`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`_______________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`________________
`
`APPLE, INC.,
`Petitioner,
`
`v.
`
`ALIVECOR, INC.,
`Patent Owner
`________________
`
`IPR2021-00971
`U.S. Patent No. 10,595,731
`________________
`
`PATENT OWNER’S RESPONSE
`
`
`
`
`
`
`
`
`
`
`
`I.
`
`II.
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`TABLE OF CONTENTS
`
`Page
`
`INTRODUCTION ........................................................................................... 1
`
`TECHNICAL BACKGROUND ..................................................................... 4
`
`A. Arrhythmias and Atrial Fibrillation ...................................................... 4
`
`B.
`
`C.
`
`D.
`
`Strokes ................................................................................................... 6
`
`Photoplethysmography (PPG) ............................................................... 7
`
`Electrocardiography (ECG) ................................................................. 11
`
`E. Myocardial Infarctions aka Heart Attacks .......................................... 14
`
`F.
`
`Epileptic Seizures ................................................................................ 15
`
`III. THE ’731 PATENT ....................................................................................... 17
`
`A. Overview ............................................................................................. 17
`
`B.
`
`C.
`
`Specification ........................................................................................ 17
`
`Prosecution History ............................................................................. 19
`
`IV. CLAIM CONSTRUCTION .......................................................................... 20
`
`A.
`
`B.
`
`C.
`
`The ITC Construed Many of the Claim Terms In the Co-
`Pending Investigation .......................................................................... 21
`
`Arrhythmia Should Be Construed Consistent With the Express
`Definition Provided in the ’731 Patent ................................................ 23
`
`Petitioner Conflates the Claims’ Separate Requirements of
`“Detect” and “Confirm” ...................................................................... 25
`
`V.
`
`LEVEL OF ORDINARY SKILL IN THE ART ........................................... 27
`
`VI. OVERVIEW OF THE PRIOR ART ............................................................. 31
`
`A.
`
`B.
`
`C.
`
`Shmueli ................................................................................................ 31
`
`Osorio .................................................................................................. 35
`
`Li 2012................................................................................................. 38
`
`D. Kleiger-2005 ........................................................................................ 39
`
`E.
`
`Chan ..................................................................................................... 39
`
`
`
`
`i
`
`
`
`
`VII. PETITIONER HAS FAILED TO PROVE UNPATENTABILITY ............. 39
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`A. Grounds 1-5 Fail Because Shmueli Does Not Render Obvious
`Arrhythmia Detection .......................................................................... 42
`
`B.
`
`C.
`
`Grounds 2-5 Fail Because Neither Shmueli Nor Osorio Renders
`Obvious Arrhythmia Detection ........................................................... 47
`
`Grounds 1-5 Fail Because Shmueli Does Not Render Obvious
`Using ECG Data To Confirm The Initial Detection Of An
`Irregular Heart Condition Using PPG Data ........................................ 51
`
`D. Grounds 2-5 Fail Because A POSITA Would Not Have Been
`Motivated To Combine Shmueli and Osorio ...................................... 57
`
`E.
`
`Ground 3 Fails Because Neither Shmueli Nor Li 2012 Disclose
`A Machine Learning Algorithm For Detecting Arrhythmias ............. 60
`
`1.
`
`2.
`
`3.
`
`Li 2012 Does Not Teach Using Machine Learning To
`Detect Arrhythmias ................................................................... 60
`
`Li 2012’s “Framework for FA Reduction Using a
`Machine Learning Approach” Is Inoperable If ECG and
`ABP Data Are Removed ........................................................... 63
`
`Petitioner Overlooks Industry Skepticism Regarding
`Machine Learning ..................................................................... 65
`
`4.
`
`Shmueli Does Not Teach Machine Learning ............................ 65
`
`VIII. CONCLUSION .............................................................................................. 67
`
`
`
`
`
`
`
`
`ii
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`EX. NO.
