges may be used.

[0062] Here, the search range may be set by, for example, the distance calculation unit 237, thereby limiting the candidate points (in this embodiment, extreme value points, inflection points, second differential extreme value points) for calculating the distance to the approximate J point; or it may be set by the distance comparison unit 238, thereby limiting the candidate points (in this embodiment, extreme value points, inflection points, second differential extreme value points) for determining the point closest in distance to the approximate J point; or the same or different search ranges may be set by the distance calculation unit 237 and the distance comparison unit 238, respectively. The search range may be set in advance, or it may be changed (adjusted) at any time. For example, an arbitrary search range may be set by user operation or automatically by the information processing device 11. Note that the search range does not necessarily have to be set; that is, even if the search range is not set, it is possible to determine the point closest in distance to the approximate J point.

[0063] In this embodiment, there are multiple types of points (in this embodiment, three types of points: extrema points, inflection points, and second differential extrema points) that are candidates for calculating the distance to the approximate point J. The distances to these multiple types of points are compared in the same way (i.e., without setting any priority), but a priority may be set. In other words, the point to be designated as point J may be selected based on the priority.

[0064] For example, a priority order may be used in which, for points of the first priority type (for instance, an extremum, but other types of points may also be used), the point closest in distance to the approximate J point is first found, and if the distance between that point and the approximate J point is less than or equal to a predetermined threshold (or less than the threshold), that point is adopted as the J point. If the distance between that point and the approximate J point exceeds the predetermined threshold (or is greater than or equal to the threshold), the same process is performed for points of the second priority type (for example, an inflection point, but other types of points may also be used), and so on, moving on to the next priority level.

[0065] Another example of priority order is distance weighting. That is, for each of the multiple types of points (in this embodiment, three types of points: extrema, inflection points, and second differential extrema), a priority order may be used in which the calculated distance is multiplied by a predetermined coefficient (for example, a different value for at least one type) and the result is compared as the final distance. Note that any other arbitrary priority order may be used.

[0066] Another example of priority is the use of a specified range set for the approximate J point. This specified range may be set in advance or may be changed (adjusted) at any time. For example, an arbitrary specified range may be set by user operation or automatically by the information processing device 11. For example, for a point of the first priority type (for example, an extremum, but other types of points may also exist), it is first determined whether one or more points exist within the specified range set for the approximate J point. If none exist, the same process is performed for a point of the second priority type (for example, an inflection point, but other types of points may also exist), and so on, moving on to the next priority. In this case, for each type of point, if one or more points exist within the specified range, one point is selected as the J point based on, for example, the distance from the approximate J point.

[0067] Such priorities may be set in advance, or they may be changed (adjusted) at any time. For example, arbitrary priorities may be set by user operation or automatically by the information processing device 11. Note that priorities do not necessarily have to be set.

[0068] [Specific Examples of Processing by the J-Point Detection Processing Unit] Specific examples of processing performed by the J-point detection processing unit 136 are shown with reference to Figures 3A-3C, 4A-4C, 5, 6A-6C, and 7A-7C. In the graphs shown in each figure, the horizontal axis represents time, and the vertical axis represents the signal level. Note that in the examples of each figure, the scales on the horizontal and vertical axes of the graphs are omitted. The level may be, for example, the amplitude level, or the power level, or any other level.

[0069] The waveforms shown in this embodiment are illustrative examples and not necessarily exact representations of actual waveforms. For example, the waveforms in each graph represent illustrative examples of characteristics (trends), and the time axis on the horizontal axis and the level axis on the vertical axis of each graph have been omitted.

[0070] <Cardiac Measurement Waveform Signal> Figure 3A shows an example of a cardiac measurement waveform signal F1 according to the embodiment. In the example in Figure 3A, the cardiac measurement waveform signal F1 for one cycle (or approximately one cycle) is shown. In the cardiac measurement waveform signal F1, as time progresses, the Q wave appears at time t1, the R wave peak at time t2, the S wave at time t3, the J point at time t4, the beginning of the T wave at time t5, the T wave peak at time t6, and the end of the T wave at time t7. Note that times t1 to t7 are in order from the past to the future.

[0071] Furthermore, the ST segment (ST portion), which is the waveform portion from time t11 to time t12, is shown. In the example of Figure 3A, time t11 represents the end of the S wave, and time t12 represents the beginning of the T wave. Note that in the example of Figure 3A, time t11 coincides with time t4, and time t12 coincides with time t5.

[0072] Thus, the ST segment is the portion from the end of the S wave to the beginning of the T wave. Point J represents the junction between the QRS wave and the ST segment in the cardiac measurement waveform signal F1.

[0073] Figure 3B shows an example of point J in the cardiac measurement waveform signal F11 according to the embodiment. In the graph shown in Figure 3B, the horizontal axis represents time, and the vertical axis represents the signal level. Note that in the example of Figure 3B, the scales on the horizontal and vertical axes of the graph are omitted. In the cardiac measurement waveform signal F11, the extreme value point at time t21 coincides with point J. Figure 3C shows an example of point J in the cardiac measurement waveform signal F12 according to the embodiment. In the graph shown in Figure 3C, the horizontal axis represents time, and the vertical axis represents the signal level. Note that in the example of Figure 3C, the scales on the horizontal and vertical axes of the graph are omitted. In the cardiac measurement waveform signal F12, the point at time t22, which is not an extreme value point, coincides with point J.

