||||||||||||||||||||||||||||||||||||||||||||||||l||||||||||||||||||||||||||
`
`USl)t)6721584B2
`
`(12) Ulllted States Patent
`Baker, Jr. ct al.
`
`(10) Patent No.:
`(45) Date of Patent:
`
`US 6,721,584 B2
`Apr. 13, 2004
`
`(54) M1+:'t‘H0t) AND /\I'l’ARA'I‘US FOR
`ESTIMATING PHYSIOIDGICAI.
`PARAMETERS USING MODEL-BASED
`ADAPTIVE FILTERING
`
`(56)
`
`(75)
`
`Invcnlors; Clark R, Baker, Jr,’ C351“) Valley‘ CA
`(US); Thomas J. Yurkey, San Ramon,
`~
`(‘A (US)
`
`(73) Assignee: Nellctir I"ttritan Bennett Incorporated,
`Pleasanton, CA (US)
`_
`‘
`_
`_
`_
`Subject to any disclaimer, the term of this
`patent
`is extended or adjusted under 35
`u.s.c. l54(b) by 0 days.
`
`_
`( s ) Notice:
`
`(21) A991’ N0‘: 09‘r876’0M
`(33
`1:i}cd:
`J1, n_ 5, 2001
`
`(65)
`
`Prior Publication Data
`US 2(JD2_t'{l[}458fJo Al Apr. 18, 2802
`
`Related U.S. Application Data
`
`(60)
`
`Colllillllalion ofapplieation No. 09/435,144, [lied on NOV. 5,
`1999, now abandoned, which is a continuation of appl icalimi
`NE W”31479} med 0" Aug 2”,
`I998’ new ML Na
`t‘i__U83_.l'r'2__ which is
`:1 continuation of application No.
`(]8,='r')b'[).5lo,
`|-‘.|¢d on J tin. T, 1996, now Pat. No. 5,853,364,
`Provisional application No. o(Jr'D(t[I,]‘)5. filed on Jun. 14,
`1-991
`Int. C1.’
`(51)
`(52) U.S. Cl.
`(58)
`F‘ Id Y‘
`_
`e
`n . earc
`
`11
`
`A613 5100
`600E323
`“WHO T”
`, _.._.,
`i
`_
`600E323, 331, 330, 336
`
`References Cited
`_.
`..
`.
`t
`Uh‘ "A””‘N’ DOCUMLN”
`4,266,554 A * M1931 Hamaguri
`4,621,643 A - ririese New et al.
`4,9-5|-U,l26 A *
`l{l)"l.99U Conlon Cl Ell.
`g0I'l€.;III':1Bl|1 cl al.
`..
`ia
`e a.
`2:199!) Robinson et al.
`5:199? Diahet al.
`
`ritirarszt
`sriarasi
`GIUK].-"3313
`fimnzs
`..
`.. 600F323
`
`oUt'Jr'323
`
`*
`:1
`_
`.
`5,494,032 A "‘
`S_.o32,2'.r’2 A ==
`_
`_
`* Wed '3!’ examlncr
`Pri‘mar_v Exr.-mi'rier—lirie 1*. Winakur
`(74) Aflorney’ Agemr or 1_.h_m_T0wnSend & Townsend &
`(jrcw LLP '
`'
`
`(57)
`
`ABSTRACT
`
`A method and apparatus for reducing the eifeets of noise on
`a system for measuring physiological parameters, such as,
`for example, a pulse oximeter. The method and apparatus of
`the invention take into account the physical limitations on
`various physiological parameters being monitored when
`weighting and averaging a series of measurements. Varying
`weights are assigned different measurements. measurements
`are rejected, and the averaging period is adjusted according
`to the reliability of the measurements. Similarly, calculated
`values derived from analyzing the measurements are also
`asgigngd varying wcjghlg and avcragcgl Qvcf
`adju_-'._‘,[a[1]c
`-
`,-
`.
`periods. More spcerfically, a general class of filters such as,
`fur examples Kalman mlcr-'5- l5 cmplolfed if: l3r0C‘*'53ini-3 [hi-3
`measurements and calculated values. The tillers use math-
`ematical models which describe how the physiological
`parameters change in time, and how these parameters relate
`1? mcasvrcrrlcni In a new environment. "1110 fi11ers_a=|ap-
`“wily modify a .5el Lit
`‘Wemgmg welghls [0 Opm-“any
`estimate the physiological parameters.
`4 Claims, 12 Drawing Sheets
`
`”
`“
`
`uiuiut Lowiiiiu
`ata:t‘~A?:s'iiti:t
`
`_|ll|I£l
`
`_
`
`“X
`"Lite 5
`_
`_
`.0 IBEI
`at
`g
`__
`pm;
`.
`BESIMIE
`me:
`
`
`
`5“‘\L
`_ _ F _ __'
`
`E
`E
`t
`EI
`
`IIn
`
`°°°‘
`
`U.S. Patent No. 89239
`
`Apple I
`APL10
`
`Apple Inc.
`APL1063
`U.S. Patent No. 8,923,941
`
`0001
`
`