`1001
`1002
`
`1003
`1004
`1005
`1006
`
`1007
`1008
`1009
`1010
`1011
`
`1012
`
`1013
`
`1014
`
`1015
`
`1016
`
`
`
`
`
`
`TABLE OF EXHIBITS
`
`DESCRIPTION
`U.S. Pat. No. 10,595,731 to Gopalakrishnan (“the ’731 Patent”)
`Excerpts from the Prosecution History of the ’731 Patent (“the
`Prosecution History”)
`Declaration of Dr. Bernard A. Chaitman
`PCT Patent Publication WO2012/140559 (“Shmueli”)
`U.S. Patent Publication 2014/0275840 (“Osorio”)
`Li Q, Clifford GD, “Signal quality and data fusion for false alarm
`reduction in the intensive care unit,” J Electrocardiol. 2012 Nov-Dec;
`45(6):596-603 (“Li 2012”)
`U.S. Patent Publication 2008/0004904 (“Tran”)
`U.S. Patent Publication 2014/0107493 (“Yuen”)
`U.S. Patent Publication 2015/0119725 (“Martin”)
`U.S. Provisional Application No. 61/794,540 (“OP”)
`Lee J, Reyes BA, McManus DD, Mathias O, Chon KH. Atrial
`fibrillation detection using a smart phone. International Journal of
`Bioelectromagnetism, Vol. 15, No. 1, pp. 26 - 29, 2013 (“Lee 2013”)
`Tsipouras MG, Fotiadis DI. Automatic arrhythmia detection based
`on time and time-frequency analysis of heart rate variability. Comput
`Methods Programs Biomed. 2004 May; 74(2):95-108 (“Tsipouras
`2004”)
`Lu S, Zhao H, Ju K, Shin K, Lee M, Shelley K, Chon KH. Can
`photoplethysmography variability serve as an alternative approach to
`obtain heart rate variability information? J Clin Monit Comput. 2008
`Feb; 22(1):23-9 (“Lu 2008”)
`Selvaraj N, Jaryal A, Santhosh J, Deepak KK, Anand S. Assessment
`of heart rate variability derived from finger-tip
`photoplethysmography as compared to electrocardiography. J Med
`Eng Technol. 2008 Nov-Dec; 32(6):479-84 (“Selvaraj 2008”)
`Lu G, Yang F, Taylor JA, Stein JF. A comparison of
`photoplethysmography and ECG recording to analyse heart rate
`variability in healthy subjects. J Med Eng Technol. 2009; 33(8):634-
`41 (“Lu 2009”)
`Suzuki T, Kameyama K, Tamura T. Development of the irregular
`pulse detection method in daily life using wearable
`
`iii
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`DESCRIPTION
`photoplethysmographic sensor. Annu Int Conf IEEE Eng Med Biol
`Soc. 2009; 2009:6080-3 (“Suzuki 2009”)
`Reed MJ, Robertson CE, Addison PS. Heart rate variability
`measurements and the prediction of ventricular arrhythmias. QJM.
`2005 Feb; 98(2):87-95 (“Reed 2005”)
`Schäfer A, Vagedes J. How accurate is pulse rate variability as an
`estimate of heart rate variability? A review on studies comparing
`photoplethysmographic technology with an electrocardiogram. Int J
`Cardiol. 2013 Jun 5; 166(1):15-29 (“Schafer 2013”)
`K. Douglas Wilkinson, “The Clinical Use of the
`Sphygmomanometer,” The British Medical Journal, 1189-90 (Dec.
`27, 1924) (“Wilkinson”)
`U.S. Pat. No. 6,095,984 (“Amano”)
`B.K. Bootsma et. al, “Analysis of R-R intervals in patients with atrial
`fibrillation at rest and during exercise.” Circulation 1970; 41:783-
`794
`Frits L. Meijler and Fred H. M. Wittkampf, “Role of the
`Atrioventricular Node in Atrial Fibrillation” Atrial Fibrillation:
`Mechanisms and Management, 2nd ed. 1997 (“Meijler”)
`Heart Diseases _ Definition of Heart Diseases by Merriam-Webster
`Acharya UR, Joseph KP, Kannathal N, Lim CM, Suri JS. Heart rate
`variability: a review. Med Biol Eng Comput. 2006 Dec;
`44(12):1031-51 (“Acharya 2006”)
`Saime Akdemir Akar, Sadık Kara, Fatma Latifoğlu, Vedat Bilgiç.
`Spectral analysis of photoplethysmographic signals: The importance
`of preprocessing. Biomedical Signal Processing and Control, 2013;
`8(1):16-22 (Akar 2013)
`U.S. Provisional Application No. 61/915,113
`U.S. Provisional Application No. 61/953,616
`U.S. Provisional Application No. 61/969,019
`U.S. Provisional Application No. 61/970,551
`U.S. Provisional Application No. 62/014516
`U.S. Patent Publication No. 2012/0203491 (“Sun”)
`U.S. Patent No. 9,808,206 (“Zhao”)
`
`iv
`
`
`
`EX. NO.