[0074] As shown in the examples in Figure 3B and Figure 3C, the waveform shape near point J does not have a fixed shape and is not necessarily an extremum.

[0075] Referring to Figures 4A, 4B, and 4C, a specific example of the processing performed by the signal splitting unit 211, the curve approximation unit 212, the fitting evaluation unit 213, and the approximate J-point storage unit 214 (for convenience of explanation, this will also be referred to as the processing of STEP 1) will be described.

[0076] Figure 4A shows an example of a division point V1 in the cardiac measurement waveform signal F21 according to this embodiment. The division point V1 is a point set on the cardiac measurement waveform signal F21. In this embodiment, as shown in Figure 4A, the signal division unit 211 sets the division point V1 between the R-wave peak point (temporal position), which is the point of the R-wave peak, and the T-wave peak point (temporal position), which is the point of the T-wave peak.

[0077] In this embodiment, the signal splitting unit 211 sets two or more splitting points. In this case, the positions of each splitting point when setting two or more splitting points may be determined, for example, in a predetermined manner.

[0078] As an example, the predetermined configuration may be a configuration in which the division points are set sequentially at points (temporal positions) shifted at predetermined intervals (temporal intervals) in the direction from the R-wave peak point (temporal position) to the T-wave peak point (temporal position), or in the opposite direction. In this configuration, for example, the initial position of the division point may be set to any position, for example, the R-wave peak point (or T-wave peak point), or a point shifted at a predetermined interval from the R-wave peak point (or T-wave peak point). Here, as a configuration in which the division points are set sequentially at points shifted at predetermined intervals, for example, a configuration may be used in which the sampling points arranged in time series are shifted one point at a time (or two or more predetermined points at a time), and the division points are set sequentially, thereby dividing the signal into two based on each division point. In the example in Figure 4A, an example of how the division point V1 is set in the direction from the R-wave peak point to the T-wave peak point is schematically shown.

[0079] As another example, the predetermined configuration may be one in which random points (temporal positions) are sequentially set as division points. In this configuration, for example, if the next random point to be determined happens to be a point that has already been set, another random point may be determined.

[0080] Here, for example, when setting a division point between the R-wave peak point and the T-wave peak point, a predetermined range (division point setting range) may also be set between the R-wave peak point and the T-wave peak point. The signal division unit 211 may then set the division point within this division point setting range.

[0081] Figure 4B shows an example of accurate fitting and division point V11 of the cardiac measurement waveform signal F21 according to the embodiment. In this example, the curve approximation unit 212 approximates the signal portion between the R-wave peak point B1 and the division point V11 (first signal portion) of the cardiac measurement waveform signal F21 with a predetermined curve C1, and approximates the signal portion between the division point V11 and the T-wave peak point B2 (second signal portion) with a predetermined curve C2.

[0082] In this example, the fitting evaluation unit 213 obtains fitting evaluation results for the curve approximation of the first signal portion and the curve approximation of the second signal portion, and the signal division unit 211 determines that the fitting evaluation results satisfy predetermined conditions. In other words, in this example, the fitting evaluation results are determined to have good numerical accuracy, and the division point V11 is adopted as the approximate J point and saved by the approximate J point saving unit 214.

[0083] Figure 4C shows an example of poorly fitting and division point V12 of the cardiac measurement waveform signal F21 according to the embodiment. In this example, the curve approximation unit 212 approximates the signal portion between the R-wave peak point B1 and the division point V12 (first signal portion) of the cardiac measurement waveform signal F21 with a predetermined curve C11, and approximates the signal portion between the division point V12 and the T-wave peak point B2 (second signal portion) with a predetermined curve C12. Here, the division point V12 shown in Figure 4C is a point at a different position from the division point V11 shown in Figure 4B.

[0084] In this example, the fitting evaluation unit 213 obtains fitting evaluation results for the curve approximation of the first signal portion and the curve approximation of the second signal portion, and the signal splitting unit 211 determines that the fitting evaluation results do not meet the predetermined conditions. In other words, in this example, the fitting evaluation results are determined to have poor numerical accuracy (i.e., poor accuracy), and the splitting point V12 is not adopted as the approximate J point, and the splitting point is reset.

[0085] Referring to Figures 5, 6A, 6B, 6C, 7A, 7B, and 7C, a specific example of the processing performed by the differentiator 231, the extremum calculation unit 232, the differentiator 233, the inflection point calculation unit 234, the differentiator 235, the second differential extremum calculation unit 236, the distance calculation unit 237, and the distance comparison unit 238 (for convenience of explanation, this will also be called the STEP 2 process) will be explained. In general terms, the STEP 1 process estimates an approximate J point, and the STEP 2 process determines the J point based on the approximate J point.

