US 6,669,632 B2
`a2) United States Patent
`(10) Patent No.:
`Dec. 30, 2003
`Nanbaetal.
`(45) Date of Patent:
`
`
`US006669632B2
`
`(54) APPARATUS AND METHOD FOR
`ELECTRONICALLY PREDICTING PLEURAL
`PRESSURE FROM PULSE WAVE SIGNALS
`
`1/1993 Hickey oe 600/486
`5,181,517 A *
`.......... 600/330
`1/1995 Yamanishi et al.
`5,385,144 A *
`
`.......00..... 600/547
`3/2002 Kimchiet al.
`6,360,123 B1 *
`OTHER PUBLICATIONS
`
`(75)
`
`.
`(*) Notice:
`
`Inventors: Shinji Nanba, Kariya (JP); Rie Ohsaki,
`Smith et al. “Pulse transit time: an appraisal of potential
`Anjo (JP); Toshiaki Shiomi, 31,
`clinical applications”, Thorax 1999; 54:452-458.*
`Terugaoka, Meitoku, Nagoya-city,
`Argodet al. “Comparison of Esophageal Pressur with Pulse
`Aichi-pref., 465-0042 (JP)
`Transit Ti
`M
`f Respiratory
`Effort for
`Scori
`(3) Asians: Deno Corporation, Kuga i); aHinessMesofReinEftSo
`Toshiaki Shiomi, Nagoya(JP)
`nal of Respiratory and Critical Care Medicine 2000;
`.
`.
`wy
`162:87-93.*
`Subject to any disclaimer, the term of this
`ae
`a,
`«
`.
`.
`.
`patent is extended or adjusted under 35
`peshikt xomura, Hemodynamics Associated with Sleep
`U.S.C. 154(b) by 84 days.
`isorders in Anesthetized
`Dogs”,
`Second
`Department of
`Internal Medicine, Hiroshima University School of Medi-
`(21) Appl. No.: 10/101,835
`cine, Hiroshima, Japan vol. 33(1) 1995 pp. 3-9.
`.
`No.:
`;
`.
`* cited by examiner
`(22)
`Filed:
`Mar. 21, 2002
`(65)
`Prior Publication Data
`
`Primary Examiner—Max F. Hindenburg
`Assistant Examiner—Patricia Mallari
`(74) Attorney, Agent, or Firm—Posz & Bethards, PLC
`US 2002/0143261 Al Oct. 3, 2002
`(57)
`ABSTRACT
`Foreign Application Priority Data
`(30)
`A pleural pressureis predicted from pleural pressure based
`‘en, 28 2002 UD)TTT anoatorsios
`upon pulse wave signals, which are continuously produced
`
`(SL) Unt. C0 eee ecccccccsecseeseeseeseereeseeseeneesees A61B 5/02 and picked up in a time sequential manner.Afirst variation
`(52) U.S. Ch. eee 600/300; 600/500; 600/529;
`Signal indicative of a condition of a variation contained in
`600/561
`the pulse wave signals is acquired based upon the pulse
`(58) Field of Search oo... eee 600/300, 485,
`wave signals. A second variation signal representative of a
`600/500-507, 529-543, 561
`condition of a variation contained in thefirst variation signal
`is acquired based uponthefirst variation signal. The pleural
`pressure is predicted based on a difference betweenthe first
`variation signal and the second variation signal.
`
`(56)
`
`References Cited
`U.S. PATENT DOCUMENTS
`
`4,074,710 A *
`
`2/1978 Tiep w..ccccececeeeeeseees 600/502
`
`26 Claims, 9 Drawing Sheets
`
`START
`
`
`}~si00
`INPUT DIGITAL SIGNAL
`EXTRACT PREURAL PRESSURE SIGNAL }~sio
`
`ACQUIRE PEAK IN ONE HEART BEAT }~ s120
`FORM FERST
`ENVELOPE
`$130
`
`
`$150
`(OT ION?
`
`
`
`a
`
`NO
`CORRECT
`
`FIRST ENVELOPE
`bec
`
`— $160
`FIRST ENVELOPE
`ACQUIRE PEAK OF
`
`
`
`
` APPLE 1011
`
`[
`FORM SECOND
`ENVELOPE
`}~ siz0
`
`CALCULATE DIFFE!
`ENCE AS PREURAL
`$180
`PRESSURE
`
`
`ACQUIRE AMPLITUDE AND BOTTOM
`}— $199
`
`DIAGNOSE SLEEP
`APNEA SYNDROME
`AND UPPER AIRWAY RESISTANCE
`SYNDROME
`
`1
`
`Apple v. Masimo
`IPR2020-01521
`
`APPLE 1011
`Apple v. Masimo
`IPR2020-01521
`
`1
`
`

