`
` UNITED STATES PATENT AND TRADEMARK OFFICE
`________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`________________
`
`FITBIT, INC.,
`Petitioner,
`
`v.
`
`VALENCELL, INC.,
`Patent Owner
`________________
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`________________
`
`
`
`
`
`PETITIONER’S OPENING BRIEF ON REMAND
`
`06938-00001/12352983.5
`
`
`
`TABLE OF CONTENTS
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`Page
`
`I.
`
`II.
`
`Introduction ...................................................................................................... 1
`
`Technical Background ..................................................................................... 1
`
`III.
`
`Procedural Background ................................................................................... 2
`
`IV. Claims 3-5 Are Obvious From Luo, Craw and Wolf ...................................... 4
`
`A.
`
`B.
`
`C.
`
`Claim 1 Has Already Been Found Invalid Based on Luo and
`Craw ...................................................................................................... 4
`
`Claim 3 Is Rendered Obvious by Luo and Craw .................................. 5
`
`Claims 4-5 Are Rendered Obvious by Luo, Craw, and Wolf ............... 8
`
`V.
`
`Claim 3 Is Obvious in View of Mault, Al-Ali, and Lee .................................. 9
`
`VI. Conclusion ..................................................................................................... 10
`
`06938-00001/12352983.5
`
`i
`
`
`
`TABLE OF AUTHORITIES
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`Page
`
`CASES
`
`Accenture Global Servs., GmbH v. Guidewire Software, Inc.,
`728 F.3d 1336 (Fed Cir. 2013) ........................................................................ 4
`
`Fitbit, Inc. v. Valencell, Inc.,
`964 F.3d 1112 (Fed. Cir. 2020) ............................................................ 3, 6, 10
`
`MaxLinear, Inc. v. CF CRESPE LLC,
`880 F.3d 1373 (Fed. Cir. 2018) ....................................................................... 5
`
`SAS Inst., Inc. v. Iancu,
`138 S. Ct. 1348 (2018) ..................................................................................... 2
`
`
`
`
`
`06938-00001/12352983.5
`
`i
`
`
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`
`I.
`
`Introduction
`
`This IPR is the last one remaining from a series of IPRs filed by Apple Inc.
`
`(“Apple”) and joined by Petitioner Fitbit, Inc. (“Fitbit”). In every other IPR, this
`
`Board found the claims of Patent Owner Valencell, Inc. (“Valencell”) unpatentable,
`
`and the Federal Circuit has since affirmed the Board’s decisions. Even in this IPR,
`
`the Board issued a final written decision finding claims 1-2 and 6-13 of U.S. Patent
`
`No. 8,923,941 (“’941 patent”) unpatentable. Valencell did not appeal that decision.
`
`Dependent claims 3-5 add little to the subject matter already found obvious
`
`and invalid in claim 1. Indeed, Valencell never previously argued these claims
`
`were patentable for any reason other than their dependency on claim 1, and the
`
`Board concluded that Fitbit’s prior art discloses the limitations recited in claim 3 in
`
`a separate IPR on claims 14-21 of the ’941 patent. In view of the petitioned
`
`grounds, the Board should find claims 3-5 invalid.
`
`II. Technical Background
`
`The ’941 patent is directed to processing signals from a PPG sensor and a
`
`motion sensor into serial data output containing physiological and motion
`
`information. ’941 patent at Abstract. This physiological information may include
`
`heart rate, and the motion information may include current running speed. Id. at
`
`23:49-54, 23:63-66. Claim 1 of the ’941 patent, which has already been
`
`invalidated, is directed to this core concept. Dependent claims 3-5 recite further
`
`06938-00001/12352983.5
`
`1
`
`
`
`
`limitations related to the organization and use of the serial data output:
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`3. The method of claim 1, wherein the serial data output is parsed out such
`
`that an application-specific interface (API) can utilize the physiological
`
`information and motion-related information for an application.
`
`4. The method of claim [3], wherein the application is configured to generate
`
`statistical relationships between subject physiological parameters and subject
`
`physical activity parameters in the physiological information and motion-
`
`related information.
`
`5. The method of claim 4, wherein the application is configured to generate
`
`statistical relationships between subject physiological parameters and subject
`
`physical activity parameters via at least one of the following: principal
`
`component analysis, multiple linear regression, machine learning, and
`
`Bland-Altman plots.