[0086] Figure 5 shows an example of the search range W1 in the cardiac measurement waveform signal F21 according to the embodiment. In this example, the search range W1 limits the range of levels, specifically representing a range that falls within ±α (where α is a positive integer) of the level of the Q-wave point B11, which is a Q-wave point. Here, any value may be used for α, for example, a predetermined percentage [%] of the maximum level (in the example of Figure 5, the level of the R-wave peak point B1) may be used.

[0087] Note that the search range W1 shown in Figure 5 is just one example for illustrative purposes and is not limited to this; various search ranges may be set. Here, the division point V21 shown in Figure 5 represents a point adopted as the approximate J point, and corresponds to, for example, the division point V11 in the example of Figure 4B.

[0088] Figure 6A shows an example of an extreme value in the cardiac measurement waveform signal F21 according to the embodiment. Figure 6A shows multiple extreme value points in the cardiac measurement waveform signal F21, and one point A1 included in the search range W1 among these multiple points is labeled with a reference numeral. In this embodiment, the extreme value points are determined by the differentiator 231 and the extreme value calculation unit 232. In the example of Figure 6A, the multiple extreme value points (extreme value points) are schematically shown using predetermined marks.

[0089] Figure 6B is a diagram showing an example of an inflection point in the cardiac measurement waveform signal F21 according to the embodiment. Figure 6B shows multiple inflection points in the cardiac measurement waveform signal F21, and among these multiple points, three points (points A2a, A2b, and A2c) included in the search range W1 are labeled with reference numerals. In this embodiment, the inflection points are determined by the differentiator 231, the differentiator 233, and the inflection point calculation unit 234. In the example of Figure 6B, multiple inflection points are schematically shown using predetermined marks.

[0090] Figure 6C is a diagram showing an example of the second derivative extremum in the cardiac measurement waveform signal F21 according to the embodiment. Figure 6C shows multiple second derivative extremum points in the cardiac measurement waveform signal F21, and one point A3 included in the search range W1 among these multiple points is labeled with a reference numeral. Figure 6C also shows the second derivative curve g21 (second derivative curve) of the cardiac measurement waveform signal F21. In this embodiment, the second derivative extremum points are determined by the differentiator 231, differentiator 233, differentiator 235, and the second derivative extremum calculation unit 236. In the example of Figure 6C, multiple points of the second derivative extremum (second derivative extremum points) are schematically shown using predetermined marks.

[0091] Figure 7A shows an example of a case in which the extremum of the cardiac measurement waveform signal F31 according to the embodiment is close to the approximate point J K11. Figure 7A shows multiple extremum points of the cardiac measurement waveform signal F31, multiple inflection points of the cardiac measurement waveform signal F31, and multiple extremum points of the second derivative of the cardiac measurement waveform signal F31. Among these points, point A11, which is closest to the approximate point J K11, is labeled. In this example, point A11 is an extremum point of the cardiac measurement waveform signal F31.

[0092] Figure 7B is a diagram showing an example in which the inflection point of the cardiac measurement waveform signal F32 is close to the approximate point J K12 according to the embodiment. Figure 7B shows multiple extreme points of the cardiac measurement waveform signal F32, multiple inflection points of the cardiac measurement waveform signal F32, and multiple extreme points of the second derivative of the cardiac measurement waveform signal F32, and among these points, point A12, which is closest in distance to the approximate point J K12, is labeled. In this example, point A12 is the inflection point of the cardiac measurement waveform signal F32.

[0093] Figure 7C is a diagram showing an example in which the extreme value of the second derivative in the cardiac measurement waveform signal F33 according to the embodiment is close to the approximate point J. Figure 7C shows multiple extreme value points in the cardiac measurement waveform signal F33, multiple inflection points in the cardiac measurement waveform signal F33, and multiple extreme value points of the second derivative in the cardiac measurement waveform signal F33, and among these points, point A13, which is closest in distance to the approximate point J K13, is labeled. In this example, point A13 is the extreme value point of the second derivative of the cardiac measurement waveform signal F33.

[0094] In this embodiment, in the examples of Figure 7A, Figure 7B, and Figure 7C, the point closest to the approximate point J is determined by the distance calculation unit 237 and the distance comparison unit 238. In the examples of Figure 7A, Figure 7B, and Figure 7C, the extremum, inflection point, and second derivative extremum are schematically shown using different marks, respectively.

[0095] As described above, the information processing device 11 according to this embodiment can detect the J point of the cardiac measurement waveform (for example, determine the position of the J point). Furthermore, the information processing device 11 according to this embodiment can accurately detect the J point of the cardiac measurement waveform by performing two-stage processing (STEP 1 processing and STEP 2 processing).

[0096] <Importance of J-point detection> Here, we will explain the importance of J-point detection. It is known that individuals with elevated J-points have a significantly increased risk of death from cardiac disease, particularly from arrhythmias, compared to those without elevated J-points.

[0097] This section explains the background technology and challenges. Specifically, the waveforms of the electrical potential or magnetic field of cardiac activity vary considerably due to individual differences or noise. Detecting feature points such as J-points often results in false detections without sophisticated algorithms. When performed visually by humans, the process is subjective and lacks reproducibility, leading to inconsistent results depending on the operator. Therefore, an objective algorithm that yields reasonable results in most cases was needed.

