200906367 九、發明說明： 【發明所屬之技術領域】 本發明係關於一種動脈硬化檢測技術，特別是關於一種 以脈波機率分析法計算血壓波增大指標作為評估動脈硬化之 檢測方法及其裝置。 【先前技術】 心血管疾病係人類發病率及死亡率的主要原因。造成心 血管疾病的主要原因為血管中層之彈性纖維板組織斷裂及破 碎造成之動脈硬化（arterial sclerosis)。動脈管壁中層組織厚度 (Intima Media Thickness ’ IM丁）、脈波傳遞速度（pu丨化BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an arteriosclerosis detecting technique, and more particularly to a method for detecting an arteriosclerosis by using a pulse wave probability analysis method for calculating a blood pressure wave increase index. [Prior Art] Cardiovascular disease is the main cause of human morbidity and mortality. The main cause of cardiovascular disease is arterial sclerosis caused by rupture and disruption of the elastic fiber sheet in the middle layer of the blood vessel. Intima Media Thickness (IM) and pulse wave velocity (pu 丨 丨)
Velocity，PWV)及脈搏波形係為醫界所普遍使用之動脈硬化 程度評估之指標。 一為了評估動脈之狀況，f知技術中有洲超音波儀器之 量測技術，例如中華民國發明公開第2〇〇5〇〇〇41號之「非侵 襲性血管機械特性檢測技術」，其係利用超音波儀器與壓力感 測單元同步監視血壓與血管管徑，並藉由適當的超音波固定 裝置與特殊雜處理技術，取得活動觸動脈血管機械性質， 並量測動脈管壁中層組織厚度。 又例如中華民國發明公開第2_5873號之「Velocity, PWV) and pulse waveform are indicators of the degree of arteriosclerosis commonly used in the medical community. In order to evaluate the condition of the arteries, the technique of measuring the ultrasonic instruments of the continent is known, for example, the "non-invasive vascular mechanical property detection technology" of the Republic of China Inventions No. 2〇〇5〇〇〇41 Ultrasonic instrument and pressure sensing unit are used to monitor blood pressure and blood vessel diameter simultaneously, and the mechanical properties of the active contact artery are obtained by appropriate ultrasonic fixation device and special miscellaneous treatment technology, and the thickness of the middle layer of the arterial wall is measured. For example, the Republic of China Invention Disclosure No. 2_5873
指數量測分析系統」，其係藉由量 U ..Ββ . 里幻不同位置動脈的血流脈波 波形時間差’以反應血管擴張曰 ，、诋私度而罝測出内皮細胞調節功 200906367 能’而作為檢測動脈硬化程度之評估依據。 在令華民國專利公告第514514號專利案中，揭露出— 種可供評估動脈硬化錢的裝置，其係在㈣者身體部位量 :出脈波’並藉由脈波速度相關資訊及脈波之振幅增大相關 資。扎與動脈&度硬化間之預定關係，而評估動脈硬化的程度。 在中華民國專利公告第546129號專利案中，揭露出— 種用於評估動脈硬化之技術，其制貞測受測者之第—部位與 第一σ卩位之脈搏波，並藉由一動脈硬化評估裝置根據該第— 部位之脈搏波振幅之增大指數與第：部位之脈搏波振幅之增 大指數相互比較，以評估該活體的動脈硬化。 中華民國專利公告第534807號專利案中，揭露出一種 立曰大扣标判疋裝置，其係利用一套囊壓力改變裝置改變一套 囊之壓力，並得出-高套囊壓力脈波與—低套囊壓力脈波， 並依據該低套囊脈波的各量值來決定一增大指數。 目前開發出許多可檢測動脈硬化以予貝測心全管疾病的方 法。其主要係依據由動脈所測量出之企壓波（bl〇〇d pressure 醫ef〇rm)來預測心血管疾病，此—方法稱為脈波分析法pwA (pulse wave analysis) ° 忒企塵波係由人體心室輸出（v_icular ⑽)所產生 之入射波（incident wave)與末梢反射之反射波（⑹如丨wave)重 疊所產生。由於所測量出之血壓波為—複合信號(_p()unding signal)，故脈波分析法能提供動脈血管硬化的資訊―。最常用 200906367 來表示動脈硬化指數為增大指標AIx (augmentation index)， 其係測量血壓波中反射波相對於總血壓波之強度。 典型的動脈血壓波概可分為波形A、B、C、D四種。第 1圖顯示波形A之動脈血壓波之波形圖及該血壓波對應之四 次微分之波形示意圖。第2圖顯示波形C之動脈血壓波之波 形圖及該血壓波對應之四次微分之波形示意圖。動脈血壓波 之波形A、B、C、D之定義分別如下： (1)波形A : 最大收縮壓 Ps (Peak systolic blood pressure)發生 在反曲點Pi (augmentation point)之後，增大指 標AIx定義為（ps — Pi)/(ps - Pd)，其中Pi為反 曲點之血壓，Pd為最低舒張壓（minimum diastolic pressure)。當增大指標AIx >0.丨2時，該血壓波 W1属於波形A，表示該血壓波之硬度較大。 (2)波形B : 最大收縮壓Ps亦發生在反曲點Pi之後的晚期收 縮壓’類似於波形A，但增大指標AIx介於〇 到0.12之間，表示該灰壓波之硬度較小。 (3)波形C : 最大收縮壓Ps發生在反曲點Pi之前，增大指標 AIx定義為（Pi - Ps)/(Ps - Pd)，所得到的增大指 標為負數’該血壓波W2屬於波形C，表示該動 脈相當具有彈性及健康。 (4)波形D : 其波形近似波形A，而增大指標AIx的定義與 波形A相同。然而，其反曲點pi無法·直接觀察 200906367 到’這是因為在血管收縮的早期，反射波已經 出現，亚且與入射波結合。顯然，增大指標Αΐχ 隨著動脈硬度由貞數（㈣c)上升至正數（波形 Α、Β及D)。許多研究顯示增大指標Αίχ在習 知脈波分析法PWA中係一個非常重要的指數’ 與動脈硬化及其他心血管問題有非常高的相關 性。 什异增大指標ΑΙχ的重點在於辨識血壓波的反曲點… 研j現4反曲點P!發生在動脈最大血液流速。換句話說， 血壓波及流速應同步測量，然後決定反曲，點Pi，計算出增大 , 技丁中，利用血壓波之微分（differentia】） 直接判斷反曲點Pi。接荽 .^ , 接者终夕的研究亦使用血壓波的微分 n ^ 曲點& ’以計算增大指標AIx。對於波形A、B及 八的Λ曲』1"1位於上升階段波形，係由血壓信號四次微 刀的第一個負5 • 、 令父越點（negative-to-positive zero crossing)來得到。對於 - 、皮形C，该反曲點Pi位於下降階段波 形對應於弟二個正至貞的零交越點。 【發明内容】 本發明所欲解決之技術問題: 雖然脈波分析法 已普遍用來檢測動脈硬化並據-以預測 心 200906367 血管疾病，然而習知脈波分析法有兩個主要的問題，首先， 脈波/刀析法中使用第四次微分的技術不能用來處理具有較高 取樣率或者純干擾的血黯。微分會產生更多雜訊波形了 可能會更難辨識出血壓波的反曲點pi。因此，在實際的應用 時’必須在電路中加入濾波器來渡除雜訊。雖然可用渡波哭 ，少雜訊、，但會增加運算的複雜性及渡掉一些有用資訊： ，-’血壓波的四次微分只能決定反曲點的位置，但無法 呈現在反曲點Pi附近的作號胜 ，，，i ^ 、' n ^ 就特欲，例如波形的寬度及形狀， 及其他具有生理重要性的波形特徵點。 緣此，實有必要針對習知的技術予以改良。本發明之主 大種不須使用微分方法即可計算出血壓波增 汁n/之^ —目的是提供—種可簡易計算出血虔波增大 才曰軚的動脈硬化檢測方法及其裝置。 i及^大 本备明之另一目的是提供— 動脈硬化檢測方法及其裝置。’具有較高的雜訊容忍力的 本發明解決問題之技術手段： 本發明為解決習知技術之 一種以脈波機率分析法評 751W狀技術手段係提供 其裝置包括有—血壓波量測單化，檢測方法及其裝置， 一血壓波正規化處理模組、一π、一血壓波數位化轉換電路、 列波模組、一增大指標計算單 200906367 元’其中該判波模組將正規化後的受測者▲壓波利用兩個 波參數來判別叙壓波的波形，其中第一個判波參數係為正手 化上升階段波形面積比值職，第二個判波參數為:規=見 升階段波形時間比值NUT。判波模組依據該判波參數判別血 壓波之波形分類後，取得該血壓波的機率分佈函數，並決定 局部高密度區(LHDR)在該機率分佈函數的位置， 2 算出該血壓波之增大指標，作為評估動脈硬化狀況。 。 本發明對照先前技術之功效： 相較於傳統的脈波分析法，本㈣具有之優點包括·· 〇)本發明脈波機率分析法可直接由血壓波之機率分佈函數 (PDF)的局部高密度區（LHDR)的特徵計算Αΐχ ; ⑺本發明脈波機率分析法具有較高_訊容忍力； (3)本發明脈波機率分析法能同步監視及定位出所有在 上之特徵點； ⑷本發明脈波解分析法可提供局部波形的詳細敘述，其可 進一步描繪出生理重要特徵； (5)树明脈波機率分析法亦可絲整合以血壓絲礎及以血 官形變為基礎的指數’以對動脈特徵作更精確的描述。本 么明脈波機率分析法在動脈血壓波的分析上是一項很實用 、方去在臨床檢測上具有很有的應用潛力。 本發明所_的具體實施例，將藉由以下之實.施例及附 10 200906367 呈圖式作進一步之說明。 【實施方式】 在本發明中，提出-種脈波機率分析法PPA (puhe爾6 probabUity analysis)的技術，利用本發明之方法可在動脈的血 塵波中找出反曲點Pi之位置及血壓波的特徵，並計算增大指 標ΑΙχ。 t曰 曰 本發明所使狀脈波機率分析法是依據血壓波之機率分 佈函數聊（pr〇bability density functi〇n)來計算出增大指標 尬。由於企壓波中反曲點pi附近的信號相較於該血墨波的 其它部份具有較小的振幅變化，亦即，血壓波中之反曲點h 在機率分佈純PDF之波形巾會呈現局部高密度區：腿 (locally high-density regi〇n)，如第 3 圖所示。 二：：令之反曲點pi是位在上升階段波形(波形A、 丄：==:=:rpDF會存在局部高 會呈現兩個主要的局部高密度區lhdr，，= :反:LP點之二於半、月__產生的切痕波 及左側局糊度//相剩她部高密度區 由以上之理論，可得到下列重點： 200906367 (υ血壓波中反曲點Pi之位置可由局部高密度區lhdr位在 機率分佈函數PDF㈣之位置而得至，卜並據以計算出增 大指標ΑΙχ。 (2) 由局部高密度區LHDR位在機率分佈函數pDF波形之上 升階段波形或下降階段波形之狀況，而可同時監_血壓 波中的生理重要資訊。 (3) 由局部高密度區lhDR之壯π ότ p &丨， 又ι_ 炙狀况可付到生理重要資訊的波 形特徵。 參閱第5圖所示’其顯示本發明以脈波機率分析法評估 動脈硬化之檢測裝置100之電路功能方塊圖’其電路主要包 括有一血壓波感測裝置1、一血壓波類比信號處理模组2、一 血壓波數位化轉換電路3、一血壓波正規化處理模植4、 波模組5、-增大指標計算單元6、一顯示裝置7。 血壓波感測裝置I县用；Λ、，1 來里測受測者血壓波，其可採用 目前普遍商用的習知血壓浊片，目丨。。 休用 „ &波感測為，例如壓脈袋、壓力感測 益、紅外線發射/接收感測器。 裝置1量測到的受測者血壓波信號S1經血 屋波類比Μ處理模組2進行類比信號之前置處理。血 類比信號處理模組2可句技古 良 大雷路22 —尨μ、+匕括有一前級濾波電路2!、一信號放 電路22、m慮波電路23等習知電路單元。 經過血壓波類比信笋虚 JU免里h組2後的受測者i壓波备再 經血壓波數位北轉換雷 ·-政曰丹 進行信號數位化的轉換.。血壓波 200906367 路1 2中可包括有一類比至數位轉換電路^-及去除波形低頻飄移部份。 复的處波 4中轉換之血壓波會駐域波正規化處理模組 j作W的正規化處I血_正規化處理—4包括有 早脾：期波形切割電路3卜血壓波振幅正規化電路42，可用 以將時間及血壓波之振幅予以正規化。 7 =之振幅正規化之後’會由判波模組5中的血摩波 上升階段波形面積比值計算單元51及血 =段波形時間比值計算單元52分別計算出正規化上= 波形面積比值職（騰malized ups加ke _ 上升階段波科㈣值NUT (騰malizedup_ketime) = 13 1 皮絲制早A 53依據該正規化上升階段波形面積比 及正規化上升階段波形時間比值NUT判別出該血壓 波之波形分類係屬第一分類C1或是第二分類c2。 2 …最後，增大指標計算單元6即可依據制出之血壓波之 ，形分類α、（：2取得該血壓波的機率分佈函數、決定局部 南密度區(LHDR)在該機率分佈函數的位置、以及計算出該血 壓波之增大指標。該計算出之結果可輸出至—顯示裝置7中 予以顯示，以作為評估動脈硬化狀況。 3 第6圖，其係本發明之流程圖。朗時參閱第$圖所示 之電路功能方塊圖對本發明之方法作進一步的說明_。首先係 200906367 以血壓波感測裝置量測受測者之血壓波（步驟Ι〇ι)，並將時間 及量測到之血壓波之振幅予以正規化（步驟1〇2)。 在血壓波之振幅正規化步驟中可採用設定最大收縮壓 等於1，設定最低舒張壓等於〇,以及設定波形周期τ等於卜 以將時間及血壓波之振幅予以正規化。在—方面，血壓波之 振幅正規化係為了便於計算後續步驟的正規化上升階段波形 面積比值NUR及正規化上升階段波形時間比值NUT。另一 方面，血壓波之振幅正規化不會影響增大指標Ah的計算， 因為增大指標ΑΙχ只是一個無單位比值。 在使用脈波機率分析法PPA來計算增大指標ΑΙχ前，要 先判斷量測到的血壓波的波形，以決定由上升階段波形或下 降階段波形之機率分佈函數(PDF)來定位反曲點。本發明利用 兩個判波參數來判別血壓波的波形： (1) 第一個判波參數係為正規化上升階段波形面積比值nur (normalized upstroke area ratio)，其定義為上升階段波形 面積與波形總面積之比值。 (2) 第二個判波參數為正規化上升階段波形時間比值NU丁 (normalized upstroke time)，其定義為血壓波之上升時間 與一個心搏周期時間之比值。Refers to the quantitative measurement and analysis system, which is based on the amount of U..Ββ. fluent blood flow pulse waveform time difference in different positions of the artery to reflect the vasodilatation 曰, 诋 度 罝 内皮 内皮 内皮 内皮 200 906 906 906 906 906 'As a basis for assessing the degree of arteriosclerosis. In the patent application No. 514514 of the Republic of China, a device for assessing arteriosclerosis money is disclosed, which is based on the amount of body parts of the body: the pulse wave' and information about the pulse wave velocity and the pulse wave. The amplitude increases the relevant capital. The degree of arteriosclerosis was assessed by the predetermined relationship between the artery and the arterial & mildening. In the Patent Case No. 546129 of the Republic of China, a technique for evaluating arteriosclerosis is disclosed, which measures the pulse wave of the first part and the first σ position of the subject, and an artery The sclerosis evaluation device compares the increase index of the pulse wave amplitude of the first portion with the increase index of the pulse wave amplitude of the first portion to evaluate the arteriosclerosis of the living body. In the Patent Case No. 534807 of the Republic of China Patent, a device for smashing a large buckle is disclosed, which uses a set of capsule pressure changing devices to change the pressure of a set of capsules, and obtains a high cuff pressure pulse wave and - a low cuff pressure pulse wave, and an increase index is determined according to each magnitude of the low cuff pulse wave. A number of methods have been developed to detect arteriosclerosis for the diagnosis of heart disease. It mainly predicts cardiovascular disease based on the embossed wave (bl〇〇d pressure medical ef〇rm) measured by the artery. This method is called pulse wave analysis (pwA). It is generated by the overlap of the incident wave generated by the human ventricular output (v_icular (10)) and the reflected wave of the peripheral reflection ((6) such as 丨wave). Since the measured blood pressure wave is a composite signal (_p()unding signal), the pulse wave analysis method can provide information on arteriosclerosis. The most commonly used 200906367 to indicate the arteriosclerosis index is the augmentation index AIx (augmentation index), which measures the intensity of the reflected wave in the blood pressure wave relative to the total blood pressure wave. Typical arterial blood pressure waves can be divided into waveforms A, B, C, and D. Fig. 1 is a waveform diagram showing the waveform of the arterial blood pressure wave of the waveform A and the waveform of the fourth differential corresponding to the blood pressure wave. Fig. 2 is a view showing a waveform diagram of the arterial blood pressure wave of the waveform C and a waveform diagram of the fourth differential corresponding to the blood pressure wave. The waveforms of arterial blood pressure waves A, B, C, and D are defined as follows: (1) Waveform A: Peak systolic blood pressure occurs after the acumentation point Pi, and the index AIx is increased. Is (ps - Pi) / (ps - Pd), where Pi is the blood pressure of the inflection point and Pd is the minimum diastolic pressure. When the index AIx > 0. 