`
`III. Procedural Background
`
`Apple Inc. (“Apple”) challenged claims 1-13 of the ’941 patent in its
`
`original IPR petition. Paper 2. The Board instituted this IPR on claims 1-2 and 6-
`
`13 after disagreeing with Apple’s proposed construction for “application-specific
`
`interface (API)” in claim 3 and its attempt to resolve the dependency of claims 4-5.
`
`Paper 10. Fitbit later joined the instituted IPR. IPR2017-01555, Paper 9.
`
`After oral arguments were already held for this IPR, the Supreme Court
`
`issued its decision in SAS Inst., Inc. v. Iancu, 138 S. Ct. 1348, 1358 (2018). In
`
`response, the Board added dependent claims 3-5 to the IPR and authorized
`
`additional briefing on this subject. Paper 39. Petitioners explained in this briefing
`
`06938-00001/12352983.5
`
`2
`
`
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`
`that claims 3-5 were invalid under both their proposed constructions and the
`
`Board’s constructions. Paper 40. In response, Valencell did not challenge the
`
`merits of petitioners’ invalidity arguments for the limitations added in claims 3-5.
`
`Paper 41. In its Final Written Decision, the Board upheld the patentability of only
`
`claims 3-5. Paper 43 at 4. The Board retained its prior constructions of these
`
`claims and “d[id] not address these claims further in this Decision.” Id. at 11-23.
`
`Fitbit appealed the Board’s patentability finding on claims 3-5. The Federal
`
`Circuit affirmed in part, vacated in part, and remanded to the Board. Fitbit, Inc. v.
`
`Valencell, Inc., 964 F.3d 1112 (Fed. Cir. 2020). On claim 3, the Court affirmed the
`
`Board’s construction of “‘application-specific interface (API)’ to mean ‘an
`
`interface which enables a particular application to utilize data obtained from
`
`hardware, such as the at least one motion sensor and the at least one PPG sensor.’”
`
`Id. at 1116. The Court observed, however, that “[t]he Board’s narrowing
`
`construction may have no significance, where, as here, the claimed ‘application-
`
`specific interface’ performs the same function as an application programming
`
`interface, i.e., ‘enabl[ing] a particular application to utilize data obtained from
`
`hardware.’” Id. at 1117 (emphasis added). The Court vacated and “remand[ed] for
`
`determination of patentability in light of the cited references.” Id. at 1118-19.
`
`On claims 4-5, the Court found that “the error that claim 4 depend from
`
`claim 3” should have been corrected, and remanded for the Board to determine the
`
`06938-00001/12352983.5
`
`3
`
`
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`
`patentability of corrected claims 4 and 5. Id. at 1120.
`
`IV. Claims 3-5 Are Obvious From Luo, Craw and Wolf
`
`The Board held independent claim 1 invalid in light of U.S. Pat. Appl. Pub.
`
`No. 2008/0200774 (“Luo”) (Ex. 1055) and U.S. Pat. Appl. Pub. No. 2008/0133699
`
`(“Craw”) (Ex. 1056). That same prior art combination also invalidates dependent
`
`claim 3 under the Board’s construction of “application-specific interface (API).”
`
`The petition adds U.S. Pat. Appl. Pub. No. 2007/01987881 (“Wolf”) (Ex. 1042) to
`
`demonstrate that dependent claims 4-5 are also invalid.
`
`Valencell never sought to distinguish the subject matter of dependent claims
`
`3-5 from Fitbit’s prior art combinations. Nor did Valencell argue it would be non-
`
`obvious to combine Wolf with Luo and Craw. Conversely, Fitbit’s positions are
`
`consistent with petitioners’ arguments throughout this IPR. Every single paragraph
`
`cited below as disclosing claims 3-5 was also cited in the original petition.
`
`A. Claim 1 Has Already Been Found Invalid Based on Luo and Craw
`
`The Board found Valencell’s independent claim 1 invalid over Fitbit’s
`
`combination of Luo and Craw. See Paper 43 at 43. Because that holding was not
`
`appealed by Valencell, the Board’s invalidation of independent claim 1 is now
`
`“final and conclusive.” See Accenture Global Servs., GmbH v. Guidewire
`
`Software, Inc., 728 F.3d 1336, 1341-42 (Fed Cir. 2013).
