TWI702936B - Pulse condition analysis method, prediction model establishment method and system based on acupoint resistance and blood pressure wave - Google Patents
Pulse condition analysis method, prediction model establishment method and system based on acupoint resistance and blood pressure wave Download PDFInfo
- Publication number
- TWI702936B TWI702936B TW108134822A TW108134822A TWI702936B TW I702936 B TWI702936 B TW I702936B TW 108134822 A TW108134822 A TW 108134822A TW 108134822 A TW108134822 A TW 108134822A TW I702936 B TWI702936 B TW I702936B
- Authority
- TW
- Taiwan
- Prior art keywords
- blood pressure
- pressure wave
- pulse
- data
- acupoint
- Prior art date
Links
Images
Landscapes
- Finger-Pressure Massage (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
一種基於穴道電阻與血壓波的脈象分析方法、預測模型建立方法及其系統,該系統包括有一穴道電阻量測器、一血壓波量測器以及一判斷模組。該穴道電阻量測器用以獲得受測者的穴道電阻資料;該血壓波量測器用以獲得該受測者的血壓波資料;該判斷模組用以接收該穴道電阻資料以及該血壓波資料,並輸出一脈象預測結果。透過上述設計,可應用於預測受測者的脈象種類、預測受測者未來健康狀況、或是進行受測者疾病風險預測。A pulse condition analysis method, prediction model establishment method and system based on acupoint resistance and blood pressure wave. The system includes an acupoint resistance measuring device, a blood pressure wave measuring device and a judgment module. The acupoint resistance measuring device is used to obtain the acupoint resistance data of the subject; the blood pressure wave measuring device is used to obtain the blood pressure wave data of the subject; the judgment module is used to receive the acupoint resistance data and the blood pressure wave data, And output a pulse prediction result. Through the above design, it can be applied to predict the type of pulse of the testee, predict the future health status of the testee, or predict the disease risk of the testee.
Description
本發明係與脈象分析方法與系統有關;特別是指一種基於穴道電阻與血壓波的脈象分析方法、預測模型建立方法及其系統。The invention relates to a pulse condition analysis method and system; in particular, it refers to a pulse condition analysis method, a prediction model establishment method and a system based on acupoint resistance and blood pressure wave.
隨著科學、醫學的發展與進步,許多的醫學研究也開始嘗試歸納患者的症狀、生理指標、臨床數據等,試圖建立出可預測疾病風險的模型,或是找出相同或不同疾病的患者之間生理狀態的關聯性。With the development and progress of science and medicine, many medical researches have also begun to try to summarize the symptoms, physiological indicators, clinical data, etc. of patients, trying to establish a model that can predict disease risk, or to find out the patients with the same or different diseases. The correlation between physiological states.
因此,如何透過量測受測者的生理數據或生命徵象進行生理狀態評估、疾病預測,藉以提供後續臨床診斷的參考數據,或提供使用者早期的生理狀態異常示警,是發明人所苦心研究的方向之一。Therefore, how to perform physiological state assessment and disease prediction by measuring the physiological data or vital signs of the subject, so as to provide reference data for subsequent clinical diagnosis, or to provide early warning of abnormal physiological state of the user, is the inventor’s painstaking research One of the directions.
過去曾有業者發展出可進行多點感測的脈診儀,透過多點感測方式,測得受測者的脈象。然而,這種多點感測的脈診儀不但單價高而且不便攜帶,難以普及;另外,根據發明人的研究成果與實驗數據顯示,單靠脈診儀所測得的脈波資料或脈搏波資料,而沒有搭配其他生理訊號作為判讀參考基礎的話,其資料量與受測者的脈象或者其他生理狀態的關聯性並不足以支撐準確的脈象預測、生理狀態判讀與疾病預測,而有需要改進的地方。In the past, an industry has developed a pulse diagnosis instrument that can perform multi-point sensing, through which the pulse condition of the subject can be measured. However, this kind of multi-point sensing pulse diagnosis instrument is not only expensive, but also not portable, and it is difficult to popularize; in addition, according to the inventor's research results and experimental data, the pulse wave data or pulse wave measured by the pulse diagnosis instrument alone Data, without other physiological signals as the reference basis for interpretation, the correlation between the amount of data and the subject’s pulse or other physiological state is not sufficient to support accurate pulse prediction, physiological state interpretation and disease prediction, and needs improvement The place.