`
`1017
`
`1018
`
`1019
`
`1020
`1021
`
`1022
`
`1023
`1024
`
`1025
`
`1026
`1027
`1028
`1029
`1030
`1031
`1032
`
`
`
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`
`EX. NO.
`1033
`
`1034
`
`1035
`
`1036
`
`1037
`
`1038
`
`1039
`
`1040
`
`1041
`
`1042
`
`
`
`
`DESCRIPTION
`Kleiger RE, Stein PK, Bigger JT Jr. Heart rate variability:
`measurement and clinical utility. Ann Noninvasive Electrocardiol.
`2005 Jan; 10(1):88-101 (“Kleiger-2005”)
`Chen Z, Brown EN, Barbieri R. Characterizing nonlinear heartbeat
`dynamics within a point process framework. IEEE Trans Biomed
`Eng. 2010 Jun; 57(6):1335-47 (“Chen 2010”)
`Karvonen, J., Vuorimaa, T. Heart Rate and Exercise Intensity During
`Sports Activities. Sports Medicine 5, 303–311 (1988) (“Karvonen
`1988”)
`Yu C, Liu Z, McKenna T, Reisner AT, Reifman J. A method for
`automatic identification of reliable heart rates calculated from ECG
`and PPG waveforms. J Am Med Inform Assoc. 2006 May-Jun;
`13(3):309-20 (“Yu 2006”)
`AliveCor v. Apple ITC Complaint Exhibit 11 (731 Infringement
`Chart)
`Tavassoli, M, Ebadzadeh, MM, Malek H. (2012). Classification of
`cardiac arrhythmia with respect to ECG and HRV signal by genetic
`programming. Canadian Journal on Artificial Intelligence, Machine
`Learning and Pattern Recognition. 3. 1-13 (“TavassoLi 2012”)
`Asl BM, Setarehdan SK, Mohebbi M. Support vector machinebased
`arrhythmia classification using reduced features of heart rate
`variability signal. Artif Intell Med. 2008 Sep; 44(1):51-64 (“Asl
`2008”)
`Yaghouby F., Ayatollahi A. (2009) An Arrhythmia Classification
`Method Based on Selected Features of Heart Rate Variability Signal
`and Support Vector Machine-Based Classifier. In: Dössel O.,
`Schlegel W.C. (eds) World Congress on Medical Physics and
`Biomedical Engineering, September 7 - 12, 2009, Munich, Germany.
`IFMBE Proceedings, vol 25/4. Springer, Berlin, Heidelberg
`(“Yaghouby 2009”)
`Dallali, A, Kachouri, A, Samet, M. (2011). Integration of HRV, WT
`and neural networks for ECG arrhythmias classification. ARPN
`Journal of Engineering and Applied Sciences. VOL. 6. 74-82
`(“Dallali 2011”)
`Sajda P. Machine learning for detection and diagnosis of disease.
`Annu Rev Biomed Eng. 2006; 8:537-65 (“Sajda 2006”)
`
`v
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`
`EX. NO.
`1043
`
`1044
`
`1045
`
`1046
`
`1047
`
`1048
`1049
`
`1050
`
`1051
`
`1052
`1053
`
`1054
`1055
`
`2001
`
`
`
`
`DESCRIPTION
`Aaron Smith. Smartphone Ownership – 2013 Update. Pew Research
`Center. June 5, 2013 (“Smith 2013”)
`C. Narayanaswami and M. T. Raghunath, “Application design for a
`smart watch with a high resolution display,” Digest of Papers. Fourth
`International Symposium on Wearable Computers, 2000, pp. 7-14
`(“Narayanaswami 2000”)
`Thong, YK, Woolfson, M, Crowe, JA, Hayes-Gill, B, Challis, R.
`(2002). Dependence of inertial measurements of distance on
`accelerometer noise, Meas. Measurement Science and Technology.
`13. 1163 (“Thong 2002”)
`AliveCor’s ITC Complaint filed on April 20, 2021 in “Certain
`Wearable Electronic Devices With ECG Capability and Components
`Thereof” ITC-337-3545-20210420 (“ITC Complaint”)
`Excerpts from Marcovitch, Harvey. Black’s Medical Dictionary.