[0098] However, while the J point is important, there were insufficient effective estimation methods compared to other points (e.g., the R-wave peak point or the T-wave peak point). For example, the J point is defined as the junction between the QRS wave and the ST segment, but attempting to determine it solely based on the local features of the waveform can be difficult due to noise or individual differences. Multiple points with similar features are often found within the time range of interest, making it challenging to determine the J point reproducibly.

[0099] Therefore, in this embodiment, we propose an information processing device 11 that performs J-point detection. The information processing device 11 can automatically perform J-point detection according to a predetermined processing procedure. In this embodiment, as STEP 1, the information processing device 11 determines the approximate position (temporal position) of the J-point from macro information (signal portion with a certain time width), and as STEP 2, it adjusts and determines the fine position (temporal position) of the J-point from local information (neighborhood of the approximate J-point). As a result, the information processing device 11 can detect the J-point even if the waveform shape near the J-point in the input signal is varied. In other words, the waveform shape near the J-point is not simpler than that of other points and can take various shapes, but in this embodiment, it is possible to robustly detect the J-point by combining macro detection based on the J-point generation principle with local detection.

[0100] Furthermore, regarding the processing of STEP 1 and STEP 2, for example, STEP 2 may be performed after the processing of STEP 1 is completed, or part or all of the processing of STEP 1 and part of the processing of STEP 2 may be performed in parallel during the same time period.

[0101] <Basis for the J-point detection method according to this embodiment> In the STEP 1 process, a division point that allows for accurate curve approximation from the R-wave peak point to the T-wave peak point is calculated, and a rough estimate of the J-point (approximate J-point) is calculated based on this. Between the R-wave peak point and the J-point, there may be an S-wave peak (S-wave peak) that is smaller than the R-wave peak point, but in most cases, the waveform can be fitted by curve approximation. Between the J-point and the T-wave peak point, the slope changes significantly with respect to the signal waveform prior to the J-point, so moving the division point to the left increases the error in the curve approximation fitting.

[0102] Thus, the slope of the waveform between the R-wave peak and point J changes significantly from the slope of the waveform between point J and point T. This tendency is a phenomenon commonly observed in many heart rate patterns. Therefore, as in this embodiment, approximation using two curves based on the division point is a good estimation method for detecting point J.

[0103] Note that the signal portion before point J (e.g., the QRS wave) is generated by ventricular depolarization. On the other hand, the signal portion after point J (e.g., the ST segment and T wave) is generated by ventricular repolarization. Since these two occur through different mechanisms, they exhibit mathematically discontinuous changes. Therefore, good estimation results can be obtained by approximating each of them with separate functions (two different functions).

[0104] Step 2 involves performing detailed positioning based on local characteristics. There is an empirical rule that the first derivative (the result of the first derivative), second derivative (the result of the second derivative), and third derivative (the result of the third derivative) tend to be zero at the connection point of the two functions mentioned above. For this reason, empirically, point J often takes the form of an extremum, an inflection point, or an extremum of the second derivative.

[0105] In this embodiment, we use terms such as first derivative, second derivative, and third derivative, but these may also be called, for example, first-order derivative, second-order derivative, and third-order derivative, respectively.

[0106] (Second Embodiment) [Information Processing Device] The information processing device 11a is configured using, for example, a computer. Figure 8 shows an example of the configuration of the J-point detection processing unit 136a of the information processing device 11a according to the second embodiment. The information processing device 11a includes an input unit 131, an output unit 132, a communication unit 133, a storage unit 134, a control unit 135, and a J-point detection processing unit 136a. The input unit 131 includes an operation unit 151. The output unit 132 includes a display unit 152.

[0107] Herein, in general terms, the configuration and operation of the information processing device 11a according to this embodiment differ from the configuration and operation of the information processing device 11 shown in Figures 1 and 2 according to the first embodiment in that it includes a J-point detection processing device 136a instead of the J-point detection processing device 136 according to the first embodiment, and is otherwise similar. For this reason, in this embodiment, the same reference numerals are used for functional blocks that are the same as those shown in Figure 1 according to the first embodiment. In this embodiment, the configuration and operation of the J-point detection processing device 136a will be described in detail, and a detailed explanation of the configuration and operation that are the same as in the first embodiment will be omitted.

[0108] <J-point detection processing unit> The J-point detection processing unit 136a comprises a signal splitting unit 211a, a curve approximation unit 212a, a fitting evaluation unit 213a, and an approximate J-point storage unit 214a. Here, the signal splitting unit 211a, the curve approximation unit 212a, the fitting evaluation unit 213a, and the approximate J-point storage unit 214a have the same functions and perform the same operations as the signal splitting unit 211, the curve approximation unit 212, the fitting evaluation unit 213, and the approximate J-point storage unit 214 shown in Figure 2, respectively.

[0109] In this embodiment, the J-point detection processing unit 136a performs one of the two processes (STEP1 process and STEP2 process) in the first embodiment (STEP1 process) and does not perform the other process (STEP2 process). In this embodiment, the approximate J-point obtained by the one process (STEP1 process) is determined to be the J-point. In this embodiment, the input signal (cardiac measurement waveform signal) is input to the signal division unit 211a, and the approximate J-point storage unit 214a outputs information about the approximate J-point as J-point information.

