`
`Apple Inc. (Petitioner)
`v.
`AliveCor, Inc. (Patent Owner)
`
`9,572,499 | IPR2021-00970 | APL2021046011
`10,595,731 | IPR2021-00971 | APL2021046012
`10,638,941 | IPR2021-00972 | APL2021046013
`
`Before Hon. Robert A. Pollock, Eric C. Jeschke, and David Cotta
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`1
`
`1
`
`APPLE 1086
`Apple v. AliveCor
`IPR2021-00971
`
`
`
`Instituted Grounds
`
`’499 IPR
`IPR2021-00970
`
`’731 IPR*
`IPR2021-00971
`
`Ground
`Obviousness over Shmueli, Osorio
`Obviousness over Shmueli, Osorio, Hu1997
`
`Claim(s) Challenged
`1-6, 10-16, 20
`
`7-9, 17-19
`
`Ground
`Obviousness over Shmueli
`Obviousness over Shmueli, Osorio
`Obviousness over Shmueli, Osorio, Li2012
`Obviousness over Shmueli, Osorio, Kleiger2005
`Obviousness over Shmueli, Osorio, Chan
`
`Claim(s) Challenged
`1, 7, 12, 13, 16, 17, 23-26, 30
`
`1, 2, 4, 7, 12-14, 16-18, 20, 23-26, 30
`3, 5, 6, 19, 21, 22
`8, 9, 10, 11, 27-29
`15
`
`’941 IPR
`IPR2021-00972
`
`Ground
`Obviousness over Shmueli, Osorio
`Obviousness over Shmueli, Osorio, Lee2013
`Obviousness over Shmueli, Osorio, Chan
`
`Claim(s) Challenged
`1, 5, 7–9, 11, 12, 16, 18, 19, 20, 22, 23
`
`2-4, 6, 13, 14, 15, 17
`10, 21
`
`* The record from IPR2021-00971 is primarily referenced throughout to address topics covered in all three IPRs
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`2
`
`2
`
`
`
`Applied Prior Art
`
`The Shmueli-Osorio Combination
`has been applied in all IPRs
`
`Shmueli
`’731
`
`Hu1997
`’499
`
`Shmueli +
`Osorio
`’499
`’731
`’941
`
`Lee2013
`’941
`
`Osorio
`
`Kleiger2005
`’731
`
`Li2012
`’731
`
`Chan
`’731
`’941
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`3
`
`3
`
`
`
`Table Of Contents
`
`The Challenged Patents
`
`Key Asserted Prior Art (Shmueli, Osorio)
`
`The Shmueli-Osorio Combination
`
`Topics for Discussion
`
`1 – Arrhythmia Detection (All IPRs)
`
`2 – “Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`3 – Atrial Fibrillation (’941 IPR)
`
`4 – Machine Learning (’499, ’731 IPRs)
`
`5 – POSITA
`
`Other Reference Slides
`
`5
`
`12
`
`21
`
`25
`
`25
`
`32
`
`42
`
`51
`
`69
`
`77
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`4
`
`4
`
`
`
`The Challenged Patents
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`5
`
`5
`
`5
`
`
`
`The Challenged Patents
`
`’941 Patent
`
`’499 and ’731 Patents
`
`’941 Patent, Fig. 7 (annotated)
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`6
`
`’499 Patent, Fig. 10 (annotated)
`
`6
`
`
`
`The ’941 Patent
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`7
`
`’941 Petition, 7 (citing APPLE-1001, Fig. 1) (annotated)
`
`7
`
`
`
`The ’941 Patent
`
`“The ’941 patent relates to cardiac monitoring to sense the presence of an intermittent
`arrhythmia in an individual.”
`
`’941 Petition, 5; APPLE-1001, Abstract, 1:17-22
`
`APPLE-1001, Claim 1; see ’941 Petition, 31-53
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`8
`
`8
`
`
`
`The ’731 and ’499 Patents
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`9
`
`’499 Patent, Fig. 10
`
`9
`
`
`
`The ’731 Patent
`
`’731 Patent, Claim 1; see ’731 Petition, 13-28
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`10
`
`10
`
`
`
`The ’499 Patent
`
`’499 Patent, Claim 1; see ’499 Petition, 28-46
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`11
`
`11
`
`
`
`Key Asserted Prior Art
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`12
`
`12
`
`12
`
`
`
`Shmueli (PCT Patent Publication WO 2012/140559)
`
`Shmueli’s Monitoring Device
`
`APPLE-1004, Figs. 1A, 1B (cited in ’731 Petition, 9)
`
`APPLE-1004, Figs. 1A, 1B (cited in ’731 Petition, 9)
`
`“…the terms ‘oxygen saturation in the blood’, ‘blood oxygen saturation”,
`‘pulse oximeter’, oximetry, SpO2, and photoplethysmography have the
`same meaning and may be used interchangeably….”
