IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`Attorney Docket No.: 50095-0024IP2
`
`Jeroen Poeze et al.
`In re Patent of:
`U.S. Patent No.: 10,631,765
`Issue Date:
`April 28, 2020
`Appl. Serial No.: 16/725,478
`Filing Date:
`December 23, 2019
`Title:
`MULTI-STREAM DATA COLLECTION SYSTEM FOR
`NONINVASIVE MEASUREMENT OF BLOOD
`CONSTITUENTS
`
`
`
`SECOND DECLARATION OF DR. THOMAS W. KENNY
`
`I hereby declare that all statements made of my own knowledge are true and
`
`that all statements made on information and belief are believed to be true. I further
`
`declare that these statements were made with the knowledge that willful false
`
`statements and the like so made are punishable by fine or imprisonment, or both,
`
`under Section 1001 of the Title 18 of the United States Code.
`
`
`
`
`
`
`
`Dated: November 7, 2021
`
`
`
`
`
`
`
`
`
`
`
`
`
`By:
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Thomas W. Kenny, Ph.D.
`
`APPLE 1047
`Apple v. Masimo
`IPR2020-01715
`
`1
`
`

`

`I. 
`II. 
`
`TABLE OF CONTENTS
`
`INTRODUCTION ........................................................................................... 1 
`GROUND 1 ..................................................................................................... 3 
`A.  Ohsaki does not teach or require that its translucent board 8 is
`“rectangular” in shape. .............................................................................. 9 
`B.  A POSITA would have recognized the benefits of Ohsaki’s teachings
`when applied to Aizawa’s sensor. .......................................................... 14 
`C.  Modifying Aizawa’s sensor to include a convex cover as taught by
`Ohsaki enhances the sensor’s light-gathering ability. ............................ 18 
`1.  Masimo ignores the well-known principle of reversibility ........... 19 
`2.  Masimo ignores the behavior of scattered light in a reflectance-
`type pulse sensor ........................................................................... 25 
`D.  A POSITA would have been motivated to select a convex cover to
`protect the optical elements. ................................................................... 35 
`E.  A POSITA would have been motivated to add a second emitter to
`Aizawa. ................................................................................................... 35 
`F.  A POSITA would have enabled the combined sensor of Aizawa,
`Inokawa, and Ohsaki to communicate wirelessly with a handheld
`computing device, based on the teachings of Mendelson-2006 ............. 37 
`G.  A POSITA would have expected success in performing the combination
` ................................................................................................................. 41 
`H.  The dependent claims 12, 18, and 29 are rendered obvious by Aizawa,
`Inokawa, Ohsaki, and Mendelson-2006 ................................................. 42 
`III.  GROUNDS 2-3 ESTABLISH OBVIOUSNESS .......................................... 43 
`IV.  CONCLUSION .............................................................................................. 43 
`
`
`2
`
`

`

`
`
`I.
`Introduction
`I have been retained on behalf of Apple Inc. to offer technical opinions
`
`1.
`
`relating to U.S. Patent No. 10,631,765 (“the ’765 Patent”) in the present case
`
`(IPR2020-01714). In this Second Declaration, I provide opinions related to Patent
`
`Owner’s Response (Paper 17) and Dr. Madisetti’s supporting declaration (Ex.
`
`2004).
`
`2.
`
`In addition to the materials listed in my First Declaration (APPLE-1003), I
`
`have also reviewed the following materials:
`
` Paper 8: Institution Decision;
`
` Paper 17: Patent Owner’s Response (“POR”);
`
` Ex. 2004: Declaration of Dr. Madisetti;
`
` Ex. 2006-2009: Transcripts of my prior depositions;
`
` APPLE-1050: Excerpts of Eugene Hecht, Optics (2nd Ed. 1990), pages
`
`79-143, 211-220;
`
` APPLE-1051: Eugene Hecht, Optics (2nd Ed. 1990);
`
` APPLE-1052: Deposition Transcript of Dr. Vijay Madisetti in IPR2020-
`
`01520, IPR2020-01537, IPR2020-01539, Day 1 (August 1, 2021);
`
` APPLE-1053: Deposition Transcript of Dr. Vijay Madisetti in IPR2020-
`
`01520, IPR2020-01537, IPR2020-01539, Day 2 (August 2, 2021);
`
`
`
`
`
`3
`
`