`
`U.S. Patent
`
`Apr. 13, 2004
`
`Sheet 1 of 12
`
`US 6,721,584 B2
`
`om
`acumsmon
`
`I'°
`
`NATURAL
`LOGARITIIII
`
`BAND PASS
`FILTER
`
`*2
`
`"'
`
`IR 5. RED
`
`IR
`IIORHALIZE
`
`
`
`
`POWER"
`SPECTRUM
`
`PATTERN
`HATCH RATE
`
`
`ADAPTIVE
`GOIIB FILTER
`
`
`
`52\
`
`5!!
`
`:2
`
`RATE
`
`
`
`I I I I I I
`
`I I
`
`I
`I - 32:
`17‘
`REJEGTIOII
`,
`I j 2|;
`}
`us on mu
`,
`I
`KALHAII SAT
`I
`I _ 24:
`I
`I
`I
`I
`L
`
`I
`
`BEST RATE T
`
`as
`
`TRIGGERS
`on £06
`
`POST PROCESS “
`
`F/G. /A
`
`I
`
`DISPLAY we
`
`'5
`
`2:
`
`za
`
`POST PROCESS
`
`DISPW SAT
`
`0002
`
`0002
`
`

`
`U.S. Patent
`
`Apr. 13, 2004
`
`Sheet 2 of 12
`
`US 6,721,584 B2
`
`om AGOUISITIOII
`
`NATURAL LOGARITHII
`
`mu PASS FILTER
`
`ETA CORRECTION
`
`'9
`
`I2
`
`'4
`
`I5
`
`I
`
`IR
`
`IIIOHIIALIZE
`
`53 I
`
`POWER
`
`zsao cnoss
`
`52\
`spscrnun
`
`ADAPTIVE
`OOIIB FILTER
`
`
`
`HIIRI-IONIC
`FILTER
`
`
`
`_
`
`NURHALIZE
`
`
`
`42
`
`E
`
`R“
`
`I
`
`BEST mm-:
`
`UIIIIIIIITEII
`
`36
`
`TRIGGEHS
`-03 E06
`
`....I"__..‘E_._..____-.._.___.._...._.__|
`
`I I
`
`I I I I I '
`
`I I I I I I I
`
`ru:9
`I
`I
`IIHIIEII
`I
`I
`I
`. I 22,
`{
`DUTLIER
`I
`,
`REJEGTION
`I
`- 1 2“
`GL5 OII
`INIT-
`
`:
`IIALIIIAII SAT
`j 241
`' BEST SAT
`:
`I"\3 O3
`L_-__ ___J
`
`POST PROCESS
`
`F/G:
`
`/5!
`
`DISPLAY RATE
`
`4
`
`4
`
`48
`
`0003
`
`27
`
`23
`
`POST PROCESS
`
`DISPLAY SAT
`
`0003
`
`

`
`U.S. Patent
`
`Apr. 13, 2004
`
`Sheet 3 of 12
`
`US 6,721,584 B2
`
`:3
`F“
`
`I6
`
`(HZ)
`
`24s3I0I2 rnsoucucv
`
`0004
`
`LI—I
`
`F/6:2.
`
`0004
`
`

`
`U.S. Patent
`
`Apr. 13, 2004
`
`Sheet 4 of 12
`
`US 6,721,584 B2
`
`\
`
`RU
`
`QC
`
`
`
`/qIlII..|I.IlI.ill|.II.III...
`
`FIG 3.
`
`0005
`
`0005
`
`

`
`U.S. Patent
`
`Apr. 13, 2004
`
`Sheet 5 of 12
`
`US 6,721,584 B2
`
`KALIMN
`
`--—c1s
`
`3
`
`E?
`
`0006
`
`0006
`
`

`
`U
`
`Patent
`
`5,M6
`
`2B
`
`M.h.at
`
`m3.33”33332.3.
`
`Cm_
`
`
`
`__.2...
`
`L_._,
`
`_..__.____._,”_....._£._._w__%_d_%T_.__Hom Fa... V.-p.5.3.5.3.:I:3.m__“as.m_“E....__.ma".._...__WW.T.:,F_E..._:_.._
`
`
`
`
`r3MM.M.__...n...
`.;e=a=ea$JJ,e¢a%fi:==.dfl;;aua:-a.
`
`
`
`
`
`
`
`§a_mT..I._.as:_h_.Il
`
`0007
`
`
`
`

`
`U.S. Patent
`
`Apr. 13, 2004
`
`Sheet 7 of 12
`
`US 6,721,584 B2
`
`
`1gP1L‘
`
`é _......w.=—
`"-:“‘
`-E3 -
`
`=
`
`-
`
`L_-;:"::_i—
`
`-
`
`%-
`
`0008
`
`
`
`—-KALIIAHc-Locx
`
`KALHMI
`
`F/6?6.
`
`:o=1vo=z:
`10:30:21
`l0=0[hZl
`
`Z0Ib9:ll
`zo=z9:u
`
`0008
`
`