`London: A. & C. Black, 2005
`U.S. Pat. No. 7,894,888 (“Chan”)
`Hu YH, Palreddy S, Tompkins WJ. A patient-adaptable ECG beat
`classifier using a mixture of experts approach. IEEE Transactions on
`Bio-medical Engineering. 1997 Sep; 44(9):891-900 (“Hu 1997”)
`Strath SJ, Swartz AM, Bassett DR Jr, et al. Evaluation of heart rate
`as a method for assessing moderate intensity physical activity.
`Medicine and Science in Sports and Exercise. 2000 Sep; 32(9
`Suppl):S465-70 (“Strath 2000”)
`Letter from Michael Amon re Conditional Stipulation dated June 4,
`2021
`Declaration of Mr. Jacob Munford
`Order Staying Case Pending Institution of And/Or Final
`Determination in Parallel ITC Matter (AliveCor Inc. v. Apple Inc.,
`6:20-cv-01112-26 (W.D. Tex. May 6, 2021)
`U.S. Provisional Application No. 61/895,995 (“Martin Provisional”)
`AliveCor’s District Court Complaint filed on May 25, 2021 in
`AliveCor, Inc. v. Apple Inc., 3:21-cv-03958 (N.D.Cal. May 25, 2021)
`(“Antitrust Complaint”)
`Declaration of Dr. Igor Efimov In Support of Patent Owner’s
`Preliminary Response
`
`
`
`vi
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`
`EX. NO.
`2002
`
`2003
`
`2004
`
`2005
`
`2006
`
`2007
`
`2008
`
`2009
`
`2010
`
`2011
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`DESCRIPTION
`B. S. Kim and S. K. Yoo, “Motion artifact reduction in
`photoplethysmography using independent component analysis,”
`IEEE Transactions on Biomedical Engineering, vol. 53, no. 3, pp.
`566-568, March 2006, doi: 10.1109/TBME.2005.869784
`Mao et al., Motion Artifact Reduction In Photoplethysmography
`For Reliable Signal Selection, arXiv, Sep 6, 2021;
`arXiv:2109.02755
`Apple’s September 10, 2021 Disclosure of Initial Invalidity
`Contentions in Response to Individual Interrogatory Nos. 19-21 of
`AliveCor’s First Set of Interrogatories to Apple, In the Matter of
`Certain Wearable Electronic Devices with ECG Functionality and
`Components Thereof, Inv. No. 337-TA-1266
`Certain Automated Storage and Retrieval Systems, Robots, and
`Components Thereof, Inv. No. 337-TA-1228, Order No. 6 Denying
`Respondents’ Motion For A Stay (Mar. 9, 2021)
`Certain Wearable Electronic Devices with ECG Functionality and
`Components Thereof, Inv. No. 337-TA-1266, Order No. 6 Setting
`Procedural Schedule (June 25, 2021)
`Respondent Apple Inc.’s Response to the Amended Complaint of
`AliveCor, Inc. Under Section 337 of the Tariff Act of 1930, As
`Amended, and Notice of Investigation, In the Matter of Certain
`Wearable Electronic Devices with ECG Functionality and
`Components Thereof, Inv. No. 337-TA-1266 (June 28, 2021)
`(Public)
`Apple’s August 18, 2021 List of Claim Terms To Be Construed, In
`the Matter of Certain Wearable Electronic Devices with ECG
`Functionality and Components Thereof, Inv. No. 337-TA-1266
`Joint Disclosure Of Proposed Claim Constructions, In the Matter
`of Certain Wearable Electronic Devices with ECG Functionality
`and Components Thereof, Inv. No. 337-TA-1266 (Sept. 13, 2021)
`Certain Wearable Electronic Devices with ECG Functionality and
`Components Thereof, Inv. No. 337-TA-1266, Order No. 12
`Construing the Terms of the Asserted Claims of the Patents at Issue
`(June 25, 2021)
`Apple’s September 24, 2021 Notice of Prior Art, In the Matter of
`Certain Wearable Electronic Devices with ECG Functionality and
`Components Thereof, Inv. No. 337-TA-1266
`
`
`
`
`
`
`
`vii
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`
`EX. NO.