`APPLE-1004, 7:25-27 (cited in ’731 Petition, 8)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`13
`
`13
`
`
`
`Shmueli (PCT Patent Publication WO 2012/140559)
`
`“Measuring ECG typically requires connecting the patient to an ECG
`measuring device via a plurality of wires connected to the patient in
`predefined places of the body.”
`
`APPLE-1004, 2:21-3:3 (cited in ’731 Petition, 7-8)
`
`“The prior art does not consider a requirement to enable a patient to perform
`ECG measurement as soon as an irregular heart activity develops and without
`requiring the ECG to be constantly wired to the patient.”
`APPLE-1004, 9:21-23(cited in ’731 Petition, 12)
`
`“The present invention…provid[es] a combined oximetry and
`electrocardiogram measuring system and a method in which the oximetry
`measurement is performed continuously and/or repeatedly, and the ECG
`measurement is triggered upon detection of an intermittent irregular heart-
`related event.”
`
`APPLE-1004, 9:23-28 (cited in ’731 Petition, 10)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`14
`
`14
`
`
`
`Shmueli (PCT Patent Publication WO 2012/140559)
`
`Shmueli’s Monitoring Technique
`
`’731 Petition, 22 (citing APPLE-1004, Fig. 7)
`
`“…Petitioner has shown sufficiently that one of ordinary skill in the art would have understood that
`Shmueli to disclose detecting arrhythmia based on PPG data and confirming the diagnosis with
`an ECG measurement.”
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`’731 Institution Decision, 45
`
`15
`
`15
`
`
`
`Osorio (U.S. Pat. App. No. 2014/0275840)
`
`APPLE-1005, Fig. 4A (annotated) (cited in ’731 Petition, 58)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`16
`
`16
`
`
`
`Osorio (U.S. Pat. App. No. 2014/0275840)
`
`“…the present disclosure relates to a
`method of detecting a pathological
`body state of a patient, comprising
`receiving a body signal of the patient;
`determining a first body data variability
`(BDV) from said body signal;
`determining an activity level of said
`patient; determining a non-pathological
`range for said first BDV, based at least
`in part on said activity level; comparing
`said first BDV to said non-pathological
`range for said first BDV; and detecting a
`pathological body state when said BDV
`is outside said non-pathological range.”
`APPLE-1005, [0003] (cited in ’731 Petition, 39)
`
`“In one embodiment, the severity may
`be measured by a magnitude and/or
`duration of a pathological state such as
`a seizure, a type of autonomic change
`associated with the pathological state
`(e.g., changes in heart rate, breathing
`rate, brain electrical activity, the
`emergence of one or more cardiac
`arrhythmias, etc.).”
`APPLE-1005, [0071] (cited in ’731 Petition, 71)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`17
`
`17
`
`
`
`Osorio (U.S. Pat. App. No. 2014/0275840)
`
`Efimov Testimony
`Q. Okay. Do you have an understanding
`of how a person of skill in the art in
`2013 would understand the term
`"pathological state"?
`
`…A
`
`. Pathological state could be also
`understood as an irregular heart
`condition.
`
`Chaitman Testimony
`Q. What does it mean to you as a
`clinician when somebody says it's
`pathological?
`
`A. “Pathological” means it falls
`outside the norm.
`
`Ex. 2017, 94:19-22
`
`APPLE-1069, 50:17-22 (cited ’731 Reply, 1-2)
`
`Efimov Testimony
`Q. Would a pathological state in 2013
`to a person of ordinary skill in the art
`include an arrhythmia?
`
`A. Yes. Arrhythmia would be
`included, yes. And also heart failure,
`diabetes, you know, metabolic condition,
`and many, many others.
`APPLE-1069, 50:17-22 (cited ’731 Reply, 1-2)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`18
`
`18
`
`
`
`Osorio (U.S. Pat. App. No. 2014/0275840)
`
`“…a certain pathological state (e.g., epileptic seizure)”
`APPLE-1005, [0037] (cited in ’731 Reply, 11)
`
`“In one embodiment, the pathological state is an epileptic event, e.g., an
`epileptic seizure”
`
`APPLE-1005, [0046] (cited in ’731 Reply, 11)
`
`“This disclosure relates to medical device systems and methods capable of
`detecting a pathological body state of a patient, which may include epileptic
`seizures, and responding to the same.”
