The Nonlinear Analysis of Heart Rate Variability
ii
i
5 (Approxiate Entropy, ApEn) ApEn ApEn ApEn ApEn ApEn ii
Abstract Chinese edicine induces several kinds of Chinese edical constitution. The physical reactivity is often the change of bowels, endocrine and physical action. These changes see as close as the regulation of the autonoic nervous syste (ANS). In order to discuss the relation between the ANS and the Chinese edical constitution, the analysis of heart rate variability was applied to assess the yin-yang constitution in Chinese edicine. Six groups of subjects including noral, yin-yang exhaustion, yin exhaustion, yin repletion, yang exhaustion and yang repletion were recruited fro the internal departent of Chinese edicine at China Medical College Hospital. ECG signals were acquired fro all subjects for period of approxiately five inutes in the supine position. In this study, the nonlinear Approxiate Entropy (ApEn) analysis ethod that can quantify the coplexity of ECG signal efficiently was adopted. We expect to exhibit connection between autonoic nervous activity and the coplexity of ECG. The results can offer efficient paraeters for clinical diagnosis. Results showed in nonlinear analysis that the Approxiate Entropy was significantly different fro yin-exhaustion groups, yin-repletion groups and yang-exhaustion groups to the norals. The ApEn values in the yin-exhaustion groups were close to those in the yang-repletion groups, and the value in the yin-repletion groups were close to those in the yang-exhaustion groups. There exists significantly different fro the yin-exhaustion and yang-repletion groups to the yin-repletion and yang-exhaustion groups. According to the Chinese edicine, the yang and yin should be balance to healthy persons, and the yang-repletion will cause the yin exhaustion. Fro the result above, the ApEn value can be used to distinguish the yin-yang constitution Keywords Autonoic nervous syste heart rate variability yin-yang constitution Approxiate Entropy (ApEn) iii
...i... ii Abstract... iii... iv... vi... vii... 1 1-1... 1 1-2... 2... 3 2-1... 3 2-2... 4 2-2.1... 4 2-3... 6 2-4... 7 2-5... 10... 15 3-1... 15 3-2... 17 3-2.1... 20 3-2.2 r... 21 3-2.3 N... 21 3-2.4 (ApEn)... 21 iv
... 23 4-1... 23 4-2... 26 4-3... 30... 31 v
2.1...3 2.2 [12]...5 2.3 [13]...7 2.4...8 2.5...9 2.6 [3]...13 2.7...13 3.1...15 3.2 APEN...20 4.1...23 4.2...24 4.3...24 4.4...25 4.5 HRV...27 4.6 SDNN...27 4.7 RMSSD...28 4.8 APEN...28 vi
2.1 [1]...14 3.1 HRV...16 4.1...26 4.2...27 4.3 HRV T-TSET...28 4.4 SDNN T-TEST...29 4.5 RMSSD T-TEST...29 4.6 SDNN T-TEST...29 vii
1-1 [1] [2] [3] ApEn 1
1-2 [3] Heart Rate SDNN [4] (Irregularity) Lyapunov exponent Correlation diension 1991 Pincus ApEn(Approxiate Entropy)[5] [6] ApEn [7] [8] EEG(Electroencephalogra)[9] [10] 2
2-1 10~15 5 LabVIEW (FCP-2201, FUKUDA DENSHI) National Instruents DAQPad-1200 2.1 [11] FUKUDA (ECG) 2.1 3
