`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`____________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`____________________
`
`
`APPLE INC.
`Petitioner,
`
`v.
`
`OMNI MEDSCI, INC.,
`Patent Owner.
`
`Patent No. 10,188,299
`____________________
`
`Case IPR2020-00175
`__________________________________________________________________
`
`Petitioner’s Reply
`
`
`
`
`
`
`
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`Table of Contents
`
`I.
`II.
`
`B.
`
`Introduction ................................................................................................ 1
`The Term “Increas[ing] Signal-to-Noise Ratio… by Increasing a Pulse
`Rate of at Least One [LED]” Reflects the Common Scientific Knowledge
`that Increasing Pulse Rate Typically Increases SNR ................................ 3
`III. Argument .................................................................................................... 6
`A.
`Lisogurski Alone Discloses a Device Configured to Increase SNR
`by Increasing the Pulse Rate of an LED .......................................... 6
`Lisogurski and Carlson Teach a System “Configured to Increase
`Signal-to-Noise Ratio by... Increasing a Pulse Rate” of an LED .... 12
`1.
`Lisogurski Teaches Increasing LED Firing Rate and that Light
`Drive Parameters Can Be Changed In Response to Increased
`Noise ...................................................................................... 12
`Carlson Teaches that Increasing LED Pulse Rate Can Increase
`SNR ........................................................................................ 14
`The Combined Teachings of Lisogurski and Carlson Teach a
`Device that Increases Pulse Rate for the Purpose of Increasing
`SNR ........................................................................................ 19
`C. Dependent Claims ........................................................................... 25
`IV. Conclusion ................................................................................................. 25
`
`
`
`2.
`
`3.
`
`
`
`
`ii
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`TABLE OF AUTHORITIES
`
` Page(s)
`
`Cases
`Apple Inc. v. Omni MedSci, Inc.,
`IPR2019-00916, Paper 23 (Oct. 31, 2020) ................................................ passim
`Apple Inc. v. Omni MedSci, Inc.,
`IPR2019-00916, Paper 39 (Oct. 14, 2020) ................................................ passim
`
`Broadcom Corp. v. Emulex Corp.,
`732 F.3d 1325 (Fed. Cir. 2013) ........................................................................ 11
`In re Keller
`642 F.2d 413 (Fed. Cir. 1981) .......................................................................... 19
`MCM Portfolio LLC v. Hewlett-Packard Co.,
`812 F.3d 1284 (Fed. Cir. 2015) .................................................................. 19, 24
`In re Merck & Co., Inc.,
`800 F.2d 1091 (Fed. Cir. 1986) .................................................................. 19, 21
`In re NuVasive, Inc.,
`841 F.3d 966 (Fed. Cir. 2016) .......................................................................... 18
`ParkerVision, Inc. v. Qualcomm Inc.,
`903 F.3d 1354 (Fed. Cir. 2018) .................................................................. 10, 11
`
`
`
`
`
`
`iii
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`I.
`
`Introduction
`The Board should find, as it did in IPR2019-00916 involving a parent of the
`
`’299 patent, that Lisogurski and Carlson make obvious devices that “increase the
`
`signal-to-noise ratio… by increasing the pulse rate” of a light emitting diode
`
`(LED). That is only aspect of the claimed devices that Omni MedSci (“Omni”)
`
`contends makes them nonobvious, but the reasoning it advances was thoroughly
`
`rejected by Board in the -916 proceeding based on a nearly identical record of
`
`evidence. The Board should reach the same conclusion in this proceeding.
`
`First, Omni contends the claim language “configured to increase signal-to-
`
`noise ratio” imposes special requirements for how the claimed devices increase
`
`signal-to-noise ratio (“SNR”) by increasing the LED pulse rate. But as the Board
`
`held in the -916 proceeding, it does not—all that claims require is that the LEDs be
`
`“capable of” having their pulse rate increased to increase SNR. See IPR2019-
`
`00916, Paper 39 (“-916 FWD”), 11-12 (Oct. 14, 2020).