[0110] As described above, the information processing device 11a according to this embodiment can detect the J point of the cardiac measurement waveform (for example, determine the position of the J point). In this embodiment, compared to the first embodiment, only the process of finding an approximate J point (processing in STEP 1) is performed, so the detection accuracy of the J point may be lower. However, the processing of the J point detection processing unit 136a can be simplified, and the processing load and processing time required for J point detection can be reduced.

[0111] [Regarding the above embodiments] The above embodiments show an example of their configuration. Here, we show an example of the configuration of the information processing device 11 according to the first embodiment, but the components related to the processing of STEP 1 are the same for the information processing device 11a according to the second embodiment.

[0112] As an example configuration, the information processing device 11 has the following configuration. The information processing device 11 includes a signal splitting unit 211 that splits the input signal of the cardiac measurement waveform at a splitting point between the R-wave peak point and the T-wave peak point. The information processing device 11 includes a curve approximation unit 212 that performs curve approximation on each signal portion split by the signal splitting unit 211. The information processing device 11 includes a fitting evaluation unit 213 that evaluates the fitting of the curve approximation performed by the curve approximation unit 212. The information processing device 11 sets the splitting point at which the evaluation result by the fitting evaluation unit 213 satisfies predetermined conditions as the approximate J point. Therefore, the information processing device 11 can detect the J point of the cardiac measurement waveform.

[0113] As an example configuration, the information processing device 11 has the following configuration. The information processing device 11 includes an approximate J-point storage unit 214 for storing approximate J-points. Therefore, the information processing device 11 can store approximate J-points, and for example, it is possible to use the stored approximate J-points.

[0114] As an example configuration, the information processing device 11 has the following configuration. The signal splitting unit 211 shifts the position of the splitting point between the R-wave peak point and the T-wave peak point at predetermined intervals. The curve approximation unit 212 performs curve approximation for each splitting point. The information processing device 11 then selects the splitting point from among the multiple splitting points that satisfies predetermined conditions as the approximate J point. Therefore, the information processing device 11 can improve the detection accuracy of the J point by determining the splitting point to be adopted using multiple candidate splitting points. Various methods can be used to shift the position of the splitting point at predetermined intervals. For example, when sampling is performed in discrete time, the sampling interval may be used as the predetermined interval, or an integer multiple of the sampling interval (here, an integer multiple of 2 or more) may be used as the predetermined interval, or a non-integer multiple of the sampling interval may be used as the predetermined interval, and the position of the splitting point may be shifted at predetermined intervals while performing interpolation calculations.

[0115] As an example configuration, the information processing device 11 has the following configuration. The signal splitting unit 211 sets the splitting points within a predetermined range set between the R-wave peak point and the T-wave peak point. Therefore, by setting the splitting points within a limited range (the predetermined range), the information processing device 11 can reduce the processing load and processing speed required for determining the splitting points to be adopted. In other words, by setting the range in advance, it becomes unnecessary to process the entire range from the R-wave peak point to the T-wave peak point.

[0116] As an example configuration, the information processing device 11 has the following configuration. The curve approximation unit 212 performs curve approximation using a quadratic curve for at least one signal portion. Therefore, the information processing device 11 can achieve good curve approximation using a quadratic curve. Note that the quadratic curve is just one example, and it is not necessary to use a quadratic curve.

[0117] As an example configuration, the information processing device 11 has the following configuration. The curve approximation unit 212 performs curve approximation using different curves for at least two signal portions. Therefore, the information processing device 11 can achieve good curve approximation by performing curve approximation using a suitable curve for each of the different signal portions.

[0118] As an example configuration, the information processing device 11 has the following configuration. The information processing device 11 includes a differential processing unit H1 that performs differential processing of an input signal. The information processing device 11 includes at least one of the following: an extremum calculation unit 232 that calculates the extremum of the input signal, an inflection point calculation unit 234 that calculates the inflection point of the input signal, and a second derivative extremum calculation unit 236 that calculates the extremum of the second derivative of the input signal. The information processing device 11 includes a distance calculation unit 237 that calculates at least one of the following: the distance between the approximate J point and the extremum point, the distance between the approximate J point and the inflection point, and the distance between the approximate J point and the extremum point of the second derivative. The information processing device 11 includes a distance comparison unit 238 that selects one point based on the distance calculated by the distance calculation unit 237. Therefore, the information processing device 11 can detect the J point of the cardiac measurement waveform with greater accuracy. Here, for example, any one of the extremum calculation unit 232, the inflection point calculation unit 234, and the second derivative extremum calculation unit 236 (that is, any two of the extremum, inflection point, and second derivative extremum) may be used, or any two may be used, or all of them may be used. The differential processing unit H1 performs the necessary number of differential operations on each of the extremum, inflection point, and second derivative extremum.