`
`U.S. Patent
`
`Apr. 13, 20114
`
`Sheet 8 of 12
`
`US 6,721,584 B2
`
`..
`
`..
`
`..,
`
`.,
`
`,.,
`
`..
`
`_
`
`,_,
`
`-_,»
`
`_,___.- - -—" ‘ '
`"
`-=‘‘:....___‘
`
`o1=9911
`1511:9911
`3119911
`
`\
`
`
`
`—--111113511519111
`
`
`
`1191:151591
`
`10:99:11
`‘ ' " ‘ " '
`-.;'~'_'_'_'_“ ' ‘ ‘ ' " ‘ ‘ ‘ '
`""‘ ‘““'---——._.____ =£9:|i
`
`.‘
`
`=z9=11
`
`.-‘-- -
`"*7
`.9 _—_—:::_’:
`
`’ 99339:"
`192941 1;
`9:39:11
`'=
`99-29-11
`" *3 19:29.11
`=a9=11_
`29:29:11
`19=z911
`
`"1
`'-’..
`
`N
`,
`S
`
`¢
`“‘
`
`2;
`-'::.'_""_—=__T:_“ ‘ ‘ " ” ' “ ' ' " " ' ' ' '
`..- : : _-_--‘"=""
`‘ ' ' ‘ ” " ‘ ‘ “
`
`I
`
`.o9z9:11
`tgg-1|
`81r=z9.-11
`11nz9:11
`" - - - ~ —(
`,__.; 9 91:29:11
`‘ ’- ,
`91=z9=11
`
`ggggg;ggQ°11=a9=11
`
`0009
`
`0009
`
`

`
`U.S. Patent
`
`Apr. 13, 2004
`
`Sheet 9 of 12
`
`US 6,721,584 B2
`
`‘E’
`1335
`:3 G:
`«:3{VIE
`21
`
`III
`
`0010
`
`0010
`
`

`
`U.S. Patent
`
`Apr. 13, 2004
`
`Sheet 10 of 12
`
`US 6,721,584 B2
`
`
`
`raeourucr(arm
`
`
`
`FIG!.9.
`
`I00 090
`
`0.80
`
`070
`
`030
`
`.50
`
`040
`
`030
`
`[120
`
`0!0
`
`0D0
`
`HGHJJNSVH
`
`0011
`
`0011
`
`

`
`U
`
`mM(DSU
`
`2B
`
`M9at
`
`..l__—n_..._$8m._:3p.....,.S2r...%.22
`
`_“J.2:
`
`
`
`
`
`
`
`
`
`
`._r_z..r........a.a..I.r.“I:.....-v.~-.....d.:..,..\n.......s....../..n./.........._....-.2285388.....93.2.32532:.853ooafiwecnenym.H1.....s>.4M_xx...._,.,.w,......fixxa_>>,._.\
`
`
`
`.asM._-H.._as5.¢...mH0
`
`42-_._2:mm,_.32Mu,mo_
`.2...urnW7mM.L.7.».ism2
`
`
`
`
`
`
`
`t.a:§!......§_=e__§_I-I2.23%.-.......$__,_e:§I1:_=2._§_.:I.
`
`
`
`
`
`0012
`
`

`
`U.S. Patent
`
`Apr.13,20ll4
`
`Sheet 12 of 12
`
`US 6,721,584 B2
`
`TH-IE
`
`FIGS/1.
`
`
`
`
`
`N200RATE“HAGFHATE
`
`
`
`
`
`0013
`
`0013
`
`