`

`U.S. Patent
`
`Dec. 30, 2003
`
`Sheet 1 of 9
`
`US 6,669,632 B2
`
`
`
`AV1dS10
`
`
`
`q}OlsViOld
`
`2
`
`
`
`

`

`U.S. Patent
`
`Dec. 30, 2003
`
`Sheet 2 of 9
`
`US 6,669,632 B2
`
`FIG. 2
`
`
`
`FIG. 3
`
`BEAT INTERVAL
`
`
`
`3
`
`

`

`U.S. Patent
`
`Dec.30, 2003
`
`Sheet 3 of 9
`
`US 6,669,632 B2
`
`FIG. 4
`
`
`
`(START)
`__L
`|~s100
`INPUT DIGITAL SIGNAL
`i
`EXTRACT PREURAL PRESSURE SIGNAL }~sito
`|ACQUIRE PEAK IN ONE HEART BEAT }~3120
`
`
`
`
`
`FORM FIRST ENVELOPE
`
`~ $130
`
`
`BODY MOTION?
`
` CORRECT
`
`FIRST ENVELOPE
`
`
`
`|ACQUIRE PEAK OF FIRST ENVELOPE L~si60
`____
`FORM SECOND ENVELOPE
`|—-St70
`CALCULATEDIFFERENCE AS PREURAL|
` oyaq
`
`PRESSURE
`
`|ACQUIRE AMPLITUDE AND BOTTOM }~siso
`
`
`DIAGNOSE SLEEP APNEA SYNDROME
`AND UPPER AIRWAY RESISTANCE
`
`SYNDROME
`
`CN)
`
`4
`
`

`

`U.S. Patent
`
`Dec. 30, 2003
`
`Sheet 4 of 9
`
`US 6,669,632 B2
`
`AdOd
`
`OSOVOF02Ol0||5|:||||||Wag=se|\fT1Sul4;id
`
`
`ryy\iH=wyANEsf
`
`
`WENNS_-~dOSANS||_3d073AN3_LSUTSgyN00|NOTLOWWWdd
`JAVK3S1Nd
`
`
`
`[98s]JWIL
`
`5
`
`
`

`

`QNOOAS
`
`
`
`
`
`
`[UlW]SWIL
`
`NS)
`
` Bt
`91viZ||TO.
`
`
`N\A}ffN
`yeNh
`
`U.S. Patent
`
`Dec. 30, 2003
`
`Sheet 5 of 9
`
`US 6,669,632 B2
`
`
`
`AdQTSANS~=4dOVSANI
`
`LSdTs
`
`
`
`bfeeNgeeseeneeeeeteeeeeeeetbeeen
`
`[
`
`A] 35VLIOA
`
`AdNSS3Ud TWansd
`
`49Sls
`
`6
`
`
`

`

`U.S. Patent
`
`Dec. 30, 2003
`
`Sheet 6 of 9
`
`US 6,669,632 B2
`
`
`
`LOls
`
`(CFuNSVAW)
`
`/
`
`
`
`Ih
`
` EANThELRThy|VVVVVVTVAv
`
`
`|
`Zgi
`9|||
`.||
`fo
`aél|
`~—_82-|
`
`[paranel
`AUNSSdddWadd
`
`
`
`(ULW]SWIL
`
`80
`
`—-||
`
`||
`
`|(C4191
`Gud)
`
`||tL
`
`v0
`
`20
`
`Q
`
`le-
`
`[OzHWO] FYNSSIYd “VENA Td
`
`
`|"Naat
`
`SuASSIUdWSSVHdOSS
`SOaan LT,
`
`7
`
`

`

`U.S. Patent
`
`Dec. 30, 2003
`
`Sheet 7 of 9
`
`US 6,669,632 B2
`
`FIG. 8
`
`0 ——____—
`
`N=111
`2
`R* =0. 1033
`
`_
`
`—-14
`
`So
`NN
`=
`
`2L
`
`od
`oa
`=>
`
`aw
`
`s a
`
`a.
`
`
`
`[V]
`
`4
`
`ta
`
`3
`
`z=
`v
`Lut
`
`4_w
`
`=
`
`—28
`
`
`
`-—28
`
`—2]
`
`-—14
`
`-7
`
`0
`
`PREDICTED PLEURAL PRESSURE [comH20]
`
`
`
` PULSEWAVESIGNAL
`
`TIME
`
`sec |
`
`8
`
`