`

`
`
` APPLE-1054: Deposition Transcript of Dr. Vijay Madisetti in IPR2020-
`
`01536, IPR2020-01538 (August 3, 2021);
`
` APPLE-1055: “Refractive Indices of Human Skin Tissues at Eight
`
`Wavelengths and Estimated Dispersion Relations between 300 and 1600
`
`nm,” H. Ding, et al.; Phys. Med. Biol. 51 (2006); pp. 1479-1489
`
`(“Ding”);
`
` APPLE-1056: “Analysis of the Dispersion of Optical Plastic Materials,”
`
`S. Kasarova, et al.; Optical Materials 29 (2007); pp. 1481-1490
`
`(“Kararova”);
`
` APPLE-1057: “Noninvasive Pulse Oximetry Utilizing Skin Reflectance
`
`Photoplethysmography,” Y. Mendelson, et al.; IEEE Transactions on
`
`Biomedical Engineering, Vol. 35, No. 10, October 1988; pp. 798-805
`
`(“Mendelson-IEEE-1988”);
`
` APPLE-1049: Eugene Hecht, Optics (4th Ed. 2002).
`
` APPLE-1058: US Patent No. 6,198,951 ("Kosuda")
`
`3.
`
`Counsel has informed me that I should consider these materials through the
`
`lens of one of ordinary skill in the art related to the ’765 Patent at the time of the
`
`earliest possible priority date of the ’765 Patent (July 3, 2008, hereinafter the
`
`“Critical Date”) and I have done so during my review of these materials. I have
`
`applied the same level of ordinary skill in the art described in my prior declaration,
`
`
`
`
`
`4
`
`

`

`
`
`which I have been informed was also adopted by the Board in the Institution
`
`Decision. APPLE-1003, [0021]-[0022]; Institution Decision, 11-12.
`
`4.
`
`I have no financial interest in the party or in the outcome of this proceeding.
`
`I am being compensated for my work as an expert on an hourly basis. My
`
`compensation is not dependent on the outcome of these proceedings or the content
`
`of my opinions.
`
`5.
`
`In writing this declaration, I have considered the following: my own
`
`knowledge and experience, including my work experience in the fields of
`
`mechanical engineering, computer science, biomedical engineering, and electrical
`
`engineer; my experience in teaching those subjects; and my experience in working
`
`with others involved in those fields. In addition, I have analyzed various
`
`publications and materials, in addition to other materials I cite in my declaration.
`
`6. My opinions, as explained below, are based on my education, experience,
`
`and expertise in the fields relating to the ’765 Patent. Unless otherwise stated, my
`
`testimony below refers to the knowledge of one of ordinary skill in the fields as of
`
`the Critical Date, or before.
`
`II. Ground 1
`As I explained at length in my first declaration, a “POSITA would have
`
`7.
`
`found it obvious to modify the [Aizawa] sensor’s flat cover…to include a
`
`lens/protrusion…similar to Ohsaki’s translucent board 8, so as to [1] improve
`
`
`
`
`
`5
`
`

`

`
`
`adhesion between the user’s wrist and the sensor’s surface, [2] improve detection
`
`efficiency, [3] and protect the elements within the sensor housing.” APPLE-1003,
`
`¶¶98-102. I further explained that a POSITA would have found it obvious in view
`
`of Inokawa to include an additional LED in Aizawa’s sensor, to [1] “improve the
`
`detected pulse wave by distinguishing between blood flow detection and body
`
`movement, in addition to [2] enabling wireless communication between the sensor
`
`and a base station”. APPLE-1003, ¶¶80-83.
`
`8.
`
`Rather than attempting to rebut my testimony on these points, Masimo and
`
`its witness Dr. Madisetti provide arguments that are factually flawed.
`
`9.
`
`Specifically, Masimo states that “Ohsaki and Aizawa employ different
`
`sensor structures (rectangular versus circular) for different measurement locations
`
`(back side versus palm side of the wrist), using different sensor surface shapes
`
`(convex versus flat) that are tailored to those specific measurement locations” and
`
`then concludes that a POSITA would not “have been motivated to combine theses
`
`references,” and would not have “reasonably expected such a combination to be
`
`successful.” POR, 1-5. Masimo also argues that “[a]dding another LED
`
`complicates Aizawa’s sensor and increases power consumption” in addition to
`
`“eliminat[ing] the ability to take and display real-time measurements.” Id.
`
`10. But Ohsaki nowhere describes its benefits as being limited to a rectangular
`
`sensor applied to a particular body location. Also, a POSITA would not have
`
`
`
`
`
`6
`
`