丨 2 is increased, the blood pressure wave W1 belongs to the waveform A, indicating that the blood pressure wave has a large hardness. (2) Waveform B: The maximum systolic pressure Ps also occurs at the late systolic pressure after the inflection point Pi, which is similar to the waveform A, but the increase index AIx is between 〇 and 0.12, indicating that the hardness of the gray pressure wave is small. . (3) Waveform C: The maximum systolic pressure Ps occurs before the inflection point Pi, and the increase index AIx is defined as (Pi - Ps) / (Ps - Pd), and the obtained increase index is a negative number 'The blood pressure wave W2 belongs to Waveform C indicates that the artery is quite elastic and healthy. (4) Waveform D: The waveform approximates waveform A, and the definition of the increase index AIx is the same as waveform A. However, its inflection point pi cannot be directly observed from 200906367 to 'this is because in the early stage of vasoconstriction, reflected waves have appeared, and are combined with incident waves. Obviously, increasing the index Αΐχ as the arterial stiffness increases from the number of turns ((4)c) to a positive number (waveforms Β, Β, and D). Many studies have shown that increasing the index Αίχ is a very important index in the conventional pulse wave analysis method, which has a very high correlation with arteriosclerosis and other cardiovascular problems. The difference between the singularity increase indicator 在于 is to identify the recurve point of the blood pressure wave... Study j now 4 recurve point P! occurs in the maximum blood flow rate of the artery. In other words, the blood pressure fluctuation flow rate should be measured synchronously, and then the recurve is determined. The point Pi is calculated to increase. In the technique, the blood pressure wave differential (differentia) is used to directly determine the inflection point Pi. In the end, the study of the recipient also uses the differential n ^ curvature point & ' of the blood pressure wave to calculate the increase index AIx. For the waveforms A, B, and eight, the distortion 1"1 is in the ascending phase waveform, which is obtained by the first negative 5 • and the negative-to-positive zero crossing of the blood pressure signal. . For - , skin C, the inflection point Pi is located in the descending phase, and the waveform corresponds to the zero crossing point of the two to the 弟. SUMMARY OF THE INVENTION The technical problem to be solved by the present invention: Although the pulse wave analysis method has been generally used to detect arteriosclerosis and to predict the heart vascular disease of 200,906,367, the conventional pulse wave analysis method has two main problems, firstly The technique of using the fourth derivative in the pulse/knife method cannot be used to treat blood sputum with a high sampling rate or pure interference. Differentiating produces more noise waveforms. It may be more difficult to identify the inflection point pi of the blood pressure wave. Therefore, in practical applications, filters must be added to the circuit to eliminate noise. Although you can use the wave to cry, less noise, but it will increase the complexity of the operation and get some useful information: , - 'The four differentials of the blood pressure wave can only determine the position of the inflection point, but can not be presented in the recurve point Pi The nearby number wins,,, i ^ , ' n ^ is a special desire, such as the width and shape of the waveform, and other physiologically important waveform feature points. For this reason, it is necessary to improve the conventional technology. The main species of the present invention can calculate the blood pressure wave increasing juice without using the differential method. The purpose is to provide an arteriosclerosis detecting method and apparatus capable of easily calculating the hemorrhagic chopping increase. i and ^ large Another purpose of this note is to provide - arteriosclerosis detection methods and devices. 'Technical means for solving the problem with high noise tolerance: The present invention provides a pulse wave probability analysis method 751W-like technical means for providing a device including a blood pressure wave measurement list , detection method and device thereof, a blood pressure wave normalization processing module, a π, a blood pressure wave digital conversion circuit, a column wave module, an increase index calculation unit 200906367 yuan 'where the determination module will be regular The measured subject ▲ pressure wave uses two wave parameters to determine the waveform of the pressure wave, the first parameter is the positive-handed rising phase waveform area ratio, and the second parameter is: = See the rising phase waveform time ratio NUT. The discriminating module discriminates the waveform of the blood pressure wave according to the determination parameter, obtains the probability distribution function of the blood pressure wave, determines the position of the local high density area (LHDR) in the probability distribution function, and 2 calculates the increase of the blood pressure wave. Large indicators, as assessed by arteriosclerosis status. . The present invention compares the effects of the prior art: Compared with the conventional pulse wave analysis method, the advantages of the present invention include: · 〇) The pulse wave probability analysis method of the present invention can directly increase the local probability of the blood