`
`APPLE-1005, [0002] (cited in ’731 Reply, 11)
`
`…
`
`APPLE-1005, Claims 1, 7 (cited in ’731 Reply, 11)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`19
`
`19
`
`
`
`Osorio (U.S. Pat. App. No. 2014/0275840)
`
`“In one embodiment, the severity may
`be measured by a magnitude and/or
`duration of a pathological state such as
`a seizure, a type of autonomic
`change associated with the
`pathological state (e.g., changes in
`heart rate, breathing rate, brain
`electrical activity, the emergence of
`one or more cardiac arrhythmias,
`etc.).”
`
`APPLE-1005, [0071] (cited in ’731 Petition, 71)
`
`“In one embodiment, the pathological
`state is an epileptic event, e.g., an
`epileptic seizure. For example, if the
`body signal is heart rate, then an
`instantaneous heart rate above the non-
`pathological heart rate range
`determined by the BDV range
`determination module 260 may indicate
`a tachycardia episode frequently seen
`with epileptic seizures originating from
`or spreading to certain brain regions,
`and an instantaneous heart rate below
`the non-pathological heart rate range
`may indicate a bradycardia episode
`occasionally seen with epileptic
`seizures originating from certain brain
`regions.”
`
`APPLE-1005, [0046] (cited in ’731 Reply, 13)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`20
`
`20
`
`
`
`The Shmueli-Osorio Combination
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`21
`
`21
`
`21
`
`
`
`The Shmueli-Osorio Combination
`
`Osorio improves Shmueli’s detection of irregular heart conditions:
`
`Improve detection accuracy
`using activity level monitoring
`
`HRV to improve detection
`
`’731 Petition, 58 (citing APPLE-1005, Fig. 4A)
`
`’731 Petition, 40 (citing APPLE-1005, FIG. 8)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`22
`
`22
`
`
`
`The Shmueli-Osorio Combination
`
`Osorio improves Shmueli’s detection of irregular heart conditions:
`
`Chaitman Testimony
`“A POSITA also would have been motivated
`to incorporate Osorio’s HRV analysis
`because it is less affected by noise.”
`APPLE-1003 (’731 IPR), ¶159 (cited in 731 IPR Petition, 48);
`APPLE-1039, 52
`
`Chaitman Testimony
`“A POSITA would have been aware that the
`normal heart rate ranges from 60-100bpm,
`and that failure to account for physical
`activity or stress, that might elevate heart
`rates above 100 bpm during normal daily
`activity would result in triggered alerts from
`physiologic and not pathologic events that
`occur during normal daily activity. Indeed,
`Osorio explicitly describes the benefits
`(e.g., improved accuracy, reliability, and
`reduced false detection) of using activity
`level to detect an irregular heart
`condition. APPLE-1005, [0029], [0036].
`With these benefits in mind, a POSITA
`would have been motivated to incorporate
`Osorio’s activity sensor and activity level
`analysis techniques into Shmueli’s heart
`monitoring device.”
`APPLE-1003 (’731 IPR), ¶151 (cited in ’731 Petition, 43)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`23
`
`23
`
`
`
`The Shmueli-Osorio Combination
`
`The Shmueli-Osorio Device
`
`Apple’s ’731 Petition
`The Shmueli-Osorio Combination would
`have:
`
`(1)
`
`Increased accuracy
`
`(2) Provided continuously monitoring
`heart rhythm with an oximetry
`sensor
`
`(3)
`
`Improved user satisfaction
`’731 Petition, 54;
`APPLE-1005, [0029]; APPLE-1003, ¶167
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`24
`
`’731 Petition, 53 (modifying APPLE-1004, FIG. 6);
`APPLE-1003 (’731 IPR), ¶167
`
`24
`
`
`
`Five Topics For Discussion
`
`1. Arrhythmia Detection (All IPRs)
`
`2.
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`3. Atrial Fibrillation (’941 IPR)
`
`4. Machine Learning (’499, ’731 IPRs)
`
`5. POSITA (All IPRS)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`25
`
`25
`
`
`
`Arrhythmia Detection (All IPRs)
`
`Apple’s ’731 Petition
`“As discussed in Section III.A.1 and
`explained by Dr. Chaitman, a POSITA would
`have understood and found obvious that
`the term “irregular heart condition” refers to
`arrhythmia.”
`
`’731 Petition, 13; APPLE-1003 (’731 IPR), ¶¶72-73
`
`AliveCor’s ’731 Preliminary Response
`“…a POSITA would not automatically
`assume that this otherwise undefined term
`refers to arrhythmia, as opposed to another
`heart condition..”
`
`’731 Response, 44
`
`’731 Institution Decision
`“…one of ordinary skill in the art would have understood Shmueli’s use of “irregular
`heart condition” as referring to—or at a minimum, encompassing—arrhythmia,
`and, thus, disclosing the detection of arrhythmia.”