`2012
`
`2013
`
`2014
`
`2015
`
`2016
`
`2017
`
`2018
`2019
`2020
`
`2021
`2022
`
`2023
`
`2024
`
`2025
`
`2026
`
`2027
`
`
`
`
`
`
`
`
`
`
`DESCRIPTION
`Complaint for Patent Infringement, AliveCor, Inc. v. Apple, Inc.,
`Case 6:20-cv-01112-ADA (W.D. Tex.) (Dec. 7, 2020)
`Complaint, iRobot Corp. v. SharkNinja Operating LLC et al., Case
`1:21-cv-10155-FDS (D. Del.) (Jan. 28, 2021)
`Oct. 12, 2021 Email authorizing Petitioner’s Reply to Patent Owner
`Preliminary Response
`Apple’s Sept. 22, 2021 Opening Claim Construction Brief, In the
`Matter of Certain Wearable Electronic Devices with ECG
`Functionality and Components Thereof, Inv. No. 337-TA-1266
`Declaration of Dr. Igor Efimov In Support of Patent Owner’s
`Response
`Transcript of March 24, 2022 Deposition of Dr. Bernard A.
`Chaitman
`Transcript of February 3, 2022 Deposition of Dr. Collin Stultz
`Heart Disease Facts, https://www.cdc.gov/heartdisease/facts.htm
`Changes in Heart Activity May Signal Epilepsy
`https://neurosciencenews.com/epilepsy-heart-rate-3827/ (Mar. 9,
`2016)
`Intentionally Omitted
`Nina Sviridova & Kenshi Sakai, Human photoplethysmogram: New
`insight into chaotic characteristics, Chaos Solitons & Fractals 77
`(Aug. 2015)
`Tania Pereira et al., Photoplethysmography based atrial fibrillation
`detection: a review, NPJ Digit Med. (Jan. 10, 2020),
`https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6954115/
`IR Efimov, et al., Optical mapping of repolarization and
`refractoriness from intact hearts, American Heart Association (Nov.
`1994)
`Josep Masip et al., Pulse oximetry in the diagnosis of acute heart
`failure, Rev Esp Cardiol (Engl Ed). (Oct. 2012)
`Eric J. Topol, High-performance medicine; the convergence of
`human and artificial intelligence, Nature Medicine, Vol. 25, 44-56
`(Jan. 2019)
`Bernard S. Chang, M.D. & Daniel H. Lowenstein, M.D.,
`Mechanisms of Disease: Epilepsy, N. Engl. J. Med., 1257-66 (2003)
`
`
`
`viii
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`EX. NO.
`2028
`
`2029
`
`DESCRIPTION
`Epilepsies: diagnosis and management, National Institute for Health
`and Care Guidance (Jan. 11, 2012)
`Carol Chen-Scarabelli et al, Device-Detected Atrial Fibrillation,
`Journal of the American College of Cardiology, Vol. 65, No. 3, 2015
`
`ix
`
`
`
`
`
`
`
`
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`I.
`
`INTRODUCTION
`
`Patent Owner AliveCor Inc. (“AliveCor”) respectfully submits this Response
`
`to Apple’s Petition.
`
`The Panel acknowledged at institution that it could not characterize the merits
`
`of the Petition as “especially ‘strong’.” The post-institution record, including the
`
`cross-examination of Apple’s expert Dr. Chaitman and the direct testimony of
`
`AliveCor’s expert, Dr. Efimov, only solidifies the Board’s initial impression of the
`
`Petition.
`
`Dr. Chaitman’s deposition testimony plainly supports AliveCor’s position. He
`
`admitted during cross-examination that the Petition’s main reference, Shmueli only
`
`uses SpO2 measurements to detect “irregular heart conditions,” which is consistent
`
`with Dr. Efimov’s testimony that a POSITA would not read Shmueli to disclose or
`
`render obvious detecting an arrythmia. Likewise, despite giving expert testimony
`
`regarding the prior art’s alleged disclosure of machine learning algorithms, Dr.
`
`Chaitman admitted that he did not consider himself an expert on that topic. This is
`
`in contrast to Dr. Efimov, who is an expert on machine learning algorithms, and
`
`opines here that the prior art fails to disclose or render obvious the claimed machine
`
`learning algorithms.
`
`
`
`1
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`
`Moreover, as explained below and supported by Dr. Efimov, neither Shmueli
`
`or Osorio—the two main references used in in each of the grounds—have anything
`
`to do with the detection of an arrhythmia. Indeed, Shmueli does not mention the term
`
`“arrhythmia” at all, a fact the Petition concedes. Shmueli instead only relates to the
`
`detection of “irregular heart conditions” detected using traditional SpO2 monitoring;
`
`a POSITA would not consider this a disclosure of detecting arrythmia.