`
`Next, Omni contends that Lisogurski’s “cardiac cycle modulation” technique
`
`does not increase SNR by increasing the pulse rate of an LED in its device. But
`
`the Board also rejected that assertion. It initially observed that Omni had admitted
`
`that Lisogurski’s cardiac cycle modulation increases SNR by modulating the pulse
`
`frequency to correlate with the cardiac cycle. -916 FWD, 28-29; IPR2019-00916,
`
`Paper 23 (“-916 Resp.”), 13, 15); see Prelim. Resp., 16 (admitting the same here).
`
`
`
`1
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`The Board also found that cardiac cycle modulation will at least some of the time
`
`increase SNR by increasing the LED pulse frequency. -916 FWD, 29-30. That
`
`aligns with the largely undisputed disclosure of Lisogurski: a device that (i) can
`
`change an LED’s light drive cycle parameters, including the LED firing rate,
`
`(Ex.1011, 25:53-55, 27:44-49, 35:24-31; Ex.1065, 59:1-5), (ii) can change those
`
`parameters in response to noise, (Resp., 15; Ex.1011, 1:67-2:3, 5:55-61, 9:46-60,
`
`27:44-49, 37:6-18; Ex.2131, ¶¶74-77), and (iii) it can change those parameters for
`
`the purpose of increasing SNR, (Prelim. Resp., 16; -916 Resp., 15; Ex.1011, 25:53-
`
`55, 27:44-49, 35:24-31). Thus, as the Board correctly found, Lisogurski’s device
`
`is capable of increasing the SNR by increasing its LED firing rate. -916 FWD, 30.
`
`That establishes that Lisogurski teaches this sole disputed element of the ’299
`
`claims as well.
`
`The Board also credited Petitioner’s explanation that Carlson makes it
`
`obvious to increase an LED pulse rate in Lisogurski’s device to increase SNR. -
`
`916 FWD, 32-34. As the Board observed, a skilled person would have considered
`
`Lisogurski and Carlson together because both describe analogous optical sensor
`
`devices and their use in performing physiological measurements. Id., 24-25. The
`
`Board also credited Petitioner’s explanation that Carlson specifically teaches the
`
`technique Omni contends is not expressly taught by Lisogurski—increasing the
`
`LED pulse rate to dynamically offset noise from ambient light, which, as Omni’s
`
`
`
`2
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`expert Dr. MacFarlane admitted, generally increases SNR. Ex.1065, 37:17-22; -
`
`916 FWD, 32-33. The record of evidence here and the Board’s findings in the -
`
`916 proceeding based on that same evidence establish that a skilled person would
`
`have found it obvious to configure Lisogurski to increase SNR by increasing its
`
`pulse rate based on Lisogurski and Carlson. And because Omni has identified no
`
`credible basis for the Board to deviate from its prior conclusion, the Board should
`
`find the challenged ’299 claims unpatentable.
`
`II. The Term “Increas[ing] Signal-to-Noise Ratio… by Increasing a Pulse
`Rate of at Least One [LED]” Reflects the Common Scientific Knowledge
`that Increasing Pulse Rate Typically Increases SNR
`Omni anchors its arguments on the requirement in independent claim 7 for a
`
`“system configured to increase signal-to-noise ratio… by increasing a pulse rate
`
`of at least one of the plurality of [LEDs].” But the specification provides scant
`
`support for this limitation. For example, the specification provides no explanation
`
`how a skilled person should or could increase the LED pulse rate to increase SNR,
`
`nor does it suggest the optical sensor must be specially configured to cause a
`
`particular type of increase in pulse rate in order to increase SNR. Instead, the
`
`specification relies entirely on the knowledge already possessed by a skilled person
`
`for this supposedly critical limitation. It is thus informative to look at the science
`
`behind variables affecting SNR.