`

`U.S. Patent
`
`Dec.30, 2003
`
`Sheet 8 of 9
`
`US 6,669,632 B2
`
`FIG. 10
`
`
`
`
`
`
`
`
`5aB
`
`ODY MOTION
`
`ere
`
`|
`
`|
`
`Hd 6
`
`H5
`
`|
`—_i
`i
`
`oeee |
`ae
`
`25
`
`30
`
`35
`
`40
`
`TIME [sec]
`
`FIG. 11
`
`DIVIDED (a:b)
`
`
`
`
`
` PULSEWAVESIGNAL[V]
`
`>ol
`
`=S
`
`os
`3
`
`9
`
`

`

`U.S. Patent
`
`Dec.30, 2003
`
`Sheet 9 of 9
`
`US 6,669,632 B2
`
`FIG. 12A
`
`
`
`POWERSPECTRUM
`
`
`
`POWERSPECTRUM
`
`
`
`4
`
`45
`
`46
`
`
`
`
`
`4
`
`45
`
`5
`
`
`
`
`
`
`
`
` NO BODY MOTION
`
`29
`
`$
`
`3.5
`
`FREQUENCY [Hz]
`
`FIG. 12B
`
`25
`
`3
`
`35
`
`FREQUENCY [42]
`
`10
`
`10
`
`

`

`US 6,669,632 B2
`
`1
`APPARATUS AND METHOD FOR
`ELECTRONICALLY PREDICTING PLEURAL
`PRESSURE FROM PULSE WAVE SIGNALS
`
`CROSS REFERENCE TO RELATED
`APPLICATION
`
`This application is based on and incorporates herein by
`reference Japanese Patent Applications No. 2001-100527
`filed Mar. 30, 2001 and No. 2002-18403filed Jan. 28, 2002.
`
`FIELD OF THE INVENTION
`
`The present inventionis related to a pleural or esophageal
`pressure predicting apparatus, a medical apparatus, a pleural
`or esophageal pressure predicting method, a diagnosing
`method and an examining method, which are capable of
`diagnosing a sleep apnea syndrome, an upper airwayresis-
`tance syndrome, andthelike, and also capable of evaluating
`a medical treatment effect, while monitoring, for instance,
`pleural or esophageal pressure conditions during sleep time.
`
`BACKGROUND OF THE INVENTION
`
`2
`tained in thefirst variation signal is acquired based upon the
`first variation signal. The pleural pressure is predicted based
`on a difference between the first variation signal and the
`second variation signal.
`This is based on a finding that there is a high correlation
`between a difference between the first variation signal (for
`example, first envelope) and the second variation signal (for
`instance, second envelope), and also the pleural pressure
`indicated by actual esophageal pressure.
`The first variation signal shows a condition of variations
`(fluctuations) of the entire pulse wave signals in which a
`large number of pulse wave signals are continuously con-
`nected to each other. This first variation signal is mainly
`varied in response to a change containedin pleural pressure.
`Amongsignal componentsofthis first variation signal, there
`are contained external disturbance components (for instance,
`signal components of autonomic nervous system for con-
`trolling blood pressure, for expanding blood vessel, and for
`compressing blood vessel) other than the pleural pressure.
`The second variation signal corresponds to such a signal
`having a lower frequency (for example, lowerthan, or equal
`to 1 Hz) than the frequencyofthefirst variation signal. Also,
`this second variation signal mainly contains the above
`external disturbance components. As a result, since the
`difference between the first variation signal and the second
`variation signal is calculated, such a signal indicative of only
`pleural pressure can be derived by eliminating the external
`disturbance components.
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`to
`Pleural pressure has close correlation with respect
`health conditions of patients, and hence may beeffectively
`used to diagnose closed type apnea syndromes and upper
`airway resistance syndromes and the like. However, it is
`practically difficult to measure such pleural pressure. Also,