2-2 (1) (Central nervous syste, CNS) (2) (Peripheral nervous, PNS) (Efferent) (Afferent division) ( ) (Soatic nervous syste) (Autonoic nervous syste) 2-2.1 (Autonoic ganglion) (Preganlionic fibers) (Postganglionic fibers) (Sypathetic) (Parasypathetic) 2.2 (Thoracolubar) (Craniosacral) 4
2.2 [12] (Sypathetic trunks) T1 L3 5
(Dual innervations), 2-3 (Partial pericardiu) (Pericardial sac) (Myocardiu) (Endocadiu) (Atriu) (Ventricle) 2.3 (Tricuspid) (Bicuspid) (Seilunar valve) 6
(Pulonary seilunar valve) 2.3 [13] (1) (2) (Stroke volue, SV) : = 2-4 (Sinoatrial SA Node) (Bundle of His) (Atrioventricular AV node) (Bundle branches) (Purkinje fibers) 2.4 7
2.4 30 c/sec, 45 c/sec [14], 110sec ;, 200 c/sec ~400 c/sec 60 s (Electrocardiography, ECG) ECG 2.5 ECG P QRS QRS coplex P 0.15 QRS 8
QRS S QRS T QRS 2.5 9
2-5 爲 爲 爲 爲 爲 [2] 10
爲 爲 爲 爲 ( ) ( ) 1 爲 11
2 爲 爲 爲 爲 爲 " " 3 " " " 4 " " 爲 2.6 ( ) 12
(a) (b) (c) (d) (e) 2.6 [3] (f) 2.6 2.7 2.7 13
2.1 2.1 [1] 14
第三章 研究方法 本研究中 線性分析方面主要包括心率變異分析 Heart Rate Variability, HRV 正常心跳間的標準差 SDNN 與相鄰 NN 區間之間差異值平方和的平 均值的平方根(RMSSD) 非線性分析方面選用近似熵 (Approxiate Entropy, ApEn) 為評估的指標 以下分別就其方法與內容作介紹 3-1 心率變異分析 在 HRV 時域分析主要的計算參數為 5 分鐘心率變異平均值與標準差 由統計的方法可計算出 線性的分析是將連續心電圖中的每一 QRS 複合波之間 隔被偵測出 相鄰的 R 波代表著心跳之週期 此時間軸間距即為 R-R Interval 而由連續的 R-R Interval 所構成的 Interval series 則代表著心率變異數 Heart Rate Variability, HRV 定義為 Noral-to-Noral (NN) interval 區間 相關參數如圖 3.1 所示 本研究即是以這心率變異數信號進行線性以及非線性的分析 圖 3.1 心電圖與其相關參數 15
3.1 SDANN 24 5 HRV SDNN RMSSD pnn50 NN50 3.1 HRV Variable Units Description SDNN s NN SDANN s 5 NN RMSSD s NN SDNN index s NN 5 NN50 count s NN 50 s pnn50 % NN NN50 16
3-2 (Approxiate entropy, ApEn) Pincus 1991 [5] ( 100 ) 爲 産 -- (1) 100 5000 1000 (2) (3) (4), r r N ( < x n) >= x(1), x(2),..., x( N) r ApEn(,r) N ApEn(,r,N) 1. < x( n) >= x(1), x(2),..., x( N) N ApEn r threshold, r 2. X ( i) = [ x( i), x( i + 1),..., x( i + + 1)], i = 1, N + 1 3. dxi [ ( ), X( j)] X i X j d [ X ( i), X ( j)] = ax [ x( i + k) x( j + k) ] k = 0, 1 Let N () i = no. of d[ X (), i X ( j)] r, C () i r 17
C () i = N ()/( i N -+ 1) 3 4 X i r r 4. (r) N 1 1 φ (r)= + ln Cr ( i) N + 1 i= 1 + 5. ApEn(, r, N) = φ ( r) φ 1 ( r) ApEn r =2 Pincus =2 r N 1 1 r = ( 0.1 ~ 0.25) SDx, SD x x( n) N 1 n 1 N N = n= 1 x( n) 6. ApEn ApEn 3.2 7. Moving Window ApEn Moving Window ApEn 200 Moving Window 50 151 ApEn 2 18
1 ) ( ) ( 1, ), ( 1) 1, (... 1) 1, ( ), ( ), ( + = = + + + + = i N i thenn j i ifc j i s j i s j i s j i C r r ), ( ), ( ), ( 1 j i s j i C j i C r r + + = + 1 ) ( ) ( 1, ) ( 1 1 1 + = = + + + + i N i thenn j i ifc r ) /( ) ( ) ( 1) /( ) ( ) ( 1 1 N i N i C N i N i C r r = + = + + 19
N 1 + 1 (r) ln C r ( i) N + 1 1 +1 (r) N i= 1 N i= 1 ln + 1 C r ( i ) ApEn(, r)= (r) +1 3.2 APEN 2 3 産 ApEn 3,r N 3-2.1 =2 =1 >2 爲 (1) >2 N 爲 N 5000 (2) N >2 r 20
=2[9] ApEn(,r) 3-2.2 r 爲 ApEn(,r,N) r r Pincus r 0.1 0.25SD x ( SD x 爲 x(i) ) [9] r=0.1sd x r=0.15sd x r=0.16sd x r=0.2sd x r=0.16sd x 3-2.3 N Pincus 爲 ApEn 僞 N=100 5000 N=200 N=1000 ApEn N=200 3-2.4 (ApEn) (1) ApEn 麽 +1 ApEn (2) 0 1 産 1 log 0 麽 ApEn 爲 産 0 =2 =3 log 産 21
ApEn ( ApEn ) ( ApEn ) (3) 2 3 産 (4) 22
4-1 LabVIEW [15] ApEn 4.1 ApEn ApEn HRV RMSSD SDNN 4.1 4.2 4.1 23