pressure wave probability distribution function (PDF) The characteristic calculation of the density region (LHDR) Αΐχ; (7) The pulse wave probability analysis method of the present invention has a higher _ signal tolerance; (3) The pulse wave probability analysis method of the present invention can simultaneously monitor and locate all the feature points above; (4) The pulse wave analysis method of the present invention can provide a detailed description of the local waveform, which can further depict the physiologically important features; (5) The tree pulse wave probability analysis method can also be integrated into the blood pressure silk foundation and the blood official shape. The index' is a more precise description of the arterial features. The Benmai pulse probability analysis method is very practical in the analysis of arterial blood pressure waves, and has great application potential in clinical testing. The specific embodiment of the present invention will be further illustrated by the following embodiments and the accompanying drawings of 200906367. [Embodiment] In the present invention, a technique of PPA (Puhe 6 probab Uity analysis) is proposed, and the position of the inflection point Pi can be found in the blood dust wave of the artery by the method of the present invention. The characteristics of the blood pressure wave, and calculate the increase index ΑΙχ. T曰 曰 The pulse wave probability analysis method of the present invention calculates the increase index 依据 according to the pr〇bability density functi〇n. Since the signal near the inflection point pi of the pressure wave has a smaller amplitude change than the other parts of the blood wave wave, that is, the recurve point h in the blood pressure wave is in the probability distribution pure PDF wave towel A local high-density regi〇n is presented, as shown in Figure 3. Two:: Let the recurve point pi be the waveform in the rising phase (waveform A, 丄:==:=: rpDF will have local high and will present two main local high-density areas lhdr, = : counter: LP point In the second half, the __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The lhdr bit in the high-density region is obtained from the position distribution function PDF (4), and the increase index is calculated accordingly. (2) The LHDR bit in the local high-density region is in the rising phase or the falling phase of the probability distribution function pDF waveform. The state of the waveform can simultaneously monitor the physiologically important information in the blood pressure wave. (3) The waveform characteristics of the physiologically important information can be paid by the strong π ό τ p & 丨, and the ι_ 炙 condition of the local high-density area lhDR. Referring to FIG. 5, it shows a circuit function block diagram of the apparatus for detecting arteriosclerosis by pulse wave probability analysis method. The circuit mainly includes a blood pressure wave sensing device and a blood pressure wave analog signal processing module. 2, a blood pressure wave digitalization Circuit 3, a blood pressure wave normalization processing module 4, a wave module 5, an increase index calculation unit 6, a display device 7. A blood pressure wave sensing device I used in the county; Λ,, 1 to measure the subject The blood pressure wave can be obtained by using the conventional blood pressure turbid film which is currently commercially available. The use of „ & wave sensing is, for example, a pressure pulse bag, a pressure sensing benefit, an infrared emission/reception sensor. The measured blood pressure wave signal S1 of the subject is pre-processed by the analogy signal processing module 2 of the blood house wave analogy processing module 2. The blood analog signal processing module 2 can be sentenced to the ancient Liangdaolulu 22 - 尨μ, +匕A conventional circuit unit such as a pre-stage filter circuit 2!, a signal discharge circuit 22, and an m-wave circuit 23 is included. After the blood pressure wave analogy, the test subject i is replaced by the h group 2 The blood pressure wave number is converted to the north. The conversion of the signal digitization is performed. The blood pressure wave is 200906367. The channel 1 2 can include a analog-to-digital conversion circuit ^- and the removal of the waveform low-frequency drift part. The blood pressure wave of the conversion will be the localized wave normalization processing module j for the normalization of W. The treatment-4 includes an early spleen: a waveform cutting circuit 3, a blood pressure wave amplitude normalization circuit 42, which can be used to normalize the time and the amplitude of the blood pressure wave. 7 = After the amplitude is normalized, the decision module 5 will be used. The blood-wave phase rising phase waveform area ratio calculating unit 51 and the blood=segment waveform time ratio calculating unit 52 respectively calculate the normalized upper=waveform area ratio value (Teng malized ups plus ke _ rising phase Poco (four) value NUT (Teng Malizedup_ketime) = 13 1 The first classification C1 or the second classification c2 is determined according to the normalized rising phase waveform area ratio and the normalized rising phase waveform time ratio NUT. 2 ... Finally, the increase index calculation unit 6 can classify α, (: 2 obtain the probability distribution function of the blood pressure wave, and determine the local south density region (LHDR) in the probability distribution function according to the generated blood pressure wave. The position and the increase index of the blood pressure wave are calculated. The calculated result can be output to the display device 7 for display as an evaluation of the arteriosclerosis condition. 