`
`US 6,721,584 B2
`
`1
`METHOD AND APPARATUS FOR
`ISSTIMATING I’HYSI0l..()GlCAI.
`PARAMlE'l‘ERS USING M()I)El.-BASED
`ADAPTIVE FILTERING
`
`RELATED APPLICATION DATA
`
`The present application is a continuation of application
`Ser. No. U9;’435,14-4. filed Nov. 5, 1999, (now abandoned),
`which is it continuation of application Ser. No. 09;’ 137,479,
`tiled Aug. 20. 1998, (now u.s. Pat. No. 6,083,172), which
`is a continuation of application Ser. No. ($8,560,510, filed
`Jun.
`'2', 1996 (now US. Pat. No. 5,853,364), which is a
`nonprovisional utility patent application based on provi-
`sional patent Application No. 6UfUO{t,195, filed Jun. 14,
`1995.
`
`BACKGROUND OF THE INVENTION
`
`The present invention relates to a method and apparatus
`which uses model-based adaptive filtering techniques to
`estimate physiological parameters. More specifically,
`the
`invention employs Kalman filtering techniques in pulse
`oximetry to estimate the oxygen saturation of hemoglobin in
`arterial blood.
`
`l’uIse oximeters typically measure and display various
`blood How characteristics including but not limited to the
`oxygen saturation of hemoglobin in arterial blood. 0xime-
`ters pass light through blood perfused tissue such as a finger
`or an ear, and photoelectrically sense the absorption of light
`in the tissue. The amount of light absorbed is then used to
`calculate the amount of the blood constituent
`(e.g.,
`oxyhemoglobin) being measu red.
`
`The light passed through the tissue is selected to be of one
`or more wavelengths that are absorbed by the blood in an
`amount representative ofthe amount of the blood constituent
`present in the blood. The amount of light passed through the
`tissue varies in accordance with the changing amount of
`blood constituent in the tissue and the related light absorp-
`tion.
`
`When the measured blood parameter is the oxygen satu-
`ration of hemoglobin, a convenient starting point assumes a
`saturation calculation based on Lambert-l3eer’s law. The
`following notation will be used herein:
`
`i(3\.-fl-1'.i(5\J==’<P('(-¥l5a(3’~)+(1-5Jl’h(3'\-JJI'(tJJ
`
`(1 J
`
`where:
`
`'J\.=wavelength;
`t-time:
`
`I-intensity of light detected;
`I‘,=intensity of light transmitted;
`s=oxygen saturation;
`[30, [S,=empirically derived absorption coellicients; and
`l(t)-a combination of concentration and path length from
`emitter to detector as a function of time.
`
`The traditional approach measures light absorption at two
`wavelengths. e.g., red and infrared (IR), and then calculates
`saturation by solving for the “ratio of ratios" as follows.
`
`1. First, the natural logarithm ol'(1) is taken ("log” will be
`used to represent the natural logarithm) for IR and Red
`
`3031=-'08n'.r($l-"..+(1-$33.1-‘
`
`(2)
`
`ll)
`
`15
`
`.
`
`3-0
`
`3-5
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`0014
`
`[fit
`
`t4't
`
`E5l
`
`2. (2) is then clilferentiated with respect to time
`
`I”
`dlogl
`In ——[3,8.-i+[l—.1l,8ylZ
`
`3. Red (3) is divided by IR (3)
`
`iilogttauitttr _ .rfl.,{.lgl+tI —sJ;‘3.tAgt
`.d'log.'t.'lug],'.r.f.t' _ .s'fi,.[.l;g1+ll —s1,8,t.l,gJ
`
`4. Solving for s
`
`s
`
`—
`
`rllogltflgl
`dlflgifhgl
`,.
`—— .1
`flllrxl
`(H
`9,‘,
`fir! at
`all
`(.1 !
`:11
`my R u'3.,t.t.-;r1—,r3,t.1,xn—
`:.t'Iog;'{.l
`1
`r I
`,
`'" ti€t.t2-.t.1—t3.-ton
`
`Note in discrete time
`
`£logJ'(.1.n
`(H
`
`‘
`
`E logl‘t.\. I3] — togrra. (.1
`
`Using log A—log B-log NB,
`
`d'|og.'t.l. n 310 -HI2. .1:
`dt
`flit. M-
`
`So, (4) can be rewritten as
`
`Jlogftagl
`___
`ff;
`rflogfilml = 1
`if!
`
`0
`
`!'i-‘L--133]
`“lg. al.xl- I R
`.iU|.il.”z!
`-'l.f;..l.”glJ
`
`where R represents the “ratio of ratios."
`Solving (4) for s using (5) gives
`
`.31-[r1)!l— Rflrlrlrnl
`5 2 -Rltfiol-‘l-IR r - fi.t.tirI1 -13.1.1»: + 3.13»)
`
`From (5) note that R can be calculated using two points
`(e.g., plethysmograph maximum and minimum). or a family
`of points. One method using a family of points uses a
`modified version of (5). Using the relationship
`
`dlog.‘
`air
`
`n‘.",u‘d.'
`I
`
`=
`
`now (5) becomes
`
`r.f|og!'t.\gJ
`
`rm. .1,-.1 — Inn. .13:
`
`Hit. .11:-l
`___
`_ W2. Mg) - fin. .1,-R1
`fin. rlml
`
`_ [-‘tr:..131-.'tI1.r\r.~_}l-‘tfi. Am
`_ lit-'1.:l-ml--‘tn.1'-mIlFt!J.Jn}
`=5?
`
`to}
`
`t?t
`
`0014
`
`