`measurements of esophageal pressure which may substitute
`the above pleural pressure measurements cannot be easily
`carried out, since sensors should be inserted from nose holes
`into esophaguses which may give strong pain to patients.
`Therefore, in actual cases, the above sleep apnea syndromes
`and upper airway resistance syndromes are diagnosed by
`monitoring respiration conditions based upon air streams
`from noses, motion of breasts, motion of venter and the like.
`
`However, in the case of such a diagnosing method by
`monitoring the respiration conditions based upon the air
`streams from the noses and the like, it is cumbersome to
`attach the sensors, or instruments. Furthermore, diagnostic
`precision of this respiration condition monitoring method is
`low. As a consequence, development of easier diagnosing
`methods with higher diagnostic precision has been strongly
`required in this medicalfield.
`
`SUMMARYOF THE INVENTION
`
`The above and other objects, features and advantages of
`the present invention will become more apparent from the
`following detailed description made with reference to the
`accompanying drawings. In the drawings:
`FIG. 1A is a schematic view of a pulse wave sensor used
`in a medical apparatus equipped with a pleural pressure
`predicting apparatus according to the first embodiment of
`the present invention, and FIG. 1B is an electric wiring
`diagram of the medical apparatus according to the first
`embodiment;
`FIG. 2 is a sectional view of the pulse wave sensor used
`in the first embodiment;
`FIG. 3 is a signal waveform chart graphically showing a
`waveform of a pulse wave signal;
`FIG. 4 is a flow chart of processing for predicting pleural
`The present invention has an object to provide a pleural
`pressure and of diagnosing patient conditions, executed in
`pressure predicting apparatus, a medical apparatus, a pleural
`the first embodiment;
`pressure predicting method, a diagnosing method, and an
`FIG. 5 is a signal waveform chart graphically showing
`examining method, capable of diagnosing a sleep apnea
`both a pulse wave signal and an envelope thereof;
`syndrome, an upper airway resistance syndrome, and the
`FIG. 6A is a signal waveform chart graphically showing
`like, while pleural pressure can be predicted by way of a
`bothafirst envelope and a second envelope, and FIG. 6B is
`simple method without giving loads on patients.
`a signal waveform chart graphically showing a pleural
`According to the present invention, a pleural or esoph-
`pressure signal acquired from a difference between the first
`ageal pressure is predicted from pleural pressure based upon
`and second envelopes;
`pulse wave signals, which are continuously produced and
`FIG. 7 is a signal waveform chart graphically showing a
`picked up in a time sequential manner. A first variation
`correlative relationship between pleural pressure measured
`signal indicative of a conditionof a variation (fluctuations of
`from a pulse wave and an actually-measured esophageal
`entire pulse wave signals) contained in the pulse wave
`pressure value;
`signals is acquired based upon the pulse wave signals. A
`second variation signal representative of a condition of a
`FIG. 8 is a graph graphically indicating a correlative
`variation (fluctuations of entire first variation signals) con-
`relationship between pleural pressure measured from a pulse
`
`11
`
`11
`
`