以中醫的觀點而言,疾病的發生起因於人體的平衡遭受破壞,例如:心腎不交、肝火犯胃等,而如何得知人體的平衡遭受破壞,則可透過觀察人體外部徵象推知,其中由中醫師對病患把脈,體察脈象變化進行脈診,一直是中醫治療中很重要的方法,可以說是中醫臨床診斷的必要手段之一;另外,利用科學儀器量測人體之皮膚電阻,並根據皮膚電阻的量測結果,對人體進行健康評估,探討皮膚電阻與疾病之關聯性,則是屬於西醫系統所研究的主題。其中,中醫與西醫分別屬於兩個觀點截然不同的醫療體系,在過去的研究當中,並沒有同時參酌受測者的脈波(或稱脈搏波、血壓波)以及穴道電阻資料,來預測受測者的脈象或者對生理狀態進行評估或疾病風險預測的技術。From the perspective of Chinese medicine, the occurrence of diseases is caused by the destruction of the balance of the human body, such as: heart and kidney failure, liver fire invading the stomach, etc. How to know that the balance of the human body is damaged can be inferred by observing the external signs of the human body. It has always been a very important method in TCM treatment by a Chinese physician to take the pulse of the patient and observe the changes in the pulse condition. It can be said to be one of the necessary methods for clinical diagnosis in TCM. In addition, scientific instruments are used to measure the skin resistance of the human body. According to the measurement results of skin resistance, the health assessment of the human body and the discussion of the correlation between skin resistance and disease are the subjects of the Western medicine system. Among them, Chinese medicine and Western medicine belong to two medical systems with completely different viewpoints. In the past research, the pulse wave (or pulse wave, blood pressure wave) and acupoint resistance data of the testee were not considered at the same time to predict the test The pulse condition of the patient or the technology for assessing the physiological state or predicting disease risk.
有鑑於此,本發明之目的在於,以科學的方式融合中、西醫的優點,一併判讀受測者的穴道電阻與血壓波資料,並基於所量測的穴道電阻資料與皮膚電阻資料預測受測者的脈象,或者更進一步地基於穴道電阻以及皮膚電阻的量測資料,判讀受測者當前的生理狀態、健康狀況或是疾病風險預測。In view of this, the purpose of the present invention is to integrate the advantages of Chinese and Western medicine in a scientific way, to interpret the testee’s acupoint resistance and blood pressure wave data together, and to predict the pressure based on the measured acupoint resistance data and skin resistance data. The pulse condition of the examinee, or further based on the measurement data of acupoint resistance and skin resistance, determines the examinee's current physiological state, health status or disease risk prediction.
緣以達成上述目的,本發明提供一種基於穴道電阻與血壓波的脈象分析方法,其包括有以下步驟:一判斷模組接收一受測者的穴道電阻資料與血壓波資料;該判斷模組以該穴道電阻資料以及該血壓波資料為輸入,基於一預測模型,進行預測結果判定,以得出對應該受測者之生理狀態的脈象預測結果。In order to achieve the above objective, the present invention provides a pulse analysis method based on acupoint resistance and blood pressure wave, which includes the following steps: a judgment module receives the acupoint resistance data and blood pressure wave data of a subject; the judgment module uses The acupoint resistance data and the blood pressure wave data are input, and the prediction result is judged based on a prediction model to obtain the pulse prediction result corresponding to the physiological state of the subject.
緣以達成上述目的,本發明另提供一種基於穴道電阻與血壓波的生理狀態的預測模型建立方法,其包括有以下步驟:獲得複數個訓練樣本,該些訓練樣本記錄有穴道電阻資料與血壓波資料;透過一分類器對該些訓練樣本進行分類訓練,以取得複數個樣本特徵;基於該些樣本特徵產生一預測模型,該預測模型用於輸入一受測者穴道電阻資料與血壓波資料,並據以得出對應該受測者之生理狀態的脈象預測結果。In order to achieve the above objective, the present invention also provides a method for establishing a prediction model based on the physiological state of the acupoint resistance and blood pressure wave, which includes the following steps: obtaining a plurality of training samples, the training samples record the acupoint resistance data and the blood pressure wave Data; the training samples are classified and trained by a classifier to obtain a plurality of sample characteristics; a prediction model is generated based on the sample characteristics, and the prediction model is used to input a testee’s acupoint resistance data and blood pressure wave data, And according to it, the result of the pulse prediction corresponding to the physiological state of the subject is obtained.
緣以達成上述目的,本發明另提供一種基於穴道電阻與血壓波的生理狀態分析系統,其包括有:一穴道電阻量測器,用以獲得一受測者的穴道電阻資料;一血壓波量測器,用以獲得該受測者的血壓波資料;一判斷模組,用以接收該穴道電阻資料以及該血壓波資料,並輸出一脈象預測結果。In order to achieve the above objective, the present invention also provides a physiological state analysis system based on acupoint resistance and blood pressure wave, which includes: an acupoint resistance measuring device to obtain acupoint resistance data of a subject; a blood pressure wave volume The detector is used to obtain the blood pressure wave data of the subject; a judging module is used to receive the acupoint resistance data and the blood pressure wave data, and output a pulse prediction result.
