United States Patent 55
`6,095,984
`Aug.1, 2000
`[45] Date of Patent:
`Amanoet al.
`
`[11] Patent Number:
`
`US006095984A
`
`[54] ARRHYTHMIA DETECTING APPARATUS
`
`[75]
`
`Inventors: Kazuhiko Amano, Suwa; Kazuo
`Uebaba, Yokohama; Hitoshi Ishiyama,
`Toride, all of Japan
`
`[73] Assignee: Seiko Epson Corporation, Tokyo,
`Japan
`
`[21] Appl. No.:
`
`08/981,349
`
`[22]
`
`PCT Filed:
`
`Apr. 17, 1997
`
`[86]
`
`PCT No.:
`
`PCT/JP97/01322
`
`§ 371 Date:
`
`Apr. 6, 1998
`
`§ 102(e) Date: Apr. 6, 1998
`
`[87]
`
`PCT Pub. No.: WO97/38626
`
`PCT Pub. Date: Oct. 23, 1997
`
`[30]
`
`Foreign Application Priority Data
`
`Apr. 17,1996
`Mar. 10,1997
`
`[JP]
`[JP]
`
`Japan cccscssssssessscssesssssseeeeeessn 8-095731
`Japan wees seeeeenee 9-055263
`
`Tint. C17 oc eeeeeecseecees A61B 5/04; A61B 5/02
`[51]
`[52] U.S. Ch oe 600/500; 600/481; 600/513;
`600/515
`[58] Field of Search occ 600/515, 513,
`600/500, 509, 481
`
`[56]
`
`References Cited
`U.S. PATENT DOCUMENTS
`
`3,996,928 12/1976 Marx w..cccccccscsecsrerseeseeesees 600/515
`4,261,370
`4/1981 Von Nettelhorst 0... 600/481
`4,338,950
`7/1982 Barlow, Jr.etal. .
`5,749,366
`5/1998 Odagiri et al. oe 600/481
`5,776,070
`7/1998 Kitazawaet al. oe eeee 600/483
`
`FOREIGN PATENT DOCUMENTS
`
`9/1996
`0 729 726
`7/1985
`60-135029
`11/1990
`2-289230
`4-285530 10/1992
`7-227383
`8/1995
`
`.
`
`European Pat. Off.
`Japan .
`Japan .
`Japan .
`Japan .
`
`Primary Examiner—Cary O’ Connor
`Assistant Examiner—Navin Natnithithadha
`
`[57]
`
`ABSTRACT
`
`The arrhythmia detecting apparatus of the present invention
`is provided with a pulse wave detecting means which
`non-invasively detects the pulse waveform, and an arrhyth-
`mia detecting means which detects arrhythmia by monitor-
`ing changes in the pulse waveform detected by the pulse
`wave detecting means. The arrhythmia detection means has
`a decision element which determines that arrhythmia has
`occurred when there is an interruption in the continuity of
`this change. Methods employed for investigating the conti-
`nuity of change in the pulse waveform include a time
`domain method employing pulse wave interval values, and
`a frequency domain method in which frequency analysis
`(FFT or wavelet transformation) is carried out on the pulse
`waveform, with continuity studied based on the results of
`this analysis. As a result of this design,it is possible to detect
`arrhythmia by means of a simpler structure and easier
`operations as comparedto an electrocardiogram and so on.
`Further,
`it
`is also acceptable to provide a body motion
`detecting means, and to determine continuity of change for
`a pulse wave componentafter the body motion component
`detected by the aforementioned means has been removed
`therefrom. In this case, since it is possible to remove the
`body motion component from the pulse waveform, arrhyth-
`mia can be detected even more accurately throughout the
`course of daily activities.
`
`37 Claims, 43 Drawing Sheets
`
`PULSE
`(PULSE WAVE COMPONENT,
`BODY MOTION COMPONENT)
`
`
`PULSE WAVE DETECTING MEANS
`
`PULSE WAVE SIGNAL
`
`BODY MOTION
`
`
`
`BODY MOTION DETECTING MEANS
`
`BODY MOTION SIGNAL
`
`
`
`
`ARRHYTHMIA DETECTING MEANS
` ARRHYTHMIA DETECTING SIGNAL
`
`1
`
`APPLE 1020
`
`1
`
`APPLE 1020
`
`

`

`U.S. Patent
`
`|‘Sls
`
`Aug. 1
`
`’
`
`2000
`
`Sheet 1 of 43
`
`NOILOWAGOE NOILOWAGOd
` (LNAINOdWO
`
`‘LNANOdWNO
`.NOILOWAGOJSAVM3S1Nd)
`
`481Nd
`
`6,095,984
`
`
`
`
`
`TWNDISONILOSLAGVIWHLAHOUY
`
`
`
`TWNOISNOILOWAGOd
`
`
`
`TWNOLSSAVMAS1Nd
`
`
`
`
`
`SNVAWONILOSLS0VIWHLAHYaY
`
`SNVAWONILOFLAG
`
`
`SNVSWONILOSLAGSAVM3S1Nd
`
`2
`
`
`
`
`