`
`’731 Institution Decision, 33-34
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`26
`
`26
`
`
`
`Arrhythmia Detection (All IPRs)
`
`AliveCor’s ’731 Response
`“…Shmueli’s disclosure of the
`heart condition genus is not a
`disclosure of the arrhythmia
`species.”
`
`’731 Response, 44
`
`’AliveCor’s ’731 Response
`Acknowledges that irregular
`heart conditions includes the
`following conditions:
`
`•Arrhythmia
`•Heart attack
`•Angina pectoris
`•Cardiomyopathy
`•Congenital heart disease
`•Coronary heart disease
`•Heart-valve defect
`
`’731 Response, 44-45 (citing APPLE 1023, APPLE 1047)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`27
`
`27
`
`
`
`Arrhythmia Detection (All IPRs)
`
`Chaitman Declaration
`“…a POSITA would have
`understood that [irregular heart
`condition] refers to arrhythmia,
`which is one of the most obvious (if
`not the most obvious) types of
`“irregular heart condition[s]” that
`can be determined
`APPLE-1003, ¶48 (citing APPLE-1016, p. 6081;
`APPLE-1020, Abstract, 44:29-32;
`APPLE-1011, Abstract;
`APPLE-1023, 2;
`APPLE-1047, 320-21;
`APPLE-1001, 1:40-42;
`APPLE-1004, 8:11-13, 15:3-5);
`(cited in ’731 Petition, 10-11)
`
`Apple’s ’731 Reply
`“If ‘irregular heart conditions’ is
`understood to include ‘numerous’
`conditions and Shmueli discloses
`detecting ‘various irregular heart
`conditions,’ then it would have
`been obvious that arrhythmia
`(which is the most common type
`of heart condition) is one
`example detected by Shmueli.”
`’731 Reply, 9 (citing APPLE-1004, 12:29-31)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`28
`
`28
`
`
`
`Arrhythmia Detection (All IPRs)
`
`Goldreich – U.S. Pat. No. 7,598,878
`“The signal that is collected from the
`SpO2 sensor may also optionally be
`used for producing other heart
`related information. For example,
`processing the signal that reflects
`the intensity of the reflected IR light
`may produce information such as
`heart rate, PWTT, irregularity of
`heart rate etc.”
`APPLE-1061, 16:54-58 (cited in ’731 Reply, 8)
`
`APPLE-1061, Claim 5 (cited in ’731 Reply, 8)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`29
`
`29
`
`
`
`Arrhythmia Detection (All IPRs)
`
`Shmueli
`“Another common solution is a Holter
`device, which is practically a small ECG
`device connected to the patient for
`typically 24 hours, recording the ECG
`signal. Hopefully, the heart-related event
`occurs during the recording time. The
`Holter does not limit the mobility range of
`the patient but has a time limit of its
`operation. For events that are not
`sufficiently frequent this solution does not
`work. Also important and highly
`disadvantageous is the Holter device uses
`electric contacts at the end of electric
`cables. Thus, the patient has to be
`constantly wired to the Holter device.”
`APPLE-1004, 2:21-3:3 (cited in ’731 Reply, 7-8)
`
`Efimov Testimony
`Another common solution is a Holter
`device, which is practically a small
`ECG device connected to the patient
`for typically 24 hours, recording the
`ECG signal. Hopefully, the heart-related
`event occurs during the recording time.
`The Holter does not limit the mobility
`range of the patient but has a time limit
`of its operation.
`APPLE-1004, 2:21-3:3 (cited in ’731 Reply, 8)
`
`Efimov Testimony
`Q. So is it fair that a Holter monitor
`was used -- one of the primary uses of
`a Holter monitor was to determine
`whether a patient may have cardiac
`arrhythmia?
`
`A.· Yes.
`
`APPLE-1069, 30:24-31:9 (cited in ’731 Reply, 8)
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`30
`
`30
`
`
`
`Arrhythmia Detection (All IPRs)
`
`Apple’s ’941 Reply
`“U.S. 2005/177051 to Almen
`discloses a wrist-watch with heart
`rate sensors, including an ECG and
`a pulse oximeter that is used to
`detect arrhythmia.”
`
`’941 Reply, 11; APPLE-1062, [0014], [0051], [0055], [0062]
`
`Efimov Testimony
`Q. Dr. Efimov, I believe you state in your
`declarations that arrhythmias can be
`intermittent; is that correct?
`
`A. Yes. Especially such as atrial
`fibrillation. Many arrhythmias could be
`intermittent.