`
`Under Federal Circuit precedent, Valencell is bound by the Board’s
`
`06938-00001/12352983.5
`
`4
`
`
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`
`invalidity holding on independent claim 1 and cannot rely on limitations of claim 1
`
`to argue dependent clams 3-5 are patentable over the cited references. Instead,
`
`Valencell must show that dependent claims 3-5 include additional limitations that
`
`materially alter the invalidity analysis. See MaxLinear, Inc. v. CF CRESPE LLC,
`
`880 F.3d 1373, 1377-78 (Fed. Cir. 2018). Valencell cannot meet this burden here.
`
`B. Claim 3 Is Rendered Obvious by Luo and Craw
`
`Claim 3 is rendered obvious by Fitbit’s combination of Luo and Craw. The
`
`Board already found sufficient reason to combine the teachings of Luo and Craw to
`
`achieve claim 1, in view of their similar technology (health monitoring) and
`
`problems addressed (device interoperability). Id. at 34, 42. The same reasons that
`
`support combining Luo and Craw for claim 1 also hold for claim 3. Ex. 1003
`
`(Sarrafzadeh Decl.) ¶¶ 90, 92-93.
`
`The combination of Luo and Craw teaches the additional limitations of claim
`
`3. Luo teaches transmitting health data to health-monitoring devices and
`
`computers, and Craw teaches that the health data is “parsed out” into a serial
`
`output of physiological and motion information consistent with a defined
`
`structure—for example, the structure shown in Fig. 7H. Ex. 1055 (Luo) ¶¶ 0033-
`
`0041; Ex. 1056 (Craw) ¶¶ 0200-0216 (describing Fig. 7H); Ex. 1003 (Sarrafzadeh
`
`Decl.) ¶92. Craw further discloses employing “interfaces” to enable a health-
`
`monitoring application to utilize this data. Ex. 1056 (Craw) ¶¶ 0048, 0149, 0202-
`
`06938-00001/12352983.5
`
`5
`
`
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`
`0203; Ex. 1003 (Sarrafzadeh Decl.) ¶¶ 92-93. For example, Craw discloses that a
`
`“data dictionary” and “string table classes” may “act as interfaces for managing,
`
`extracting, and displaying information from binary information streams.” Ex. 1056
`
`(Craw) ¶ 0256 (emphasis added); Ex. 1003 (Sarrafzadeh Decl.) ¶92. In turn,
`
`Craw’s health-monitoring application may, for example, display the health data on
`
`a graphical user interface. Ex. 1056 (Craw) ¶ 208 (describing Fig. 9A).
`
`Craw’s “interfaces” each qualify as an “application-specific interface (API)”
`
`under the Board’s narrowing construction. As the Federal Circuit observed, “the
`
`claimed ‘application-specific interface’ performs the same function as an
`
`application programming interface, i.e., ‘enabl[ing] a particular application to
`
`utilize data obtained from hardware.’” Fitbit, 964 F.3d at 1117. Valencell’s expert
`
`similarly testified in his deposition that the ’941 patent “essentially refers to an
`
`API” that is “used as [an] application-specific interface.” Ex. 1069 (Pollonini Tr.)
`
`at 127:8-24. Craw’s “interfaces” are also used by specific applications. For
`
`example, Craw teaches that a health-monitoring application, such as the blood
`
`pressure application in Figure 9A, uses the disclosed interfaces in order to display
`
`extracted health data. Ex. 1056 (Craw) ¶¶ 203, 208 (describing Fig. 9A). Craw’s
`
`interfaces thus “enabl[e] a particular application to utilize data obtained from
`
`hardware”—that is all that is required under the Board’s construction of
`
`“application-specific interface (API)” and the Federal Circuit’s guidance.