`

`U.S. Patent
`
`60E80¢90§E0ELOE
`
`¢Ola
`
`Sheet 2 of 43
`
`6,095,984
`
`2000
`
`NoncuaansYlIdLWNILYSANODG/¥
`
`Aug.
`
`’1
`
`Le
`
`WVU|QNIAAITdW¥
`
`
`LINOYIDONIdVHSYOSNAS
`
`
`WHOIZAVM381d4YNOISSAV.4S"1NdJAYM4S1Nd
`
`
`
`
`
`LINOY19
`
`
`
`1g
`
`voeZ0§
`
`AV1dS10
`
`ndd
`
`NOILOWAdOS
`
`WHOSSAVM
`
`INSW313LINOYIDINIdVHS
`
`
`
`NOILOWAdOd
`ONIASTTdWYTWNOISNO|LOWAGO
`
`
`
`LINDYYOSNSS
`
`ole
`
`AINANGAYS
`
`INIQIAIG
`
`Lingalo
`
`
`
`NOILVTTI0S0
`
`LINDY)
`
`LLe
`
`3
`
`
`
`
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 3 of 43
`
`6,095,984
`
`FIG. 3
`
`
`
`4
`
`

`

`
`
`TLLplOl2266281828998SB8rI6EC6SEt860000
`
`
`
`(ZH)AQNANQAYS
`
`6,095,984
`
`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 4 of 43
`
`2Srrm
`
`©L
`
`0
`
`pyOla
`
`09
`
`Fant|dV
`
`wo
`
`5
`
`

`

`6,095,984
`
`
`
`(ZH)AONANDAYS
`
`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 5 of 43
`
`G‘DIS
`
`
`
`eeenemLIENey'
`
`v
`
`AGALILTdWV
`
`
`
`ZiLLpZOL2426628182p8998SB8rIBEE67SEL
`
`860000
`
`6
`
`€
`

`

`U.S. Patent
`
`9Dld
`
`Aug.1, 2000
`
`Sheet 6 of 43
`
`6,095,984
`
`2446628 (ZH)
`clitvlOl
`182p8998SB8yIGEE67SEL
`
`AQNANDAYS
`
`860
`00°0
`
`St
`
`AqNL|dV
`
`7
`
`

`

`U.S. Patent
`
`ZOlas
`
`O08
`
`6,095,984
`
`
`
`(ZH)AQNANDSYS
`
`Aug.1, 2000
`
`Sheet 7 of 43
`
`cLtbvlOlLL66L8ILVEO9BGBerLEE662SBL860000
`aaa|iiOL
`
`JGNL|dW
`
`©
`
`8
`
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 8 of 43
`
`6,095,984
`
`FIG. 8
`
`SAI
`
`START
`
`DETECTION OF
`BODY MOTION
`
`
`
`SA2Ss
` BODY MOTION
`
`NO
`PRESENT ?
`
`
`
`DETECTION OF
`PULSE WAVE
`
`SAT
`
`SA8
`
`
`
`BODY MOTION
`PRESENT ?
`
`
`
`A/D CONVERSION
`
`SAg
`PROCESSING OF PULSE
`
`
`WAVE DETECTION
`RECTANGULAR WAVE
`PROCESSING OF
`
`PULSE WAVE
`
`
`
`A/D CONVERSION
`PROCESSING OF BODY
`MOTION DETECTIO
`
`
`
`
`
`
`FFT PROCESSING
`
`EXTRACTION OF PULSE
`WAVE COMPONENT
`
`SA10
`
`CALCULATION
`OF PULSE RATE
`
`END
`
`9
`
`

`

`U.S. Patent
`
`Aug. 1UYUoseyeonVVVN
`
`2000
`
`Sheet 9 of 43
`
`6,095,984
`
`~a——AJNINDSYS
`
`
`
`d6dla
`
`10
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 10 of 43
`
`6,095,984
`
`AONANDAYS
`
`AQNANDAYS
`
`
`
`AONSNDSAYS
`
`
`
`(NOILOWAGO8)as)|YSNOd
`(AAVM3S1Nd)au)|yamod
`
`
`VOlSls
`
`dolOlas
`
`11
`
`
`
`V0!‘Ola
`
`11
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 11 of 43
`
`6,095,984
`
`FIG. 11
`
`START
`
`BODY MOTION DETECTION (fgo)
`
`PULSE WAVE DETECTION (fmg)
`
`SBI
`
`SB?
`
`PROCESSING TO SUBTRACT BODY|-°28
`MOTION EROM PULSE WAVE
`fy = fme-tse
`
`
`SB4
`
`SB5
`
`
`
`
`DETERMINATION OF FREQUENCY
`OF MAXIMUM SPECTRUM FROM
`AMONG SUBTRACTION RESULT,
`(fMmax)
`
`fy
`
`PULSE WAVE = fumax
`
`12
`
`12
`
`