`
`Osorio suffers from similar deficiencies. Osorio is directed to detecting
`
`epileptic seizures (a neurological event), which is not the same as nor does it render
`
`obvious detecting arrhythmias (a cardiac condition). In fact, its detection mechanism
`
`is tailored to the detection of these neurological conditions, and a POSITA would
`
`not have considered such detection mechanisms to render obvious detecting an
`
`arrythmia.
`
`Given these fundamental differences, a POSITA would never have combined
`
`the teachings of Shmueli and Osorio. While Shmueli is directed to detecting heart
`
`conditions, Osorio is directed to detecting neurological conditions. Moreover,
`
`Apple’s motivation to combine these teachings is premised entirely on the
`
`assumption that Shmueli discloses detecting an arrythmia; it does not, and without
`
`that teaching, Apple’s motivation argument fails.
`
`
`
`2
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`
`Shmueli also fails to teach “confirming” detection of an arrhythmia using an
`
`ECG. Shmueli teaches initiating a ECG after an SpO2 measurement is used to detect
`
`an irregular heart condition. There is no disclosure that Shmueli uses the ECG data
`
`to confirm an irregular heart condition. Instead, Shmueli merely teaches searching
`
`for correlations between the SpO2 and ECG measurements and using those
`
`correlations to update detection parameters. Dr. Chaitman admitted in cross-
`
`examination that the detection in Shmueli is performed based on the SpO2
`
`measurement and the SpO2 measurement only, and that the ECG “detection
`
`parameters” are merely used to detect the presence of an ECG waveform, not an
`
`irregular heart condition in that waveform. Shmueli contains no express or inherent
`
`disclosure of confirmation and does not render confirmation obvious.
`
`The prior art likewise fail to teach the required “machine learning” limitations.
`
`With respect to Shmueli, the Petition half-heartedly argues that Shmueli’s
`
`“correlation” element somehow, without explanation, renders the claims obvious.
`
`This is supported by Dr. Chaitman, who is admittedly not an expert in this area. As
`
`explained by Dr. Efimov, a POSITA would not consider this disclosure to be
`
`“machine learning,” but instead a rules-based algorithm—a type of algorithm Dr.
`
`Chaitman conceded is not machine learning.
`
`
`
`3
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`
`The Petition alternatively relies on Li for this limitation. Li, however, does
`
`not remedy Shmueli’s deficiencies. While Li discusses machine learning, it does so
`
`divorced from the context of the claims, which expressly require a machine learning
`
`algorithm “trained to detect arrythmias.” Li, in fact teaches the opposite: the use of
`
`machine learning to minimize false positives. It makes no mention of arrythmias,
`
`and gives no disclosure on how machine learning could be applied to detecting
`
`arrythmias. And, as Dr. Efimov testifies, a POSITA would not have found such
`
`machine learning algorithms obvious in light of Li’s disclosure.
`
`Thus, while the Petition may have presented sufficient evidence to pass the
`
`“reasonable likelihood” standard at institution, the post-institution evidence firmly
`
`establishes that the Petition has failed to carry the burden to establish the higher
`
`“preponderance” standard necessary to render the claim unpatentable.
`
`II. TECHNICAL BACKGROUND
`
`A. Arrhythmias and Atrial Fibrillation
`
`Cardiovascular diseases are considered to be one of the leading causes of
`
`death in the World. Ex. 1001, 1:34-39; Ex. 2019 (“Heart disease is the leading cause
`
`of death for men, women, and people of most racial and ethnic groups in the United
`
`States.”). One common cardiac condition is arrhythmia, which occurs when the
`
`electrical activity of the heart is irregular and/or is faster or slower than normal,
`
`
`
`4
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`known as tachyarrhythmia and bradyarrhythmia, respectively. Id., 1:40-44; Ex.
`
`2016, ¶ 3, 4. One of the most common forms of cardiac arrhythmia is atrial
`
`fibrillation, referred to as AFib, which occurs when electrical conduction through
`
`the atria of the heart is irregular and disorganized, leading to irregular activation of
`
`ventricles. Id., 1:44-49; Ex. 2016, ¶ 5. AFib is associated with atrial blood clot
`
`formation, which can lead to clot migration and stroke. Id., 1:49-51; Ex. 2016, ¶ 6,
`
`7. Specifically, Afib is “an abnormal cardiac rhythm characterized by a disorganized
`
`atrial activity.” Ex. 2023. Afib can be recognized in an ECG as “an irregularly
`
`irregular rhythm lasting more than 30s, with no discernible P-waves preceding the
`
`QRS complex.” Id.