`
`
`
`3
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`Pulsing a signal is a standard technique used to enhance a signal’s
`
`detectability in the presence of noise, such as ambient light. Ex.1003, ¶47. In an
`
`optical sensor, an LED is pulsed, and each time it is, a detector measures the
`
`amount of light reflected back from the sample and determines how the sample is
`
`changing (e.g., how the volume of blood in tissue is changing over time). Ex.1003,
`
`¶¶41, 43-44. It was well-known that, in the presence of noise, increasing the rate at
`
`which an LED pulses (and the sampling rate, which is the rate at which the signal
`
`is measured) will generally increase the SNR. At his deposition, Omni’s expert
`
`Dr. MacFarlane admitted this was a well-known scientific fact:
`
`Q… Why is it that changing the pulse rate of an LED would
`change the signal-to-noise ratio?
`A. There are a number of reasons… why that might happen.
`Generally speaking, the faster the modulation, the faster the pulse
`rate, the lower the background noise.
`That's a general statement that describes something that -- that's
`a general statement of -- of truth. There are… counterexamples, but
`generally speaking…, as you have a faster or an increased pulse
`rate, you see a lower noise environment.
`
`Ex.1065, 37:13-38:3.1 Dr. MacFarlane also testified the increasing the pulse rate
`
`of an LED would typically increase SNR:
`
`
`
`1 All emphases added, unless otherwise noted.
`
`
`
`4
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`Q. Will you agree that in general, when you increase the pulse
`rate of an LED, you will increase the signal-to-noise ratio, though that
`won't always happen?
`* * *
`THE WITNESS: Yes.
`
`Id., 39:12-17.
`
`In its sur-reply, Omni may try to deny this scientific truth. For example,
`
`after his cross-examination was complete, Omni’s counsel spoke to Dr.
`
`MacFarlane about his answers. Ex.1065, 85:9-21. On redirect, Dr. MacFarlane
`
`then tried to change his answer to the second question (the question at 39:12-17,
`
`quoted in the previous paragraph), testifying that “I'd like to change my answer
`
`from a yes to a no.” Id., 81:17-18. That change in answer, made after discussions
`
`with counsel, lacks credibility. Moreover, Dr. MacFarlane’s original answer was
`
`consistent with his prior testimony, where he stated “[g]enerally speaking, the
`
`faster the modulation, the faster the pulse rate, the lower the background noise…
`
`[T]hat’s a general statement… of truth.” Id., 37:17-22. Dr. MacFarlane made no
`
`attempt square his changed answer with his prior testimony. And when asked on
`
`re-cross to identify scenarios when increasing pulse rate did not increase SNR, he
`
`could only postulate there might be hypothetical scenarios he could imagine, but
`
`did not identify any actual ones that would. Ex.1065, 83:5-10. Dr. MacFarlane’s
`
`testimony thus shows that a skilled person would generally expect that increasing
`
`
`
`5
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`the pulse rate of an LED will increase SNR, and that only in unusual circumstances
`
`would it not. Thus, as the Board observed in the -916 proceeding, the critical point
`
`is that increasing an LED’s pulse rate will increase SNR at least some of the time.
`
`-916 FWD, 29-30.
`
`III. Argument
`A. Lisogurski Alone Discloses a Device Configured to Increase SNR
`by Increasing the Pulse Rate of an LED
`Lisogurski alone discloses a device that increases the firing (pulse) rate of at
`
`least one of its LEDs and thereby meets the limitation of a “system configured to
`
`increase signal-to-noise ratio… by increasing a pulse rate of at least one [LED].”
`
`Based on the same record of evidence, the Board previously found that
`
`Lisogurski’s cardiac cycle modulation meets this element, (-916 FWD, 29-30), and
`
`Omni has presented no reason why the Board should reach a different conclusion
`
`here.
`
`Initially, it is undisputed that Lisogurksi alone discloses a device configured
`
`to increase the firing rate (“pulse rate”) of at least one of its LEDs. See Ex.1011,
`
`25:52-55. As Lisogurksi states:
`
`[T]he sampling rate may represent the amount of time between “on”
`periods. For example…, decreasing the duration of the “off” periods
`(i.e., increasing the emitter firing rate) relates to an increased
`sampling rate.
`
`
`
`6
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`Ex.1011, 35:24-31. Omni admits this, stating that “Lisogurski discloses a pulse
`
`oximeter having an adjustable ‘firing rate.’” Resp., 22; see id., 2. Even Dr.