本發明之效果在於,藉由讀取受測者的穴道電阻資料與血壓波資料,便可得到對應受測者之脈象預測結果,而所述的脈象預測結果可應用於了解受測者當前生理狀態是否正常、預測受測者未來健康狀況、或是進行受測者疾病風險預測。The effect of the present invention is that by reading the testee’s acupoint resistance data and blood pressure wave data, the pulse condition prediction result corresponding to the test subject can be obtained, and the pulse condition prediction result can be used to understand the subject’s current physiology Whether the state is normal, predict the future health status of the subject, or predict the risk of the subject’s disease.
為能更清楚地說明本發明,茲舉一實施例並配合圖式詳細說明如後。請配合圖1所示,本發明之基於穴道電阻與血壓波的生理狀態分析系統基本上包括有一穴道電阻量測器10、一血壓波量測器20以及一判斷模組30,該穴道電阻量測器10用以獲得一受測者的穴道電阻資料;該血壓波量測器20用以獲得該受測者的血壓波資料;該判斷模組30用以接收該穴道電阻資料以及該血壓波資料,並輸出一脈象預測結果。於一實施例中,該判斷模組30係以該穴道電阻資料以及該血壓波資料作為輸入,基於一預測模型,進行預測結果判定,以得出對應受測者之生理狀態的脈象預測結果。In order to explain the present invention more clearly, an embodiment is given in conjunction with the drawings in detail as follows. Please cooperate with FIG. 1, the physiological state analysis system based on the acupoint resistance and blood pressure wave of the present invention basically includes an acupoint
其中,所述預測模型係由預先訓練而得的。請配合圖2所示,該預測模型的建立方法包括有以下步驟:首先,係先獲取複數個訓練樣本該些訓練樣本記錄有穴道電阻資料與血壓波資料。其中,所述訓練樣本的來源可以是從醫療體系所提供的,例如從醫院、診所、醫療研究機構的臨床研究或診斷數據,或者是如醫師、護理師等護理人員以觸診或藉由儀器或其他方式量測而得。另外,所述的訓練樣本也可以是自多個確診罹患疾病的病人的生理狀態檢測結果收集而來的,其中,較佳者,所述病人的生理狀態檢測結果包括有經中醫師對病人進行脈診的診斷結果,藉以蒐集出包括有至少一種脈象資料的檢測結果,其中,所述的脈象可以但不限於包括有浮脈、散脈、沉脈、伏脈、遲脈、緩脈、華脈、弦脈、緊脈、結脈等。Wherein, the prediction model is obtained by pre-training. As shown in Figure 2, the method of establishing the prediction model includes the following steps: First, a plurality of training samples are obtained. These training samples are recorded with acupoint resistance data and blood pressure wave data. Wherein, the source of the training samples may be provided by the medical system, for example, clinical research or diagnostic data from hospitals, clinics, and medical research institutions, or nurses such as doctors, nurses, etc. who use palpation or equipment Or measured in other ways. In addition, the training samples may also be collected from the physiological state test results of multiple patients who have been diagnosed with the disease. Preferably, the patient’s physiological state test results include a medical examination conducted by a Chinese physician on the patient. The diagnosis result of pulse diagnosis is used to collect test results including at least one kind of pulse condition data, wherein the pulse condition may include, but is not limited to, floating pulse, scattered pulse, sinking pulse, Fu pulse, delayed pulse, slow pulse, Hua Pulse, xuan pulse, tight pulse, knot pulse, etc.
