CONVERGENT ENGINEERING, INC. et al.

Patent Trials and Appeals Board

Dec 24, 2021

2021001883 (P.T.A.B. Dec. 24, 2021)

UNITED STATES PATENT AND TRADEMARK OFFICE
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
14/401,657 11/17/2014 Tammy Y. Euliano 049648/501165 6448
826 7590 12/24/2021
ALSTON & BIRD LLP
ONE SOUTH AT THE PLAZA
101 SOUTH TRYON STREET
SUITE 4000
CHARLOTTE, NC 28280-4000
EXAMINER
STEINBERG, AMANDA L
ART UNIT PAPER NUMBER
3792
NOTIFICATION DATE DELIVERY MODE
12/24/2021 ELECTRONIC
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
usptomail@alston.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte TAMMY Y. EULIANO, NEIL RUSSELL EULIANO II, and
SHALOM DARMANJIAN
Appeal 2021-001883
Application 14/401,657
Technology Center 3700
Before JENNIFER D. BAHR, CARL M. DEFRANCO, and
LISA M. GUIJT, Administrative Patent Judges.
BAHR, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the
Examiner’s decision to reject claims 1–5, 9–11, 13, 14, 16, 17 and 20–22.2

1 We use the word “Appellant” to refer to “applicant” as defined in 37
C.F.R. § 1.42(a). Appellant identifies the real parties in interest as the
University of Florida Research Foundation, Inc. and Convergent
Engineering, Inc. Appeal Br. 3.
2 The Examiner includes claim 6 on the list of claims that are rejected.
Non-Final Act. 1 (Office Action Summary, item 7). However, the Examiner
does not, in fact, set forth a rejection of claim 6. See Non-Final Act. 10–24.
Thus, there is no rejection of claim 6 before us for review.
Appeal 2021-001883
Application 14/401,657

2
We have jurisdiction under 35 U.S.C. § 6(b). We AFFIRM IN PART and
enter NEW GROUNDS OF REJECTION.
CLAIMED SUBJECT MATTER
Appellant’s invention is directed to a “non-invasive system and
method for predicting and/or determining preeclampsia in a patient.”
Spec. 5:3–4. Claim 1, reproduced below, is representative of the claimed
subject matter.
1. A preeclampsia detection system comprising:
one or more PPG sensors configured to non-invasively
capture photoplethysmography (PPG) signals from a patient;
one or more ECG sensors configured to non-invasively
capture electrocardiogram (ECG) signals from the patient; and
a memory storing a preeclampsia recognizer; and
a processor configured to execute the preeclampsia
recognizer to:
receive PPG signals from the one or more PPG
sensors,
receive ECG signals from the one or more ECG
sensors,
identify corresponding pulses between the received
PPG signals and the received ECG signals, wherein the
corresponding pulses comprise a first R-wave peak, a
PPG signal dip, a dicrotic notch, and a second R-wave
peak,
determine, using the received PPG signals and the
received ECG signals and based at least in part on the
identified corresponding pulses, signal values
comprising: an elapsed time between the identified the
first R-wave peak and the second R-wave peak, an
elapsed time between the first R-wave peak and the
dicrotic notch, the height of the dicrotic notch, and the
height of the percussion wave peak of the PPG,
generate, using the determined signal values based
on the received PPG signals and the received ECG
Appeal 2021-001883
Application 14/401,657

3
signals, a feature set comprising derivative features of the
signal values, wherein the feature set comprises
parameters from two or more different physiologic
classes,
create a multidimensional feature vector using the
generated feature set, and
determine, using the multidimensional feature
vector, whether preeclampsia exists in the patient,
wherein the preeclampsia recognizer comprises a
classifier trained using patient data to generate the
determination regarding the existence of preeclampsia.
REFERENCES
The prior art relied upon by the Examiner is:
Name Reference Date
Almog US 6,340,346 B1 Jan. 22, 2002
Farringdon US 2005/0113703 A1 May 26, 2005
Sharrock US 6,994,675 B2 Feb.7, 2006
Sogin US 2008/0071151 A1 Mar. 20, 2008
Martin US 2010/0016694 A1 Jan. 21, 2010
Gill US 2012/0065528 A1 Mar. 15, 2012
Najarian US 2012/0123232 A1 May 17, 2012
REJECTIONS
Claims 1, 4, and 10 stand rejected under 35 U.S.C. § 103(a) as being
unpatentable over Sogin and Gill.
Claims 1, 4, 5, 16, 17, 20, and 22 stand rejected under
35 U.S.C. § 103(a) as being unpatentable over Martin, Gill, and Najarian.
Claim 21 stands rejected under 35 U.S.C. § 103(a) as being
unpatentable over Martin, Gill, Najarian, and Sharrock.
Claims 2, 3, 13, and 14 stand rejected under 35 U.S.C. § 103(a) as
being unpatentable over Martin, Gill, Najarian, and Farringdon.
Appeal 2021-001883
Application 14/401,657