`
`MacFarlane admitted that Lisogurski discloses this:
`
`Q… Do you agree that Lisogurski describes a device that is
`configured to increase the emitter firing rate in some circumstances?
`* * *
`THE WITNESS: I believe so.
`
`Ex.1065, 59:1-5. When Lisogurski increases the LED firing rate, this typically
`
`will increase the SNR. Omni’s expert Dr. MacFarlane admitted that “[g]enerally
`
`speaking, the faster the modulation, the faster the pulse rate, the lower the
`
`background noise.” Ex.1065, 37:17-22. Thus, Lisogurski discloses that its device
`
`will increase SNR by increasing the pulse rate of an LED, other than in unusual
`
`circumstances.
`
`As Omni has admitted and as the Board previously found, Lisogurski also
`
`discloses increasing the LED firing rate for the purpose of increasing SNR. -916
`
`FWD, 29-30. Lisogurski describes using cardiac cycle modulation to vary how the
`
`LEDs are illuminated, and cardiac cycle modulation is a technique that is intended
`
`to increase SNR. Ex.1011, 25:49-55. And Omni admitted that “Lisogurski teaches
`
`[] different techniques for improving SNR…, [including] by modulating the light
`
`signal to correlate with ‘physiological pulses’ such as a ‘cardiac pulse,’ e.g.,… ‘[]
`
`cardiac cycle modulation.’” Prelim Resp., 16; -916 Resp., 15.
`
`
`
`7
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`Lisogurski also explains that cardiac cycle modulation can increase the LED
`
`firing rate (and the sampling rate) to become or remain synchronous with a cardiac
`
`cycle. Ex.1011, 25:49-55; 31:11-24, 31:39-55. When Lisogurski’s device applies
`
`that modulation technique, depending on the individual’s cardiac cycle, the device
`
`will increase its LED pulse rate. Ex.1003, ¶136 (“Cardiac cycle modulation
`
`schemes would adjust the brightness of a light in a way that is synchronized to the
`
`cardiac cycle and similar to or faster than the cardiac pulse rate.”). Consistent with
`
`Dr. Anthony’s explanation, and as the Board found in the -916 proceeding, this
`
`could occur when a person’s heart rate increases, which would result in
`
`Lisogurski’s device increasing its LED’s firing rate to remain synchronous with the
`
`faster cardiac cycle. -916 FWD, 28-29. Omni has also admitted this, stating
`
`“[v]arying the LED firing rate to remain synchronous with… the heart rate will
`
`both decrease and increase the firing rate.” Resp., 13; see id., 16. This increase in
`
`the firing rate can increase SNR by reducing noise, and such noise can represent
`
`1%-4% of the computed PPG signal. Ex.1011, 42:50-54; see id., 25:66-26:14
`
`(modulating LED drive signal to correlate to respiratory cycle can increase SNR).
`
`Lisogurski explains that “increasing the sampling rate for a portion of the
`
`cardiac cycle may result in more accurate and reliable physiological
`
`information.” Ex.1011, 33:46-52. Although Lisogurski does not use the term
`
`“SNR” in this passage, a skilled person would have understood that the
`
`
`
`8
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`physiological information is “more accurate and reliable” because its SNR is
`
`higher. As Dr. Anthony explains:
`
`Lisogurski states that increasing the sampling rate ‘may result in more
`accurate and reliable physiological information.’ Ex.1011
`(Lisogurski), 33:56-58. Therefore, Lisogurski explains that the LED
`firing rate can be increased to increase signal-to-noise ratio.
`
`Ex.1003, ¶156.
`
`Thus, the skilled person considering Lisogurski’s cardiac cycle modulation
`
`would have understood that when Lisogurki’s device increases the LED firing rate
`
`in this manner, it would increase the signal relative to the noise (e.g., ambient
`
`light) in the measurement, and thus would increase SNR. See Ex.1065, 37:17-22,
`
`39:12-17.