`
`06938-00001/12352983.5
`
`6
`
`
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`
`
`But even if more were required, Craw discloses tailoring the “application-
`
`specific interface (API)” to a specific application. For example, Craw discloses
`
`using “a decipherable dictionary and an interpreter” in order to “extract
`
`information and act on it,” and further teaches that “[a]cting on the received
`
`information may depend on the goal of the application.” Ex. 1056 (Craw) ¶ 48
`
`(emphasis added). Craw thus recognizes that “medical devices may include one or
`
`more types of software, and the medical devices and software may be configured
`
`to operate upon a particular subset of physiological data.” Id. ¶ 4 (emphasis
`
`added). Accordingly, not only are Craw’s “interfaces” used by “a particular
`
`application,” they can be specifically tailored to each “particular application.” The
`
`Board’s narrowing construction of “application-specific interface (API)” thus has
`
`no significance on the ultimate success of the petitioned grounds.
`
`The Board’s finding in IPR2017-00321 that Craw discloses the additional
`
`limitations of claim 3 further confirms the unpatentability of claim 3. IPR2017-
`
`00321 involved the same parties and claims 14-21 of the ’941 patent. Valencell
`
`proposed substitute claim 22, which added the exact language of claim 3 of the
`
`’941 patent to independent claim 14. IPR2017-00321, Paper 44 at 51. In
`
`analyzing substitute claim 22, the Board construed “application-specific interface
`
`(API)” to have the same meaning as in this IPR. Id. at 62. After then reviewing
`
`the disclosure in Craw, including the specific disclosure cited in the original
`
`06938-00001/12352983.5
`
`7
`
`
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`
`petition in this IPR (id. at 72 (citing Craw ¶¶ 200-216)), the Board “[was]
`
`persuaded that Craw teaches or suggests that ‘the output data is parsed out such
`
`that an application-specific interface (API) can utilize the physiological
`
`information and motion-related information for an application,” and rejected claim
`
`22. Id. at 74. Valencell did not appeal this decision.
`
`C. Claims 4-5 Are Rendered Obvious by Luo, Craw, and Wolf
`
`Luo, Craw, and Wolf render obvious the additional limitations of claims 4-5.
`
`As to claim 4, Luo teaches generating relationships between “subject physiological
`
`parameters and subject physical activity parameters” such as determining that a “a
`
`heart rate jump” is “normal” because “the subject is going through a activity
`
`change from the resting condition to run condition.” Ex. 1055 (Luo) ¶ 0031
`
`(discussing “the integration of the physiological signal monitoring and physical
`
`activity monitoring”); see also Ex. 1003 (Sarrafzadeh Decl.) ¶ 99.
`
`Wolf satisfies the remaining limitations of claims 4 and 5 by teaching the
`
`“generat[ing] statistical relationships” limitation of claim 4 and the “machine
`
`learning” limitation of claim 5. For example, Wolf discloses “Bayesian filters 605,
`
`consisting of statistical filter subroutines,” that allow the system to “use voice and
`
`other interactions to update [their] ability to provide correct detection of different
`
`health related incidents”—that is, to “learn[] standard behavior patterns.” Ex.
`
`1042 (Wolf) ¶¶ 0019, 0023, 0062, 0063, 0070-0073 (emphasis added); see also Ex.
`
`06938-00001/12352983.5
`
`8
`
`
`
`
`1003 (Sarrafzadeh Decl.) ¶¶ 100-103. The combination of Luo, Craw, and Wolf
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`thus teaches all limitations of claims 4-5.
`
`It would be obvious to combine Luo and Craw with Wolf to arrive at claims
`
`4-5. Wolf teaches a health monitoring device very similar to the device of Luo, but
`
`with the added advantage of a cognitive inference engine to reduce uncertainties
`
`regarding the health state of the monitored subject. Ex. 1042 (Wolf) ¶0024; Ex.
`
`1003 (Sarrafzadeh Decl.) ¶104. It would thus have been obvious to combine the
`
`method suggested by Luo and Craw with an application like the cognitive engine
`
`of Wolf to improve health state determinations weighing more learned information.
`
`As with claim 3, Valencell never previously argued claims 4-5 were valid
`
`for any reason other than their dependency from independent claim 1. Valencell
`
`did not address the additional limitations of claims 4-5 in its preliminary patent
`
`owner response or supplemental briefing. Paper 6 at 27-30; Paper 41 at 5-7. Nor
`
`did Valencell contest the obviousness of combining Luo and Craw with Wolf. Id.