[0119] As an example, the information processing device 11 has the following configuration. The information processing device 11 includes a differential processing unit H1 that performs differential processing of the input signal. The information processing device 11 includes an extremum calculation unit 232 that calculates the extremum of the input signal. The information processing device 11 includes an inflection point calculation unit 234 that calculates the inflection point of the input signal. The information processing device 11 includes a second derivative extremum calculation unit 236 that calculates the extremum of the second derivative of the input signal. The information processing device 11 includes a distance calculation unit 237 that calculates the distance between the approximate J point and the extremum point, the distance between the approximate J point and the inflection point, and the distance between the approximate J point and the extremum point of the second derivative. The information processing device 11 includes a distance comparison unit 238 that compares these distances and selects one point. Therefore, the information processing device 11 can detect the J point of the cardiac measurement waveform with greater accuracy.

[0120] As an example configuration, the information processing device 11 has the following configuration. The differential processing unit H1 (differentiator 231) performs differentiation of the input signal and calculates the first derivative result. The extremum calculation unit 232 calculates the extremum based on the first derivative result. Therefore, the information processing device 11 can improve the detection accuracy of the J point by using the extremum point as a candidate for the J point.

[0121] As an example configuration, the information processing device 11 has the following configuration. The differential processing unit H1 (differentiator 233) performs the second derivative of the input signal and calculates the second derivative result. The inflection point calculation unit 234 calculates the inflection point based on the second derivative result. Therefore, the information processing device 11 can improve the detection accuracy of the J point by using the inflection point as a candidate for the J point.

[0122] As an example configuration, the information processing device 11 has the following configuration. The differential processing unit H1 (differentiator 235) performs the third derivative of the input signal and calculates the third derivative result. The second derivative extremum calculation unit 236 calculates the second derivative extremum based on the third derivative result. Therefore, the information processing device 11 can improve the detection accuracy of the J point by using the point of the second derivative extremum as a candidate for the J point.

[0123] As an example configuration, the information processing device 11 has the following configuration. The distance calculation unit 237 calculates the distance based on a predetermined priority order. Therefore, by calculating the distance between a candidate point J and an approximate point J based on the priority order, the information processing device 11 can reduce the processing load and processing speed, for example, when calculating the distance to select a point to be point J.

[0124] As an example configuration, the information processing device 11 has the following configuration. The distance comparison unit 238 selects one point based on a predetermined priority order. Therefore, by using a priority order for selecting the point to be point J in the information processing device 11, it is possible to reduce the processing load and processing speed when selecting the point to be point J, for example.

[0125] As an example configuration, the information processing device 11 has the following configuration. The cardiac measurement waveform is the waveform of an electrocardiogram. Therefore, the information processing device 11 can detect point J of the electrocardiogram waveform.

[0126] As an example configuration, the information processing device 11 has the following configuration. The cardiac measurement waveform is the waveform of a magnetocardiogram. Therefore, the information processing device 11 can detect point J of the magnetocardiogram waveform.

[0127] Furthermore, a program to realize the function of any component in any device described above may be recorded on a computer-readable recording medium, and that program may be loaded into a computer system and executed. Here, "computer system" includes hardware such as operating systems and peripheral devices. "Computer-readable recording medium" refers to portable media such as flexible disks, magneto-optical disks, ROMs, CDs (Compact Disc)-ROMs (Read Only Memory), and storage devices such as hard disks built into computer systems. Moreover, "computer-readable recording medium" also includes volatile memory within a computer system that acts as a server or client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line, which holds the program for a certain period of time. Such volatile memory may be, for example, RAM (Random Access Memory). The recording medium may be, for example, a non-temporary recording medium.

[0128] Furthermore, the above program may be transmitted from a computer system that stores this program in a memory device or the like to another computer system via a transmission medium or by transmission waves within the transmission medium. Here, the "transmission medium" for transmitting the program refers to a medium that has the function of transmitting information, such as a network like the Internet or a communication line like a telephone line. Also, the above program may be for the purpose of realizing a part of the functions described above. Furthermore, the above program may be one that can realize the functions described above in combination with a program already recorded in the computer system, a so-called differential file. A differential file may also be called a differential program.

[0129] Furthermore, the functions of any component in any device described above may be implemented by a processor. For example, each process in the embodiment may be implemented by a processor that operates based on information such as a program, and a computer-readable recording medium that stores information such as a program. Here, the processor may be implemented by implementing the functions of each part in separate hardware, or by implementing the functions of each part in integrated hardware. For example, the processor includes hardware, and the hardware may include at least one of a circuit that processes digital signals and a circuit that processes analog signals. For example, the processor may be configured using one or more circuit devices or one or both of one or more circuit elements mounted on a circuit board. An IC (Integrated Circuit) may be used as the circuit device, and a resistor or capacitor may be used as the circuit element.

[0130] Here, the processor may be, for example, a CPU. However, the processor is not limited to a CPU, and various types of processors may be used, such as a GPU (Graphics Processing Unit) or a DSP (Digital Signal Processor). The processor may also be, for example, a hardware circuit using an ASIC (Application Specific Integrated Circuit). Furthermore, the processor may be composed of, for example, multiple CPUs, or of hardware circuits using multiple ASICs. The processor may also be composed of, for example, a combination of multiple CPUs and hardware circuits using multiple ASICs. Furthermore, the processor may include, for example, one or more amplifier circuits or filter circuits that process analog signals.