`

`U.S. Patent
`
`Sheet 12 of 43
`Aug.1, 2000
`FIG. 12
`
`6,095,984
`
`START
`
`
`SCI
`
`FIND FREQUENCY COMPONENTS
`
`FROM RESULT OBTAINED AFTER
`
`
`FFT PROCESSING OF BODY MOTION
`
`
`SIGNAL,
`IN ORDER OF LARGEST
`
`
`LINE SPECTRUM (fs)
`
`
`SC2
`
`
`YES
`
`= fs
`
`SC3
`
`4
`
`SC4
`
`fsi
`
`(FUNDAMENTAL WAVE)
`
`= fs2/2
`
`
`SC5
`
`FIND FREQUENCY COMPONENTS
`
`FROM RESULT OBTAINED AFTER
`
`
`FFT PROCESSING OF PULSE WAVE
`
`
`SIGNAL,
`IN ORDER OF LARGEST
`
`LINE SPECTRUM (fm)
`
`
`
`NO
`
`SC9
`
`PULSE WAVE = fim
`
`13
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 13 of 43
`
`6,095,984
`
`FIG. 13
`
`POWER
`
`SECOND HIGHER HARMONIC WAVE
`
`
`
`FUNDAMENTAL WAVE
`
`
`
`1Hz
`
`2Hz
`
`3Hz
`
`4Hz
`
`14
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 14 of 43
`
`6,095,984
`
`FIG. 14
`
`START
`
`SD 1
`
`D75
`
`LARGEST LINE SPECTRUM (fs)
`SD3
`LARGEST LINE SPECTRUM (fm)
`
`FIND FREQUENCY COMPONENTS FROM
`RESULT OBTAINED AFTER FFT PROCESSING
`OF BODY MOTION SIGNAL,
`IN ORDER OF
`
`fs1
`
`(FUNDAMENTAL WAVE)
`
`= fs/HMC
`
`FIND FREQUENCY COMPONENTS FROM
`RESULT OBTAINED AFTER FFT PROCESSING
`OF PULSE WAVE SIGNAL,
`IN ORDER OF
`
`SD9
`
`es
`
`NO
`
`SD12
`
`PULSE WAVE = fm
`
`15
`
`15
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 15 of 43
`
`6,095,984
`
`FIG. 15
`
`START
`
`SE
`
`BODY MOTION DETECTION
`
`BODY MOTION
`PRESENT ?
`
`YES
`
`SWITCHING SIGNAL OFF
`
`[on
`
`
`
`TIME DURATION SUM Rt
`
`
`
`
`SWITCHING SIGNAL ON
`
`
`
`RECTANGULAR WAVE
`
`
`PROCESSING OF PULSE
`
`WAVE DETECTION
`
`
`SE3
`
`SE4
`
`SE5
`
`SE6
`
`SE7
`
`SE8
`
`PULSE WAVE DETECTION
`
`BODY MOTION DETECTION
`
`FFT PROCESSING
`
`EXTRACTION OF PULSE
`WAVE COMPONENT
`
`
`
`,
`CALCULATION OF PULSE RATE
`
`SE13
`
`16
`
`16
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 16 of 43
`
`6,095,984
`
`FIG. 16
`
`START
`
`SF]
`
`BODY MOTION DETECTION
`
`SF2
`
`NO BODY MOTION
`
`PRESENT ?
`
`YES
`
`PULSE WAVE DETECTION
`
`SF 3
`
`SF4
`
`SF5
`
`YES
`BODY MOTION
`
`PRESENT ?
`
`A/D CONVERSION
`PROCESSING OF PULSE
`
`WAVE DETECTION
`
`
`A/D CONVERSION
`PROCESSING OF BODY
`MOTION DETECTION
`
`RECTANGULAR WAVE
`PROCESSING OF
`PULSE WAVE
`
`‘0
`
`
`
`
`
`
`
`
`CONSTANCY
`PRESENT ?
`
`EXTRACTION OF PULSE
`WAVE COMPONENT
`
`PULSE RATE CALCULATED
`
`SF 12
`
`END
`
`17
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 17 of 43
`
`6,095,984
`
`FIG. 17
`
`aaai|
`|
`
`
`
`18
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 18 of 43
`
`6,095,984
`
`FIG. 19
`
`
`
`19
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 19 of 43
`
`6,095,984
`
`FIG. 20
`
`f1
`
`f?
`
`f3
`
`BODY MOTION
`DETECTING MEANS
`
`SECOND WAVELET
`TRANSFORMING MEANS
`
`PULSE WAVE
`DETECTING MEANS
`
`FIRST WAVELET
`TRANSFORMING MEANS
`
`FIRST FREQUENCY
`CORRECTING MEANS
`
`SECOND FREQUENCY
`CORRECTING MEANS
`
`MASK MEANS
`
`f7
`
`£8
`
`£9
`
`DECIDING MEANS
`
`NOTIFYING MEANS
`
`RECORDING MEANS
`
`FREQUENCY CALCULATING
`MEANS
`
`SUMMING MEANS
`
`£11
`SECOND NOTIFYING MEANS
`
`re
`
`
`FOURTH NOTIFYING MEANS
`
`THIRD NOTIFYING MEANS
`
`f15
`
`
`
`4
`
`f5
`
`f6
`
`10
`
`f13
`
`f14
`
`20
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 20 of 43
`
`6,095,984
`
`FIG. 21
`
`144
`
`21
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 21 of 43
`
`6,095,984
`
`FIG. 22A
`
`
`
`FIG, 22B
`
`
`
`22
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 22 of 43
`