`

`US 6,669,632 B2
`
`4
`As a result, since the amount of light absorbed by the
`blood capillary 1b changes,
`the amount of such light
`detected by the light receiving element 11 changes, so that
`this change in the light receiving amount is outputted from
`this pulse wave sensor 1 to the data processing unit 5 as
`pulse wave information (that is, sensor output A correspond-
`ing to voltage signal indicative of pulse wave). The exem-
`plary waveform pattern of this sensor output signal A is
`shown in FIG. 3.
`
`10
`
`3
`wave and an actually-measured pleural pressure by an
`esophageal pressure sensor;
`FIG. 9 is a signal waveform chart graphically showing a
`variation condition of pulse waves caused by body motion;
`FIG. 10 is a signal waveform chart showing a correction
`method for removing an adverse influence of the body
`motion from the pulse waves;
`FIG. 11 is a signal waveform chart showing graphically a
`pleural pressure predicting method, which uses an amplitude
`ratio line, according to the second embodiment of the
`present invention; and
`FIGS. 12A and 12B are graphs graphically showing a
`variation condition of frequency components contained in
`pulse waves, caused by body motion, according to the third
`embodiment of the present invention.
`DETAILED DESCRIPTION OF THE
`PREFERRED EMBODIMENTS
`
`(First Embodiment)
`Referring first to FIGS. 1A and 1B, a medical apparatus
`is provided with a pulse wave sensor 1, a drive circuit 3, a
`data processing unit 5, and a display device 7. The pulse
`wave sensor 1 is used by being mounted on such a place of
`a humanbody 1a, for example, a rear side portion ofa wrist,
`the movement of which is small. The drive circuit 3 elec-
`
`the data processing unit 5 has a
`On the other hand,
`detecting circuit 15, an A/D converter (ADC) 17 and a
`microcomputer 19. The detecting circuit 15 amplifies the
`sensor output A. The A/D converter 17 converts a detected
`analog signal B indicative of the pulse wave amplified by the
`detecting circuit 15 into a digital signal C. The microcom-
`puter 19 processes the digital signal C to predict the pleural
`pressure and the like.
`The microcomputer 19 corresponds to such an electronic
`circuit provided with a CPU, a ROM, a RAM, andthe like,
`which are well known in this technical field. A computer
`program used to process the digital signal C obtained from
`the ADC 17 is stored in this microcomputer 19. This
`program is made based upon an algorithm by which the
`signal indicative of the pulse wave detected by the pulse
`wave sensor 1 is processed to predict the pleural pressure,
`trically drives the pulse wave sensor 1. The data processing
`and to diagnose and examineboth the sleep apnea syndrome
`unit 5 measures human pulse waves based upon a measure-
`and the upper airway resistance syndrome.
`ment result outputted from the pulse wave sensor 1 to
`The microcomputer 19 is programmed to execute the
`calculate a pulse interval, a pulse fluctuation, and the like.
`processing shown in FIG. 4.
`The display device 7 displays thereon a processed result
`(1) Major Processing Operation:
`obtained from the data processing unit 5.
`As shown in FIG. 4, the microcomputer 19 executes an
`As shownin FIG. 2, the pulse wave sensor 1 is an optical
`input of the pulse wave at step S100. For instance, the digital
`type reflection mode sensor equipped with a light emitting
`signal C is inputted into the microcomputer 19. At step $110
`element (for instance, green LED) 9, a light receiving
`subsequentto this step $100, the microcomputer 19 executes
`element (for example, photodiode) 11, a transparent lens
`a digital filtering processing in order to extract a pleural
`body 13, and the like. The transparent lens body 13 may
`pressure signal. In order to extract the pleural pressure signal
`penetrate therethrough light, and may receive light in high
`whichis reflected on the pulse wave from this digital signal
`efficiency.
`C, both such noise having frequencies higher than, or equal
`This pulse wave sensor 1 has a higher S/N ratio of 1/60
`to 3 Hz due to external disturbance noise, and also signals
`than that of the conventional pulse wave sensor, and ampli-
`having frequencies lower than, or equal to 0.1 Hz (namely,
`fies all of frequencies of pulse waves, namely, performs a
`lower frequencies from that of pleural pressure signal),
`DC amplification. This S/N ratio impliesaratio of a signal
`which are caused by body motion (motion artifact), are cut
`component “S” to a total light receiving amount “N” in
`off.
`conjunction with a volume changeof bloods. As a result, this
`pulse wave sensor 1 may acquire not only a pulse wave
`number, but also an activity of an autonomic nerve system
`which controls a heart beat number, and furthermore, may
`acquire most of such information contained in pulse waves
`such as a respiration condition and a blood vessel condition.
`Whenlightis irradiated from the light emitting element 9
`to the human body 1a by supplying electric drive power
`from the drive circuit 3, a portion of this light is lightened
`on a blood capillaries 1b which runsinside the human body
`la. Then,
`the light
`is absorbed by mainly hemoglobin
`contained in blood flowing through each blood capillary 1b.
`The remaining light
`is repeatedly scattered, and then, a
`portion of this scattering light enters into the light receiving
`element 11. At this time, since the amount of hemoglobin in
`the blood capillary 1b is changed in a wave motion manner
`by pulsatory moveofthe blood, light which is to be absorbed
`by the hemoglobin is also changed in a wave motion manner.
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`Then, the microcomputer 19 executes processing capable
`of extracting a feature of the pulse waveform acquired in the
`previous step $110, and for producing a numeral value from
`this extracted feature. In this case, a method for extracting a
`feature of a waveform (pulse wave waveform) of a pulse
`wave signal by employing a variation (fluctuation) of this
`pulse wavesignal is described.
`As indicated in FIG. 5, peaks of the pulse wave of one
`heart beat are acquired at step $120. It should be understood
`that FIG. 5 shows a temporal change of signal outputs of
`pulse waves, and an ordinate indicates a magnitude of an
`output of a pulse wave signal from a reference value (0). At
`the subsequent step $130, the respective peaks acquired in
`the previous step $120 are connected to each other to form
`a first envelope as shown in FIG. 5.
`At
`the next step $140,
`the microcomputer 19 checks
`whether or not body motion is present.
`In the case of
`
`12
`
`12
`
`