`
`APPLE-1069, 23:25-24:4 (cited in ’941 Reply, 8)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`31
`
`31
`
`
`
`Five Topics For Discussion
`
`1. Arrhythmia Detection (All IPRs)
`
`2. “Confirmation” of Arrhythmia Detection (’731, ’941
`IPRs)
`
`3. Atrial Fibrillation (’941 IPR)
`
`4. Machine Learning (’499, ’731 IPRs)
`
`5. POSITA (All IPRS)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`32
`
`32
`
`
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`’731 Patent
`
`’941 Patent
`
`…
`
`…
`
`’731 Patent, Claim 1 (cited in ’941 Petition, 51)
`
`’941 Patent, Claim 1 (cited in ’941 Petition, 51)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`33
`
`33
`
`
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`Efimov Testimony
`“PPG monitoring is …historically has
`been found to be less reliable in
`detecting arrhythmias, especially atrial
`arrhythmias, such as atrial
`fibrillation….Compared to an ECG, heart
`rate estimation is more challenging when
`using a PPG signal.”
`Ex. 2016, ¶16 (cited in ’731 POR, 10-11; ; ’941 POR, 9); Ex. 2023, 2, 9
`
`Pereira
`“ECG remains the gold standard for the
`electrophysiological definition and
`recognition of arrhythmias, including AF
`diagnosis”
`
`Ex. 2023, 3 (cited in ’731 POR, 11; ’941 POR, 10)
`
`Chaitman Testimony
`“A POSITA would have recognized that Shmueli’s focus on
`enabling ECG measurements “as soon as” an irregular heart
`condition is detected enables ECG data to be used to
`confirm the detection of the irregular heart condition
`using PPG data, thereby improving detection accuracy
`compared to prior art heart monitoring devices.”
`Ex. 2016, ¶16 (cited in ’731 POR, 10-11)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`34
`
`34
`
`
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`APPLE-1004, FIG. 7 (cited in ’731 Petition, 27; ’941 Petition, 14)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`35
`
`35
`
`
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`Shmueli
`“The software program proceeds to element 38 to derive
`from the SpO2 measurement physiological parameters
`such as pulse rate, pulse amplitude, pulse shape, rate of
`blood flow, etc. Then, the software program scans the
`derived physiological parameters to detect various
`irregularities of the heart condition. The scanning for an
`irregular heart condition preferably uses heart-irregularity
`detection parameters (element 39) stored in the memory
`unit 28. When an irregular heart condition is detected
`(element 40) the software program continues to element
`41. However, the SpCh measurement (element 37)
`preferably continues and optionally also the derivation of
`physiological parameters as well as the detection of
`irregular heart conditions (element 38).”
`
`APPLE-1004, 13:14-22 (cited in ’731 Reply, 16)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`36
`
`36
`
`
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`Chaitman Testimony
`“…a POSITA would have understood that the software at element
`50, element 39 and element 38 causes the processing device to
`confirm the presence of the arrhythmia based on the ECG
`data, by searching for correlations between the PPG and
`ECG data, modifying detection parameters, and confirming the
`presence of arrhythmia.
`
`APPLE-1003, ¶112 (cited in ’731 Petition, 27);
`APPLE-1003, ¶154 (cited in ’941 Petition, 68)
`
`APPLE-1004, FIG. 7 (cited in ’731 Petition, 27; ’941 Petition, 14)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`37
`
`37
`
`
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`Chaitman Testimony
`“…the software proceeds to element 51 to determine a set of stop
`conditions (element 52), such as whether ‘the irregular heart condition has
`stopped.’ APPLE-1004, 13:22-29. Shmueli discloses that when the
`software program detects that “the irregular heart condition has
`stopped” (element 51), the software program notifies the user that the ECG
`measurement has stopped (element 53) and stops the ECG measurement
`(element 54).”
`
`APPLE-1003 (’731 IPR), ¶113 (cited in ’731 Petition); APPLE-1004, 13:22-29
`
`APPLE-1004, FIG. 7 (cited in ’731 Petition, 27; ’941 Petition, 14)
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`38
`
`38
`
`
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`AliveCor’s ’731 Response
`“...all detection of irregular
`heart conditions in Shmueli is
`by the SpO2 measurement.”
`
`’731 Response, 52
`
`Apple’s ’731 Reply
`“…Shmueli’s ECG data still
`has a role. As shown in FIG.
`7, Shmueli’s ECG analysis
`(element 50) leads to new
`detection parameters
`(element 39) used for more
`accurate detection of the
`irregular heart condition
`(element 38) with SpO2/PPG
`data.”