而後,再透過一分類器對該些訓練樣本進行訓練,以取得複數個樣本特徵。舉例而言,請參照圖3所示,可基於多筆已知脈象結果(例如弦脈、平脈等)的訓練樣本(例如訓練樣本1、訓練樣本2、訓練樣本3、…)進行訓練,接著,再基於該些樣本特徵產生一預測模型,該預測模型用於輸入受測者的穴道電阻資料與血壓波資料,並據以得出對應受測者之生理狀態的脈象預測結果。其中,所述的預測模型係隨著訓練樣本的不同而可進行不同的脈象預測或脈象判讀,而上述的脈象預測結果除了可用來預測受測者的脈象之外,當所使用的訓練樣本有納入一些確診罹患疾病之患者的資料時,亦可使用於判斷受測者是否罹患疾病或是疾病風險預測。舉例而言,當訓練樣本是從心血管疾病之患者所量測的生理資訊而得,所述生理資訊可以是但不限於穴道電阻值、血壓波值,除了可判斷受測者的脈象種類之外,亦可一併判讀受測者是否罹患心血管疾病,或者是否為潛在罹患心血管疾病的高危險群;假設訓練樣本是糖尿病患者所量測之生理資訊而得,除了可判斷受測者的脈象種類之外,亦可判讀受測者是否罹患糖尿病,或者是否為潛在罹患糖尿病的高危險群。也就是說,根據不同的訓練樣本,可建立判讀不同脈象種類的預測模型,甚至是建立判斷受測者是否罹患疾病或者疾病風險預測的預測模型,例如可以是但不限於製作預測慢性腎病、大腸直腸癌的預測模型。Then, the training samples are trained through a classifier to obtain a plurality of sample features. For example, referring to Figure 3, training can be performed based on multiple training samples (such as
於本實施例中,所述預測模型係用來預測受測者的脈象種類。請一併配合本案圖4及圖5所示,於進行本發明的脈象分析方法之前,係以穴道電阻量測器與血壓波量測器對受測者1進行量測,取得受測者1的穴道電阻資料與血壓波資料。In this embodiment, the prediction model is used to predict the pulse type of the subject. Please cooperate with Figure 4 and Figure 5 of this case together. Before performing the pulse analysis method of the present invention, the
其中,受測者1可藉由智慧型手機40自行量測,例如,智慧型手機40可連接貼片形式的穴道電阻量測器10,而後,再將穴道電阻量測器10接觸受測者1的經絡穴位處的皮膚而得,例如,可量測人體左右各十二條經脈的24個代表穴位點,而於一實施例中,所述的穴道電阻資料也可以是量測人體單一側(左側或右側)的十二條經脈的十二個穴位點,或者是量測手三陰經中的至少一者、手三陽經中的至少一者、足三陽經中的至少一者以及足三陰經中的至少一者,不過較佳者,則是量測全部十二條經脈的十二個穴位點。另外,於一實施例中,所述的穴道電阻量測器10也可以是獨立的一套設備,並於量測完受測者1的穴道電阻資料後,再透過有線或是無線傳輸的方式傳送給智慧型手機40。Among them, the
另外,所述的血壓波(或稱脈波、脈膊波)資料可以是但不限於使用光電測量法,例如可以光體積變化描述波形(photoplethysmography,簡稱 PPG)取得的PPG訊號,舉例而言,於操作上,可以是但不限於利用受測者1之智慧型手機40上的鏡頭貼近受測者1的身體,例如手指、手腕或其他部位,當脈搏跳動時,血管會產生微小波動,再以鏡頭記錄流經量測部位之血液光強度的週期變化,這些光訊號的周期性變化可代表受測者1血壓、脈搏搏動的訊號,再將這些光訊號轉化成電訊號,以獲取該使用者的血壓波資料。但於其他應用上,血壓波資料的取得方式並不以此為限,亦可以透過如Apple Watch等穿戴式電子裝置量測受測者1的血壓波資料。In addition, the blood pressure wave (or pulse wave, pulse wave) data can be, but not limited to, the use of photoelectric measurement methods, such as PPG signals obtained by photoplethysmography (PPG), for example, In operation, it can be, but not limited to, using the lens on the
於智慧型手機40接收到受測者1的穴道電阻資料與血壓波資料後,可傳送給判斷模組30進行判讀,以基於預測模型進行預測結果判定,得出對應受測者1之生理狀態的脈象預測結果。例如於一實施例中,智慧型手機40可連線至一雲端伺服器50,受測者1可以將穴道電阻資料與血壓波資料上傳至雲端伺服器50,由該雲端伺服器50的判斷模組30進行預測結果判定,而後再將預測結果回覆予智慧型手機40,並於智慧型手機40的螢幕顯示,以供受測者1參考。於本實施例中,預測結果的判定結果係根據經絡儀理論以及王唯工教授所提出的血液循環共振理論中的C1-C10當中至少幾個特徵以及時域及/或頻域上的特徵,再藉由以支援向量機為例的分類器分類而得的預測結果,其中,就本實施例進行脈象分類的實施例而言,脈象判讀結果可能有數種可能,例如是平脈的機率為85%、弦脈的機率為10%、細脈的機率為3%、沉脈的機率為1%,請配合圖6所示,回覆至智慧型手機40顯示示的內容,可被設定成僅顯示可能性最高的脈象(例如僅顯示預測結果為平脈),或者除了顯示預測結果之外,也一併顯示各種脈象的可能性,並可一併顯示其機率,以供受測者1或醫護人員參考。After the
其中,於判斷模組30接收到穴道電阻資料以及血壓波資料後,可以是但不限於透過演算法提取訊號於時域及/或頻域上的特徵,例如可根據王唯工教授所提出的血液循環共振理論,再藉由如支援向量機(SVM)將受測者1所提供的穴道電阻資料與血壓波資料進行分類,以得出預測結果。Wherein, after the