`

`
`
`understood those benefits as being so limited. For example, Ohsaki teaches that
`
`“the detecting element and the sensor body 3 may be worn on the back side of the
`
`user’s forearm” or wrist. Nowhere does Ohsaki teach that its sensor can only be
`
`worn on a particular body location. APPLE-1009, [0030], [0008]-[0010], Abstract.
`
`In its summary of invention and claim preambles, Ohsaki explains that the object
`
`of its invention is “to provide a human pulse wave sensor which is capable of
`
`detecting the pulse wave of a human body stably and has high detection
`
`probability.” APPLE-1009, [0007], claims 1-8. Thus, Ohsaki’s disclosure should
`
`not be narrowly understood as applying to a single location or a single
`
`embodiment.
`
`11. Aizawa similarly reveals an embodiment in which its sensor is located on
`
`the palm side of the wrist (see APPLE-1006, FIG. 2, 0002, 0009), but does not
`
`limit its sensor to being applied to just the palm side of the wrist. A POSITA,
`
`based on Aizawa and Ohsaki’s disclosure, would have understood that the sensors
`
`in Aizawa and Ohsaki, when combined in the manner explained in my earlier
`
`declaration, would have been applicable to various locations on a human body and
`
`would have improved the performance of the sensor by providing the benefits
`
`described in these disclosures. Indeed, a POSITA would understand that the
`
`claimed benefits of the detector arrangement and the convex cover would have
`
`been useful and beneficial for measurements on many other locations.
`
`
`
`
`
`7
`
`

`

`
`
`12.
`
`In addition to the above, and as shown in Ohsaki’s FIG. 2 (reproduced
`
`below), Ohsaki attributes the prevention of slippage afforded by use of its board
`
`(and related improvements in signal quality) to the fact that “the convex surface of
`
`the translucent board…is in intimate contact with the surface of the user’s skin”
`
`when the sensor is worn. APPLE-1003, ¶¶71, 98-102, 159; APPLE-1009, [0015],
`
`[0017], [0025], FIGS. 1, 2, 4A, 4B (all emphasis added unless otherwise noted).
`
`APPLE-1009, FIG. 2 (annotated).
`
`
`
`13. Ohsaki’s discussion of these benefits does not mention or suggest that they
`
`relate to the shape of the exterior edge of Ohsaki’s board (whether circular,
`
`rectangular, ovoid, or other). Instead, when describing the advantages associated
`
`with its board, Ohsaki contrasts a “convex detecting surface” from a “flat detecting
`
`surface,” and explains that “if ‘the translucent board 8 has a convex
`
`
`
`
`
`8
`
`

`

`
`
`surface…variation of the amount of the reflected light…that reaches the light
`
`receiving element 7 is suppressed.’” APPLE-1003, ¶¶100-102; APPLE-1009,
`
`[0015], [0025].
`
`14. From this and related disclosure, a POSITA would have understood that a
`
`protruding convex cover would reduce the adverse effects of user movement on
`
`signals obtainable by the photodetectors within Aizawa’s sensor, which like
`
`Ohsaki’s light receiving elements, are positioned to detect light reflected from user
`
`tissue. APPLE-1003, ¶¶98-102, 154-161; APPLE-1009, [0015], [0017], [0025],
`
`FIGS. 1, 2, 4A, 4B; see also APPLE-1006, [0012], [0013], [0023], [0024], [0026],
`
`[0030], [0034], FIGS. 1(a), 1(b). A POSITA would expect that these benefits
`
`would apply to the pulse wave sensor of Aizawa, as well as to other wearable
`
`physiological monitors.
`
`15.
`
`In addition, as I explain with respect to the prior art figures reproduced
`
`below, a POSITA would have found it obvious to improve Aizawa’s sensor based
`
`on Ohsaki’s teachings, and would have been quite capable of making any
`
`inferences and creative steps necessary to achieve the benefits obtainable by
`
`modifying Aizawa’s cover to feature a convex detecting surface. APPLE-1008,
`
`¶¶14-15, FIG. 1; APPLE-1024, [0012], [0024], [0033], [0035], FIG. 6. In fact,
`
`Ohsaki nowhere depicts or describes the shape of its board as “rectangular”,
`
`“longitudinal”, or having “long rectangular shape”, or a “a pronounced
`
`
`
`
`
`9
`
`