`
`Valencell’s failure to contest the obviousness of limitations in claims 4-5
`
`demonstrates that these claims are obvious over the combination of Luo, Craw, and
`
`Wolf.
`
`V. Claim 3 Is Obvious in View of Mault, Al-Ali, and Lee
`
`The Board already found that Fitbit’s combination of Mault and Al-Ali
`
`renders obvious claim 1 of the ’941 patent. Paper 43 at 76. Valencell did not
`
`06938-00001/12352983.5
`
`9
`
`
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`
`appeal this finding and is thus bound by the Board’s prior decision.
`
`The original petition adds Lee et al., “A Mobile Care System With Alert
`
`Mechanism” (Sept. 2007) (“Lee”) (Ex. 1059) in order to render obvious the
`
`additional limitations of claim 3. Lee discloses using a “Bluetooth application
`
`programming interface (BT API)” to enable a “mobile phone application program”
`
`to receive and process “physiological parameter measurement data” over a “serial
`
`port transmission protocol.” Ex. 1059 (Lee) at 509- 511; see also Ex. 1003
`
`(Sarrafzadeh Decl.) ¶¶ 157-61. Thus, Lee discloses “serial data output” (data
`
`transmitted using Lee’s “serial port transmission protocol”) that is parsed out such
`
`that an “application-specific interface (API)” (Lee’s “BT API”) can utilize the
`
`physiological information and motion-related information for an application (Lee’s
`
`“mobile phone application program”). Lee discloses “application-specific
`
`interface (API)” because Lee’s “mobile phone application program” is a
`
`“particular application” using the data. See Fitbit, 964 F.3d at 1117. Finally, it
`
`would be obvious to combine Lee with Mault and Al-Ali to allow communicating
`
`physiological information and motion-related information to a computing device,
`
`such as a mobile device. Ex. 1003 (Sarrafzadeh Decl.) ¶¶ 162-63.
`
`VI. Conclusion
`
`Fitbit respectfully requests that the Board find remaining claims 3-5 of the
`
`’941 patent unpatentable.
`
`06938-00001/12352983.5
`
`10
`
`
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
` By: /Jim Glass/
`James M. Glass (Reg. No. 46,729)
`QUINN EMANUEL URQUHART &
`SULLIVAN LLP
`51 Madison Avenue, 22nd Floor
`New York, NY 10010
`Email: jimglass@quinnemanuel.com
`Phone: 212-849-7142
`Fax: 212-849-7100
`
`Counsel for Petitioner Fitbit, Inc.
`
`
`
`
`
`
`Date: October 5, 2020
`
`
`
`
`
`
`
`06938-00001/12352983.5
`
`11
`
`
`
`IPR2017-00319
`U.S. Patent No. 8,923,941
`
`
`
`
`
`CERTIFICATE OF SERVICE (37 C.F.R. §§ 42.6(E), 42.105(A))
`
`The undersigned hereby certifies that the foregoing document was served in
`
`its entirety on October 5, 2020 upon the following parties via Electronic Mail.
`
`Justin B. Kimble
`Jeffrey R. Bragalone
`Daniel F. Olejko
`T. William Kennedy
`Jonathan H. Rastegar
`Marcus Benavides
`R. Scott Rhoades
`Sanford E. Warren, Jr
`
`BRAGALONE CONROY PC
`2200 Ross Ave.
`Suite 4500 – West
`Dallas, TX 75201
`JKimble-IPR@bcpc-law.com
`
`jbragalone@bcpc-law.com
`dolejko@bcpc-law.com
`bkennedy@bcpc-law.com
`jrastegar@bcpc-law.com
`mbenavides@bcpc-law.com
`srhoades@wriplaw.com
`swarren@wriplaw.com
`
`Date: October 5, 2020
`
`
`
`
`
`
`
` By: /Jim Glass/
`James M. Glass (Reg. No. 46,729)
`QUINN EMANUEL URQUHART &
`SULLIVAN LLP
`51 Madison Avenue, 22nd Floor
`New York, NY 10010
`Email: jimglass@quinnemanuel.com
`Phone: 212-849-7142
`Fax: 212-849-7100
`
`Counsel for Petitioner Fitbit, Inc.
`
`06938-00001/12352983.5
`
`12
`
`