US008923958B2
`
`(12) United States Patent
`Gupta et al.
`
`(10) Patent N0.:
`(45) Date of Patent:
`
`US 8,923,958 B2
`Dec. 30, 2014
`
`(54)
`
`SYSTEM AND METHOD FOR EVALUATING
`AN ELECTROPHYSIOLOGICAL SIGNAL
`
`(75)
`
`Inventors: Sunny Gupta, AmherstvieW (CA);
`Mohsen Naj a? Yazdi, Kingston (CA)
`
`(58) Field of Classi?cation Search
`None
`See application ?le for complete search history.
`
`(56)
`
`References Cited
`
`(73)
`
`(*)
`
`(21)
`
`(22)
`
`(65)
`
`(63)
`
`(60)
`
`(51)
`
`(52)
`
`Assignee: Analytics For Life Inc. (CA)
`
`U.S. PATENT DOCUMENTS
`
`Notice:
`
`Subject to any disclaimer, the term of this
`patent is extended or adjusted under 35
`U.S.C. 154(b) by 0 days.
`
`App1.No.: 13/605,364
`
`Filed:
`
`Sep. 6, 2012
`
`Prior Publication Data
`US 2013/0096394 A1
`Apr. 18, 2013
`Related US. Application Data
`Continuation-in-part
`of
`application
`PCT/US2012/024037, ?led on Feb. 6, 2012.
`Provisional application No. 61/462,640, ?led on Feb.
`4, 2011.
`
`No.
`
`Int. Cl.
`A6IB 5/0402
`A6IB 5/04 76
`
`(2006.01)
`(2006.01)
`(Continued)
`
`US. Cl.
`CPC ........... .. A61B 5/04012 (2013.01); A61B 5/044
`(2013.01); A613 5/0452 (2013.01); A613
`5/0245 (2013.01); A613 5/4848 (2013.01);
`A613 5/046 (2013.01); A613 5/0464
`(2013.01); A613 5/02028 (2013.01); A613
`5/055 (2013.01); A613 6/032 (2013.01); A613
`5/7253 (2013.01); A613 5/7203 (2013.01);
`A613 5/04011 (2013.01); A613 5/04525
`(2013.01); G06K 9/0051 (2013.01); G06K
`9/00536 (2013.01); A613 5/4884 (2013.01);
`A613 5/7239 (2013.01); A613 5/7232
`(2013.01)
`USPC ......................................... .. 600/509; 600/544
`
`4,723,536 A
`6,325,761 B1
`
`2/1988 Rauscher et al.
`12/2001 Jay
`(Continued)
`OTHER PUBLICATIONS
`Efron, Bradley, et al., “Least Angle Regression,” Annals of Statistics,
`vol. 32, N0. 2, 2004, pp. 407-499.
`
`(Continued)
`
`Primary Examiner * Michael Kahelin
`(74) Attorney, Agent, or Firm * Meunier Carlin & Curfman,
`LLC
`
`ABSTRACT
`
`(57)
`A method of evaluating an electrophysiological signal is dis
`closed. A mathematical reconstruction over at least one cycle
`of the electrophysiological signal is used to identify an abnor
`mal substrate. A non-transitory computer readable medium is
`also disclosed. The nontransitory computer readable medium
`has stored thereon instructions for identifying a pathological
`substrate from a mathematical reconstruction of an electro
`physiological signal, Which, When executed by a processor,
`causes the processor to perform steps comprising using a
`mathematical reconstruction over many cycles of the electro
`physiological signal to identify a pathological state. A system
`for evaluating an electrophysiological signal includes a pro
`cessor con?gured to identify a pathological condition from a
`mathematical reconstruction of the electrophysiological sig
`nal. The system also includes a data input coupled to the
`processor and con?gured to provide the processor With the
`electrophysiological signal. The system further includes a
`user interface coupled to either the processor or the data input.
`
`15 Claims, 24 Drawing Sheets
`(11 of 24 Drawing Sheet(s) Filed in Color)
`
`06
`
`04
`
`02
`
`-0.2
`
`-0.4
`
`-0.6
`
`-0.8
`
`1
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US 8,923,958 B2
`Page 2
`
`(51)
`
`Int. Cl.
`A6IB 5/04
`A6IB 5/044
`A6IB 5/0452
`A6IB 5/0245
`A6IB 5/00
`A6IB 5/046
`A6IB 5/0464
`A6IB 5/02
`A6IB 5/055
`A6IB 6/03
`G06K 9/00
`
`(2006.01)
`(2006.01)
`(2006.01)
`(2006.01)
`(2006.01)
`(2006.01)
`(2006.01)
`(2006.01)
`(2006.01)
`(2006.01)
`(2006.01)
`
`(56)
`
`References Cited
`
`U.S. PATENT DOCUMENTS
`
`2004/0059203 A1
`
`3/ 2004 Guerrero et al.
`
`2004/0230105 A1 11/2004 Geva et al.