`
`’731 Response, 52
`
`’731 Reply, 15-16;
`’941 Reply, 20
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`39
`
`39
`
`
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`AliveCor’s ’731 Sur-Reply
`“Shmueli’s ‘correlation’ of ECG and
`SpO2 measurements is used to
`update detection parameters.”
`
`’731 Response, 52
`“And those detection parameters
`are merely used to detect the
`presence of an ECG signal.”
`’731 Sur-Reply, 18
`
`Apple’s ’731 Petition
`“…a POSITA would have
`understood that the software at
`element 50, element 39 and
`element 38 causes the processing
`device to confirm the presence
`of the arrhythmia based on the
`ECG data.”
`
`’731 Petition, 27; ’941 Petition, 51-53
`
`APPLE-1004, FIG. 7 (cited in ’731 Petition, 27; ’941 Petition, 14)
`
`APPLE-1004, FIG. 7 (cited in ’731 Petition, 27; ’941 Petition, 14)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`40
`
`40
`
`
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`Shmueli
`“When the software program detects
`that a condition for stopping the ECG
`measurement is met (element 51)
`using stop conditions (element 52)
`preferably stored in the memory unit 28
`the software program preferably notifies
`the user that the ECG measurement
`has stopped (element 53) and stops
`the ECG measurement (element 54).”
`APPLE-1004, 14:22-29 (cited in ’731 Reply, 15; ’941 Reply, 20)
`
`Shmueli
`Examples of conditions for stopping the
`ECG measurement:
`• The irregular heart condition has
`stopped.
`• The heart condition returned to
`normal.
`• A predefined period elapsed.
`• A predefined number ofheart beats
`were counted.
`APPLE-1004, 14:22-29 (cited in ’731 Reply, 15; ’941 Reply, 20)
`
`APPLE-1004, FIG. 7 (cited in ’731 Petition, 27; ’941 Petition, 14)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`41
`
`41
`
`
`
`Five Topics For Discussion
`
`1. Arrhythmia Detection (All IPRs)
`
`2.
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`3. Atrial Fibrillation (’941 IPR)
`
`4. Machine Learning (’499, ’731 IPRs)
`
`5. POSITA (All IPRS)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`42
`
`42
`
`
`
`Atrial Fibrillation (’941 IPR)
`
`Claim 1
`
`…
`
`Chaitman Testimony
`“By 1977, both detecting possible atrial
`fibrillation using irregular pulse rhythms or
`heartbeats and techniques to quantitatively
`characterize irregularities were well-known.”
`APPLE-1003 (’941 IPR), 27 (cited in ’941 Petition, 2)
`
`Claim 2
`
`’941 Patent, Claims 1 and 2
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`43
`
`43
`
`
`
`Atrial Fibrillation (’941 IPR)
`
`Apple’s ’941 Petition
`“…both Shmueli and Lee 2013
`disclose detecting arrhythmia
`using PPG data, further confirming
`their combinability.”
`
`’941 Reply, 26
`
`Chaitman Testimony
`“A POSITA would have therefore found it
`obvious to implement the Shmueli-
`Osorio- Lee-2013 device without
`invention such that it executes Lee-
`2013’s software application to detect AF
`using PPG data collected by oximetry
`sensor 13”
`APPLE-1003 (’941 IPR), ¶154 (cited in ’941 Petition, 68)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`44
`
`44
`
`
`
`Atrial Fibrillation (’941 IPR)
`
`Lee-2013
`“Atrial fibrillation is the most common
`sustained arrhythmia. Over 3 million
`Americans are currently diagnosed, and the
`prevalence of AF is increasing with the aging
`of the U.S. population (Go et al., 2001).”
`
`APPLE-1011, 1 (cited in ’941 Petition, 66)
`
`Efimov Testimony
`Q. Would you agree that atrial fibrillation is
`the most common cardiac arrhythmia
`present?
`
`A. Statistically, yes. This is what is referred
`to.
`
`APPLE-1069, 23:5-9 (cited in ’941 Reply, 10)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`45
`
`45
`
`
`
`Atrial Fibrillation (’941 IPR)
`
`’AliveCor’s ’731 Response
`“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).”
`
`’731 Response, 5 (citing APPLE-1016, 1)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`46
`
`46
`
`
`
`Atrial Fibrillation (’941 IPR)
`
`Motivation to Combine
`
`Chaitman Testimony
`“The combined Shmueli-Osorio-
`Lee-2013 device would have
`also provided an improvement
`over Lee-2013’s technique
`using a mobile device to detect
`AF since the combined device
`leverages the wrist-mounted
`form factor of the Shmueli’s
`heart monitoring device
`without requiring the user to
`carry a separate mobile device.”