`6,095,984
`
`tl
`
`AV1dS1d
`
`WVd
`
`YAWNNS
`
`AQNANGAYS
`
`YOLVINITVI
`
`4¥4dzZZNg
`
`‘
`
`ay
`
`él
`
`Lt
`
`Ol
`
`0€1
`
`ds LNAWA14
`
`
`
`NOIS1930
`
`ADNANDIYS
`
`YOLISNN09
`
`YAWYOASNVYL
`
`LATSAVM
`
`
`
`AS1NdSdASSsadd
`
`
`
`YOSNASSAVM
`
`23
`
`23
`
`
`
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 23 of 43
`
`6,095,984
`
`801
`
`
`
`isdld
`
`AYOWSN
`
`
`LINIUIDINIGIAIG
`
`LINOUIDOINILIALAG
`
`ADNANGAYS
`
`SSOUDOYSZ
`
`ONION
`
`4511s
`
`veDIA
`
`QWbennnnnerrrrrrriess
`c0l201101
`31V9SYOLVISNVULAMOWSWY3LYAANO9yqquooau!TaiwuvdW444n9NOILONMSasva|!
`
`
`
`
`
`
`eMZNLM!
`
`AMOWANNCo
`
`(NOSK
`
`HW
`
`24
`
`24
`
`
`
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 24 of 43
`
`6,095,984
`
`FIG. 25A wi
`
`FIG. 25B
`
`FIG. 25C
`
`FIG. 25D cs
`
`25
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 25 of 43
`
`6,095,984
`
`Gel
`
`9c
`
`YOLVEVdWODdd
`
`
`
`:|——hr.YOLV
`INDTVO~A¥OWSWLt+H/1(+)
`
`
`26
`
`26
`
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 26 of 43
`
`6,095,984
`
`FIG. 27
`
`12
`
`4.0Hz ~ 3. 5Hz
`3. 5Hz ~ 3. OHz
`3.0Hz ~ 2. 5Hz
`
`2.5Hz ~ 2. 0Hz
`
`2.0Hz ~ 1. 5Hz
`
`
`
`
`
`
`
`
`
`
`1.5Hz ~ 1. 0Hz
`
`1.0Hz ~ 0. 5Hz
`
`
`! Ma8
`
`! Ma?
`Ma6
`
`! Ma5
`
`Ma4
`
`Ma3
`
`
`
`
`27
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 27 of 43
`
`6,095,984
`
`cOA©LL
`
`<<
`
`FIG. 28B
`
`FIG. 28E
`
`FIG. 28F
`
`WoWUPWateaa
`
`COP-OLOSFOON—
`
`HouUHWa
`
`WakabStUetra
`
`NLOMLOMOe—
`
`COP-OLOSFON—
`
`WoTTbWewou
`
`NMOMLONS-Or
`
`28
`
`28
`
`
`
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 28 of 43
`
`6,095,984
`
`FIG. 29
`
`
`
`29
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 29 of 43
`
`6,095,984
`
`ASTINallgol
`
`
`aniHL
`
`aI¥01934Y09HANYOISNVYL
`
`ONaw
`
`AONANDSU4LSILSTSAVKSUT
`JYNS$$saud
`
`
`
`YOSNASSAYM
`
`VIIvO!
`
`0€“Od
`
`SS1Ad
`
`
`
`
`
`
`
`OWNAONANDAY4GNOIISLSTSAVMGNODSS
`
`ZI
`
`EERE!
`
`
`
`NOISII30}gy)
`
`
`
`YOLOTYYODAWYOASNYYL
`
`eSI9
`
`
`
`
`
`
`
`AV1dS1WvewAWWNSeyroa;
`
`HOLWINDTI
`
`NOILV8S1S5900¥
`
`30
`
`irT¥377Ng
`
`LIza)
`
`GH
`
`
`
`30
`
`
`
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 30 of 43
`
`6,095,984
`
`
`
`FIG. 31A
`MH
`
`FIG. 31C
`MH’
`
`FIG. 31B TH
`
`f=
`
`1. 3Hz
`
`31
`
`31
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 31 of 43
`
`6,095,984
`
`Te
`
`Tf
`
`2Sa2zSa2zSSa
`
`cOOmOL—OSIMAN
`
`hoaTPUPdPaove
`HaeWeTEHUtt
`Waweawoudbal
`
`NILAMLOMMAID
`DOOOOOrN
`NIMLOMOre
`
`COr-OLlOWON
`
`COP-COLOSTOIC
`
`COP-COLOSTONr—
`
`COP-COLOSTOONr—
`
`
`
`Or-LOIMOOO+t
`
`WOUUETPakate
`WoWabaeaWEat
`WomheaeTna
`
`SSSSSSSz=
`
`COr-OLOTONE-
`
`Te
`ee
`HunaaWoW
`
`DOMNANO
`
`OO000rlor
`OCOM—NNAON
`
`Wua)ubflHPUntl
`
`TMONATMONOE
`OOOOOOOO
`TONTMONOG
`
`Wawe
`WoWWeaaeWEUbail
`
`NOMTADOOLO
`OOOO0--MOt
`NIestNOOOr—
`
`riewmwe
`Hadiwen
`howabUWab
`
`COP-COLOTON
`COM-COLOSTCOON
`
`SSSSSSz2=z
`SSSSSz2z=z
`SSS22S2B
`
`MNIMCOLOANIM-LOW
`
`NabWPLnaCOPR-OWOTON
`COrn-OLOTONr—
`HouWEWu
`COr-OWOTON
`UPHabeataw
`
`22z2zaSSaSS
`22SSazzaS
`SSSSSsaz2za
`
`32
`
`on©LL
`
`ial
`
`32
`
`
`
`
`
`
`
`
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 32 of 43
`
`6,095,984
`
`
`
`LtiTHIN
`
`01Of!
`
`CeOls
`
`
`
`AONANDYSLTaAVIA3sindaunssaud
`
`
`
`
`
`
`
`401949N09YAWYOASNVYLYOSNASJAYM
`
`
`
`
`
`61
`
`ZI
`
`¢
`
`OW
`
`aa
`
`NOLLOWAGOg
`
`YOLvuydas
`
`NorTs1930
`
`NaW313
`
`AV1dS1dWVUYAWWNS
`
`AdNANGAY
`
`YOLVINITVI
`
`4AZZNE
`
`33
`
`33
`
`
`
`