`

`longitudinal directionality”. APPLE-1009, [0001]-[0030]; FIGS. 1, 2, 3A, 3B, 4A,
`
`4B. The following annotated FIG. 1(b) from Aizawa shows the results of the
`
`proposed combination:
`
`
`
`
`
`APPLE-1006, FIG. 1(b)(annotated).
`16. Also, contrary to Masimo’s contentions, a POSITA would not have been
`
`dissuaded from achieving those benefits by a specific body location associated
`
`with Ohsaki’s sensor. POR, 12-14, 20-26. Actually, a POSITA would have
`
`understood that a light permeable convex cover would have provided improved
`
`adhesion as described by Ohsaki in a sensor placed, e.g., on the palm side of the
`
`wrist, or other locations on the body. APPLE-1009, [0025], Claim 3 (stating that
`
`“the detecting element is constructed to be worn on a user’s wrist or a user’s
`
`forearm” without specifying a back or front of the wrist or forearm), FIGS 4A, 4B,
`
`Claims 4-8; see also APPLE-1019, 91.
`
`
`
`
`
`10
`
`

`

`
`
`17. A POSITA would also have understood that certain locations present
`
`anatomical features that provide for easy measurement of large reflected light
`
`signals and others present anatomical features that reduce the amplitude of the
`
`reflected light signals. Because of this, a POSITA would be motivated to search
`
`for features from other references that can provide improved adhesion, improved
`
`light gathering, reduced leakage of light from external sources, and protection of
`
`the elements within the system in order to successfully detect a pulse wave signal
`
`from many locations.
`
`18. For these and other reasons explained below, Masimo’s arguments should be
`
`rejected. The sections below address the arguments with respect to Ground 1
`
`presented in Masimo’s POR and explain, in more detail, why those arguments fail.
`
`A. Ohsaki does not teach or require that its translucent board 8
`is “rectangular” in shape.
`I demonstrated that a POSITA would have modified Aizawa in view of
`
`19.
`
`Ohsaki such that Aizawa’s cover “would include a convex surface, improving
`
`adhesion between a subject’s wrist and a surface of the sensor.” APPLE-1003,
`
`¶¶98; Pet., 31-35. Ohsaki (at [0025]) explains that the “convex surface of the
`
`translucent board 8” provides this improved adhesion. See id.
`
`20. Masimo states that it is not the “convex surface” that improves adhesion
`
`(i.e., prevents slippage) in Ohsaki, but instead a supposed “longitudinal shape” of
`
`“Ohsaki’s translucent board [8].” See POR, 20-26. But the cited portion of Ohsaki
`
`
`
`
`11
`
`

`

`
`
`does not include any reference to board 8. See APPLE-1009, [0019]. Instead,
`
`Ohsaki associates this “longitudinal” shape to a different component: “detecting
`
`element 2.” See id. (“it is desirable that the detecting element 2 is arranged so that
`
`its longitudinal direction agrees with the longitudinal direction of the user's arm”).
`
`Ohsaki never describes the “translucent board 8” as “longitudinal,” and nowhere
`
`describes that the “translucent board 8” and “detecting element 2” have the same
`
`shape. See generally APPLE-1009. As illustrated in Ohsaki’s FIG. 2 (reproduced
`
`below), translucent board 8 (annotated yellow) is not coextensive with the entire
`
`tissue-facing side of detecting element 2 (annotated green).
`
`
`
`
`
`
`
`12
`
`