舉例而言,於判斷模組30接收到資料後,於進行分類或預測前,所述的血壓波資料係經過時域及/或頻域上的訊號處理,例如在本實施例中,所述時域上的訊號處理包括有但不限於:對PPG訊號進行前處理,以除去訊號中的雜訊,例如;移除運動偽影(motion artifact),使用的手段可以是但不限於,利用週期移動平均濾波器(Periodic Moving Average Filter, PMAF)消除雜訊而獲得平均波形,例如可以分割原始PPG訊號為多個波形,並以每多個波形為一組執行一次PMAF,以獲得平均波形,再藉由動態時間校正(Dynamic Time Warping)對平均波形與原始PPG訊號波形進行比較,以除去與原始波形具有顯著差異的雜訊。前述對PPG訊號的前處理可以提升後續在提取PPG訊號時域上的特徵的準確率,但並非為必要的步驟。於消除雜訊後,可得到如圖7所示PPG訊號在時域上的波形,並進一步得到脈波收縮峰值(Pulse Wave Systolic Peak, PWSP)、脈波開始(Pulse Wave Begin, PWB)以及脈波舒張峰值(Pulse Wave Diastolic Peak, PWDP),並根據以上三個點,請一併配合圖7及圖8所示,依據下列的公式計算出PPG訊號在時域上的八個特徵。其中,公式(1)用於計算出PWSP和PWDP之間的波幅(Systolic Peak – Diastolic Peak Amplitude)。公式(2)用以計算PWDP和PWE之間的波幅(Diastolic Peak – Pulse Wave End Amplitude)。公式(3)為配合圖8所示計算PWSP的波峰夾角(Angle),其中請配合圖8所示,a為PWSP至動脈切跡(Dicrotic Notch)的距離,b為PWSP至PWB的距離,c為PWB至PWDP。公式(4)計算PWSP和PWB之間的波幅,以獲得脈波波幅(Pulse Wave Amplitude, PWA)。公式(5)計算PWSP和PWB的距離,以得出收縮相位(Systolic Phase)。公式(6)計算PWSP和PWE的距離,以得出舒張相位(Diastolic Phase)。公式(7)計算PWSP和PWDP的距離,以獲得脈波傳播時間(Pulse Propagation Time)。公式(8)用以計算相鄰兩個PWSP的距離,以獲得心跳間隔(Interbeat Interval)。For example, after the
而在頻域上的訊號處理方面,包括有藉由快速傅立葉轉換將時域上的PPG訊號轉換為頻域,並根據王唯工教授所提出的血液循環共振理論,提取C1-C10中的至少幾個特徵,例如取得C1-C7等七個特徵,其中,C1-C7分別代表肝經、腎經、脾經、肺經、胃經、膽經、膀胱經的特徵值。於後,再藉由如支援向量機(SVM)將經訊號處理後的穴道電阻資料與血壓波資料進行分類,以獲得預測結果。其中,前述建立預測模型的方法所使用的訓練樣本中的血壓波資料同樣經過如同前述的時域及/或頻域上的訊號處理,於後再透過分類器對該些經訊號處理後的訓練樣本進行訓練。 In terms of signal processing in the frequency domain, the PPG signal in the time domain is converted to the frequency domain by fast Fourier transform, and based on the blood circulation resonance theory proposed by Professor Wang Weigong, at least a few of C1-C10 are extracted Features, such as obtaining seven features such as C1-C7, where C1-C7 respectively represent the characteristic values of the liver meridian, kidney meridian, spleen meridian, lung meridian, stomach meridian, gallbladder meridian, and bladder meridian. Then, the signal-processed acupoint resistance data and blood pressure wave data are classified by means of, for example, a support vector machine (SVM) to obtain prediction results. Among them, the blood pressure wave data in the training samples used in the aforementioned method of establishing a predictive model is also subjected to the aforementioned signal processing in the time domain and/or frequency domain, and then the signal processed training is performed through the classifier Samples for training.
請配合下表一以及圖7所示,下表為透過隨機決策森林(random decision forests)來評估上述PPG訊號之時域與頻域之特徵,表格中所呈現的權重越高,表示該特徵影響判斷結果的重要性越高,當前的排序係以重要性順序由高至低依序排列。而表一特徵後所示數字(1)~(7)分別與公式(1)~(7)相對應。 表一: Please cooperate with Table 1 and Figure 7. The table below is the use of random decision forests to evaluate the time domain and frequency domain characteristics of the PPG signal. The higher the weight presented in the table, the greater the influence of the feature The higher the importance of the judgment result, the current ranking system is arranged in descending order of importance. The numbers (1)~(7) shown after the characteristics of Table 1 correspond to formulas (1)~(7) respectively. Table I:
另外,於一實施例中,智慧型手機40與雲端伺服器50的連線機制可以是在智慧型手機40上安裝有行動應用程式,智慧型手機40再透過行動應用程式將穴道電阻資料與血壓波資料上傳至雲端伺服器50進行預測結果判定。另外,於一實施例中,請配合圖9所示,所述行動應用程式可設計為提供有多個檢測選項供受測者1選擇,受測者1可點選其欲檢測的檢測項目,於後,智慧型手機40便可將受測者1的穴道電阻資料、血壓波資料以及受測者1所點選的檢測項目資訊上傳至雲端伺服器50,再由雲端伺服器50根據檢測項目資訊選擇對應的預測模型,再以受測者1的穴道電阻資料、血壓波資料套入該受測模型以取得得出對應受測者1之生理狀態的預測結果,並回應至智慧型手機40供受測者1參考。In addition, in one embodiment, the connection mechanism between the
另外,對所測得之穴道電阻資料與血壓波資料進行判讀的判斷模組並不以建置於雲端伺服器50的判斷模組30為限,於一實施例中,所述的判斷模組亦可內建於智慧型手機當中,或者使用智慧型手機的處理器執行判斷模組的功能,而在智慧型手機內可儲存有預測模型,藉此,使用者便可操作智慧型手機將穴道電阻資料與血壓波資料輸入智慧型手機所儲存的預測模型,並得出對應的脈象預測結果。In addition, the judgment module for interpreting the measured acupoint resistance data and blood pressure wave data is not limited to the
另外,請配合下表二所示,為發明人使用五次交叉驗證(5-fold cross-validation)方式來驗證本發明之基於穴道電阻與血壓波的脈象分析方法的準確率。其中,第一、二、三組為對照組,第四、五、六組為實驗組,由第一至第三組的驗證結果可以看出,當只有使用穴道電阻特徵或者是只有使用血壓波特徵(頻域或時域)時,其預測的準確率只有約62.5%、65%。反觀,當應用本發明之分析方法時,例如,第四組使用八個血壓波時域特徵以及十二個穴道電阻特徵時,其準確率則達到72.5%;第五組使用七個血壓波頻域特徵以及十二個穴道電阻特徵時,其準確率則達到75%;第六組使用八個血壓波時域特徵、七個血壓波頻域特徵等十五個血壓波特徵以及十二個穴道電阻特徵時,其準確率達90%。由此可見,運用本發明基於穴道電阻與血壓波的脈象分析方法在預測的準確率上,相較於只依據穴道電阻特徵或是只依據血壓波特徵而言,確實可顯著提升預測準確率。
表二
請配合圖10所示,為本發明之脈象分析系統與商用儀器ARDK(自動反射診斷系統, Auto-Reflex-Diagnostic-Kinetics )量測之多個穴道之肌電圖振幅的折線圖,由圖式可看出,本發明之脈象分析系統在量測穴道電阻上,與商用儀器ARDK有相當高的關聯性,而可佐證本發明之脈象分析系統具有高準確度。Please cooperate with Figure 10, which is a broken line graph of the electromyographic amplitude of multiple acupoints measured by the pulse analysis system of the present invention and the commercial instrument ARDK (Auto-Reflex-Diagnostic-Kinetics). It can be seen that the pulse analysis system of the present invention has a very high correlation with the commercial instrument ARDK in measuring the acupoint resistance, which can prove that the pulse analysis system of the present invention has high accuracy.
請配合圖11至圖13所示,本發明之脈象分析系統與ANS Watch 腕式生理監視器(型號TS-0411)量測PPG訊號之折線圖。其中,圖11至圖13分別比對了關於心律(HR)、心跳間期標準偏差(SDNN)以及交感/副交感平衡指標(LF/HF),由圖式可看出,本發明之脈象分析系統在量測PPG訊號上,與ANS Watch 腕式生理監視器具有相當高的關聯性,而可佐證本發明之脈象分析系統具有高準確度。Please cooperate with Figure 11 to Figure 13, the pulse analysis system of the present invention and the ANS Watch wrist-type bio-monitor (model TS-0411) to measure the line graph of the PPG signal. Among them, Figure 11 to Figure 13 respectively compare the heart rhythm (HR), the standard deviation of the heartbeat interval (SDNN) and the sympathetic/parasympathetic balance index (LF/HF). It can be seen from the diagram that the pulse analysis system of the present invention In measuring the PPG signal, it has a very high correlation with the ANS Watch wrist-type bio-monitor, which can prove that the pulse analysis system of the present invention has high accuracy.