`

`U.S. Patent
`
`beSls
`
`LNL!LSans{iN LIAIYTI
`f61colLoL
`
`
`
`
`AMNIWAFOVUSAV
`
`INILVINDTV9
`
`
`
`YSLNNOD4nvMasind
`
`Aug. 1, 2000
`
`Sheet 33 of 43
`
`6,095,984
`
`LiNdyto
`
`NO!
`
`YAdVHS
`
`HW
`
`34
`
`34
`
`
`
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 34 of 43
`
`6,095,984
`
`ONTMCONCG
`
`COrMm-OLNNSON—
`
`Wouttraaa
`
`NIMr=STNDMLD
`
`2222222=
`
`COP-OLOSNTOMON—
`
`Haduwwuy
`
`Td
`
`COM-OLONTOAI—
`
`WUEUhaTe
`
`222222Sz
`
`HORTEUDtRURUEut
`
`OMLOMOst
`™M-LNMOOO
`
`22fSS2222B
`SS
`
`COP-OLOTON
`COR-COLOSON
`
`HouudWeaue
`
`HonWateatt
`
`CO-NNMLOOD
`COONNMOO
`
`
`
`WoHPEEWPabTh
`
`HothWe
`
`NLOMLOAMOANIOD
`NLOMLOMMOO
`
`22222222
`SSSSzaaza2a
`
`COM-COLNSFMONI—
`COr-OLlOTMN—
`
`NIMOOLONIP-LOW
`NOMLAIS-OO
`
`ee
`
`FIG. 35E
`
`35
`
`35
`
`
`
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 35 of 43
`
`6,095,984
`
`VW
`
`cw
`
`oW
`
`LW
`
`©
`
`_— is
`
`¢ t
`
`oO
`—_
`
`©
`—_
`
`Oo
`_
`
`>
`
`—
`
`—
`
`SW
`
`NJ
`-=
`
`oO
`
`LW
`
`OW
`
`©
`
`jo
`
`©
`
`ON
`
`=
`
`NI
`
`=>
`
`SW
`
`NI
`>
`
`<=
`
`oo
`l
`
`—_—
`
`a
`L
`
`<
`L
`
`(2)OH
`
`Jt
`
`
`
`9€“ODIs
`
`(Z)LH
`
`(Z)OH
`
`(2)
`
`x=
`
`(2)1H“e—
`
`os
`
`Jt
`
`VI
`
`ma
`L
`
`<
`l
`
`a
`[
`
`(Z)OH
`
`(Z)LH
`
`(2)LH
`
`Pr
`
`(Z)OH
`
`Jt
`
`al
`
`jet
`
`Jl
`
`al
`
`36
`
`36
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 36 of 43
`
`6,095,984
`
`FIG. 37
`
`
`
`37
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 37 of 43
`
`6,095,984
`
`FIG. 38A
`
`FIG. 38C
`
`200
`
`204
`
`201
`
`205
`
`206
`
`38
`
`207
`
`38
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 38 of 43
`
`6,095,984
`
`FIG. 39
`
`
`
`39
`
`

`

`U.S. Patent
`
`Aug. 1, 2000
`
`Sheet 39 of 43
`
`6,095,984
`
`
`LINDYIOONIGIAIC
`
`C)T7AMOWaN
`
`
`LINDYIDINILISLAG
`
`$SO¥9O04dZ
`Or'OIs
`
`itvai1LINNpyEMoftIM
`
`GN
`
`AMOWSW(awGNOD3Sor
`
`eyeeCM}
`
`
`
`
`
`TaTWavdy344ng31SNOILONN4asva
`
`Ol
`
`ALL444211
`
`
`
`LINDYIDONILIALAGLINDY)
`
`
`JONL|TdW¥YOLVEVWO.31¥9
`
`EOL
`
`AQNANDAYS
`
`
`
`801ISH
`
`ONIONIY
`
`431114
`
`Hi
`
`40
`
`HOLVISNVYLAMOWAINY3LUSANO9yaaa’
`
`
`
`
`
`
`
`D7eee—EeeeEeeEee7
`
`40
`
`
`
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 40 of 43
`
`6,095,984
`
`FIG. 41
`
`RUNNING PITCH,
`
`PULSE RATE
`
`(TI MES/MIN)
`
`220
`
`120 140 160 180 200 220 240 260 280
`RUNNING SPEED (METERS/MIN)
`
`A1
`
`41
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 41 of 43
`
`6,095,984
`
`FIG. 42A
`
`FIG. 42B
`
`42
`
`42
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 42 of 43
`
`6,095,984
`
`FIG. 438A
`
`NUMBER OF
`ARRHYTHMIA
`
`NUMBER OF
`ARRHYTHMIA
`
`NUMBER OF
`ARRHYTHMIA
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12 (MONTH)
`
`FIG. 43B
`
`SUN. MON. TUE. WED.
`
`THU.
`
`FRI.
`
`SAT.
`
`(DAY OF
`THE WEEK)
`
`FIG. 43C
`
`0
`
`2
`
`4
`
`6
`
`8
`
`10
`
`12
`
`14
`
`16
`
`18 20 22 24 (CLOCK
`TIME)
`
`43
`
`43
`
`

`

`U.S. Patent
`
`Aug.1, 2000
`
`Sheet 43 of 43
`
`6,095,984
`
`FIG. 44
`
`1997/04/01
`
`10:28:15
`
`1997/04/01
`
`10:59:21
`
`19997/04/01
`
`11:40:18
`
`
`
`
`
`
`
`
`
`44
`
`