`

`
`
`APPLE-1009, FIG. 2 (annotated)
`
`21. Masimo argues that translucent board 8 has a “very pronounced longitudinal
`
`directionality”. Masimo concludes, without explanation or citation to any
`
`disclosure in Ohsaki, that the translucent board 8 has a “rectangular” shape that is
`
`allegedly incompatible with Aizawa. See POR, 16-17. But this extremely narrow
`
`requirement asserted by Masimo cannot be supported by any specific element of
`
`the disclosure provided by Ohsaki. Ohsaki never describes translucent board 8, or
`
`any other component, as “rectangular.” In fact, the words “rectangular” and
`
`“rectangle” do not appear in Ohsaki’s disclosure. See generally APPLE-1009.
`
`Massimo’s assertions that the board must be “rectangular”, “longitudinal”, or
`
`having “long rectangular shape”, or “a pronounced longitudinal directionality” are
`
`simply incorrect.
`
`22. The POR incorrectly assumes that because Ohsaki’s light emitting element
`
`and the light receiving element are arranged in a longitudinal structure, Ohsaki’s
`
`translucent board must have a rectangular structure. APPLE-1009, [0009], [0019];
`
`POR, 1-3, 15-17. A POSITA would have known and understood that an elliptical
`
`or circular sensor or board configuration can have a longitudinal structure or
`
`appearance under a cross-sectional view. An example illustrating such an
`
`understanding is shown below in U.S. Patent No. 6,198,951 (“Kosuda”)’s FIGS. 3
`
`and 4. APPLE-1058, 8:42-56.
`
`
`
`
`
`13
`
`

`

`
`
`APPLE-1058, FIGS 3 and 4
`
`
`
`23. Attempting to confirm its false conclusion, Masimo asserts that “Ohsaki
`
`illustrates two cross-sectional views of its board that confirm it is rectangular.”
`
`POR, 15 (citing Ex. 2004, [39]-[42]). Masimo identifies these “two cross-
`
`sectional views” as FIGS. 1 and 2, and infers the supposed “rectangular shape” of
`
`the translucent board 8 based on FIG. 1 showing the “short” side of the device, and
`
`FIG. 2 showing the “long” side of the same device. See POR, 15-17. But,
`
`according to Ohsaki, FIG. 2 is “a schematic diagram,” not a cross-sectional view,
`
`and Ohsaki never specifies that FIGS. 1 and 2 of Ohsaki are different views of the
`
`
`
`
`
`14
`
`

`

`
`
`same device. APPLE-1009, [0013]. Accordingly, nothing in Ohsaki supports
`
`Masimo’s inference that the “translucent board 8” must be “rectangular” in shape.
`
`See, e.g., APPLE-1009, [0013], [0019], [0025], FIG. 2. Further, even if it is
`
`possible for the translucent board 8 to be “rectangular,” Ohsaki certainly does not
`
`teach or include any disclosure teaching or “requiring” this or any other particular
`
`shape of the board’s exterior edge. See id.
`
`24. Further, Dr. Madisetti and I have repeatedly indicated that a POSITA would
`
`not have interpreted figures from the references at issue as precise or limiting
`
`representations of the concepts described. See, e.g., Ex. 2006, 73:19-21 (Dr.
`
`Kenny: “this figure in [the patent document reference] is not intended as a
`
`precision engineering drawing of a[n] optimized system”); APPLE-1052, 79:19-
`
`80:2 (Dr. Madisetti: “I believe that to a POSA, these figures are not detailed optical
`
`diagrams”).
`
`25. The POR presents multiple arguments with respect to Ground 1 that are
`
`premised on Ohsaki requiring the translucent board 8 to be “rectangular.” See
`
`POR, 20-28. Because Ohsaki neither teaches nor requires such shape for the
`
`translucent board 8, these arguments fail.
`
`26.
`
`In my opinion, a POSITA would have been fully capable of modifying
`
`Aizawa to feature a light permeable protruding convex cover to obtain the benefits
`
`attributed to such a cover by Ohsaki.
`
`
`
`
`
`15
`
`