3 Fig. 6 is a flow chart of the present invention. The method of the present invention will be further described with reference to the circuit function block diagram shown in FIG. _. First, the system 200069367 measures the blood pressure wave of the subject by the blood pressure wave sensing device (step Ι〇ι), and the time and The amplitude of the measured blood pressure wave is normalized (step 1〇2). In the amplitude normalization step of the blood pressure wave, the maximum systolic pressure can be set equal to 1, the minimum diastolic pressure is set equal to 〇, and the set waveform period τ is equal to In order to normalize the amplitude of the time and blood pressure waves, the amplitude of the blood pressure wave is normalized in order to facilitate the calculation of the subsequent step of the normalized rising phase waveform area ratio NUR and normalization. On the other hand, the normalization of the amplitude of the blood pressure wave does not affect the calculation of the increase index Ah, because the increase index is only a unitless ratio. The increase is calculated using the pulse rate analysis method PPA. Before the indicator, the waveform of the measured blood pressure wave is first determined to determine the inflection point by the probability distribution function (PDF) of the rising phase waveform or the falling phase waveform. The present invention uses two determination parameters to discriminate blood pressure. Waveform waveform: (1) The first decision parameter is the normalized upstroke area ratio, which is defined as the ratio of the waveform area of the rising phase to the total area of the waveform. (2) The second The decision wave parameter is a normalized upstroke time, which is defined as the ratio of the rise time of the blood pressure wave to the time of a heart cycle.
信號正規化使得判波參數只取決於血壓波，不受血壓高 低及心跳的影響。要注意的是，血壓波波形C的上升階段波 开> 並/又有反曲點’其血壓有很急速的上升，因此，Nur及NXJT 14 200906367 的值應該相對地比較小。在另— D的上升階段波形具有反曲點， 而有較大的NUR及NUT。 方面’血壓波波形A、B及 會有較長的時間到達波峰， ^因此，在本發明步驟中包括有計算取得一正規化上升階 &波形面積比值NUR之步驟（步驟1()3)，以及計算取得—正 規化上升階段波形時間比值NUT之步驟（步驟1〇4)。 第7圖顯示正規化上升階段波形面積比值nur與正值 及負值ΑΙχ之對應關係示意圖，第8圖顯示正規化上升階严 波形時間比值NUT與正值及負值ΑΙχ之對應㈣示意圖 於正值增大指標AIx’NUR的平均值及標準差為〇39±〇〇4 ; 對於負值增大指標AIx，NUR的平均值及標準差為〇.丨5 ± 0.03。另一方面，NUT的平均值及標準差為〇23 ± 〇〇3 ;對 於負值增大指標AIx，NUR的平均值及標準差為〇1 ± 〇〇2。 由統計理論顯示，NUR及NUT之正值與負值增大指標ΑΙχThe normalization of the signal makes the decision parameters only depend on the blood pressure wave, and is not affected by the high blood pressure and heart rate. It should be noted that the rising phase of the blood pressure wave waveform C is turned on > and/or there is an inflection point' whose blood pressure has a rapid rise, and therefore, the values of Nur and NXJT 14 200906367 should be relatively small. In the rising phase of the other - D, the waveform has an inflection point and a larger NUR and NUT. Aspect 'Blood wave waveforms A, B and will have a longer time to reach the peak, ^ Therefore, the step of the present invention includes the step of calculating a normalized rising order & waveform area ratio NUR (step 1 () 3) And the step of calculating the acquired-normalized rising phase waveform time ratio NUT (step 1〇4). Figure 7 is a schematic diagram showing the correspondence between the waveform area ratio nur and the positive value and the negative value 正规 in the normalized rising phase, and Fig. 8 shows the correspondence between the normalized rising step and the strict waveform time ratio NUT and the positive and negative values (4) The mean value and standard deviation of the value increase index AIx'NUR are 〇39±〇〇4; for the negative value increase index AIx, the mean and standard deviation of NUR are 〇.丨5 ± 0.03. On the other hand, the mean and standard deviation of the NUT are 〇23 ± 〇〇3; for the negative value increase indicator AIx, the mean and standard deviation of the NUR are 〇1 ± 〇〇2. According to statistical theory, the positive and negative values of NUR and NUT increase the indexΑΙχ