`
`APPLE-1003 (’941 IPR), ¶152 (cited in ’941 Petition, 66)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`47
`
`47
`
`
`
`Atrial Fibrillation (’941 IPR)
`
`Apple’s Key Position
`A POSITA would have found it obvious to incorporate Lee-2013’s teachings of detecting
`atrial fibrillation using PPG data into the Osorio-Shmueli combination
`
`Apple’s ’941 Petition
`“Shmueli and Osorio each describes techniques for generally detecting
`arrhythmias, but do not address detection of specific types of arrhythmias,
`such as AF.”
`
`’941 Petition, 66
`
`Lee2013
`“Atrial fibrillation is the most common
`sustained arrhythmia.”
`APPLE-1011, 26 (cited in ’941 Petition)
`
`Chaitman Testimony
`“Given the prominence of AF, a
`POSITA would have recognized that
`incorporating AF detection into the
`Shmueli-Osorio device provides a new
`capability for classifying an arrhythmia
`as AF using PPG data.”
`APPLE-1003, ¶152 (cited in ’941 Petition, 66)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`48
`
`48
`
`
`
`Atrial Fibrillation (’941 IPR)
`
`AliveCor’s ’941 Response
`“…Lee 2013 teaches not to use a separate
`ECG.”
`
`’941 Response, 56
`
`Lee-2013
`“Given the ever-growing popularity of cell
`phones and smartphones, a smartphone-
`based AF detection application provides
`patients and their caregivers with access to an
`inexpensive and easy-to-use monitor for AF
`outside of the traditional health care
`establishment. Because the application does
`not involve a separate ECG sensor and
`instead employs built-in hardware, it is
`both novel and cost-effective. We believe
`this package will lead to better acceptance and
`more widespread use than existing out-of-
`hospital arrhythmia monitors.”
`APPLE-1011, 29 (cited in ’941 Response, 56-57)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`49
`
`49
`
`
`
`Atrial Fibrillation (’941 IPR)
`
`AliveCor’s ’941 Sur-Reply
`“Apple presents no argument that
`the references, alone or in
`combination, would confirm the
`presence of atrial fibrillation using
`an ECG sensor.”
`
`’941 Sur-Reply, 19-20
`
`Chaitman Testimony
`“Given the prominence of AF, a
`POSITA would have recognized that
`incorporating AF detection into the
`Shmueli-Osorio device provides a new
`capability for classifying an arrhythmia as
`AF using PPG data.”
`
`APPLE-1003 (’941 IPR), ¶152 (cited in ’941 Petition,66)
`
`Chaitman Testimony
`The combination “…improves the accuracy of
`AF detection provided by Lee-2013 alone
`since the Shmueli-Osorio-Lee-2013 device
`uses ECG data to confirm AF detection
`based on PPG data.”
`APPLE-1003 (’941 IPR), ¶153 (cited in ’941 Petition,67)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`50
`
`50
`
`
`
`Five Topics For Discussion
`
`1. Arrhythmia Detection (All IPRs)
`
`2.
`
`“Confirmation” of Arrhythmia Detection (’731, ’941 IPRs)
`
`3. Atrial Fibrillation (’941 IPR)
`
`4. Machine Learning (’499, ’731 IPRs)
`
`5. POSITA (All IPRs)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`51
`
`51
`
`
`
`Machine Learning (’499, ’731 IPRs)
`
`Apple’s Key Positions
`Machine learning was well-known:
`
`1) Using machine learning to detect arrhythmia was well-known
`
`2) The ’499 and ’731 patents only recite broad categories of well-known machine
`learning methods
`
`3) Machine learning was known to offer significant benefits
`
`’731 Reply, 18-20; ’499 Reply, 18-20
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`52
`
`52
`
`
`
`Machine Learning (’499, ’731 IPRs)
`
`1) Using machine learning to detect arrhythmia was well-known
`
`Asl 2008
`“This paper presents an effective cardiac
`arrhythmia classification algorithm using
`the heart rate variability (HRV) signal. The
`proposed algorithm is based on the
`generalized discriminant analysis (GDA)
`feature reduction scheme and the support
`vector machine (SVM) classifier.”
`
`APPLE-1039, Abstract (cited in ’731 Petition, 69; ’731 Reply, 20; and
`in APPLE-1003 (’499 IPR), ¶214)
`
`Dallali 2011
`“Integration of HRV, WT and neural
`networks for ECG arrhythmias
`classification.”
`APPLE-1041, p.74 (cited in ’731 Petition, 67; ’731 Reply, 20; and in
`APPLE-1003 (’731 IPR), ¶¶117, 259)
`
`Yaghouby 2009
`“Some examples of these automatic arrhythmia
`detection and classification techniques are neural
`networks [1,2], wavelet transforms [3], support
`vector machines [4], fuzzy logic [5] and the rule-
`based algorithms [6].”