`

`
`
`B. A POSITA would have recognized the benefits of Ohsaki’s
`teachings when applied to Aizawa’s sensor.
`27. Masimo argues that “Ohsaki indicates that its sensor’s convex board only
`
`improves adhesion when used on the back (i.e., watch) side of the wrist,” that
`
`“Aizawa requires its sensor be positioned on the palm side of the wrist,” and
`
`concludes that “[a] POSITA seeking to improve adhesion of Aizawa’s sensor
`
`would not incorporate a feature that only improves adhesion at a different and
`
`unsuitable measurement location.” POR, 29. But Ohsaki does not say that its
`
`sensor can only be used at a backside of the wrist. Instead, at most, Ohsaki
`
`discloses such an arrangement with respect to a preferred embodiment. APPLE-
`
`1009, [0019] (“it is desirable that the detecting element 2 is arranged so that its
`
`longitudinal direction agrees with the longitudinal direction of the user’s arm”).
`
`28.
`
`Indeed, Ohsaki’s claim language reinforces that Ohsaki’s description would
`
`not have been understood as being that limited. For example, Ohsaki explains that
`
`“the detecting element 2…may be worn on the back side of the user's forearm” as
`
`one form of modification. APPLE-1009, [0030], [0028] (providing a section titled
`
`“[m]odifications”). The gap between the ulna and radius bones at the forearm is
`
`even greater than the gap between bones at the wrist, which is already wide enough
`
`to easily accommodate a range of sensor sizes and shapes, including circular
`
`shapes. In addition, Ohsaki’s claim 1 explicitly states that “the detecting element
`
`is constructed to be worn on a back side of a user’s wrist or a user’s
`
`
`
`
`16
`
`

`

`
`
`forearm.” See also APPLE-1009, Claim 2. As another example, Ohsaki’s
`
`independent claim 5 states that “the detecting element is constructed to be worn on
`
`a user’s wrist or a user’s forearm,” without even mentioning a backside of the
`
`wrist or forearm. See also APPLE-1009, Claims 6-8. A POSITA would have
`
`understood this language to directly contradict Masimo’s assertion that Ohsaki
`
`discloses a “very limited benefit” and that “Ohsaki repeatedly specifies that its
`
`sensor ‘is worn on the back side of a user’s wrist corresponding to the back of the
`
`user’s hand.’” POR, 38. Indeed, from this and related description, a POSITA
`
`would have understood that Ohsaki’s benefits are provided when the sensor is
`
`placed, for example, on either side of the user’s wrist or forearm. APPLE-1009,
`
`[0025], FIGS. 4A, 4B.
`
`29. Section B.2 of the POR presents several arguments with respect to Ground 1
`
`that are premised on Ohsaki requiring the detecting element to be worn on a back
`
`side of a user’s wrist or a user’s forearm. See POR, 28-41. Because Ohsaki
`
`neither teaches nor requires such a location for the translucent board 8, these
`
`arguments fail.
`
`30. Moreover, even assuming for the sake of argument that a POSITA would
`
`have understood Aizawa’s sensor as being limited to placement on the palm side of
`
`the wrist, and would have understood Ohsaki’s sensor’s “tendency to slip” when
`
`arranged on the front side as informing consideration of Ohsaki’s teachings with
`
`
`
`
`
`17
`
`

`

`
`
`respect to Aizawa, that would have further motivated the POSITA to implement a
`
`light permeable convex cover in Aizawa’s sensor, to improve detection efficiency
`
`of that sensor when placed on the palm side. POR, 28-41; APPLE-1009, [0015],
`
`[0017], [0023], [0025], FIGS. 1, 2, 3A, 3B, 4A, 4B.
`
`31. When describing advantages associated with its translucent board, Ohsaki
`
`explains with reference to FIGS. 4A and 4B (reproduced below) that “if the
`
`translucent board 8 has a flat surface, the detected pulse wave is adversely affected
`
`by the movement of the user’s wrist,” but that if the board “has a convex
`
`surface…variation of the amount of the reflected light…that reaches the light
`
`receiving element 7 is suppressed.” APPLE-1003, ¶¶98, 100; APPLE-1009,
`
`[0015], [0017], [0025].
`
`
`
`APPLE-1009, FIGS. 4A, 4B.
`
`
`
`
`18
`
`