本發明創新地採用人體的穴道電阻與血壓波資料作為判讀受測者脈象種類的依據,成功地融合中醫與西醫的優點,並提供使用者可以便利的進行脈象分析,甚至基於脈象分析結果進行生理狀態或疾病預測的評估,不需要使用笨重、龐大的量測儀器,也不需要特定的環境或場所才能進行評估,任何人不論何時、何地,只要有如智慧型手機等行動裝置並搭配穴道電阻量測器,都可以利用本發明所提供的系統進行脈象的預測,或者更進一步地進行生理狀態的分析或疾病的預測。而且上述簡單設備所建構出的脈象分析系統可得到堪比目前市面上昂貴的商用檢測儀器,足見本發明具有突破的創新。除此之外,本發明所提供的脈象分析方法、預測模型建立方法與系統除了可應用於預測脈象之外,亦可適用於其他應用,可以是但不限於預測疾病等,例如預測心血管疾病、糖尿病、肝病等其他疾病。例如,發明人將本發明的脈象分析系統應用在預測心衰竭上,基於127個測試對象,使用本發明的脈象分析系統評估出確實罹患心衰竭之對象的準確率也高達98%,由此可見,本發明確實亦可應用於預測或評估病患是否有罹患心衰竭等心血管疾病。The invention innovatively uses the human body’s acupoint resistance and blood pressure wave data as the basis for interpreting the type of pulse of the subject, successfully combines the advantages of traditional Chinese medicine and western medicine, and provides users with convenient pulse analysis, and even physiological analysis based on the results of pulse analysis. The evaluation of the state or disease prediction does not require the use of bulky and large measuring instruments, and does not require a specific environment or place to perform the evaluation. Anyone needs a mobile device such as a smartphone and acupoint resistance no matter when and where. The measuring device can use the system provided by the present invention to predict the pulse condition, or further analyze the physiological state or predict the disease. Moreover, the pulse condition analysis system constructed by the above-mentioned simple equipment can obtain commercial detection instruments that are comparable to those currently on the market, which shows that the present invention has a breakthrough innovation. In addition, the pulse condition analysis method, prediction model establishment method and system provided by the present invention can be applied to predict pulse condition as well as other applications, such as but not limited to predicting diseases, such as predicting cardiovascular diseases. , Diabetes, liver disease and other diseases. For example, the inventor applied the pulse analysis system of the present invention to predict heart failure. Based on 127 test subjects, the accuracy rate of using the pulse analysis system of the present invention to estimate that the subject is indeed suffering from heart failure is also as high as 98%. Indeed, the present invention can also be applied to predict or assess whether a patient has cardiovascular diseases such as heart failure.
以上所述僅為本發明較佳可行實施例而已,舉凡應用本發明說明書及申請專利範圍所為之等效變化,理應包含在本發明之專利範圍內。The above are only the preferred and feasible embodiments of the present invention. Any equivalent changes made by applying the specification of the present invention and the scope of the patent application should be included in the patent scope of the present invention.
[本發明] 1:受測者 10:穴道電阻量測器 20:血壓波量測器 30:判斷模組 40:智慧型手機 50:雲端伺服器 [this invention] 1: Subject 10: Acupoint resistance measuring device 20: Blood pressure wave measuring device 30: Judgment module 40: smart phone 50: Cloud server
圖1為本發明一實施例之脈象分析系統的方塊圖。 圖2為本發明一實施例之預測模型建立方法的流程圖。 圖3為本發明應用之類神經網路示意圖。 圖4本發明一實施例之脈象分析方法的流程圖。 圖5為本發明一實施例應用有脈象分析系統的示意圖。 圖6為智慧型手機的示意圖,揭示智慧型手機的螢幕。 圖7為PPG訊號在時域上的八個特徵的示意圖。 圖8為PWSP的角度示意圖。 圖9為智慧型手機的示意圖,揭示智慧型手機的螢幕。 圖10為本發明之脈象分析系統與商用儀器ARDK量測之多個穴道之肌電圖振幅的折線圖。 圖11至圖13分別為本發明之脈象分析系統與ANS Watch 腕式生理監視器量測PPG訊號之折線圖。 Figure 1 is a block diagram of a pulse analysis system according to an embodiment of the present invention. FIG. 2 is a flowchart of a method for establishing a prediction model according to an embodiment of the present invention. Figure 3 is a schematic diagram of a neural network such as the application of the present invention. Fig. 4 is a flowchart of a pulse analysis method according to an embodiment of the present invention. Figure 5 is a schematic diagram of an embodiment of the present invention applying a pulse condition analysis system. Figure 6 is a schematic diagram of a smart phone, revealing the screen of the smart phone. Figure 7 is a schematic diagram of the eight characteristics of the PPG signal in the time domain. Figure 8 is a schematic diagram of the angle of PWSP. Figure 9 is a schematic diagram of a smart phone, revealing the screen of the smart phone. Figure 10 is a broken line diagram of the electromyographic amplitude of multiple points measured by the pulse analysis system of the present invention and the commercial instrument ARDK. Figures 11 to 13 are respectively the line graphs of the PPG signal measured by the pulse analysis system of the present invention and the ANS Watch wrist-type bio-monitor.