`
`In its Response, Omni argues that Lisogurski does not meet this limitation
`
`because cardiac cycle modulation varies the firing rate for the purpose of
`
`remaining synchronous with cardiac cycle and not with the intent to increase SNR.
`
`Resp., 11-13. But Omni has already admitted that one of the purposes of cardiac
`
`cycle modulation is to increase SNR. Prelim. Resp., 13; -916 Resp., 15.
`
`Moreover, Omni’s assertions about why Lisogurski increases LED firing rate are
`
`irrelevant. The claims cover a device that increases SNR by increasing the pulse
`
`rate of an LED, even if the device does not intend to increase SNR by doing so.
`
`The Federal Circuit “explained long ago that ‘apparatus claims cover what a device
`
`
`
`9
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`is, not what a device does.’” ParkerVision, Inc. v. Qualcomm Inc., 903 F.3d 1354,
`
`1361 (Fed. Cir. 2018); see Application of Michlin, 256 F.2d 317, 320 (C.C.P.A.
`
`1958). A device that performs a specified action and causes the result specified by
`
`the claim satisfies the claim requirements, irrespective of whether the device
`
`intended to do so or not. See ParkerVision, 903 F.3d at 1361 (“[A] prior art
`
`reference may anticipate or render obvious an apparatus claim... if the reference
`
`discloses an apparatus that is reasonably capable of operating so as to meet the
`
`claim limitations, even if it does not meet the claim limitations in all modes of
`
`operation.”).
`
`Omni also argues that Lisogurski does not meet this limitation because even
`
`though cardiac cycle modulation sometimes increases SNR by increasing the LED
`
`firing rate, it sometimes increases SNR by decreasing the firing rate. Resp., 11-13.
`
`But all that matters is that Lisogurski sometimes increases SNR by increasing LED
`
`firing rate, which Omni admits that it does. See Resp., 13 (“[Lisogurksi] merely
`
`teaches that heart rate tracking may improve SNR (whether the firing rate
`
`increases, decreases, or stays the same).”). That Lisogurski also sometimes
`
`increases SNR by decreasing LED firing rate (e.g., when heart rate decreases) is
`
`irrelevant. See -916 FWD at 30 (“Thus, a light source is ‘configured’ to increase
`
`signal-to-noise by increasing LED pulse rate when it is ‘capable’ of doing so, i.e.,
`
`when the increased pulse rate allows the device to operate in a frequency range
`
`
`
`10
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`having less environmental noise. This is true, even if the device does not always
`
`do so.”). “It is well settled that an [apparatus] that ‘that sometimes, but not always,
`
`embodies a claim[] nonetheless’” satisfies the claim. Broadcom Corp. v. Emulex
`
`Corp., 732 F.3d 1325, 1333 (Fed. Cir. 2013); see ParkerVision, 903 F.3d at 1361.
`
`Omni also argues that increasing LED firing rate to remain synchronous
`
`with an increased heart rate (a human’s heart rate can vary from 0.5 to 3 Hz) would
`
`not necessarily reduce interference from ambient light or environmental noise
`
`(which typically is from 0.5 to 10 Hz). Resp., 13-15. Again, that is irrelevant to
`
`the claims, which do not require increasing SNR every time pulse rate is increased.
`
`As the Board recognized in the -916 proceeding, “whether [SNR] increases with
`
`increasing pulse rate depends on an external factor—the noise spectrum in the
`
`environment within which the device operates. But the claim is directed to the
`
`device itself, regardless of the environment in which it operates.” -916 FWD, 30.
`
`Omni’s argument also is inconsistent with Lisogurski’s teachings, which provide
`
`that cardiac cycle modulation increases SNR in the presence of Gaussian noise of
`
`0-5 Hz. Ex.1011, 41:46-52.
`
`Omni also points out that Lisogurski discloses other techniques for
`
`increasing SNR, but Lisogurski’s disclosure of multiple techniques for increasing
`
`SNR is irrelevant, given that it is undisputed that cardiac cycle modulation
`
`
`
`11
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`increases SNR too. Resp., 17-18. Therefore, Lisogurski alone meets this claim
`
`limitation.
`
`B.