[0131] While embodiments of this disclosure have been described in detail above with reference to the drawings, the specific configuration is not limited to these embodiments and may include designs and other elements that do not depart from the gist of this disclosure.

[0132] [Note] Examples of configurations 1 to 16 are shown.

[0133] [Configuration Example 1] An information processing device comprising: a signal division unit that divides an input signal of a cardiac measurement waveform at a division point between the R-wave peak point and the T-wave peak point; a curve approximation unit that performs curve approximation on each signal portion divided by the signal division unit; and a fitting evaluation unit that evaluates the fitting of the curve approximation performed by the curve approximation unit, wherein the division point is defined as the approximate J point when the result of the evaluation by the fitting evaluation unit satisfies predetermined conditions.

[0134] [Configuration Example 2] The information processing apparatus according to [Configuration Example 1], further comprising an approximate J-point storage unit for storing the approximate J-point.

[0135] [Configuration Example 3] The information processing apparatus according to [Configuration Example 1] or [Configuration Example 2], wherein the signal splitting unit shifts the position of the splitting point between the R-wave peak point and the T-wave peak point at a predetermined interval, the curve approximation unit performs the curve approximation for each of the splitting points, and among the plurality of splitting points, the splitting point for which the evaluation result satisfies the predetermined condition is set as the approximated J point.

[0136] [Configuration Example 4] The signal splitting unit sets the splitting point within a predetermined range set between the R-wave peak point and the T-wave peak point, as described in any one of [Configuration Example 1] to [Configuration Example 3].

[0137] [Configuration Example 5] The information processing device according to any one of [Configuration Example 1] to [Configuration Example 4], wherein the curve approximation unit performs the curve approximation using a quadratic curve for at least one of the signal portions.

[0138] [Configuration Example 6] The information processing apparatus according to any one of [Configuration Example 1] to [Configuration Example 5], wherein the curve approximation unit performs the curve approximation using two different curves for at least two of the signal portions.

[0139] [Configuration Example 7] An information processing apparatus according to any one of [Configuration Example 1] to [Configuration Example 6], comprising: a differential processing unit that performs differential processing of the input signal; at least one of the following: an extremum calculation unit that calculates the extremum of the input signal; an inflection point calculation unit that calculates the inflection point of the input signal; and a second differential extremum calculation unit that calculates the extremum of the second differential of the input signal; a distance calculation unit that calculates at least one of the distance between the approximate J point and the extremum point, the distance between the approximate J point and the inflection point, and the distance between the approximate J point and the extremum of the second differential; and a distance comparison unit that selects one point based on the distance calculated by the distance calculation unit.

[0140] [Configuration Example 8] The information processing apparatus according to [Configuration Example 7], wherein the differential processing unit performs differentiation of the input signal to calculate the first derivative result, and the extreme value calculation unit calculates the extreme value based on the first derivative result.

[0141] [Configuration Example 9] The information processing apparatus according to [Configuration Example 7] or [Configuration Example 8], wherein the differential processing unit performs the second derivative of the input signal to calculate the second derivative result, and the inflection point calculation unit calculates the inflection point based on the second derivative result.

[0142] [Configuration Example 10] The information processing apparatus according to any one of [Configuration Example 7] to [Configuration Example 9], wherein the differential processing unit performs the third derivative of the input signal to calculate the third derivative result, and the second derivative extremum calculation unit calculates the second derivative extremum based on the third derivative result.

[0143] [Configuration Example 11] The distance calculation unit calculates the distance based on a predetermined priority order, as described in any one of [Configuration Example 7] to [Configuration Example 10].

[0144] [Configuration Example 12] The distance comparison unit selects one point based on a predetermined priority order, as described in any one of [Configuration Example 7] to [Configuration Example 11].

[0145] [Configuration Example 13] The cardiac measurement waveform is an electrocardiogram waveform, as described in any one of [Configuration Example 1] to [Configuration Example 12].

[0146] [Configuration Example 14] The information processing device according to any one of [Configuration Example 1] to [Configuration Example 12], wherein the cardiac measurement waveform is a magnetocardiogram waveform.

[0147] It is also possible to provide a method of processing performed by an information processing device. [Configuration Example 15] An information processing method comprising: a signal splitting unit splits the input signal of a cardiac measurement waveform at a splitting point between the R-wave peak point and the T-wave peak point; a curve approximation unit performs curve approximation for each signal portion split by the signal splitting unit; a fitting evaluation unit evaluates the fitting of the curve approximation performed by the curve approximation unit; and the splitting point at which the result of the evaluation by the fitting evaluation unit satisfies predetermined conditions is set as the approximate J point.

[0148] It is also possible to provide a program that is executed by a computer (computer program). [Configuration Example 16] A program for a computer that enables the following functions: a function to divide the input signal of a cardiac measurement waveform at a dividing point between the R wave peak point and the T wave peak point; a function to perform curve approximation on each of the divided signal portions; a function to evaluate the fitting of the curve approximation; and a function to set the dividing point as the approximate J point when the result of the evaluation satisfies predetermined conditions.