`

`6,095,984
`
`1
`ARRHYTHMIA DETECTING APPARATUS
`BACKGROUND OF THE INVENTION
`1. Field of the Invention
`
`The present invention relates to an arrhythmia detecting
`apparatus suitable for detecting arrhythmiaat rest and during
`daily activity, from the pulse.
`2. Description of the Related Art
`The pulse generated by the contraction of the heart may
`be obtained in the form of a pulse wave (pulse waveform).
`This pulse waveform can be broadly separated into a “prin-
`cipal wave”, which arises from the contraction of the heart
`which causes blood to be sent
`into the arteries, and a
`“dicrotic wave”, which derives from closure of the heart
`valve. In a healthy individual, there is a regular repetition in
`the contraction of the heart, such that the pulse waveform
`exhibits a constant rhythm.
`However, when the heart is weak due to an abnormality
`in the circulatory system, an arrhythmic pulse may occur.
`While these types of disorders in the pulse (hereinafter,
`referred to as “arrhythmia”) can be caused by smoking,they
`are also frequently seen in cardiac diseases such as cardiac
`valve disease, myocardial infarction, cardiomyopathy and so
`on. Accordingly, arrhythmia detection would be useful for
`the diagnosis of some abnormalities in the circulatory sys-
`tem.
`
`The danger of bradycardia (in which the pulse rate is less
`than 40 beats/min), caused by symptoms of complete atrio-
`ventricular block or Adums-Stokes syndrome,
`is well
`known. Moreover, it is also known that symptoms of sick
`sinus syndrome cause bradycardia or tachycardia (in which
`the pulse rate is 150 beats/min or more). Further, extrasys-
`tole can cause intermittent pulse (arrhythmia), so that its
`occurrence during exercise may be considered quite dan-
`gerous. Thus, in view of these circumstances, considerable
`focus has been placed on detection of arrhythmia.
`It has been the conventional to detect arrhythmia such as
`described above by employing an electrocardiogram in
`which electrodes are attached to the subject’s chest, and an
`electrocardiac potential is detected via the electrodes.
`An electrocardiographyis a large device, and requires the
`attachment of electrodes to the subject. Accordingly,
`the
`subject’s movementsare restrained since he cannotleave the
`examination room for the duration of procedure. While
`portable electrocardiogram recording devices have been
`developed in recent years, these too are problematic since
`they are difficult to use, and do not enable a person without
`specialized knowledge to detect arrhythmia on his own
`easily. In other words, until now,it has not been possible to
`detect accurately arrhythmias during in one’s normal daily
`activities.
`
`SUMMARYOF THE INVENTION
`
`The present invention was conceived in consideration of
`the above-described circumstances, and has as its objective
`the provision of a portable arrhythmia detecting apparatus
`capable of accurately detecting arrhythmia by meansof a
`simple operation which can be performed during the course
`of normal daily activities.
`In order to resolve the problems discussed above,
`present invention is basically provided with:
`a pulse wave detecting means which non-invasively
`detects the pulse waveform in the body;
`a body motion component excluding means which
`excludes the body motion component from the pulse
`waveform detected by the pulse wave detecting means;
`and
`
`the
`
`5
`
`15
`
`20
`
`50
`
`55
`
`60
`
`65
`
`2
`a notifying means which provides notice of information
`showing the pure pulse waveform, which is the wave-
`form obtained after the body motion component
`excluding means has excluded the body motion com-
`ponent.
`Alternatively, the arrhythmia detecting apparatus according
`to the present invention may be provided with:
`a pulse wave detecting means which non-invasively
`detects the pulse waveform in the body;
`a body motion component excluding means which
`excludes the body motion component from the pulse
`waveform detected by the pulse wave detecting means;
`an arrhythmia detecting apparatus which detects arrhyth-
`mia by monitoring changes in the pure pulse waveform
`obtained after the body motion component excluding
`means has excluded the body motion component; and
`a notifying means which provides notice of the results of
`the detection by the arrhythmia detecting means.
`The above-described structure permits arrhythmia to be
`detected based on pulse waveforms which can be obtained
`in a non-invasive manner. Thus, arrhythmia can be detected
`by means of a simple operation, without requiring the
`attachment of electrodes to the test subject’s chest.
`Moreover, the design of the arrhythmia detecting apparatus
`according to the present
`invention is simpler and more
`compact than an electrocardiograph, making it possible to
`realize an arrhythmia detecting apparatus with excellent
`portability. Additionally, it is noted here that while the pulse
`waveform detected by the pulse wave detecting means may
`be considered to include a body motion component which
`arise from daily activity, it is an easy matter to distinguish
`between the body motion component and the pulse wave
`component provided that
`the body motion is within the
`normal daily range. Accordingly, it is possible to obtain a
`pure pulse waveform, thereby allowing accurate detection of
`arrhythmia during the course of daily activities. Moreover,
`it goes without saying that by providing notice of informa-
`tion showing the pure pulse waveform, the individual thus
`notified is made aware of the presence or absence of
`arrhythmia.
`if a body motion detecting means which
`Moreover,
`detects body motion and outputs a body motion waveform is
`provided, and changes in the pulse wave component
`obtained after excluding the body motion component from
`the pulse waveform are monitored, then it is possible to
`cancel out this body motion component. Therefore,
`it is
`possible to accurately detect arrhythmia even when the
`subject is exercising.
`Further, the presence or absence of continuity over a time
`or frequency domain mayalso be investigated. When check-
`ing continuity over a time domain, the difference between a
`reference value and the pulse wave interval is monitored.
`However, in order to more accurately detect arrhythmia,it is
`also acceptable to update the pulse wave interval.
`In the case where checking continuity over a frequency
`domain, arrhythmia is detected by carrying out frequency
`analysis on the waveform, and employing the result which
`corresponds to the pulse wave component. In this case, it is
`also acceptable to more accurately detect arrhythmia by
`changing the frequency region on which frequency analysis
`is performed, in response to the body’s state of activity.
`Moreover,
`it is also acceptable to carry out detection of
`arrhythmia only when there is constancy in the activity, so
`that arrhythmia detection is carried out only during regular
`exercise (running or other intense exercise, for example).
`With respect to the frequency analysis method, FFT (fast
`Fourier transform), wavelet
`transformation, or another
`45
`
`45
`
`

