Patent Application
20230263402
This application claims priority of Taiwanese Invention Patent Application No. 111106429, filed on Feb. 22, 2022.
The disclosure relates to hemodynamic analysis, and more particularly to a method and a processing device of hemodynamic analysis for detecting a particular syndrome.
Conventional hemodynamic analysis may be utilized to facilitate detection of certain cardiovascular diseases such as hypertension, atherosclerosis, heart failure, etc.
Although hemodynamic analysis is commonly used in modern medicine, the same is not utilized in traditional Chinese medicine. An object of the disclosure is to provide a method and a processing device for hemodynamic analysis that may facilitate detection of syndromes from the perspective of traditional Chinese medicine, such as poor Qi-blood circulation.
Therefore, according to an aspect of the disclosure, a method for detecting a particular syndrome based on hemodynamic analysis is to be performed by a processor. The method includes steps of: obtaining a piece of hemodynamic data that represents a hemodynamic waveform and that is related to a testee; performing first moving average (MA) filtering on the hemodynamic waveform to obtain a first filtered waveform that corresponds to the hemodynamic waveform; using a sliding window algorithm to determine multiple troughs of the first filtered waveform that are each a diastolic nadir of the hemodynamic waveform in order to determine multiple waveform segments of the first filtered waveform that are each between adjacent two of the troughs; determining smoothness of the waveform segments; and determining a relation between the hemodynamic waveform and a particular syndrome based on the smoothness of the waveform segments, and generating a detection result indicating a possibility of the testee being afflicted with the particular syndrome based on the relation thus determined.
According to an aspect of the disclosure, a system for detecting a particular syndrome based on hemodynamic analysis includes a hemodynamic sensor and a processing device. The hemodynamic sensor is adapted to be positioned on a testee. The hemodynamic sensor includes a first connection module and a hemodynamic sensing module that is electrically connected to the first connection module. The hemodynamic sensing module is configured to detect a hemodynamic status of the testee in order to generate a piece of hemodynamic data that represents a hemodynamic waveform and that is related to the testee. The processing device is configured to communicate with the hemodynamic sensor. The processing device includes a storage module, a second connection module, a processor and an output module. The storage module stores an application program. The second connection module is configured to communicate with the first connection module. The processor is electrically connected to the storage module and the second connection module. The output module is electrically connected to the processor. The processor is configured to, upon reading and executing the application program stored in the storage module, obtain the piece of hemodynamic data from said hemodynamic sensor through said second connection module. The processor is further configured to perform first moving average (MA) filtering on the hemodynamic waveform to obtain a first filtered waveform that corresponds to the hemodynamic waveform. The processor is further configured to use a sliding window algorithm to determine multiple troughs of the first filtered waveform that are each a diastolic nadir of the hemodynamic waveform, in order to determine multiple waveform segments of the first filtered waveform that are each between adjacent two of the troughs. The processor is further configured to determine smoothness of the waveform segments. The processor is further configured to determine a relation between the hemodynamic waveform and a particular syndrome based on the smoothness of the waveform segments. The processor is further configured to generate a detection result indicating a possibility of the testee being afflicted with the particular syndrome based on the relation thus determined. The processor is further configured to control the output module to output the detection result.
Other features and advantages of the disclosure will become apparent in the following detailed description of the embodiment (s) with reference to the accompanying drawings, of which:
Before the disclosure is described in greater detail, it should be noted that where considered appropriate, reference numerals or terminal portions of reference numerals have been repeated among the figures to indicate corresponding or analogous elements, which may optionally have similar characteristics.
The hemodynamic sensor 110 is configured to be wearable by a testee (e.g., on the body of the testee). The hemodynamic sensor 110 includes a first connection module 111, and a hemodynamic sensing module 112 adapted to be positioned on the testee. The first connection module 111 supports at least one communication protocol. The at least one communication protocol may include but is not limited to the Internet Protocol (IP) and/or a short-distance wireless communication protocol such as, for example, a Bluetooth® Protocol or a near-field communication (NFC) protocol. The hemodynamic sensing module 112 is configured to detect a hemodynamic status of the testee in order to generate a piece of hemodynamic data that represents a hemodynamic waveform and that is related to the testee. Specifically, the hemodynamic sensing module 112 is configured to detect the mechanical action of the heart and blood flow of the testee, and to generate hemodynamic data based on the mechanical action thus detected. According to an embodiment of the disclosure, the hemodynamic sensor 110 may be a photoplethysmogram (PPG) sensor, and the hemodynamic data generated by the PPG sensor may be a PPG signal. The hemodynamic sensor 110 is further configured to output the hemodynamic data thus generated to the processing device 120 through the first connection module 111.