[0149] 11, 11a... Information processing device, 131... Input unit, 132... Output unit, 133... Communication unit, 134... Storage unit, 135... Control unit, 136, 136a... J-point detection processing unit, 151... Operation unit, 152... Display unit, 211, 211a... Signal division unit, 212, 212a... Curve approximation unit, 213, 213a... Fitting evaluation unit, 214, 214a... Approximate J-point storage unit, 231, 233, 235... Differentiator, 232... Extreme value calculation unit, 234... Inflection point calculation unit, 236... Second differential extreme value calculation unit, 237... Distance calculation unit, 238... Distance comparison unit , A1, A11...point (extreme point), A2a, A2b, A2c, A12...point (inflection point), A3, A13...point (secondary differential extreme point), B1...R wave peak point, B2...T wave peak point, B11...Q wave point, C1, C2, C11, C12...curve, F1, F11, F12, F21, F31, F32, F33... Cardiac measurement waveform signal, g21... Quadratic differential curve, H1... Differential processing section, K11, K12, K13... Approximate J point, t1 to t6, t11, t12, t21, t22... Time, V1, V11, V12, V21... Division point, W1... Search range

Claims

1. An information processing device comprising: a signal division unit that divides an input signal of a cardiac measurement waveform at a division point between the R-wave peak point and the T-wave peak point; a curve approximation unit that performs curve approximation on each signal portion divided by the signal division unit; and a fitting evaluation unit that evaluates the fitting of the curve approximation performed by the curve approximation unit, wherein the division point is defined as the approximate J point when the result of the evaluation by the fitting evaluation unit satisfies predetermined conditions.

2. The information processing apparatus according to claim 1, further comprising an approximate J-point storage unit for storing the approximate J-point.

3. The information processing apparatus according to claim 1, wherein the signal splitting unit shifts the position of the splitting point between the R-wave peak point and the T-wave peak point at a predetermined interval, the curve approximation unit performs the curve approximation for each of the splitting points, and among the plurality of splitting points, the splitting point for which the evaluation result satisfies the predetermined condition is defined as the approximate J point.

4. The information processing apparatus according to claim 1, wherein the signal splitting unit sets the splitting point within a predetermined range set between the R-wave peak point and the T-wave peak point.

5. The information processing apparatus according to claim 1, wherein the curve approximation unit performs the curve approximation using a quadratic curve for at least one of the signal portions.

6. The information processing apparatus according to claim 1, wherein the curve approximation unit performs the curve approximation using two different curves for at least two of the signal portions.

7. An information processing apparatus according to claim 1, comprising: a differential processing unit for performing differential processing of the input signal; at least one of the following: an extremum calculation unit for calculating the extremum of the input signal; an inflection point calculation unit for calculating the inflection point of the input signal; and a second differential extremum calculation unit for calculating the extremum of the second derivative of the input signal; a distance calculation unit for calculating at least one of the distance between the approximate J point and the extremum point, the distance between the approximate J point and the inflection point, and the distance between the approximate J point and the extremum of the second derivative; and a distance comparison unit for selecting one point based on the distance calculated by the distance calculation unit.

8. The information processing apparatus according to claim 7, wherein the differential processing unit performs differentiation of the input signal to calculate the first derivative result, and the extreme value calculation unit calculates the extreme value based on the first derivative result.

9. The information processing apparatus according to claim 7, wherein the differential processing unit performs the second derivative of the input signal to calculate the second derivative result, and the inflection point calculation unit calculates the inflection point based on the second derivative result.

10. The information processing apparatus according to claim 7, wherein the differential processing unit performs the third derivative of the input signal to calculate the third derivative result, and the second derivative extreme value calculation unit calculates the second derivative extreme value based on the third derivative result.

11. The information processing apparatus according to claim 7, wherein the distance calculation unit calculates the distance based on a predetermined priority order.

12. The information processing apparatus according to claim 7, wherein the distance comparison unit selects one point based on a predetermined priority order.

13. The information processing device according to any one of claims 1 to 12, wherein the cardiac measurement waveform is an electrocardiogram waveform.

14. The information processing apparatus according to any one of claims 1 to 12, wherein the cardiac measurement waveform is a magnetocardiogram waveform.

15. An information processing method comprising: a signal splitting unit splits the input signal of a cardiac measurement waveform at a splitting point between the R-wave peak and the T-wave peak; a curve approximation unit performs curve approximation on each signal portion split by the signal splitting unit; a fitting evaluation unit evaluates the fitting of the curve approximation performed by the curve approximation unit; and the splitting point at which the result of the evaluation by the fitting evaluation unit satisfies predetermined conditions is defined as the approximate J point.

16. A program for a computer to implement the following functions: a function to divide the input signal of a cardiac measurement waveform at a dividing point between the R-wave peak and the T-wave peak; a function to perform curve approximation on each of the divided signal portions; a function to evaluate the fitting of the curve approximation; and a function to set the dividing point as the approximate J point when the result of the evaluation satisfies predetermined conditions.

Images