`
`AFib, however, is not the only kind of cardiac arrhythmia. As recognized in
`
`the Suzuki paper cited by Petitioner, “[t]here are 8 kinds of arrhythmia” recognized
`
`by the widely used Minnesota Code Manual of Electrocardiographic Finding (June
`
`1982). Ex. 1016, 1. Of these, only “atrial or junctional premature beat;” “ventricular
`
`premature beat;” “atrial fibrillation / atrial flutter;” “supraventricular tachycardia
`
`intermittent;” “sick sinus syndrome;” “sinus tachycardia;” and “sinus bradycardia”
`
`are “detectable on the basis of RR intervals.” Id.; Ex. 2016, ¶ 8.
`
`There are other kinds of arrhythmia as well. For example, Osorio recognizes
`
`another kind of arrhythmia referred to as “respiratory sinus arrhythmia.” Ex. 1005,
`
`
`
`5
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`[0043]. Respiratory sinus arrhythmia is a “cardiac vagal reflex” where heart rate
`
`increases during inhalation (breathing in) and decreases during exhalation (breathing
`
`out). See Ex. 2020; Ex. 2023. Unlike the arrhythmia defined in the patent, respiratory
`
`sinus arrhythmia is not “a cardiac condition in which the electrical activity of the
`
`heart is irregular or is faster (tachycardia) or slower (bradycardia) than normal.” Ex.
`
`2016, ¶ 9.
`
`B.
`
`Strokes
`
`While arrhythmia themselves are often asymptomatic, they increase the
`
`likelihood of “palpitations, shortness of breath, fainting, chest pain or congestive
`
`heart failure,” but above all are a root cause of blood clots and stroke. Ex. 1001,
`
`1:47-51; Ex. 2016, ¶ 10. In AFib, the chaotic heart rhythm and lack of synchronized
`
`atrial contraction cause blood to collect in the heart and form clots, when these clots
`
`escape the heart they often travel via the arteries to the brain, causing strokes. Ex.
`
`2016, ¶ 11; see also Ex. 2029. Because clots and strokes are the immediate concern
`
`associated with AFib, they are sometimes treated before of the underlying
`
`arrhythmia. Ex. 1001, 1:56-58; Ex. 2016, ¶ 12. This is particularly true because long-
`
`undetected arrhythmia such as persistent AFib are so difficult to treat, and permanent
`
`AFib in many cases is untreatable. Ex. 2016, ¶ 12. However, if the early stage
`
`paroxysmal AFib is diagnosed early, it can often be readily treated by medications
`
`
`
`6
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`and/or ablation. Id. One in five strokes are associated with AFib and one-third of
`
`cardiac arrhythmias hospitalizations are due to AFib-related complications. Ex.
`
`2023.
`
`C.
`
`Photoplethysmography (PPG)
`
`A photoplethysmography (PPG) sensor can measure “a pulse pressure signal
`
`resulting from the propagation of blood pressure pulses along arterial blood vessels.”
`
`Ex. 2023; Ex. 2016, ¶ 13. PPG is “a pulse pressure waveform that originates from
`
`the heart contraction and propagates through the vascular tree.” Id. PPG waveforms
`
`have typical morphological components corresponding to landmark events in the
`
`cardiac cycle. Id.
`
`During the contraction of the left ventricle, blood is
`ejected out of the heart and propagates along the arterial
`tree, this corresponds to the initial positive slope of a PPG
`pulse. The systolic peak marks the maximum of the
`waveform. A decrease in amplitude following the systolic
`peak is marked by a local minimum, or the dicrotic notch,
`which corresponds to the closing of aortic valves
`separating the systolic and diastolic phases.
`
`Id. A PPG waveform for a healthy patient, including the diastolic notch and systolic
`
`peak, can be seen in the below image:
`
`
`
`7
`
`
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`Ex. 2022 at 55. When a longer time period is used for measuring the signal, the peaks
`
`are more compressed:
`
`
`
`
`
`8
`
`
`
`
`
`Id.