`
`Lisogurski and Carlson Teach a System “Configured to Increase
`Signal-to-Noise Ratio by... Increasing a Pulse Rate” of an LED
`Even if the Board were to find that Lisogurski alone does not disclose this
`
`claim element, the combination of Lisogurski and Carlson would render a device
`
`including that element obvious. Omni’s primary argument for why Lisogurski and
`
`Carlson do not do so together is that neither Lisogurski alone nor Carlson alone
`
`teaches a device that, while in operation, actively increases the pulse rate of an
`
`LED for the purpose of increasing SNR. Omni’s arguments rely on
`
`mischaracterizations of what Lisogurski and Carlson each teach and a legally
`
`erroneous formulation of obviousness.
`
`The Board considered this same question, the same body of evidence, and
`
`essentially the same arguments in the -916 proceeding, and found that Lisogurski
`
`and Carlson together taught this claim element. -916 FWD, 32-35. The Board
`
`should do the same here.
`
`1.
`
`Lisogurski Teaches Increasing LED Firing Rate and that
`Light Drive Parameters Can Be Changed In Response to
`Increased Noise
`Lisogurski teaches that its device can detect changes in background noise
`
`and ambient light, and that in response, the device can alter the light drive
`
`parameters or modulation techniques used.
`
`
`
`12
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`Lisogurski explains that “the system may alter the cardiac cycle modulation
`
`technique based on the level of noise, ambient light, or other suitable reasons.”
`
`Ex.1011, 9:46-48. One way the system can alter the cardiac cycle modulation
`
`technique is by switching between a first and a second mode of operation.
`
`Ex.1011, 36:48-53, 37:6-8. For example, Lisogurski describes switching between
`
`modes in response to background noise: “the system may detect a condition where
`
`a second mode is required… [f]or example, the system may detect a change in
`
`background noise, a change in ambient light, a change in the available power, other
`
`suitable changes.” Ex.1011, 37:8-11. Lisogurski explains that when it changes the
`
`cardiac cycle modulation technique in response to noise, that change can be done
`
`“to improve the signal-to-noise ratio.” Ex.1011, 9:50-52; see Prelim. Resp., 16
`
`(admitting cardiac cycle modulation increases SNR); -916 Resp., 15.
`
`When altering cardiac cycle modulation, Lisogurski can change various light
`
`drive parameters, including: “light brightness, duty cycle, firing rate,” and other
`
`modulation parameters. Ex.1011, 25:52-55; see id., 1:67-2:3, 27:46-49. Thus, one
`
`of the parameters than can be adjusted is the firing rate (“pulse rate”) of an LED.
`
`Lisogurski describes examples where the LED firing rate (along with the detector
`
`sampling rate) can be increased. Ex.1011, 35:29-31. Lisogurski explains that
`
`while “cardiac cycle modulation techniques [are] generally related to the cardiac
`
`cycle, [they] may not necessarily be precisely correlated to the cardiac cycle and
`
`
`
`13
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`may be related to external triggers (e.g., respiration), user input, [or] other suitable
`
`techniques.” Ex.1011, 5:41-47.
`
`Thus, Lisogurski teaches a device that is configured to (i) detect changes in
`
`the noise level, (ii) vary cardiac cycle modulation (e.g., by changing the light drive
`
`parameters) in response to changes in the noise, (iii) vary cardiac cycle modulation
`
`for the purpose of increasing SNR, and (iv) increase LED firing rate (which is one
`
`of the light drive parameters) in some circumstances. Based on the evidence, the
`
`Board found Lisogurski taught these same elements in the -916 proceeding. -916
`
`FWD, 28, 33.
`
`As in the -916 proceeding, Omni has argued that Lisogurski does not
`
`explicitly disclose increasing LED firing rate for the purpose of increasing SNR.
`
`Even if this distinction were accepted, configuring Lisogurski to do so would have
`
`been obvious based on Carlson.
`
`2.