`2008/0027341 A1
`1/2008 Sackner et a1.
`2009/0292180 A1 11/2009 Mirow
`
`OTHER PUBLICATIONS
`
`Mallat, Stéphane G, et al., “Matching Pursuits With Time-Frequency
`Dictionaries,” IEEE Transactions on Signal Processing, vol. 41, No.
`12, Dec. 1993, pp. 3397-3415.
`International Search Report, dated Jun. 15, 2012, received in connec
`tion With corresponding International Patent Application No. PCT/
`US2012/024037.
`Ki H. Chon, “Accurate Identi?cation of Periodic Oscillations Buried
`in White or Colored Noise Using Fast Orthogonal Search”, IEEE
`Transactions on Biomedical Engineering, vol. 48, No. 6, Jun. 2001.
`Marica KleniZ and Thomas J. Osler, A Child’s Gardent of Fractional
`Derivative‘. The College Mathematics Journal, vol. 31, No. 2, 2000,
`pp. 82-887.
`
`2
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 1 0f 24
`
`US 8,923,958 B2
`
`302
`
`Receiving
`Moduie
`
`Pronessing
`Mnciuie
`
`106
`
`168
`
`3H)
`
`HG. “E
`
`3
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 2 0f 24
`
`US 8,923,958 B2
`
`=11” Start /2@0
`
`Receive signal? from an ECG or f 202
`other etectrophysmtogacai
`sigma!
`
`.
`
`;
`
`_.
`
`<
`
`204
`Create impiicitorthogonet <f
`functions from non~orthogonet é—m
`candidates
`t
`Choose the candidate compiex
`exponentiats pair or ether candidate
`term that causes the greatest
`reductien 5n least angie and mean
`square error (MSE)
`
`{,f 2536
`
`2(38
`
`l
`
`L
`
`212
`
`N0 //"
`
`_\/ \\~\
`
`\\~\ N0 /// certain
`£5 3
`prede?ned
`\
`?x
`my 0?
`
`smati twee and ieast
`terms f/f/ \‘\~{i?ted?/K
`
`engte
`/"
`,
`, ’_P//
`\asfthf/e/
`Yes
`
`Yes
`
`/.//{)0 remaining
`\\ N0
`/,/
`candidatgs
`//
`
`tail mse and ieast angie
`//
`\\ reduction
`threshotd? //
`/
`\ Fee
`
`Recursiveiy caicuiate J 214
`weights of setected non
`orthoqonal candidates
`
`4
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 3 0f 24
`
`US 8,923,958 B2
`
`1GB
`
`Jill]
`
`EDD
`
`
`
`Signal Amplitude
`
`1
`
`|
`
`I
`
`|
`
`|
`
`I.
`
`|
`
`|
`
`|
`
`l
`
`|
`
`U
`
`SUI]
`
`1UUU
`
`15GB
`
`ZUUU 25GB SUCH]
`TIFFIE!
`
`SEIIIU
`
`4|][lU 45GB
`
`PEG. 3
`
`5
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 4 0f 24
`
`US 8,923,958 B2
`
`mustiE?
`
`Time
`
`F56. 5
`
`m
`
`PEG. 6
`
`6
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 5 0f 24
`
`US 8,923,958 B2
`
`0.3 mV
`
`
`
`smObth heart_i_sha'ped
`
`
`
`Lijnit Cycl-é Q ~
`
`7
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`U S. Patent
`
`Dec. 30, 2014
`
`Sheet 6 0f 24
`
`US 8,923,958 B2
`
`M} Maggy
`
`8
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 7 0f 24
`
`US 8,923,958 B2
`
`
`
`
`
`356 smug? $21 8%;2 132.1111 .
`
`
`
`
`
`
`
`
`
`
`
`£2.51 wwgu 8:52:qu 51:50 . . . . ..
`
`
`
`
`
`
`
`Gum :Eton $95 azm E22 1‘.
`
`9
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 8 0f 24
`
`US 8,923,958 B2
`
`1902
`
`/
`
`-t
`
`-
`
`.
`
`1
`
`.
`
`-
`
`.
`
`-
`
`-
`
`;/ Use Modi?ed Matching Pursuit (MM?) algerithm to Gem/e a mime! ever at least 100 ‘\
`'\ consecutive secerads of an Eiectrophysioiogical signai to identity a pathslogicai substrate /
`/
`
`\~
`
`Deriving an M MP model of the iiiectmphysioiogicai
`Signai
`
`An i/O expansion of the Eiectrophysiological signai in which
`at least one the terms of the i/O expansion are fractionaliy
`differentiable and integrable
`
`A subspace of the Electrophysioiogical signal
`
`Noniinear terms tetra-spending to the Eiectrcphvsioiogicai
`signal
`
`Piuraiitv of terms from a search through candidate terms
`
`Piureiity sf terms from an MMP transformatisn of the
`Electrophysiolcgicai slgnai
`
`identificath of Low energy terms from an MM?