`

`6,095,984
`
`3
`method may be suitably employed. In the case of FFT, an
`appropriate range of variationis first estimated for the pulse
`frequency. A determination that arrhythmia has occurred is
`then made if a frequency spectrum projecting into the
`estimated range of change is not present in the FFT results.
`When employing wavelet transformation, the continuity
`of the pulse wave analysis data obtained by performing
`wavelet transformation on the pulse waveform is analyzed in
`each frequency region. A determination that arrhythmia has
`occurred is made if an abnormal portion is detected. In this
`case, it is also acceptable to correct the pulse wave analysis
`data based on the various corresponding frequencies so that
`the power density per frequency becomes constant. Cor-
`rected pulse wave data is thereby generated, and the conti-
`nuity of the corrected pulse wave data is analyzed in each
`frequency region. If an abnormal portion is detected, then a
`determination is made that arrhythmia has occurred.
`Wavelet transformation may also be performed on the
`pulse waveform and the body motion waveform,
`respectively, to obtain pulse wave analysis data and body
`motion analysis data. Then, an analysis of continuity may be
`carried out on the result obtained after subtracting the body
`motion analysis data from the pulse wave analysis data. It is
`of course also acceptable. to carry out the aforementioned
`subtraction after correcting the wavelet
`transformation
`results so that the power density per frequency becomes
`constant, or to carry out correction after the subtraction
`operation. In addition, as another method for excluding the
`body motion componentfrom the pulse waveform, a method
`may be employed in which the frequency component cor-
`responding to body motion is excluded from the pulse wave
`analysis data or the pulse wave correction data. In addition,
`wavelet transformation may be carried out in synchroniza-
`tion with the period of the pulse waveform.
`When a decision is madethat arrhythmia has occurred, the
`user may be notified of this fact, and the time of the event
`may be recorded. It is also acceptable to provide notice to the
`user after associating the information (a histogram,
`for
`example) corresponding to the recorded time of an arrhyth-
`mia event with physiological rhythmsin the body.It is also
`acceptable to calculate the arrhythmia frequency informa-
`tion by calculating the numberof times that a determination
`of arrhythmia is made within a specified period of time. In
`this case, notice is provided whenthis arrhythmia frequency
`information exceeds a specific value determined in advance.
`It is also acceptable to add the numberof times that an
`arrhythmia determination has been made,
`to generate
`arrhythmia sum information. In this case, notice is then
`provided when the arrhythmia sum information exceeds a
`specific value determined in advance. In addition, it is also
`acceptable to provide notice when both the arrhythmia
`frequency information and the arrhythmia sum information
`exceed the respective specific values determined in advance
`therefor,or, alternatively, when at least one of these exceeds
`the specific value determined in advance therefor. By means
`of these various designs,it is thereby possible for the user to
`be aware of his own physical condition.
`Arrhythmia detection processing and notice processing
`may becarried out in parallel. The pulse wave detecting
`means may be composed of a pressure pulse wave sensor
`which employs pressure to detect
`the arterial pulse.
`Alternatively,
`the pulse wave detecting means may be
`designed so as to receive the light reflected when the
`detection site on the body is irradiated with light having a
`wavelength of 300 to 700 nm, and detect the received light
`signal as a pulse waveform. It is also acceptable that the
`pulse wave detecting means be designed to receive trans-
`
`10
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`60
`
`4
`mitted light when the detection site is irradiated with light
`having a wavelength of 600 to 1000 nm, and detect the
`received light signal as the pulse waveform.
`BRIEF DESCRIPTION OF THE FIGURES
`
`FIG. 1 is a block diagram showing an overview of the
`structure of an arrhythmia detecting apparatus founded on
`the basic technical concept of the present invention.
`FIG. 2 is a block diagram showing an example of the
`structure of an arrhythmia detecting apparatus according to
`the first preferred embodimentof the present invention.
`FIG. 3 is a slant view showing an example of the outer
`appearance of the same apparatus.
`FIG. 4 is a block diagram showing an example of the
`results of frequency analysis of the fingertip plethysmogram
`when arrhythmia is not present.
`FIG. 5 is a block diagram showing an example of the
`results of frequency analysis of the fingertip plethysmogram
`when arrhythmia is present.
`FIG. 6 is a block diagram showing an example of the
`results of frequency analysis of the radius artery pulse wave
`when arrhythmia is not present.
`FIG. 7 is a block diagram showing an example of the
`results of frequency analysis of the radius artery pulse wave
`when arrhythmia is present.
`FIG. 8 is a flow chart showing an example of the pulse
`waveform detection operation according to the first pre-
`ferred embodiment of the present invention.
`FIG. 9A shows an example of the signal obtained by
`adding frequency f, and frequency f,. FIG. 9B is a graph
`showing an example of the result obtained after performing
`FFT on the aforementioned added signal.
`FIGS. 10A, 10B and 10C are graphs showing an example
`of the result obtained after performing FFT on the signals
`output from the pulse wave sensor and the body motion
`sensor when the subject is exercising.
`FIG. 11 is a flow chart showing an example of the method
`for specifying the pulse wave componentin the arrhythmia
`detecting apparatus according to a first preferred embodi-
`ment of the present invention.
`FIG. 12 is a flow chart showing an example of the method
`for specifying the pulse wave component
`in this same
`apparatus.
`FIG. 13 is a graph showing an example of the result
`obtained when FFT is performed on the signal output from
`the body motion sensor.
`FIG. 14 is a flow chart showing an example of the method
`for specifying the pulse wave componentin the arrhythmia
`detecting apparatus according to a first preferred embodi-
`ment of the present invention.
`FIG. 15 is a flow chart showing an example of the method
`for specifying a pulse wave component in this same appa-
`ratus.
`
`FIG. 16 is a flow chart showing an example of the
`operation to detect the pulse waveform in the arrhythmia
`detecting apparatus according to a second preferred embodi-
`ment of the present invention.
`FIG. 17 is a cross-sectional view showing an example of
`installation in the case where a piezo element is employed as
`the notifying means.
`FIG. 18 shows a face chart which may be employed as a
`modification in each of the embodiments of the present
`invention.
`
`FIG. 19 showsthe structure of a system employing the
`arrhythmia detecting apparatus according to each of the
`embodiments of the present invention.
`46
`
`46
`
`