`

`US 6,669,632 B2
`
`5
`presence of the body motion, the processing is advanced to
`a step S150. To the contrary, in the case of no body motion,
`the processing is advanced to another step S160.
`At step $150, since the body motion is present, the first
`envelope after the body motion (after end of body motion)
`is corrected by executing a methodof correcting an envelope
`to remove an adverseinfluence of this body motion from the
`first envelope formed at step $130. In the case that the body
`motion is present, pulse wave signals appearing during a
`time duration of the body motion indicated by a dot and dash
`line in FIG. 5 are cut off and are not used.
`
`On the other hand, at step S160, peaks of the first
`envelope which is obtained when the body motion is not
`present at step S130 are acquired, or peaks of the first
`envelop whichis obtained by the correction made in the case
`that the body motion is present at step S150 are acquired. At
`the next step S170, the respective peaksofthe first envelope
`are connected to each other to form a second envelope
`indicated by broken line in FIG. 5. At the subsequent step
`$180, the microcomputer 19 calculates a difference between
`the first envelope and second envelope, and then sets this
`difference as a pleural pressure signal.
`At the subsequent step $190, both an amplitude and a
`bottom of this pleural pressure signal are acquired. This
`amplitude of the pleural pressure signal indicates a variation
`of pleural pressure, and this bottom thereof shows pleural
`pressure whenair is sucked.
`the microcomputer 19
`At
`the subsequent step S200,
`diagnoses both the sleep apnea syndrome and the upper
`airway resistance syndrome by employing both the ampli-
`tude and the bottom of the pleural pressure signal acquired
`in the previous step $190. Then, this processing is once
`finished. The pleural pressure and diagnostic results are
`displayed on the display device 7.
`(2) Method for Predicting Pleural Pressure:
`In this embodiment, the pleural pressure is predicted as
`follows.
`
`As described above, the pulse wave signal is detected in
`which waveformsof individual signals correspondingto the
`respective pulse waves are continued, and peaks (namely,
`upper peaks in this case) of respective waveforms of this
`pulse wave signal are connected to each other in order to
`acquire the first envelope. As shown in FIG. 6A,thisfirst
`envelope has such a waveform whichis varied in connection
`with a changein time. It should be noted that an ordinate of
`this figure shows a voltage [V] and an abscissa thereof
`represents time [minute].
`Next, peaks (upper peaks in this case) of the respective
`waveformsofthe varied first envelope are connected to each
`other to acquire the second envelope. As shown in this
`figure, while this second envelope has waveforms which are
`varied in connection with a change in time, a frequency of
`this second envelope is lower than the frequency ofthe first
`envelope, and also represents a small variation.
`Next,
`the second envelop is subtracted from the first
`envelop to obtain a difference signal corresponding to the
`pleural pressure signal, as represented in FIG. 6B. It should
`be understood that a pleural exponent of an ordinate of this
`figure implies a value indicative of pressure, and an abscissa
`thereof shows time [minute].
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`6
`It is confirmed by the inventors that the waveform ofthis
`difference signal is largely changed (namely, amplitude of
`waveform is increased) during each apnea condition indi-
`cated with circles in FIGS. 6A and 6B. As a result, the apnea
`condition can be diagnosed based uponthe difference signal.
`As a consequence, both the sleep apnea syndromeand the
`upper airway resistance syndrome can be diagnosed from
`the difference signal.
`Also, it is confirmed by the inventors that a correlative
`relationship shown in FIG. 7 holds between an actually-
`measured esophageal pressure value indicative of actual
`pleural pressure, and the predicted pleural pressure value
`(difference between the first envelope and the second enve-
`lope. As a consequence, this difference may be regarded as
`such a signal indicative of pleural pressure (pleural pressure
`signal).
`Furthermore, it is confirmed by the inventors that such a
`correlative relationship (correlative coefficient R=0.85,
`R?=approx. 0.7) shownin FIG.8 holds betweenthe actually-
`measured esophageal pressure value indicative of actual
`pleural pressure, and the pleural pressure predicted by the
`difference between the first envelope and the second enve-
`lope. As a consequence, this difference may be regarded as
`such a signal indicative of pleural pressure (pleural pressure
`signal). It should be noted that a total data numberN in this
`testing is 111.
`(3) Method of Detecting Body Motion:
`Next, the method of detecting the body motion executed
`in the step $140 will now be explained.
`As to a measuring portion of a pulse wave by the pulse
`wave sensor 1, such a portion having small motion as a
`wrist, an arm, a foot, and a forehead is suitable in view of
`a measurementstability, and furthermore, in viewsof attach-
`ability of this pulse wave sensor 1. The optimum measuring
`portion is a rear-sided portion of a wrist in view of the
`attachability of the pulse wave sensor1.
`In general, a pulse wave is measured at a finger tip
`portion. While pulse waves are measured at such a fingertip
`portion, waveforms of these pulse waves are strongly
`changed due to temperatures of this finger tip portion, and
`the mounting characteristic (depression pressure of detect-
`ing unit), and motion. As a result, this measuring operation
`at the finger tip portion is not always properly carried out.
`Also, such a place whose motionis large as an inner portion
`of a wrist where a tendonis present is not preferable.
`However, even when pulse waves are measured on a rear
`side of a wrist, there are somepossibilities that wave heights
`of the pulse waves are changed. In most cases, a change of
`bodyattitudes (body motion) may occur in connection with
`these wave heights. As a result, a body motion signal is
`superimposed on these pulse waves. Accordingly, detection
`of such a body motion is important.
`in this
`As a consequence, as shown in FIG. 9,
`embodiment, the amplitude of the first envelope is mea-
`sured. When this amplitude of the envelope is increased (for
`example, 3 times, or more larger than average value of
`previously-measured amplitudes),
`the microcomputer 19
`determines that body motionstarted. Also, in such a casethat
`the large amplitude of the envelope is returned to be
`decreased (for instance, 1.2 times, or more smaller than
`
`13
`
`13
`
`