`transformation from the Eiectrophysioiogical signal
`
`Characterization of a pathoiogical substrate
`
`irratisnai fractional subspace derivative frem the MMP
`reconstruction 0f the Electrophysioiogicai signal
`
`At ieast one Lyapunov exponent preduced from the
`Eiectrophysioiogical signal
`
`1020
`
`,
`
`.
`
`A
`
`.
`
`'/ Drspiay on a user interface one or more lndlCBtOi'S based on the i/O expansmn to
`\
`represent at least a portion of the Eiectrephysioiogieei sigma!
`
`.
`
`.
`
`FEG. 13A
`
`1604
`
`1095
`
`1008
`
`1010
`
`1612
`
`1014
`
`1616
`
`“918
`
`\
`
`‘i,
`},/
`
`10
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 9 0f 24
`
`US 8,923,958 B2
`
`Appiy an Modified Mat-citing Pursuit (MM?) transformation to the /
`Eiectroptwsioiogicaf signai to generate a plurality of terms corresponding to the
`Etect-“ophysioisgicei signal
`
`1028
`
`Form a reconstructed Electroohysioiogieai signs! whereby a noise component is
`separated from the Electrophysioiogicat signai using a subset of the piureiity of terms, /
`corresponding to the Electrophysioiogical signai which have at ieast a desired
`compression ratio
`
`Fi?. 198
`
`11
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 10 0f 24
`
`US 8,923,958 B2
`
`@132 “N
`
`g“? 34 a r wwwwwwwwwwwwwww m m
`
`1142
`
`i
`‘
`8289423
`P
`m we Q
`“J
`3
`3136
`
`{W}
`f”
`
`usm
`
`was x
`
`ME
`{mm
`mm}? WWW {A
`i
`mpmaa
`g
`amazi
`j
`
`WTERFAL‘E mmmm w “"
`
`PEG. ii
`
`3146 "Mk
`
`;1 34 “x
`
`p PRUC'EQGQFZ
`
`1’! 42
`‘
`\_
`
`@133 w
`
`{EAPTURE
`owxca
`"/7
`
`:1 40
`
`usER
`fNTERFAQE
`
`3144 "’j
`
`PEG. [L
`
`12
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 11 0124
`
`US 8,923,958 B2
`
`F36, £15
`
`13
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 12 0f 24
`
`US 8,923,958 B2
`
`HG, W:
`
`14
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 13 0f 24
`
`US 8,923,958 B2
`
`-
`;
`
`<mw OE
`
`
`
`@Um commmmwmm nmi @m? Nw Eek mwmm
`
`
`
`
`
`15
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 14 0f 24
`
`US 8,923,958 B2
`
`namqu
`
`mm? .QE
`
`‘‘‘‘‘‘‘‘
`
`
`
`m? mam“ woow
`
`
`
`
`
`
`
`a *0 2mm uswEFSmwmb $538 “mi “$me 9: E“ 6on swoo?w m mEucm 3 gen 2 sowmnmocmu wmmaz
`
`
`
`
`
`
`
`16
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`U S. Patent
`
`Dec. 30, 2014
`
`Sheet 15 0f 24
`
`US 8,923,958 B2
`
`FIG. 165
`
`17
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 16 0f 24
`
`US 8,923,958 B2
`
`FIG. 16C
`
`18
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 17 0f 24
`
`US 8,923,958 B2
`
`
`
`100 m 700 censecutive seconds (If ECG data is gathered on each 0f singie $2 iead nr 3 iead .................. . .anngnn?i wads. ............. . .
`
`
`
`
`
`:
`
`
`
`Signa¥ is normaiized and baseiine wander are remeveé usmg a modi?ed moving average fiiter
`
`f
`
`\
`
`é
`
`
`
`Mndi?ed Matching Pursnii (MW) algorithm w find a nniseiess mode? of an ECG
`
`?
`
`5
`
`Deiayed phase space recnnstruction is used in
`
`
`
`
`
`mm 318 8min 6? 3 km or 12 193d 50g; m a 5 ........... . . .01? ninhef dimensivnai $936? ............ .
`
`>
`
`@798
`
`V
`
`
`
`The snace~tirne domain is divided into 30 regions and the density of the signai is camputed {01' each
`
`h
`
`i710
`
`1712
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`These if) quantities are fed into nested sinusoidai Gaussians in gnnerate a risk measure for adverse .................. ..¢i%n%¢n¥¢m¢<>m.es . . . . . . . . . . . . . . . _ _ _
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`HG, 1?
`
`19
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`US. Patent
`
`Dec. 30, 2014
`
`Sheet 18 0f 24
`
`US 8,923,958 B2
`
`Fig. 18
`
`20
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`21
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`22
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`23
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`24
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`25
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`26
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`27
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`28
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`29
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`30
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`31
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`32
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`33
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`34
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`35
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`36
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`

`

`37
`
`Analytics For Life, Inc. v. 8825319 Canada Limited
`IPR-01742
`X2022
`
`