`
`US 6,721,584 B2
`
`Now deline
`Then
`
`describes a cluster of points whose slope of y versus x will
`give R.
`
`-"("J‘=l [(-5 >3-In l‘f(l1»-ll-1nl.li('1=i\-at
`
`."lt‘J'lI(‘:-i’V.nl"(l:ti‘-nllll-'1:-3‘-rkl
`
`_\'(tJ-Rrtrt
`
`(31
`
`The optical signal through the tissue can be degraded by
`both noise and motion artifact. One source of noise is
`ambient
`light which reaches the light detector. Another
`source of noise is electromagnetic coupling from other
`electronic instruments. Motion of the patient also introduces
`noise and affects the signal. For example,
`the contact
`between the detector and the skin, or the emitter and the
`skin, can be temporarily disrupted when motion causes
`either to move away from the skin. In addition, since blood
`is a fluid, it responds differently than the surrounding tissue
`to inertial effects, thus resulting in momentary changes in
`volume at the point to which the oximeter probe is attached.
`Motion artifact can degrade a pulse oximetry signal relied
`upon by a physician, without the physician’s awareness.
`This is especially true if the monitoring of the patient is
`remote, the motion is too small to be observed, or the doctor
`is watching the instrument or other parts of the patient, and
`not the sensor site.
`In one oximeter system described in U.S. Pat. No. 5,025,
`791, an accelerometer is used to detect motion. When
`motion is detected, readings influenced by motion are either
`eliminated or indicated as being corrupted.
`In a typical
`oximeter, measurements taken at the peaks and valleys ofthe
`blood pulse signal are used to calculate the desired charac~
`teristic. Motion can cause a false peak,
`resulting in a
`measurement having an inaccurate value and one which is
`recorded at the wrong time. In U.S. Pat. No. 4,802,486,
`assigned to Nellcor, the assignee of the present invention,
`the entire disclosure of which is incorporated herein by
`reference, an EKG signal is monitored and correlated to the
`oximeter reading to provide synchronization to limit
`the
`effect of noise and motion artifact pulses on the oximetcr
`readings. This reduces the chances of the oximeter locking
`onto a periodic motion signal. Still other systems, such as the
`one described in U.S. Pat. No. 5,078,136, assigned to
`Nellcor, the entire disclosure of which is incorporated herein
`by reference, use signal processing in an attempt to limit the
`effect of noise and motion artifact. The ‘I36 patent,
`for
`instance, uses linear interpolation and rate of change tech-
`niques to analyze the oximeter signal.
`Each of the above-described techniques for compensating
`for motion artifact has its own limitations and drawbacks. It
`is therefore desirable that a pulse oximetry system be
`designed which more effectively and accurately reports
`blood—oxygen levels during periods of motion.
`SUMMARY ()1-' T} [E INVENTION
`
`According to the present invention, a method and appa-
`ratus are provided for reducing the effects of motion artifact
`and noise on a system for measuring physiological
`parameters, such as. for example, a pulse oximeter. The
`method and apparatus of the invention take into account the
`physical
`limitations on various physiological parameters
`being monitored when weighting and averaging a series of
`Samples or measurements. Varying weights are assigned
`different measurements. Optionally, measurements are
`rejected if unduly corrupt. The averaging period is also
`
`ill
`
`15
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`4
`adjusted according to the reliability of the measurements.
`More specifically, a general class of filters is employed in
`processing the measurements. The filters use mathematical
`models which describe how the physiological parameters
`change in time. and how these parameters relate to mea-
`surcment
`in a noisy environment. The filters adaptively
`modify a set of averaging weights and averaging times to
`optimally estimate the physiological parameters.
`In a specific embodiment, the method and apparatus of the
`present invention are applied to a pulse oximeter which is
`used to measure the oxygen saturation of hemoglobin in
`arterial blood. The system takes the natural logarithm of the
`optical oximetry data and then bandpass filters the data to get
`absorption-like data. The bandpass filter strongly attenuates
`data below 0.5 Hz and above l0 Hz in an attempt to remove
`as much out-of-band noise as possible. This liltered data is
`then processed through two algorithms: a rate calculator and
`a saturation calculator.
`
`The system calculates the heart rate of the patient one of
`three ways using the oximetry data. An adaptive comb filter
`(ACF) is employed to track the slowly varying heart rate.
`The tracking of the heart rate by the ACE is quite robust
`through noisy environments, however, the ACF is not a good
`heart rate tinder. As a result, the system periodically calcu-
`lates the power spectrum of one of the wavelengths and uses
`it to find andfor verify the heart rate. In cases of arrhythmia
`or suddenly changing heart rates,
`the system employs a
`pattern matching technique that recognizes sequences of
`crests and troughs in the data and calculates an average heart
`rate period over a set number of samples.
`The system then employs the calculated heart rate to
`digitally comb litter the data so that only the energy at
`integer multiples of the heart rate are allowed through the
`litter. The comb filter frequency varies as the heart rate
`varies, attenuating motion energy not at the heart rate or
`multiples thereof. To remove noise energy at integer mul-
`tiples of the heart rate, the system adaptively signal averages
`full cycles of past plethysrrtographs,
`i.e., pleths, using a
`Kalman filter to limit the rate of change in the pleth shape
`or size.
`
`The system then calculates two saturations, one with the
`pleth cycle data which has been comb filtered as described
`above, and one with raw data from the output of the band
`pass filter. Both saturations are calculated using time based
`signals and using an adaptive Kalman filter which continu-
`ously weights all data according to an estimate of the current
`noise, and limits the rate of change of saturation to a defined
`limit {currently 1.3 saturation points per second). Data
`points that result in a saturation calculation {prior to weight-
`ing and averaging} which is obviously not physiologically
`possible (e.g., negative saturation, or a saturation greater
`than 100%) are deemed invalid and are not used and are
`rejected in an “outlier rejection” step in both saturation
`calculations. The system then arbitrates between the two
`saturation values based on rules described below to deter-
`mine the best saturation. For example, the arbitration may be
`based on such factors as the noise level or the age of the
`saturation value. The best saturation may also be a weighted
`average of the different saturation values.
`According to a specific embodiment of the invention, a
`method for reducing noise elifects in a system [or measuring
`a physiological parameter is provided. A plurality of mea-
`surements is generated corresponding to at least one wave-
`length of electromagnetic energy transmitted through living
`tissue. Selected measurements are compared with at least
`one expected measurement characteristic. A variable weight
`
`0015
`
`0015
`
`