`

`US 6,669,632 B2
`
`7
`average value of amplitudes measured before body motion
`is commenced), the microcomputer 19 determines that the
`body motion disappeared.
`(4) Method of Correcting Envelope:
`When a body motion occurs, since a body motion signal
`caused by the occurrence of this body motion is superim-
`posed on a pulse wave signal, an adverse influence thereof
`is required to be eliminated. As a consequence,
`in this
`embodiment, at a time instant when a body motion signal
`appears, a pleural pressure signal is corrected at step S150
`in FIG. 4 based upon waveheights of pulse waves existing
`before/after this time instant under stable conditions.
`
`Morespecifically, as shown in FIG. 10, an average wave
`height Havb of wave heights H1 to H3 ofthree heart beats
`to five heart beats before such a time period during which
`body motion appears is calculated based upon the following
`equation (1):
`
`Havb=(H1+H2+H3)/3
`
`(4)
`
`Similarly, an average wave height Havaof wave heights
`Hé4to H6of three heart beats to five heart beats after body
`motion disappears is calculated based upon the following
`equation (2):
`
`10
`
`15
`
`20
`
`25
`
`Hava=(H4+H5+H6)/3
`
`(2)
`
`8
`
`(6) Calibration Method:
`The pleural pressure signal produced from the pulse wave
`indicates not an absolute value of the pleural pressure, but a
`relative value of the same. As a result, when an absolute
`value is desired, a calibration should be carried out.
`As this calibration manner, for example, while constant
`pressure (for instance, 20 cmH,0)is applied to a mouth, or
`a nose of a patient by employing a respirator such as CPAP,
`the pleural pressure whichis actually predicted by the pulse
`waves may be calibrated. Also, while a pressure control
`valve is provided at an air intake port of an air intake mask,
`constant negative pressure is produced whenair is sucked.
`Then, the pleural pressure value may be calibrated which is
`predicted by pulse waves appearing under this negative
`pressure. Furthermore, while a pressure sensor is mounted
`on the above air intake mask, the pulse wave prediction
`value may be calibrated based upon a pressure value
`obtained form this pressure sensor. It should be understood
`that while these calibrations are carried out,
`the pleural
`pressure value is normalized as the average wave height of
`the pulse waves. That is, when the sensor depression force
`is changed, the pulse waves are increased, or decreased.
`Otherwise,
`the pleural pressure signal
`is increased, or
`decreased in proportionalto this depression pressure change.
`Therefore, this value is divided by the average wave height
`to be corrected.
`
`Then, as defined in the following equation (3), while the
`average wave height Havb before the body motion is used as
`a reference, the pleural pressure signal is corrected based
`upon an increase/decrease ratio of the average wave height
`Hava after the body motion:
`
`corrected envelope after appearance of body motion=(envelope
`acquired in step $130)x(Havb/Hava)
`
`3)
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`As described above, the microcomputer 19 processes the
`pulse wave signals acquired by the pulse wave sensor 1, so
`that the difference is calculated between the first envelope
`indicative of the variation of the pulse wave signals and the
`second envelope representative of the variation of thefirst
`envelope signal. As a result, the pleural pressure can be
`correctly predicted by this calculated difference. As a
`consequence, in comparison with the conventional method
`for directly measuring the esophageal pressure, the pleural
`pressure can be predicted by wayof the very simple manner
`Thus, after the body motion appears, such an envelope of
`without giving the loads to the patient.
`the pulse wave signal which is obtained by performing the
`Furthermore, when the above difference is calculated,it is
`correction at step $150 of FIG. 4 may be used asthefirst
`checked whether or not the body motion is present. When
`envelope. That
`is,
`the above envelope obtained in the
`equation (3) is used as the first envelope.
`such a body motion appears, the correction is carried out by
`(5) Diagnosing Method:
`eliminating the adverse influence caused by this body
`motion from the pulse wave signal. As a consequence,there
`Diagnosing at step S200 is executed as follows with
`is an advantage that the pleural pressure can be correctly
`respect to both the sleep apnea syndromeand also the upper
`acquired in view of this correction.
`airway resistance syndrome.
`Moreover, while the pleural pressure acquired from the
`At a time instant when the envelope produced by con-
`pulse waves is employed, such a diagnostic operation as to
`necting the bottoms of the pleural pressure signal shown in
`the sleep apnea syndrome and the upper airway resistance
`FIG. 7 decreases gradually and then reaches a predetermined
`value (for example, -13 cmH,O),
`the microcomputer 19
`syndrome can be carried out by executing a computerized
`diagnosis,
`the cumbersome diagnosing works can be
`determinesthis state as a closed type apnea syndrome. In the
`reduced, and the diagnostic precision can be improved.
`case that
`the total number of this reaching operation
`(Second Embodiment)
`becomes larger than, or equal to 5 times per one hour, the
`In this second embodiment, a first amplitude ratio line
`microcomputer 19 determines this state as a sleep apnea
`(respiration curve) is employed instead of the above first
`syndrome.
`envelop used in the first embodiment.
`Also, under such a case that the total time of closed type
`In accordance with this embodiment, as shown in FIG. 11,
`apnea is smaller than, or equal to 5 times per hour which is
`amplitudes (L) of the respective waveformsof a pulse wave
`checked by a sleep polygraph, when the pleural pressure
`signal are divided by an average waveheightto be corrected.
`prediction value predicted by the above method is decreased
`Such points that the amplitudes are subdivided by a prede-
`lower than, or equal to a predetermined value, there are
`termined ratio (for example, arbitrary ratio of a:b) are
`manypossibilities that the upper airway resistance syndrome
`65
`
`occurs. Therefore, the diagnosis executed by employing the connected to each other to acquireafirst amplitude ratio line
`(respiration curve) indicative of a variation condition of the
`pulse wave sensor 1 is additionally carried out in conjunc-
`tion with the sleep polygraph andthelike.
`pulse wave signals.
`
`14
`
`14
`
`