的數據分佈不重疊’彼此分得很開。證明NUr及Νυτ具有 足夠的能力將血壓波分類為2類，第1類為波形a、Β及D 型（正值增大指標ΑΙχ)，而另一類為波形c(負值增大指標 ΑΙχ)。因此’可利用NUR及NUT的中間值來判定反曲點是 在上升階段波形或下降階段波形。更精確的，當NUR > 〇.27(或 者NUT > 0.165)時，反曲點在上行區段；當NUR < 〇.27(或 者NUT <0.165)時，反曲點在下行區段。 在取得正規化上升階段波形面積比值NUR及 '正規化上 15 200906367 升階段波形時間比值NUT後，即可利用正規化上升階段波形 面積比值NUR或正規化上升階段波形時間比值Νυτ判別出 血壓波之波形（步驟105)。當NUR > 0.27(或NUT > 0.165)表 不反曲點Pi係位於上升階段波形，其血壓波之波形為波形A、 B或D。而NUR < 0.27(或NUT < 0.165)表示反曲點PUiL於 下降階段波形，其血壓波之波形為波形C。 亦即，在本發明之方法中，可將前述血壓波之波形概分 為第一分類（即含蓋波形A、B或D)及第二分類（即波形〇， 其中第一分類具有反曲點位於該血壓波之上升階段波形之特 徵，第二分類之反曲點位於該血壓波之下降階段波形。 經過前述步驟後，若判別出血壓波為波形A、B或D型， 可付到其上升階段波形的機率分佈函數（pDF)(步驟1〇6)，然 後決定局㉝高密度區（LHDR)在該上升階段波形t機率分佈函 數_)的正確位置χΑ (步驟1〇7)。χΑ係表示局部高密度區 (LHDR)在上升階段波形之機率分佈函數（pDF)的位置，此一 位置即為反曲點Pi之位置。如此，可在χΑ步驟1〇8據以計 异出增大指標 AIx =(Ps — Pi)/(Ps — pd)=(卜 χΑ)/〇 _ 〇) $ — χΑ、。 ^判別出血壓波為波形c，可得到其下降階段波形的機 2佈函數_)(步驟⑽），然後蚊局部高密度區（LHDR) ιιο\ $Ί皮形之機帛分佈函數(PDF)的正確位置xC (步驟 。XC係表示局部高密度區（LHDR)在下降階段波形之機 16 200906367 率分佈函數（PDF)的位置，此一位置即為反曲點Pi之位置。 如此，可在步驟111據以計算出增大指標AIx=(Pi-Ps)/(Ps-Pd) =(xC — 1)/(1 — 0) =(xC — 1)。 在使用微分技術來計算增大指標AIx時的最大問題通常 係雜訊影響。為評估本發明脈波機率分析法PPA的雜訊容忍 度（noise tolerance)，本發明將模擬白雜訊（simulated white noise)加入波形A之壓力波形中以調變40db至20db信號與 雜訊比（signal-to-noise ratio，SNR)。如第 9A 圖至第 9C 圖所 示，其分別顯示在調變40db，30db及20db信號與雜訊比後， 對波形A的正規化振幅及機率的影響之波形圖。當SNR下 降時，利用習知微分技術來計算增大指標AIx的結果並不好， 這是由於被雜訊干擾的血壓波在經過四次微分後，產生許多 的零交越點，而使反曲點Pi無法判斷。相反的，即使SNR 下降20db，本發明在上升階段波形機率分佈函數（PDF)的局 部高密度區（LHDR)仍可分辨出反曲點Pi。因此，本發明之脈 波機率分析法PPA比習知脈波分析法PWA有更佳的雜訊容 忍力，換句話說，脈波機率分析法PPA允許利用品質較差的 血壓波來計算增大指標AIx。目前事實證明，本發明之脈波 機率分析法PPA亦不需配合濾波器來處理信號，而能夠更有 效地避免濾波處理可能造成波形資訊的漏失。 請參閱第10A圖至第10C圖分別顯示不同血壓波形， 其上升階段波形的機率分佈函數及下降階段波形的機率分佈 17 200906367 函數之波形圖。第i案例（第10A圖）及第2案例（第i〇b圖） 之血壓波形為具有明顯反曲點的波形A，而第3案例（第 圖）屬於反曲點非常不明顯的波形D。在這三個案例中，反曲 點，重博波形（dicrotic wave)，預輸出波形（pre_ejecti〇n wave)(不 能明顯看出，但事實上存在）在上升階段波形及下降階段波形 的機率⑥、度函數(PDF)有其對應的局部高密度區⑽⑽）。 這表示兩個重要的意義：（1)脈波機率分析法PPA可由 局邛间雄度區（LHDR)的特徵在血壓波波形D之波形中識別出 :曲點，(2)在同時顯示上升階段波形及下降階段波形的機率 刀佈函數下，本發明之脈波機率分析法能夠同步監視及定位 出具有生理重要性的波形特徵點。 而且，該三種血壓波形的反曲點位於相近的位置，但在 局部的波形有列㈣狀及列的振幅。在第丨案例中，反 曲點附近的波形振幅變化很小。而在第2案例及第3案例中， 反曲點變得越來越不明顯，而其反曲點附近的波形振幅有很 大的變化。這表示這些個案具有报接近的Αΐχ，而不必然有 、，L的動脈硬化私度。在第丨案例，lhdr^應了機率密度 函數(PDF)的反曲點，比較尖及狹窄。這顯示可進一步 描反曲點的特微，簡A，B及D波形分別出來，其中lhdr =較尖及狹窄的表示AS B型，而L職比較平坦及寬闊為 」因此提i、更多的資訊作為分析動脈血壓波形。ppA的 亦可應用於將其他生理重要點特徵化。利用p職及咖 18 200906367 計异出ΑΙχ的對照表列在表1。 表一：利用脈波分析法(PWA)及脈波機率分析法(ppA) 項目 脈波分析法(PWA) 脈波機率分析法（PPA) | 評估方式 利用血壓波形的四 |次微分 利用血壓波形的機率分| 佈函數（PDF) 濾'波 |需要濾浊 ~~~~~- ___ 由PDF的局^卩高密度區 (LHpR)判斷 判斷反曲點 由零交越點的四次 ___ 微分判 雜訊容忍力 較差 較H 同步定位出反|無 曲點的能力 丁人 f 土 I '— 字反曲點特徵|無 _ ---- 有 率分析法PPA的;波’但不影響本發明脈波機 所有波形_波以辨識分析法可應用於 理因素方面包括受測者之年=二在影響動_皮之生 動 '個人身高及性別。人類 51康、食物、心跳、運 素，因為交感神經及副交也是-項重要因 其他生理表現。同時’與疾病；;二=響心跳’血壓’及 勺二因素亦會影響動脈 19 200906367 血壓波的位線，這此因去可 3 二因素了犯疋動脈硬化、動脈粥狀硬化 f w—L、糖尿病及心臟衰竭。該 會被藥物所影響，例如硝酸 土渡也 · •夂甘油（mtr〇glycerin)及血管緊縮素 Π㈣⑼議η Π)。除了以特徵點的位置來計算增大指桿尬 二其：順數’脈波機率分析法ρρΑ簡化了以時間為基礎 舁，例如反射時間及上行時間，以進-步將動脈 的材負特性描續出來。 以上之實施例說明，僅為本發明之較佳實施例說明，凡 習於此項技術者當可依據本發明之上述實施例說明而作其它 種種之i：良及又化。然而這些依據本發明實施例所作的種種 改良及變化’當仍屬於本發明之發明精神及界^之專利範圍 【圖式簡單說明】 第1圖顯示波形A之動脈血驗之波形圖及該血壓波對應之 四次微分之波形示意圖； 第2圖顯示波形c之動脈血壓波之波形圖及該血壓波對應之 四次微分之波形示意圖； 第3圖顯不如果血壓波中之反曲點pi是位在上升階段波形（波 形A、B及D)，則上升階段波形的機率分佈函數PE)f 會呈現局部高密度區LHDR之特徵； 第4圖顯示如果血壓波中之反曲點是位在下降階段波形（波 20 200906367 形C)，則下降階段波形的機率分佈函數PDF會呈現局 部高密度區LHDR之特徵。 第5圖係本發明以脈波機率分析法評估動脈硬化之檢測裝置 之電路功能方塊圖； 第6圖係本發明之流程圖； 第7圖顯示本發明中所採用的正規化上升階段波形面積比值 NUR與正值及負值ΑΙχ之對應關係示意圖； 第8圖顯示本發明中所採用的正規化上升階段波形時間比值 NUT與正值及負值ΑΙχ之對應關係示意圖； 第9Α圖至第9C圖分別顯示在調變40db，30db及20db信號 與雜訊比後，對波形A的正規化振幅及機率的影響之 波形圖；及 第10A圖至第10C圖分別顯示不同血壓波形，其上升階段波 形的機率分佈函數及下降階段波形的機率分佈函數之 波形圖。 【主要元件符號說明】 100 以脈波機率分析法評估動脈硬化之檢測裝置 1 血壓波感測裝置 2 血壓波類比信號處理模組 21 前級濾波電路 22 信號放大電路 - 21 200906367 23 後級濾波電路 3 血壓波數位化轉換電路 31 類比至數位轉換電路 32 數位濾波電路 4 血壓波正規化處理模組 41 單一週期波形切割電路 42 血壓波振幅正規化電路 5 判波模組 51 血壓波正規化上升階段波形面積比值計算單元 52 血壓波正規化上升階段波形時間比值計算單元 53 血壓波波形判別單元 6 增大指標計算單元 7 顯示裝置 C1 血壓波第一分類 C2 血壓波第二分類The data distribution does not overlap and is very open to each other. It is proved that NUR and Νυτ have sufficient ability to classify blood pressure waves into two categories, the first type is waveform a, Β and D type (positive value increase index ΑΙχ), and the other type is waveform c (negative value increase index ΑΙχ) . Therefore, the intermediate value of NUR and NUT can be used to determine whether the inflection point is a waveform in the rising phase or in the falling phase. More precisely, when NUR > 〇.27 (or NUT > 0.165), the inflection point is in the up segment; when NUR < 〇.27 (or NUT < 0.165), the inflection point is in the down zone segment. After obtaining the normalized rising phase waveform area ratio NUR and the 'normalized upper 15 200906367 liter phase waveform time ratio NUT, the normalized rising phase waveform area ratio NUR or the normalized rising phase waveform time ratio Νυτ can be used to determine the blood pressure wave. Waveform (step 105). When NUR > 0.27 (or NUT > 0.165) indicates that the inflection point Pi is in the rising phase waveform, the waveform of the blood pressure wave is waveform A, B or D. NUR < 0.27 (or NUT < 0.165) indicates that the inflection point PUiL is in the falling phase waveform, and the waveform of the blood pressure wave is the waveform C. That is, in the method of the present invention, the waveform of the blood pressure wave can be roughly divided into a first classification (ie, a covered waveform A, B, or D) and a second classification (ie, a waveform 〇, wherein the first classification has a recurve The point is located in the waveform of the rising phase of the blood pressure wave, and the inflection point of the second classification is located in the falling phase waveform of the blood pressure wave. After the foregoing steps, if the blood pressure wave is determined to be the waveform A, B or D type, the point can be paid The probability distribution function (pDF) of the rising phase waveform (step 1〇6), and then determines the correct position of the local high-density region (LHDR) of the waveform t probability distribution function _) (step 