4.2 r 0.1 0.25SD x X Y i i+1 4.3 4.3 24
Excel 4.4 4.4 Yang[9] 200 1000 25
4-2 t-test HRV SDNN RMSSD ApEn 4.1 4.2 4.5 4.6 4.7 4.8 p 0.05 4.3 4.4 4.5 4.6 (HRV) 4.3 (SDNN) 4.4 RMSSD 4.5 ApEn 4.6 4.1 RMSSD HRV (sec) SDNN 36.151±9.718 885.832±34.276 40.812±4.678 16.614 ±8.378 702.634 ±97.745 19.307 ±9.605 21.510±4.814 881.727±92.034 35.598±10.480 56.533 ±12.812 933.089±175.627 47.720±8.837 54.992 ±18.109 889.962±113.316 47.212±13.676 40.587±5.526 826.854±39.537 55.992±8.124 26
4.2 ApEn 0.782±0.038 0.776 ±0.031 0.818±0.021 0.705±0.047 0.708±0.029 0.815±0.048 1200 1100 1000 900 800 700 600 500 HRV, HRV 4.5 HRV 70 SDNN 60 50 40 30 20 10 0, SDNN 4.6 SDNN 27
80 70 60 50 40 30 20 10 0 RMSSD, RMSSD 4.7 RMSSD 0.90 ApEn 0.85 0.80 0.75 0.70 0.65 0.60, ApEn 4.8 APEN HRV 0.0008* 4.3 HRV T-TSET 0.4778 0.0021* 0.9085 0.0079* 0.4797 0.9238 0.0034* 0.8756 0.5703 0.0067* 0.0085* 0.1539 0.1351 0.1723 * p<0.05 28
SDNN 0.0002* 4.4 SDNN T-TEST 0.0775 0.0060* 0.2287 2.56E-5* 0.0258* 0.2434 0.0004* 0.0787 0.9311 0.0008* 1.16E-6* 0.0008* 0.0718 0.1458 * p<0.05 RMSSD 0.0008* 4.5 RMSSD T-TEST 0.0032* 0.1792 0.0033* 8.3E-06* 0.0001* 0.0255* 0.0003* 0.0010* 0.8473 0.2854 1.92E-5 3.96E-6* 0.0096* 0.0624 * p<0.05 ApEn 0.7431 4.6 SDNN T-TEST 0.0378* 0.0070 0.0032* 0.0037* 0.0001* 0.0008* 0.0004* 8.95E-7* 0.8966 0.1515 0.0773 0.8470 0.0004* 0.0002* * p<0.05 29
4-3 爲 HRV SDNN [4] 4.3 4.4 4.5 4.2 ApEn ApEn 0.75 ApEn 4.2 ApEn ApEn ApEn 30
[1] 1996 [2] 1999 [3] 1996 [4] 2000 [5] S. M. Pincus, Approxiate Entropy: A Coplexity Measure For Biological Tie Series Data, Bioengineering Conference, 1991, Proceedings of the 1991 IEEE Seventeenth Annual Northeast, pp. 35 36, 1991. [6] M. Akay, Nonlinear Bioedical Signal Processing-Dynaic Analysis and Processing, Institute of Electrical and Electronics Engineers, Inc.3Park Avenue, 17 Floor, New York, NY 10016-5997, 2001. [7] S. C. Schuckers and P. Raphisak, Distinction of Arrhythias with the Use of Approxiate Entropy, Coputers in Cardiology, pp. 347 350, 1999 [8] S. C. Schuckers, Approxiate Entropy as a Measure of Morphologic Variability for Ventricular Tachycardia and Fibrillation, Coputers in Cardiology, pp.265-268, 1998. [9] Y. Fusheng, H. Bo, T. Qingyu, and T.C. Chen, Approxiate Entropy and Its Preliinary Application In The Field of EEG And Cognition, Engineering in Medicine and Biology Society, 1998. Proceedings of the 20th Annual International Conference of the IEEE, pp. 2091 2094, vol.4, 1998. [10] L. Quanzheng and G. Xiaorong, Subsection Approxiate Entropy and Its Application in sleep staging, Engineering in Medicine and Biology, 1999. 21st Annual Conference and the 1999 Annual Fall Meeting of the Bioedical 31
Engineering Society] BMES/EMBS Conference, 1999. Proceedings of the First Joint, vol.2, P 900, 1999. [11] C. C. Chiu, S. J. Yeh, and R. C. Lin, Data acquisition and validation analysis for Finapres signals, Chinese Journal of Medical and Biological Engineering, 15: 47-58, 1995. [12] ( ) 1999 [13] 1998 [14] 1988 [15] 1997 32