`
`Carlson Teaches that Increasing LED Pulse Rate Can
`Increase SNR
`Omni devotes a substantial portion of its brief to mischaracterizing Carlson’s
`
`teachings. Resp., 19-23, 26-31. According to Omni, Carlson shows a pulse
`
`oximeter where a person makes design choice as to the pulse rate of an LED when
`
`creating the oximeter, and that Carlson does not describe a device that can actively
`
`increase the LED’s pulse rate during operation. Resp., 26-31. Omni’s
`
`interpretation of Carlson is factually incorrect.
`
`
`
`14
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`As explained in the Petition, Carlson teaches techniques for increasing the
`
`SNR of a pulsoximeter to improve its performance in the presence of noise, such as
`
`ambient light from sunlight or fluorescent lights. Pet., 21-22; see Ex.1009, [0002]
`
`(Carlson’s goal is increasing pulsoximetry performance “in terms of quality and
`
`robustness of the measurement signal versus environmental disturbances and
`
`energy consumption”), [0067]-[0068] (describing indoor and outdoor lighting).
`
`Carlson also explains that, whether used indoors or outdoors, the amount of
`
`ambient light varies over time. Ex.1009, [0007] (“environmental optical radiation
`
`strongly varies as a function o[f] time and place where the pulsoximeter is used,
`
`e.g. day versus night, indoor versus outdoor”), [0067]-[0068]. Carlson, thus,
`
`makes clear that its device is designed to address changing environmental
`
`conditions (e.g., where the amount of ambient light is continually varying).
`
`To handle interference from ambient light when it is present, Carlson
`
`explains that its device “temporarily modulate[s] the amplitude of the optical
`
`radiation of, e.g., the LED at a carrier frequency fc in order to shift the power
`
`spectrum of the pulsoximeter signals into a higher frequency range where
`
`environmental optical radiation is unlikely.” Ex.1009, [0020]; see id., [0065]. In
`
`this passage, Carlson describes shifting the frequency of an LED’s emission “to a
`
`higher frequency range,” thus indicating that the LED previously emitted pulses at
`
`a lower frequency—i.e., that Carlson is switching between at least two different
`
`
`
`15
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`frequencies. Ex.1009, [0020]. Carlson also states that its device temporarily
`
`makes this adjustment, id., [0020], which means the device will change how its
`
`LEDs pulse based on the presence and characteristics of ambient light at any
`
`particular moment in time, id., [0068]. Based on his reading of Carlson, Dr.
`
`Anthony explains that “Carlson teaches that increasing the modulation frequency
`
`of the pulsed LEDs improves the signal-to-noise ratio.” Ex.1003, ¶161; see
`
`Ex.1009, [0069].
`
`Omni incorrectly portrays Carlson as disclosing a device that only emits
`
`light continuously or constantly (i.e., not pulsed) but in the presence of ambient
`
`light can switch to pulsing an LED at a single, fixed pulse rate. Resp., 19, 27-28;
`
`id., 21. That interpretation makes no sense in the context of Carlson’s invention.
`
`For example, under Omni’s reading, Carlson’s device would consume excessive
`
`battery power, something that is entirely at odds with the stated purpose of its
`
`mobile, battery-powered device: to balance signal quality with energy
`
`consumption. Ex.1009, [0002] (describing “increasing the technical
`
`performance… versus environmental disturbances and energy consumption”),
`
`[0048]; see Ex.1011, 10:23-24, 36:38-47 (teaching use of modulation when
`
`battery-powered to optimize power consumption).
`
`Omni also asserts that Carlson shows only shifting from unmodulated light
`
`to light modulated at a fixed frequency, and that Carlson does not teach shifting the
`
`
`
`16
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`frequency (pulse rate) of the light pulses. Resp., 27-28. Omni’s argument is
`
`inconsistent with Carlson’s claims, which specify a device that includes “a light
`
`source amplitude modulating means to modulate the frequency of the emitted
`
`light” and that the means further can “shift the frequency of the emitted light.”
`
`Ex.1009, claims 10-13. Thus, contrary to Omni’s arguments, Carlson discloses
`
`shifting the frequency of the pulses between two different pulse frequencies. Read
`
`correctly, Carlson describes switching from a lower frequency pulse rate to a
`
`higher one to avoid interference caused by ambient light. Ex.1003, ¶161.