`

`US 6,669,632 B2
`
`9
`Furthermore, such points that the amplitudes are subdi-
`vided by a predetermined ratio (for example, arbitrary ratio
`of a:b) are connected to each other to acquire a second
`amplitude ratio line indicative of a variation of the first
`amplitude ratio line.
`Thereafter, the second amplitude ratio line is subtracted
`form the first amplitude ratio to obtain a difference signal
`(namely, pleural pressure signal). Then, pleural pressure
`may be predicted based upon this pleural pressure signal in
`the similar manneras in the first embodiment.
`
`Also, in this second embodiment, since the pleural pres-
`sure signal acquired by executing the above method corre-
`sponds to the pleural pressure, the pleural pressure can be
`predicted from this pleural pressure signal similar to the first
`embodiment. Also,
`in accordance with this second
`embodiment, the diagnosing operation asto either the sleep
`apnea syndrome or the upper airway resistance syndrome
`can be carried out.
`It should also be noted that various methodsother than the
`
`above method of the second embodiment maybecarried out.
`That is, (1) while peaks of the respective waveformsof the
`first amplitude ratio line are connected to each otherin order
`to obtain the second envelope, the pleural pressure signal
`may be acquired based upon the difference betweenthefirst
`amplitude ratio line and the second envelope. Alternatively,
`(2) while such points that the amplitudes of the respective
`waveformsofthe first envelope are subdivided by a prese-
`lected ratio are connected to each otherin order to obtain the
`
`second amplitude ratio line, the pleural pressure signal may
`be acquired based upon the difference between the first
`envelope and the second amplituderatio line. Also, although
`precision is lowered, a f