`

`
`
`32. Contrary to Masimo’s arguments, a POSITA would not have understood
`
`these benefits of a convex surface over a flat surface to be limited to one side or the
`
`other of the user’s wrist. APPLE-1009, [0023]-[0025]. Instead, a POSITA would
`
`have understood that, by promoting “intimate contact with the surface of the user’s
`
`skin,” a light permeable convex cover would have increased adhesion and reduced
`
`slippage of Aizawa’s sensor when placed on either side of a user’s wrist or
`
`forearm, and additionally would have provided improvements in signal quality.
`
`APPLE-1009, [0015], [0017], [0025]; FIGS. 1, 2, 4A, 4B, Claims 3-8; see also
`
`APPLE-1019, 87, 91. Indeed, a POSITA would have known that modifying
`
`Aizawa’s flat plate to feature a convex protruding surface, as taught by Ohsaki,
`
`would have furthered Aizawa’s stated goal of “improv[ing] adhesion between the
`
`sensor and the wrist” to “thereby further improve the detection efficiency.”
`
`APPLE-1006, [0013], [0026], [0030], [0034].
`
`33. Further, the POSITA would have been fully capable of employing inferences
`
`and creative steps when improving Aizawa based on Ohsaki’s teachings, and
`
`would have expected success when applying those teachings. Indeed, a POSITA
`
`would have understood that adding a convex protrusion to Aizawa’s flat plate
`
`would have provided an additional adhesive effect that would have reduced the
`
`tendency of that plate to slip.
`
`
`
`
`
`19
`
`

`

`
`
`C. Modifying Aizawa’s sensor to include a convex cover as
`taught by Ohsaki enhances the sensor’s light-gathering ability.
`34. Masimo contends that the combined sensor “would direct light away from
`
`the detectors and thus decrease light collection and optical signal strength—not
`
`increase signal strength as Petitioner incorrectly asserts.” See, e.g., POR, 41-42.
`
`To the contrary, a POSITA would have understood that a cover featuring a convex
`
`protrusion would improve Aizawa’s signal-to-noise ratio by causing more light
`
`backscattered from tissue to strike Aizawa’s photodetectors than would have with a
`
`flat cover. APPLE-1019, 52, 86, 90; APPLE-1051, 84, 87-92, 135-141; APPLE-
`
`1057, 803-805; APPLE-1016, FIGS. 1(a)-1(b).
`
`35. Against this, Masimo and Dr. Madisetti assert that “a convex surface
`
`would…direct[] light away from the periphery and towards the center of the
`
`sensor,” but, in so doing, fail to articulate a coherent position—e.g., whether
`
`Masimo’s position is that “all” light or only “some” light is directed “to” or
`
`“towards the center.” POR, 46; Ex. 2004, ¶88.
`
`36. For example, Dr. Madisetti testified during deposition that “as I describe in
`
`my Declaration...if you have a convex surface...all light reflected or otherwise
`
`would be condensed or directed towards the center.” APPLE-1052, 40:4-11; see
`
`also id., 127:22-128:18; Ex. 2004, 86-87 (“A POSITA Would Have Understood
`
`That a Convex Cover Directs Light To The Center Of The Sensor”). However,
`
`during the same deposition, Dr. Madisetti further stated that that a convex cover
`
`
`
`
`20
`
`

`

`
`
`would redirect light “towards the center,” which could be “a general area at which
`
`the convex surface would be redirecting…light” or “a point,” while contrasting the
`
`phrase “to the center” from “towards the center.” APPLE-1052, 105:12-107:1,
`
`133:19-135:11.
`
`37.
`
`In contrast, and as explained in more detail below, Dr. Kenny has
`
`consistently testified that a POSITA would have understood that a convex cover
`
`improves “light concentration at pretty much all of the locations under the
`
`curvature of the lens,” and for at least that reason would have been motivated to
`
`modify Aizawa’s sensor to include a convex cover as taught by Ohsaki. Ex. 2006,
`
`164:8-16.
`
`1. Masimo ignores the well-known principle of
`reversibility
`38. The well-known optical principle of reversibility dispels Masimo’s claim
`
`that “a convex cover condenses light towards the center of the sensor and away
`
`from the periphery,” when applied to Aizawa. POR, 42; APPLE-1051, 87-92;
`
`APPLE-1049, 106-111. Based on the principle of reversibility, “a ray going from
`
`P to S will trace the same route as one from S to P.” APPLE-1051, 92, 84;
`
`APPLE-1049, 101, 110; APPLE-1054, 80:20-82:20. Importantly, the principle
`
`dictates that rays that are not completely absorbed by user tissue will propagate in
`
`a reversible manner. In other words, every ray that completes a path through tissue
`
`from an LED to a detector would trace an identical path through that tissue in
`
`
`
`
`21
`
`