`“In this paper we present an arrhythmia
`classification method using Heart Rate Variability
`(HRV) signal features and Support Vector
`Machine (SVM) Classifier.”
`APPLE-1040, 1928 (cited in ’731 Petition, 67; ’731 Reply, 20; and in
`APPLE-1003 (’731 IPR), ¶¶27, 117, 259)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`53
`
`53
`
`
`
`Machine Learning (’499, ’731 IPRs)
`
`1) Using machine learning to detect arrhythmia was well-known
`
`Tavassoli 2012
`Classification of cardiac arrhythmia with respect
`to ECG and HRV signal by genetic
`programming.”
`APPLE-1038, Abstract (cited in ’731 Petition, 69; ’731 Reply, 20; and in
`APPLE-1003 (’731 IPR), ¶263)
`
`Tsipouras 2004
`“We have developed an automatic arrhythmia
`detection system, which is based on heart rate
`features only. Initially, the RR interval duration
`signal is extracted from ECG recordings and
`segmented into small intervals. The analysis is
`based on both time and time—frequency (t—f)
`features. Time domain measurements are
`extracted and several combinations between the
`obtained features are used for the training of a
`set of neural networks.”
`
`APPLE-1012, Abstract (cited in ’731 Petition, 69; ’731 Reply, 20; and in
`APPLE-1003 (’731 IPR), ¶¶26, 35, 117, 263)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`54
`
`54
`
`
`
`Machine Learning (’499, ’731 IPRs)
`
`2) The ’499 and ’731 patents only recite broad categories of well-known
`machine learning methods
`
`’731 Patent
`“Any number of machine learning algorithms
`or methods may be trained to identify atrial
`fibrillation or other conditions such as
`arrhythmias. These may include the use of
`decision tree learning such as with a random
`forest, association rule learning, artificial
`neural network, inductive logic programming,
`support vector machines, clustering,
`Bayesian networks, reinforcement learning,
`representation learning, similarity and metric
`learning, sparse dictionary learning, or the
`like.”
`
`APPLE-1001 (’731 IPR), 9:67-10:9 (cited in ’731 Reply, 19)
`
`Apple’s ’731 Reply
`“The types of learning listed in the patent
`were all generic categories known in the
`art—the claimed ‘machine learning
`algorithm’ is nothing more than generic
`functional language that adds no inventive
`concept.”
`
`’731 Reply, 19
`
`Efimov Testimony
`Q: And those were all machine learning
`algorithms or structures that were already
`known by December of 2013; correct?
`
`…A
`
`: Well, based on this text, which I couldn’t
`recall, yes, they’d already been known.
`APPLE-1069, 170:9-14 (cited in ’731 Reply, 20)
`
`DEMONSTRATIVE EXHIBIT - NOT EVIDENCE
`
`55
`
`55
`
`
`
`Machine Learning (’499, ’731 IPRs)
`
`3) Machine learning was known to offer significant benefits
`
`Sajda 2006
`“Machine learning offers a principled approach for
`developing sophisticated, automatic, and
`objective algorithms for analysis of high-
`dimensional and multimodal biomedical data….
`recent developments in machine learning … have
`made significant impacts in the detection and
`diagnosis of disease in biomedicine.”
`APPLE-1042, Abstract (cited in ’731 Petition, 68; ’731 Reply, 20;
`and in APPLE-1003 (‘’731 IPR), ¶260))
`
`Yaghouby 2009
`“BSVM is a classification algorithm based on
`SVM which is able to solve the multi-class
`classification problems. here, five types of the
`most life threatening cardiac arrhythmias … can be
`discriminated by BSVM and selected features with
`the average accuracy of 99.78%.”
`
`Efimov Testimony
`“there are two uses of machine learning, I
`would say, in medicine in terms of benefits.
`One is to assist a physician or cardiologist
`or radiologist in looking through a lot of
`data… And another one is to basically work
`in real life, in the world, on the wearable
`device, for example, and basically alert the
`user or potential patient that, let’s say, an
`arrhythmia was detected”
`
`APPLE-1069, 185:17-24 (cited in ’731 Reply, 20)
`
`Dallali 2011
`“The classification accuracy of the ANNs
`introduced classifier up to 90.5% was achieved,
`and a 99.5% of sensitivity.”
`
`APPLE-1040, 1928 (cited in ’731 Petition, 67; ’731 Reply, 20;
`and in APPLE-1003 (’731 IPR), ¶¶117, 259)
`DE