10:穴道電阻量測器 10: Acupoint resistance measuring device
20:血壓波量測器 20: Blood pressure wave measuring device
30:判斷模組 30: Judgment module
Claims (10)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108134822A TWI702936B (en) | 2019-09-26 | 2019-09-26 | Pulse condition analysis method, prediction model establishment method and system based on acupoint resistance and blood pressure wave |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
TW108134822A TWI702936B (en) | 2019-09-26 | 2019-09-26 | Pulse condition analysis method, prediction model establishment method and system based on acupoint resistance and blood pressure wave |
Publications (2)
Publication Number | Publication Date |
---|---|
TWI702936B true TWI702936B (en) | 2020-09-01 |
TW202112304A TW202112304A (en) | 2021-04-01 |
Family
ID=73643888
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
TW108134822A TWI702936B (en) | 2019-09-26 | 2019-09-26 | Pulse condition analysis method, prediction model establishment method and system based on acupoint resistance and blood pressure wave |
Country Status (1)
Country | Link |
---|---|
TW (1) | TWI702936B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113080892A (en) * | 2021-03-22 | 2021-07-09 | 北京大学深圳研究生院 | Detection data processing method and system for risk prediction of cardiovascular and cerebrovascular diseases |
CN113229798A (en) * | 2021-05-18 | 2021-08-10 | 平安科技(深圳)有限公司 | Model migration training method and device, computer equipment and readable storage medium |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105530314A (en) * | 2015-12-30 | 2016-04-27 | 珠海欧雷伯电子科技有限公司 | Cloud health internet platform system based on cloud intelligent health mobile phone |
-
2019
- 2019-09-26 TW TW108134822A patent/TWI702936B/en active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105530314A (en) * | 2015-12-30 | 2016-04-27 | 珠海欧雷伯电子科技有限公司 | Cloud health internet platform system based on cloud intelligent health mobile phone |
Non-Patent Citations (1)
Title |
---|
Christoph Fischer , Benno Dömer , Thomas Wibmer , Thomas Penzel , "An Algorithm for Real-Time Pulse Waveform Segmentation and Artifact Detection in Photoplethysmograms", IEEE Journal of Biomedical and Health Informatics ( Volume: 21 , Issue: 2 , March 2017 ) ,Page(s): 372 – 381. * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113080892A (en) * | 2021-03-22 | 2021-07-09 | 北京大学深圳研究生院 | Detection data processing method and system for risk prediction of cardiovascular and cerebrovascular diseases |
CN113229798A (en) * | 2021-05-18 | 2021-08-10 | 平安科技(深圳)有限公司 | Model migration training method and device, computer equipment and readable storage medium |
CN113229798B (en) * | 2021-05-18 | 2023-08-22 | 平安科技(深圳)有限公司 | Model migration training method, device, computer equipment and readable storage medium |
Also Published As
Publication number | Publication date |
---|---|
TW202112304A (en) | 2021-04-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Charlton et al. | Breathing rate estimation from the electrocardiogram and photoplethysmogram: A review | |
US9220437B2 (en) | Diagnostic classifications of pulse signal waveform data | |
JP2021519621A (en) | A method for estimating blood pressure and arterial wall sclerosis based on photoplethysmographic (PPG) signals | |
Alafeef et al. | Smartphone-based respiratory rate estimation using photoplethysmographic imaging and discrete wavelet transform | |
JP2014193349A (en) | Device and method for monitoring physiological condition | |
Javaid et al. | Elucidating the hemodynamic origin of ballistocardiographic forces: Toward improved monitoring of cardiovascular health at home | |
JP2008295517A (en) | Analysis system and method of pulse diagnosis in doctor of chinese medicine | |
TWI702936B (en) | Pulse condition analysis method, prediction model establishment method and system based on acupoint resistance and blood pressure wave | |
Wang et al. | Quantitative comparison of the performance of piezoresistive, piezoelectric, acceleration, and optical pulse wave sensors | |
Vakulenkoa et al. | New application of blood pressure monitor with software environment Oranta-AO based on arterial oscillography methods | |
Elgendi et al. | A six-step framework on biomedical signal analysis for tackling noncommunicable diseases: Current and future perspectives | |
Shin et al. | Complementary photoplethysmogram synthesis from electrocardiogram using generative adversarial network | |
TW581671B (en) | Method and device for detecting ""yin"" and ""yang"" | |
WO2023026158A1 (en) | Methods and systems for engineering conduction deviation features from biophysical signals for use in characterizing physiological systems | |
Li et al. | The correlation study of Cun, Guan and Chi position based on wrist pulse characteristics | |
TW200841860A (en) | Analysis system and method for pulse diagnosis of Chinese medicine | |
Ab Hamid et al. | Methods of extracting feature from photoplethysmogram waveform for non-invasive diagnostic applications | |
Phan et al. | Noninvasive, wearable multi biosensors for continuous, long-term monitoring of blood pressure via internet of things applications | |
Yang et al. | Blood pressure estimation from photoplethysmogram and electrocardiogram signals using machine learning | |
More et al. | Development of non-invasive diagnostic tool for diseases using Photo Plethysmography | |
CN111345791B (en) | Pulse wave measuring device | |
Yamada et al. | Preliminary study of an objective evaluation method for pulse diagnosis using radial artery pulse measurement device | |
Wu et al. | An Ambulatory Blood Pressure Monitoring System Based on the Uncalibrated Steps of the Wrist | |
US20230129313A1 (en) | Method of detecting parameters indicative of activation of sympathetic and parasympathetic nervous systems | |
Tiloca | A machine learning approach for non-invasive blood pressure estimation |