`
`US 6,721,584 B2
`
`5
`is assigned to each of the selected measurements based on
`the comparison. thereby generating a plurality of differently
`weighted measurements for each wavelength. A first number
`of weighted measurements is averaged to obtain a filtered
`measurement.
`the first number varying according to the
`manner in which weights are assigned to a plurality of
`successive weighted measurements. A plurality of filtered
`measurements are thus generated for each wavelength. The
`filtered measurements for each wavelength are then com-
`bined and calculations resulting therefrom are adaptively
`filtered using variable weights based on comparing the
`calculations to an expected calculation. A second number of
`the weighted calculations are averaged to obtain a filtered
`calculation,
`the second number varying according to the
`manner in which weights are assigned to a plurality of
`successive weighted calculations.
`A further understanding of the nature and advantages of
`the present invention may be realized by reference to the
`remaining portions of the specification and the drawings.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`FIGS. la and 1b are block diagrams illustrating the data
`flow in a pulse oximetry system designed according to two
`specific embodiments of the invention;
`FIG. 2 shows the frequency response of an infinite
`impulse response (IIR) filter employed by a specific embodi-
`ment of the invention;
`FIG. 3 shows a sensorloximeter combination for use with
`the present invention in which the transmission character-
`istics of the sensor are identified by a calibration resistor;
`FIG. 4 is a graph comparing the performance of a classic
`least squares algorithm to that of the Kalman algorithm;
`FIG. 5 is a graph comparing the inputs and outputs of the
`Kalman cardiac gated averaging filter;
`FIG. 6 is a graph illustrating the improvement in satura-
`tion calculation gained by enhancing the pulse shape with
`the Kalman cardiac gated averaging filter;
`FIG. 7 is a graph illustrating the weighting and aging of
`pulses by one embodiment of a Kalman cardiac gated
`averaging filter;
`FIG. 8 is a graph illustrating the improvement in satura-
`tion calculation gained by employing both the Kalman
`cardiac gated averaging filter and the Kalman saturation
`algorithm;
`FIG. 9 is a frequency domain graph depicting the response
`of a comb filter;
`FIG. 10 is a graph showing the validity measure for data
`pulses in relation to the relative strengths of several signal
`harmonics; and
`FIG. 11 is a graph showing the pulse rate reported by the
`adaptive comb filter employed by the present invention as
`compared to the pulse rate reported by a prior art system.
`Dl_1SCRII"l'I()N OF TIIIL I’RI;iI"F.RRI_".D
`EMBODIMENT
`
`FIG. lo shows the flow of data according to one embodi-
`ment of the present invention. A separate platform collects
`the oxirnetry data (step 10) and passes it to processors 50 and
`52 of the present invention. A preferred platform is described
`in US. Pat. No. 5,348,004 assigned to Nellcor, the entire
`disclosure ofwhich is incorporated herein by reference. The
`data is first pre-processed (steps 12 and 14), and is then
`passed to a saturation calculation algorithm (box 50). The
`algorithm described herein employs an improved Kalman
`
`Ill
`
`15
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`6
`It will be understood that other
`filter method (step 24).
`saturation calculation techniques may be employed. The
`pulse rate calculation method {box 52) and a cardiac gated
`averaging technique also using a Kalman filter (step 16) are
`discussed below.
`
`the processing
`According to a preferred embodiment.
`technique employs the following pre-processing. The natu-
`ral logarithm of the IR and Red wavelength data is taken
`(step 12), and then the data is band pass filtered with an
`infinite impulse response (IIR) filter that has a high pass
`cutoff frequency at 0.5 Hz, i.e., 30 beats per minute. and a
`low pass rolIolT from 10 to 20 Hz (step 14}. FIG. 2 shows the
`frequency response of an HR filter employed by a specific
`embodiment of the invention.
`After the oximetry data has been filtered, it is processed
`by a saturation calculation algorithm (box 50). According to
`a preferred embodiment of the invention depicted in FIG.
`In,
`two saturation values are calculated in parallel by
`saturation calculator 50. One saturation value is calculated
`using a harmonic filter I7 and a Kalman-filter-based cardiac
`gated averaging {(_'GA)
`technique (step 16) (described
`below) to obtain a more reliable data stream. Kalman CGA
`16 is gated by triggers based on the pulse rate which are
`supplied by pulse rate calculator 52.
`In a specific
`embodiment, the data is put through a harmonic filter (step
`17) before it
`is averaged in step 16. llarmonic filter 17
`digitally filters the [R and red waveforms such that only
`energy at
`integer multiples of the heart rate is allowed
`through the filter. The response of harmonic filter 1'7 varies
`with the heart rate which is supplied by pulse rate calculator
`52 to attenuate motion and noise energy not at the heart rate.
`In one embodiment, only one of the IR and red waveforms
`is filtered by harmonic filter 17. In this embodiment, the
`subsequent filtering by Kalman CGA 16 andfor the satura-
`tion calculation algorithm described below applies the same
`weighting and averaging to both the IR and red data streams
`on the basis of the single filtered data stream.
`Both saturation values are calculated in the following
`manner. The data pulses (either directly from the band pass
`filter or from steps 16 and 17) are nonnalizcd (step 18) and
`then "whitened” (step 20). Normalizing downweights large
`pulse amplitudes so that each pulse has roughly the same
`average amplitude. Normalizing step 18 assumes that from
`one sample to the next, noise energy should look substan-
`tially the same statistically. As a result, samples exhibiting
`large amounts of noise are down weighted,
`thus
`de-emphasizing outliers. Whitening step 2|] involves taking
`the derivative of the nonnalized data, thereby emphasizing
`the higher hannonics of the pleth so that its energy is more
`evenly distributed between them. Data points resulting in an
`impossible saturation calculation are rejected (step 22) and
`the resulting data are used to calculate the saturation values
`using a Kalman filter technique described below (step 24).
`The best saturation value is then chosen [step 26) according
`to confidence levels associated with each, and, after some
`post processing (step 27). the saturation value is output to
`the display (step 28). Post processing 27, which will be
`discussed in greater detail below, uses available metrics with
`regard to the saturation value to determine its reliability and
`determine whether and how it is to be displayed. In specific
`preferred embodiments of the present invention, the initial
`saturation value calculated by each calculation path in
`saturation calculator 50 may be calculated by the well
`known classic least squares (CLS) technique as indicated by
`step 21. The use of this technique occurs on initialization of
`saturation calculator 50 only.
`The pulse or heart rate is calculated in pulse rate calcu-
`later 52 in the following manner. After the pre-processing
`
`0016
`
`0016
`
`