`

`
`
`reverse, if the positions of the LED emitting the ray and the receiving detector
`
`were swapped. APPLE-1051, 92.
`
`39. To illustrate the relevance of this principle with reference to the annotated
`
`version of Inokawa FIG. 2 below, I have illustrated two example ray paths from
`
`the LEDs (green) to the detector (red):
`
`
`
`POR, 48 (red annotations added by Petitioner).
`
`40. As a consequence of the principle of reversibility, a POSITA would have
`
`understood that if the LED/detector configuration were swapped, as in Aizawa, the
`
`two example ray would travel identical paths in reverse, from a central LED (red)
`
`to surrounding detectors (green). A POSITA would have understood that, for these
`
`rays, any condensing/directing/focusing benefit achieved by Inokawa’s cover
`
`(blue) under the original configuration would be identically achieved under the
`
`reversed configuration:
`
`
`
`
`
`
`
`22
`
`

`

`
`
`
`
`41. Even when factoring in additional scattering that may occur when light is
`
`reflected within human tissue, reversibility holds for each of the rays that are not
`
`completely absorbed; consequently, “if we’re concerned with the impact of the
`
`lens on the system, it’s absolutely reversible.” APPLE-1053, 209:19-21, 207:9-
`
`209:21 (“one could look at any particular randomly scattered path…and the
`
`reversibility principle applies to all of the pieces [of that path] and, therefore,
`
`applies to the aggregate”).
`
`42. An example of reversibility in a situation with diffuse light, such as is
`
`present when LEDs illuminate tissue, is shown below from Hecht’s Figure 4.12.
`
`43.
`
`In this figure 4.12a, collimated light is incident on a smooth surface, and
`
`exhibits specular reflection, in which parallel light rays encounter and are reflected
`
`from the surface and remain parallel. A POSITA would certainly understand
`
`specular reflection. In the case of the reflection as shown in Figure 4.12b, the
`
`random roughness of the surface scatters the incoming rays into many directions,
`
`and the resulting light would appear to be diffuse. However, even in this
`
`circumstance, the principle of reversibility applies–each individual ray can be
`
`
`
`
`23
`
`

`

`
`
`reversed such that a ray travelling to the surface and scattered in a random
`
`direction can be followed backwards along exactly the same path.
`
`44.
`
`In more detail, and as shown with respect to the example paths illustrated
`
`below (which include additional scattering within tissue), each of the countless
`
`photons travelling through the system must abide by Fermat’s principle. APPLE-
`
`1049, 106-111. Consequently, even when accounting for various random
`
`redirections and partial absorptions, each photon traveling between a detector and
`
`an LED would take the quickest—and identical—path along the segment between
`
`each scattering event, even if the positions of the detector and LED were swapped.
`
`
`
`
`
`
`45. To better understand the effect of a convex lens on the propagation of light
`
`rays towards or away from the different LEDs or detectors, the first and last
`
`segment of the light path may be representative of the light propagation of the
`
`various light rays. In the figures above, starting at the upper left, there is a pink-
`
`colored light ray emerging from the green LED and passing through the convex
`
`
`
`
`24
`
`

`

`
`
`lens and entering the tissue. On the lower left, there is a pink-colored light ray
`
`leaving the tissue and entering the convex lens. As drawn, these rays are the same
`
`in position and orientation, except that the direction is exactly reversed. This
`
`illustration is consistent with the Principle of Reversibility as applied to this pair of
`
`possible light rays. According to the principle of reversibility, the upper light path
`
`from the LED to the first interaction with a corpuscle is exactly reversed. This
`
`same behavioral pattern applies to all of the segments of the many light paths that
`
`cross the interface at the surface of the convex lens. Importantly, in this example,
`
`the convex lens does not refract the incoming ray in a different direction from the
`
`outgoing ray, e.g., in a direction towards the center different from the outgoing ray.
`
`As required by the principle of reversibility, this incoming ray follows the same
`
`path as the outgoing ray, except in the reverse direction. This statement is true for
`
`every segment of these light paths that crosses the interface between the tissue and
`
`the convex lens. Any ray of light that successfully traverses a path from the LED
`
`to the detector, that path already accounts for the random scattering as that
`
`scattering is what allowed the ray to go from the LED to a detector along the path
`
`to thereby be subsequently detected by the detector. A POSITA would have
`
`underst