`

`6,095,984
`
`5
`FIG. 20 is a functional block diagram for the arrhythmia
`detecting apparatus according to the third through fifth
`preferred embodiments of the present invention.
`FIG. 21 is a slant view of the arrhythmia detecting
`apparatus according to the third embodiment of the present
`invention.
`
`FIGS. 22A and 22B are explanatory figures showing the
`state of the arrhythmia detecting apparatus accordingto this
`embodimentat the time of use.
`
`10
`
`FIG. 23 is a block diagram showingthe electrical struc-
`ture of the arrhythmia detecting apparatus accordingto this
`embodiment.
`
`FIG. 24 is a block diagram showingthe detailed structure
`of the wavelet transform element according to this embodi-
`ment.
`
`15
`
`FIGS. 25A, 25B, 25C and 25D are timing charts for the
`wavelet transform element according to this embodiment.
`FIG. 26 is a block diagram showingthe detailed structure
`of the decision element according to this same embodiment.
`FIG. 27 showsthe details of memory 124 according to
`this same embodiment.
`
`FIGS. 28A, 28B, 28C, 28D, 28E and 28F are timing
`charts for explaining the operation of this embodiment.
`FIG. 29 is an explanatory figure showingthe state of the
`arrhythmia detecting apparatus according to the fourth pre-
`ferred embodiment of the present invention at the time of
`use.
`
`FIG. 30 is a block diagram showingthe electrical struc-
`ture of the arrhythmia detecting apparatus accordingto this
`same embodiment.
`
`FIGS. 31A, 31B and 31C are waveform diagrams show-
`ing the body motion waveform and the pulse waveform
`according to this same embodiment.
`FIGS. 32A, 32B, 32C, 32D, 32E and 32F are timing
`charts for explaining the operation of this same embodiment.
`FIG. 33 is a block diagram showingthe electrical struc-
`ture of the arrhythmia detecting apparatus according to the
`fifth preferred embodiment of the present invention.
`FIG. 34is a circuit diagram showingthe structure of body
`motion separator 19 according to this same embodiment.
`FIGS. 35A, 35B, 35C, 35D and 35Eare timing charts for
`explaining the operation of this same embodiment.
`FIG. 36 is a block diagram showing the structure of the
`filter bank in a modification.
`
`FIG. 37 is a circuit diagram of the photoelectric pulse
`wave sensor according to a modification.
`FIGS. 38A and 38Bare figures for explaining the state of
`the photoelectric pulse wave sensor when in use, according
`to a modification.
`
`FIG. 39 is a slant view showing a modification in which
`an apparatus connected to a photoelectric pulse wave sensor
`is attached to a pair of eyeglasses.
`FIG. 40 is a circuit diagram for the first wavelet trans-
`forming means according to a modification.
`FIG. 41is a figure for explaining the relationship between
`beat number and pitch in the modification.
`FIGS. 42A and 42B shows the pulse waveform, 42A
`showing a normal pulse wave and 42B showingan arrhyth-
`mia pulse wave.
`FIGS. 43A, 43B and 43C are figures showing an example
`of the notifying means according to the present invention,
`wherein 43A, 43B and 43C show monthly, weekly and
`hourly histograms, respectively.
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`6
`FIG. 44 is a figure showing an example of the time stamp
`which is recorded in the present invention.
`
`DESCRIPTION OF THE PREFERRED
`EMBODIMENTS
`
`Preferred embodiments of the present invention will now
`be explained with reference given to the figures.
`Before explaining the preferred embodiments of the
`present invention, however, a discussion will first be made
`of the basic technical concepts of the present invention.
`Ordinarily, there is a difference between exercising and
`non-exercising pulse rate. In either case, however, provided
`the individual
`is healthy,
`it
`is unusual for the pulse to
`fluctuate sharply. Rather, the change over time in the pulse
`waveform is continuous(regular). In contrast, while various
`types of pulse waveforms maycharacterize an arrhythmia, in
`all cases,
`the continuity of change is disrupted, or cut.
`Accordingly, by detecting irregular change in the pulse
`waveform,it is possible to detect the occurrence of arrhyth-
`mia. This fact comprises the basic technical concept of the
`present invention.
`FIG. 1 showsan overview of the structure of an arrhyth-
`mia detecting apparatus based onthis technical concept. The
`arrhythmia detecting apparatus in this figure is provided
`with a pulse wave detecting means which detects the pulse
`wave in a non-invasive manner and outputs a pulse wave
`signal (pulse waveform); a body motion detecting means
`which detects body motion and outputs a body motion signal
`(body motion waveform); and an arrhythmia detecting
`means which detects the presence or absence of an arrhyth-
`mia event based on the pulse wave signal and body motion
`signal. Body motion exerts an influence on the pulse
`detected by the pulse wave detecting means. Accordingly,
`not only the pulse wave component, but also a body motion
`component, are present in the pulse wave signal which is
`output from the pulse wave detecting means. Therefore, the
`arrhythmia detecting means monitors the pure pulse wave-
`form which is expressed as a pulse wave component