`
`US 6,721,584 B2
`
`H!
`
`15
`
`30
`
`35
`
`40
`
`7
`described above, data from one channel, e.g., the IR channel,
`are normalized (step 29) by the downwcighting of data
`corresponding to large pulse amplitudes so that each pulse
`has roughly the same average amplitude. The data are then
`sent
`to two different algorithms for calculation of the
`patient’s pulse rate. According to one algorithm, the deriva-
`tive of the data is taken (step 30) as described above, and the
`fundamental frequency of the pulse rate is tracked using an
`adaptive comb filter (AC1-') 32 as discussed below. AC1" 32
`supplies its pulse rate directly to harmonic filter 17 as
`described above. ACF 32 also provides the trigger
`for
`Kalman CGA16 after the data is unwhitened by integration
`(step 34} and the triggers for Kalmart CGA are generated
`(step 36). Alternatively, the triggers for Kalrnan CGA 16
`may be derived from, for example, an ECG waveform. Act‘
`32 is a robust pulse rate tracker, but not a good pulse rate
`linder. Therefore,
`the frequency power spectrum of the
`nonnalized data is calculated periodically (step 38) to deter-
`mine whether ACF 32 is tracking the fundamental rather
`than a super- or subharmonic of the pulse rate.
`The normalized data is also supplied to a pattern matching
`algorithm 40 which recognizes sequences of crests and
`troughs in the data and calculates an average period of the
`pleth over a set number of samples. This algorithm is
`preferably used primarily to track the pulse rate for an _
`arrhythmic pulse rate during periods where no motion is
`detected. Abest rate algorithm 42 then arbitrates between the
`pulse rates calculated by AC F 32 (as updated by frequency
`power spectrum 38} and pattern matching algorithm 4|]
`using confidence levels associated with each, which are
`based on various metrics. After post processing (step 44), the
`pulse rate is output
`to the display {step 46). As with
`saturation calculator 50, post processing 44 uses available
`metrics to determine the reliability of the pulse rate and to
`determine whether and how it is to be displayed.
`FIG. lb shows the flow of data according to a second
`embodiment of the present invention. The system operates
`the same as the system of FIG. In except that after the data
`is band pass filtered by HR filter 14,
`it undergoes an
`additional processing step in eta correction processor 15
`before it is sent to either saturation calculation algorithm 50
`or pulse rate calculation algorithm 52. Like other aspects of
`the present invention already described, eta correction pro-
`cessor 15 scrves to reduce the elfeets of motion and other
`noise artifact. The operation of eta correction processor 15
`is based on an analysis ofthe signal intensity received for the
`different wavelengths, without separately measuring the
`motion signal for each wavelength, without providing feed-
`back to cancel the motion signal, and without attempting to
`mathematically eliminate the motion signal individually for
`each wavelength. Instead, processor 15 mathematically rec-
`ognizes the presence of the motion signal and recognizes a
`few key characteristics of the motion signal. First, although
`the magnitude of the elfect of motion on the signal intensity
`for each wavelength will be dilferent,
`the change in the
`logarithm of the motion component will be approximately
`the same (for signals obtained at approximately the same
`time). This allows the motion component to be cancelled out
`in a ratiometric equation. Second, it
`is assumed that the
`blood pulse signal is not aifected by motion. This second
`assumption is more of an approximation, since the blood
`pulse signal is somewhat affected by motion, which can
`actually change the blood volume characteristics at any
`point in the patient. Eta correction processor 15 recognizes
`that the intensity signal for each of the wavelengths includes
`a time-varying motion term, and that