The processing device 120 may be a computing system such as a smart phone, a personal computer (PC) , a laptop computer, a tablet computer, an ultra-mobile PC (UMPC), a personal digital assistant (PDA) , or a cloud server. The processing device 120 includes a storage module 121, a second connection module 122, an output module 124, and a processor 123 that is electrically connected to the storage module 121, the second connection module 122 and the output module 124. The storage module 121 stores an application program. The second connection module 122 also supports the at least one communication protocol (e.g., the Internet Protocol and/or the short-distance wireless communication protocol), and is configured to communicate with the first connection module 111 of the hemodynamic sensor 110 through the at least one communication protocol, so that the processing device 120 may receive the hemodynamic data from the hemodynamic sensor 110 and analyze the hemodynamic data thus received. The processor 123 is configured to implement a method of hemodynamic analysis for detecting a particular syndrome, e.g., poor Qi-blood circulation from the perspective of traditional Chinese medicine, by reading and executing the application program stored in the storage module 121. The output module 124 is controlled by the processor 123 to visually and/or audibly output a detection result generated by the processor 123 implementing the method. In an embodiment where the processing device 120 is a laptop computer, the storage module 121 is the memory, the second connection module 122 is the communication module, the processor 123 is the central processing unit (CPU), and the output module 124 is the screen and the speaker of the laptop computer.
In Step 21, the processor 123 obtains a piece of hemodynamic data that is generated by the hemodynamic sensor 110 which is worn on a testee. The piece of hemodynamic data represents a hemodynamic waveform and is related to the testee. In an embodiment where the hemodynamic sensor 110 is a PPG sensor, the piece of hemodynamic data obtained by the processor 123 is a PPG signal.
In Step 22, the processor 123 performs first moving average (MA) filtering on the hemodynamic waveform to obtain a first filtered waveform that corresponds to the hemodynamic waveform. In an embodiment, the processor 123 performs the first MA filtering by performing zero-phase digital filtering on the piece of hemodynamic data with an infinite impulse response (IIR) Butterworth bandpass filter, and the first filtered waveform is obtained by using a filtering criterion that is a frequency range from 0.5 Hz to 15 Hz for the Butterworth bandpass filter.
In Step 23, the processor 123 determines multiple troughs of the first filtered waveform that are each a diastolic nadir of the hemodynamic waveform, which is representative of a diastole during a heartbeat interval, in order to determine multiple waveform segments of the first filtered waveform that are each between adjacent two of the troughs and that each correspond to a pulse cycle (also known as cardiac cycle). In particular, in some embodiments, the processor 123 uses a sliding window algorithm to determine the troughs, so as to determine the waveform segments. Specifically, the processor 123 performs the sliding window algorithm by defining a window that has a particular length (e.g., 10 seconds), and applying the window multiple times respectively on multiple portions of the first filtered waveform to find, each time the window is applied on a portion of the first filtered waveform, a lowest one of data points on the first filtered waveform within the window to serve as one of the troughs, wherein the data points each have a first differential value equaling zero. The window is initially applied at the beginning of the first filtered waveform, and then is gradually moved toward the end of the first filtered waveform until the end is reached. It is known that a normal pulse cycle falls within a range between about 0.3 seconds and 1.5 seconds. If the troughs thus determined do not have intervals in said range, the length of the window is adjusted (e.g., reduced by 0.5 seconds) and Step 23 is repeated with the window having the adjusted length.
After Step 23, a first procedure for detecting the particular syndrome and a second procedure for evaluating vascular elasticity and deep sleep quality with respect to last night's sleep are performed. The first procedure includes Steps 24-26, and the second procedure includes Steps 27-29. It is noted that although the first and second procedures are illustrated in
In Step 24, the processor 123 determines, for each of the waveform segments, a systolic peak within the waveform segment that is a crest within the waveform segment and nearest a starting point (also referred to as “onset”) of the waveform segment, so that the waveform segment may be divided into a first portion and a second portion based on the systolic peak. Specifically, the first portion is from the onset of the waveform segment to the systolic peak of the waveform segment, and the second portion is the rest of the waveform segment.
In Step 25, the processor 123 determines smoothness of the waveform segments. According to some embodiments, Step 25 may include sub-steps 41-47 illustrated in
Referring to
In Sub-step 42, the processor 123 obtains a subtracted waveform by subtracting one of the first filtered waveform and the second filtered waveform from the other of the first filtered waveform and the second filtered waveform. The subtracted waveform includes multiple subtracted waveform segments respectively corresponding to the waveform segments of the first filtered waveform.
In Sub-step 43, for each of the subtracted waveform segments of the subtracted waveform, the processor 123 calculates a standard deviation value of magnitudes of data points on the subtracted waveform segment.
In Sub-step 44, the processor 123 calculates a mean value of the standard deviation values calculated for the subtracted waveform segments.
In Sub-step 45, the processor 123 compares the mean value thus calculated with a threshold value, in order to determine whether the mean value exceeds the threshold value.