`
`Reading Carlson as teaching switching among different pulse frequencies
`
`rather than from continuous light emission to pulsed light emission also is
`
`consistent with common sense. Carlson describes interference from ambient light
`
`and teaches that such light varies over time depending on where a user is.
`
`Ex.1009, [0007] (“environmental optical radiation strongly varies as a function o[f]
`
`time and place where the pulsoximeter is used, e.g. day versus night, indoor versus
`
`outdoor”). For example, Carlson recognizes that interference may occur at one
`
`frequency when indoors and at others when outdoors. Ex.1009, [0067] (“artificial
`
`light…is going up to approximately 120 Hz”), [0068] (“sunlight can have a
`
`dramatic influence, e.g. if a person is walking through streets with relatively quick
`
`changing conditions between sunlight and shadow. Another serious possibility is
`
`caused by a tree avenue when driving along the trees.”). Because the nature of
`
`
`
`17
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`ambient light changes over time, it would be natural to interpret Carlson as
`
`describing multiple pulse frequencies. See Ex.1065, 34:5-35:7 (Dr. MacFarlane
`
`agreeing that “Because sunlight variations and the weather are constantly changing
`
`environmental conditions, it would be impractical, as a matter of common sense, to
`
`have the user manually reconfigure the LED pulse rate as conditions change”).
`
`Omni’s interpretation is at odds with that understanding.
`
`Omni also incorrectly argues that, in the Institution Decision, the Board
`
`relied on an argument that Petitioner did not make when the Board found that
`
`Carlson’s disclosure of switching from unmodulated to modulated light taught
`
`increasing the pulse rate. Resp., 21-23. The Board relied on the same evidence
`
`and examples from Carlson that were discussed in the Petition and in Omni’s
`
`Preliminary Response. Compare Inst. Dec., 50-52 (discussing Carlson at Figs 7c
`
`& 8, [0064]-[0065], [0067]-[0069]) with Pet., 50-51 (discussing Carlson at Figs 7c
`
`& 8, [0067]-[0069]) and Prelim. Resp., 18-21 (discussing Carlson at Figs 7c & 8,
`
`[0064]-[0065], [0067], [0069]). It is of course permissible for the Board to
`
`consider and rely upon the portions of the references cited in the papers. It also is
`
`proper for the Board to rely on portions of the references cited in the reply and sur-
`
`reply briefs. In re NuVasive, Inc., 841 F.3d 966, 972 (Fed. Cir. 2016).
`
`
`
`18
`
`
`
`IPR2020-00175
`
`
`
`Petitioner’s Reply
`
`3.
`
`The Combined Teachings of Lisogurski and Carlson Teach
`a Device that Increases Pulse Rate for the Purpose of
`Increasing SNR
`Omni’s primary argument for why Lisogurski and Carlson do not render this
`
`limitation obvious rests on a legally erroneous foundation. Omni asserts that the
`
`art does not teach “increas[ing] signal-to-noise ratio by… increasing a pulse rate
`
`of” an LED because neither Lisogurski alone nor Carlson alone teaches a device
`
`that actively increases an LED’s pulse rate for the purpose of increasing SNR
`
`while the device is in operation. Omni ignores that “[n]on-obviousness cannot be
`
`established by attacking references individually where the rejection is based upon
`
`the teachings of a combination of references.” In re Merck & Co., Inc., 800 F.2d
`
`1091, 1097 (Fed. Cir. 1986); see In re Keller, 642 F.2d 413, 425 (Fed. Cir. 1981)
`
`(the test for obviousness is “what the combined teachings of the references would
`
`have suggested to those of ordinary skill in the art”); MCM Portfolio LLC v.
`
`Hewlett-Packard Co., 812 F.3d 1284, 1294 (Fed. Cir. 2015) (“Moreover, [t]he test
`
`for obviousness is not whether the features of a secondary reference may be bodily
`
`incorporated into the structure of the primary reference….”); see -916 FWD, 35
`