`this time-varying
`motion term is proportional for each of the wavelengths. In
`
`45
`
`50
`
`55
`
`60
`
`65
`
`0017
`
`8
`addition, each wavelength signal occurs close enough in
`time with one another that
`the motion should not vary
`noticeably, and can he assumed to be the same for each
`signal. The output from eta correction processor 15 is an IR
`or red signal which has significantly less motion noise than
`the signals fed into processor 15. If the data include infor-
`mation [rom a third wavelength, the output of processor 15
`is both an IR signal and a red signal depleted of motion
`noise. A more detailed description of the operation of eta
`correction processor 15 is described in a commonly
`assigned, copending U.S. patent application Ser. No. 08;‘490,
`315 for METHOD AND APPARATUS FOR REMOVING
`ARTIFACTAND NOISE FROM PULSE OXIMETRY, filed
`Jun. 14, 1995, the entire disclosure of which is incorporated
`herein by reference.
`The method for calculation of blood oxygen saturation
`(step 24) described below uses a Kalman filter. The method
`first transforms the pre-processed data into quantities cor-
`responding to the oxyhemoglobin and total hemoglobin
`concentrations using appropriate extinction coellicienls. The
`instantaneous ratio of these two transformed quantities gives
`the saturation.
`It will be understood from the equation
`immediately following equation (4) above that the instan-
`taneous saturation value may he calculated directly by using
`the extinction ooeflicients, or from the ratio of ratios as
`shown in the equation immediately following equation (5).
`According to a preferred embodiment, the method does not
`search for maxima or minima like a pulse searching algo-
`rithm {allhough maxima or rninima could be used and
`Kalman filtered if desired). Using instantaneous ratios (Le,
`a time based algorithm) rather than maximafminima ratios
`(i.e., an event based algorithm) keeps the code from being
`event driven and having to qualify data as it arrives. Thus,
`the preferred method of the present invention is simpler to
`implement
`than a pulse—searching event—based saturation
`calculation algorithm.
`The extinction coefficients are determined with reference
`to the wavelength or wavelengths being transmitted by the
`Ll£Ds in the particular sensor attached to the patient. In a
`preferred embodiment,
`the sensor includes a means for
`generating a signal which corresponds to at least one of the
`wavelengths being transmitted by the sensor’s I.EDs. The
`oximeter monitor receives the signal and determines the
`proper extinction cocflicients based on the wavelength or
`wavelengths indicated by the signal. This avoids the need to
`recalibrate an oximeter to match the transmission charac-
`teristics of a particular probe. In a preferred embodiment, the
`means for generating the signal is an electrical impedance
`element such as, for example, a resistor, the value of which
`corresponds to the wavelengths of the LEDs. A preferred
`embodiment of a sensorloximeter combination is shown in
`FIG. 3. Oximetry system 6|] includes a sensor 61 and an
`oximetcr monitor 62. Sensor 61 includes LEDs 63 and 64
`typically having wavelength emission characteristics in the
`infrared and red ranges of the spectrum, respectively. Pho-
`todiode sensor 65 receives the light transmitted by Ll3Ds 63
`and 64. Resistor 66 (or a similar electrical
`impedance
`reference) is chosen to correspond to a specific wavelength
`or combination of wavelengths as specified by a table
`relating impedance values to wavelengths. Once decoding
`means 67 determines the impedance value of resistor 66,
`appropriate extinction coefficients are generated which cor-
`respond to the transmission characteristics of the particular
`sensor 61. Thus, the oximcter may be used with a variety of
`sensors having LEDs which emit varying wavelengths of
`light without recalibration.
`Sensor 61 may be detachably coupled to oximeter monitor
`62