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`
`The “primary clinical application of PPG is arterial blood oxygen saturation
`
`(SpO2) estimation through pulse oximetry,” where “SpO2 is defined as the
`
`percentage of oxygen saturation in the arterial blood.” Ex. 2023. It is only recently—
`
`well after the priority dates of the AliveCor patents, that new applications of PPG
`
`“emerged for the continuous estimation of valuable cardiovascular parameters in
`
`ambulatory settings.” Id.; Ex. 2016, ¶ 14.
`
`A PPG signal has two main components: a DC component “which represents
`
`light reflected/transmitted from static arterial blood, venous blood, skin and tissues;”
`
`and an AC component “which arises from modulation in light absorption due to
`
`
`
`9
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`changes in arterial blood volume.” Id.; Ex. 2016, ¶ 15. PPG measurement can be
`
`carried out using two modes: transmission and reflectance. Id. In transmission mode,
`
`“the light transmitted through the medium is detected by a photodetector (PD),
`
`which is positioned in the opposite site of the light source.” Transmission mode PPG
`
`measurements are “limited to the extremities of the body, such as the fingertips,”
`
`and the location of the device “can interfere with daily routine movements.” Id. In
`
`reflectance mode, “the PD detects light that is back scattered or reflected from
`
`tissues, bone, and/or blood vessels, which means the light source and PD are
`
`positioned on the same side.” Id. Reflectance mode PPG can be measured at the
`
`wrist, such as on a smartwatch. Id.
`
`PPG monitoring is reliable in measurements of oxygen saturation and average
`
`heart rate, but historically has been found to be less reliable in detecting arrhythmias,
`
`especially atrial arrhythmias, such as atrial fibrillation. See Ex. 2023 (noting in 2017
`
`that while “PPG can be an alternative to ECG for AF detection, it remains that in
`
`real-world applications, PPG-based AF detection could be limited by a number of
`
`factors.”). Compared to an ECG, heart rate estimation is more challenging when
`
`using a PPG- signal. Id. In particular, motion artifacts, caused by the user’s physical
`
`activity (e.g., arm movement), can create noisy signals resulting significantly
`
`reduced PPG-signal quality. Ex. 2016, ¶ 16, 17; Ex. 2002. As a result, it is difficult
`
`
`
`10
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`to obtain a clean signal and extract HR from contaminated PPG. Id. Therefore,
`
`increasing the accuracy and robustness of PPG-based heart rate estimation remains
`
`at the forefront of research in this area even to this day. Id.; see also Ex. 2003.
`
`As Petitioner’s expert Dr. Stultz testified in the co-pending ITC investigation,
`
`ECG—and not PPG—is the “gold standard” for arrhythmia detection. Ex. 2018 at
`
`62:9-21; see also Ex. 1006 (“Gold standard data sets and subsets of critical ECG
`
`arrhythmia alarms”); Ex. 2023 (“ECG remains the gold standard for the
`
`electrophysiological definition and recognition of arrhythmias, including AF
`
`diagnosis.”); Ex. 2016, ¶ 18. Dr. Stultz explained that “continuous PPG” is a
`
`“suboptimal replacement” which “does not always result in a reliable signal and does
`
`not provide any information about whether p-waves are present or absent in the
`
`underlying ECG.” Ex. 2018 at 62:9-21; see also Ex. 2023 (“Compared to ECG, PPG-
`
`based AF detection is more challenging”). These p-waves are “needed to diagnose
`
`Afib.” Ex. 2018 at 62:9-21; Ex. 2016, ¶ 18..
`
`D. Electrocardiography (ECG)
`
`“In conventional clinical practice, electrocardiography (ECG) at hospital is
`
`used for diagnosis of arrhythmia.” Ex. 1016; see also Ex. 1001 at 1:52-54. An
`
`electrocardiogram, as its name implies, is a measurement of the electrical activity of
`
`the heart. Typical ECGs, including small Holter ECG devices, require connecting
`
`
`
`11
`
`
`
`
`electrodes to the patient’s chest to measure heart activity. See Ex. 1016 at 1; Ex.
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`2016, ¶ 19.
`
`The ECG waveform has several important features:
`
`Ex. 2016, ¶¶ 20-22.
`
`There are several important aspects to the ECG waveform relevant to this
`
`proceeding. First is the P-wave, which reflects atrial depolarization (activation). Id.
`
`
`
`
`
`12
`
`
`
`
`
`IPR2021-00971
`
`U.S. Patent No. 10,595,731
`PATENT OWNER’S RESPONSE
`
`
`Next is the QRS complex, which reflects the depolarization (activ