1〇7). The χΑ indicates the position of the probability distribution function (pDF) of the local high-density area (LHDR) in the rising phase, and this position is the position of the inflection point Pi. In this way, the indicator AIx = (Ps - Pi) / (Ps - pd) = (Bu) / _ _ 〇) $ - χΑ, can be calculated according to the steps 1〇8. ^ Discriminate that the blood pressure wave is the waveform c, and the machine 2 function of the falling phase waveform is obtained (step (10)), and then the mosquito local high density area (LHDR) ιιο\ $Ί skin shape machine distribution function (PDF) The correct position xC (step. XC indicates the position of the local high-density area (LHDR) in the falling stage waveform 16 200906367 rate distribution function (PDF), which is the position of the inflection point Pi. Step 111 is calculated to calculate the increase index AIx = (Pi - Ps) / (Ps - Pd) = (xC - 1) / (1 - 0) = (xC - 1). Use the differential technique to calculate the increase index The biggest problem in AIx is usually the influence of noise. In order to evaluate the noise tolerance of PPA of the pulse wave probability analysis method of the present invention, the present invention adds simulated white noise to the pressure waveform of waveform A. To adjust the signal-to-noise ratio (SNR) of 40db to 20db. As shown in Figure 9A to Figure 9C, it shows the signal and noise ratio of 40db, 30db and 20db respectively. Waveform of the effect on the normalized amplitude and probability of waveform A. When the SNR decreases, using conventional differential techniques It is not good to calculate the result of increasing the index AIx. This is because the blood pressure wave interfered by the noise generates a lot of zero crossing points after four times of differentiation, and the inflection point Pi cannot be judged. On the contrary, even The SNR decreases by 20 db, and the local high-density area (LHDR) of the waveform probability distribution function (PDF) of the present invention can still distinguish the inflection point Pi. Therefore, the pulse wave probability analysis method PPA of the present invention is better than the conventional pulse wave analysis. The PWA has better noise tolerance. In other words, the pulse probability analysis method PPA allows the use of poor quality blood pressure waves to calculate the increase index AIx. It has been proved that the pulse wave probability analysis method PPA of the present invention does not. It is necessary to cooperate with the filter to process the signal, and it is more effective to avoid the loss of waveform information caused by the filtering process. Please refer to Figures 10A to 10C to show the different blood pressure waveforms, the probability distribution function of the rising phase waveform and the falling phase waveform. Probability distribution 17 200906367 Waveform of the function. The blood pressure waveform of the i-th case (Fig. 10A) and the second case (figure i-b) is waveform A with a significant inflection point, and the 3 Case (figure) is a waveform D that is very inconspicuous at the inflection point. In these three cases, the inflection point, the dicrotic wave, and the pre-ejecti〇n wave (not obvious) However, there is a probability that the waveform of the waveform and the falling phase in the rising phase, the degree function (PDF) has its corresponding local high-density region (10) (10)). This represents two important meanings: (1) Pulse wave probability analysis method PPA can be identified in the waveform of blood pressure wave waveform D by the characteristics of the inter-burst male area (LHDR): curvature point, (2) simultaneous rise Under the probability knife function of the phase waveform and the descending phase waveform, the pulse wave probability analysis method of the present invention can synchronously monitor and locate waveform feature points having physiological importance. Moreover, the inflection points of the three blood pressure waveforms are located at similar positions, but the local waveforms have column (four) shapes and column amplitudes. In the case of Dijon, the amplitude of the waveform near the inflection point changes little. In the second case and the third case, the inflection point becomes less and less obvious, and the amplitude of the waveform near the inflection point changes greatly. This means that these cases have close reports of sputum, not necessarily, and L's arteriosclerosis is private. In the case of Dijon, lhdr^ responded to the inflection point of the probability density function (PDF), which was sharp and narrow. This shows that the waveforms of the recursive points can be further described, and the waveforms of Jane A, B, and D are respectively shown, where lhdr = sharper and narrower indicates AS B type, while L is relatively flat and wide "so i, more The information is used to analyze the arterial blood pressure waveform. ppA can also be used to characterize other physiologically important points. Using the p job and the coffee 18 200906367 The comparison table of the differences is listed in Table 1. Table 1: Pulse wave analysis (PWA) and pulse wave probability analysis (ppA) Project pulse wave analysis (PWA) Pulse wave probability analysis (PPA) | Evaluation method utilizes blood pressure waveforms for four-time differential use of blood pressure waveforms Probability points | Cloth function (PDF) Filter 'wave|Need to filter ~~~~~- ___ Judging by the high-density area of the PDF (LHpR), judging the inflection point by the zero crossing point four times ___ Differential judgment of noise tolerance is worse than H. Synchronization locates the opposite | No ability to bend points Ding people f Earth I '- Word recurve points characteristics | No _ ---- Rate analysis method PPA; Wave 'but does not affect All waveforms of the pulse wave machine of the present invention can be applied to the rational factors including the age of the subject = two in the influence of the dynamic 'personal height and gender. Humans 51, food, heartbeat, and transportation, because sympathetic and accessory sex are also important because of other physiological manifestations. At the same time 'and disease;; two = ringing heartbeat 'blood pressure' and two factors will also affect the arterial line of 200906367 blood pressure wave, which can be caused by the two factors of atherosclerosis, atherosclerosis fw-L , diabetes and heart failure. This will be affected by drugs such as nitrous oxide, mtr〇glycerin and angiotensin (4) (9) η Π). In addition to the calculation of the position of the feature points, the increase of the finger 尬2: the singular 'pulse wave probability analysis method ρρΑ simplifies the time-based 舁, such as the reflection time and the up time, to advance the arterial material negative characteristics Continue to show. The above embodiments are merely illustrative of the preferred embodiments of the present invention, and those skilled in the art can make various other modifications in accordance with the above embodiments of the present invention. However, the various modifications and changes made in accordance with the embodiments of the present invention are still within the scope of the invention and the scope of the invention. [Fig. 1 shows a waveform diagram of the arterial blood of waveform A and the blood pressure wave. Corresponding to the four-differential waveform diagram; Figure 2 shows the waveform of the arterial blood pressure wave of waveform c and the waveform of the four differentials corresponding to the blood pressure wave; Figure 3 shows that if the recurve point pi in the blood pressure wave is When the waveform is in the rising phase (waveforms A, B, and D), the probability distribution function PE)f of the rising phase waveform will exhibit the characteristics of the local high-density region LHDR; Figure 4 shows that if the inflection point in the blood pressure wave is in the position In the falling phase waveform (wave 20 200906367 shape C), the probability distribution function PDF of the falling phase waveform will exhibit the characteristics of the local high density region LHDR. Figure 5 is a block diagram showing the circuit function of the apparatus for detecting arteriosclerosis by the pulse wave probability analysis method; Figure 6 is a flow chart of the present invention; Figure 7 is a view showing the waveform area of the normalized rising phase used in the present invention. Schematic diagram of the correspondence between the ratio NUR and the positive and negative values; Figure 8 is a schematic diagram showing the correspondence between the normalized rising phase waveform time ratio NUT and the positive and negative values used in the present invention; 9th to 9th The graphs show the waveforms of the effects on the normalized amplitude and probability of waveform A after adjusting the signal and noise ratios of 40db, 30db and 20db respectively; and the 10A to 10C graphs show the different blood pressure waveforms, respectively. Waveform diagram of the probability distribution function of the waveform and the probability distribution function of the waveform of the falling phase. [Explanation of main component symbols] 100 Detection device for assessing arteriosclerosis by pulse wave probability analysis 1 Blood pressure wave sensing device 2 Blood pressure wave analog signal processing module 21 Pre-stage filter circuit 22 Signal amplifying circuit - 21 200906367 23 Post-stage filter circuit 3 blood pressure wave digital conversion circuit 31 analog to digital conversion circuit 32 digital filter circuit 4 blood pressure wave normalization processing module 41 single cycle waveform cutting circuit 42 blood pressure wave amplitude normalization circuit 5 decision wave module 51 blood pressure wave normalization rising stage Waveform area ratio calculation unit 52 Blood pressure wave normalization rising phase waveform time ratio calculation unit 53 Blood pressure wave waveform discrimination unit 6 Increase index calculation unit 7 Display device C1 Blood pressure wave first classification C2 Blood pressure wave second classification
Aix 增大指標Aix increase indicator
Pi 反曲點Pi recurve
Pd 最低舒張壓 S1 血壓波信號 W1 血壓波 W2 血壓波 22Pd minimum diastolic blood pressure S1 blood pressure wave signal W1 blood pressure wave W2 blood pressure wave 22