When it is determined in Sub-step 45 that the mean value exceeds the threshold value, the process goes to Sub-step 46 where the processor 123 determines that a percentage (also referred to as “not-smooth percentage”) of the waveform segments that are not smooth in the entirety of the waveform segments meets or exceeds a threshold percentage (that is, at least the threshold percentage of the waveform segments of the first filtered waveform are not smooth) ; otherwise, the process goes to Sub-step 47 where the processor 123 determines that the not-smooth percentage is smaller than the threshold percentage. In an embodiment, the threshold value used in Sub-step 45 is 0.005, and the threshold percentage is 50%, but the disclosure is not limited thereto.
Returning to
Part of a first filtered waveform that is derived from a hemodynamic waveform that is related to a healthy person (i.e., someone who does not have poor Qi-blood circulation) is exemplarily illustrated in
Part of a first filtered waveform that is derived from a hemodynamic waveform that is highly related to poor Qi-blood circulation is exemplarily illustrated in
Reference is now made to the second procedure illustrated in
In Step 28, the processor 123 determines a confirmation result by, for each of the waveform segments of the first filtered waveform, determining whether the waveform segment includes a dicrotic notch and a dicrotic pulse based on the approximate curve of the waveform segment. According to some embodiments, the dicrotic notch and the dicrotic pulse may also be determined based on an article entitled “A Robust PPG Time Plane Feature Extraction Algorithm for Health Monitoring Application” by Abhishek Chakraborty et al. (e.g., by using an algorithm configured based on the article). The confirmation result may indicate that none of the waveform segments includes the dicrotic notch and the dicrotic pulse, or that at least a portion of the waveform segments includes the dicrotic notch and the dicrotic pulse.
In Step 29, the processor 123 generate an evaluation result with respect to at least one of the vascular elasticity and the deep sleep quality (with respect to last night's sleep) related to the testee based on the confirmation result determined in Step 28, and controls the output module 124 to visually and/or audibly output the evaluation result.
Specifically, when the confirmation result indicates that none of the waveform segments includes the dicrotic notch and the dicrotic pulse, the evaluation result generated by the processor 123 indicates vascular sclerosis (or low vascular elasticity).
On the other hand, when the confirmation result indicates that at least a portion of the waveform segments includes the dicrotic notch and the dicrotic pulse, the processor 123 locates, for each dicrotic notch in the waveform segments, a notch point which is a turning point at the bottom of the dicrotic notch, calculates a mean value of the magnitude(s) of the notch point(s) thus located for the dicrotic notch(es) , and evaluates the deep sleep quality based on the mean value thus calculated. In addition, the processor 123 determines, for each dicrotic pulse in the waveform segments, a slope of a leading edge of the dicrotic pulse, and evaluates the vascular elasticity based on the slope(s) thus determined for the dicrotic pulse(s). In some embodiments, the processor 123 compares the mean value of the magnitude(s) of notch point(s) with a predetermined threshold. The processor 123 may determine a poor deep sleep quality when the mean value exceeds the predetermined threshold, and determine a good deep sleep quality otherwise. In some embodiments, the processor 123 determines whether a majority of the slope(s) (e.g., more than 75%) of the leading edge(s) of the dicrotic pulse(s) has a positive value. The processor 123 may determine a good vascular elasticity when a majority of the slope(s) has a positive value (that is, when there are multiple dicrotic pulses, most leading edges thereof are literally rising upward) , and determine poor vascular elasticity otherwise.
Each of
According to some embodiments, in Step 29, the processor 123 may further determine, for each of the waveform segments, whether multiple dicrotic notches exist in the waveform segment, and determine a poor deep sleep quality when at least a portion of the waveform segments each include multiple dicrotic notches.
It is noted that alterations may be made to the method shown in
To sum up, by using the method and the system for hemodynamic analysis according to the disclosure, a particular syndrome (e.g., poor Qi-blood circulation) from the perspective of traditional Chinese medicine, as well as vascular elasticity and deep sleep quality, can be determined based on a piece of hemodynamic data generated by a hemodynamic sensor, providing a more objective way to make diagnostic decisions. The method and the system can be used not only in medical facilities but also at home, allowing common people to understand his/her health condition, and providing auxiliary information when seeking medical treatment.
In the description above, for the purposes of explanation, numerous specific details have been set forth in order to provide a thorough understanding of the embodiment(s). It will be apparent, however, to one skilled in the art, that one or more other embodiments may be practiced without some of these specific details. It should also be appreciated that reference throughout this specification to “one embodiment,” “an embodiment,” an embodiment with an indication of an ordinal number and so forth means that a particular feature, structure, or characteristic may be included in the practice of the disclosure. It should be further appreciated that in the description, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of various inventive aspects, and that one or more features or specific details from one embodiment may be practiced together with one or more features or specific details from another embodiment, where appropriate, in the practice of the disclosure.
While the disclosure has been described in connection with what is (are) considered the exemplary embodiment(s), it is understood that this disclosure is not limited to the disclosed embodiment(s) but is intended to cover various arrangements included within the spirit and scope of the broadest interpretation so as to encompass all such modifications and equivalent arrangements.
| Number | Date | Country | Kind |
|---|---|---|---|
| 111106429 | Feb 2022 | TW | national |