4
Claims 9–11 stand rejected under 35 U.S.C. § 103(a) as being
unpatentable over Martin, Gill, Najarian, and Almog.
OPINION
Obviousness—Sogin and Gill
The Examiner’s rejection is predicated in pertinent part on: (1) the
Examiner’s finding that Sogin discloses a processor configured to execute a
preeclampsia recognizer to receive PPG and ECG signals, generate a feature
set comprising derivative features of the PPG and ECG signals, create a
multidimensional feature vector using the generated feature set, and
determine, using the multidimensional feature vector, whether preeclampsia
exists in the patient; and (2) the Examiner’s determination that it would have
been obvious
to modify the preeclampsia diagnosis of Sogin to include
specific timing features derived by the combination of ECG and
PPG, including blood pressure, because Sogin teaches that
[preeclampsia][3] is a cardiac related illness related to
hypertension, and because Gill teaches that these specifically
calculated features allow for accurate and low-cost calculation
of important parameters for assessment of cardiac functionality.
See Non-Final Act. 10–11.
The Examiner’s finding and reasoning are flawed for the reasons set
forth by Appellant on pages 20–21 and 23 of the Appeal Brief. Sogin
discloses using ECG and PPG (i.e., pulse oximetry) signals in assessing

3 The Examiner at times uses the hyphenated “pre-clampsia” and at other
times uses “preeclampsia.” See, e.g., Non-Final Act. 10, 12. Sogin uses
“pre-eclampsia.” See, e.g., Sogin ¶ 3. For consistency, we use
“preeclampsia” herein as used in the present application.
Appeal 2021-001883
Application 14/401,657

5
physiological parameters that are indicators of obesity. Sogin ¶ 62. Further,
Sogin discloses providing for the diagnosis of the hypertensive status of a
subject as one of preeclampsia (PE), chronic hypertension (CHTN),
preeclampsia superimposed on chronic hypertension (PE+CHTN), or
normo-tension (i.e., normal), based on the measurement of two factors: the
level of soluble fms-like tyrosine kinase 1 (sFlt-1) and an obesity factor.
Sogin ¶ 32. However, Sogin makes clear that the determination of whether
or not the subject has preeclampsia is based solely on the level of sFlt-1, and
the determination of whether or not the subject has chronic hypertension is
based solely on the value of the obesity factor. Sogin ¶ 57. The level of
sFlt-1 is determined by assaying a blood sample; it is not based on an ECG
or a PPG signal. See Sogin ¶¶ 60, 128. Sogin discloses that one or more
other biomarkers known in the art to be associated with preeclampsia can be
selected, in addition to or as an alternative for sFlt-1, for measurement by
sampling bodily fluid. Sogin ¶¶ 72–73. In other words, Sogin does not
disclose diagnosing preeclampsia (i.e., determining whether preeclampsia
exists in a patient) based on the value of the obesity factor and, thus, does
not disclose a preeclampsia recognizer that determines whether preeclampsia
exists in a patient using a multidimensional feature vector comprising a
feature set comprising derivative features of PPG and ECG signal values.
Gill discloses assessing cardiac functionality by determining an
interval indicative of pre-injection interval (PEI) from the time from a
detected predetermined feature of an ECG signal to detected predetermined
features of a signal, such as a PPG signal, indicative of changes in arterial
volume. Gill ¶ 28; see also Gill ¶ 46. However, even if Gill’s technique for
processing PPG and ECG signals to assess cardiac functionality, as an
Appeal 2021-001883
Application 14/401,657

6
obesity factor, were incorporated into Sogin, it is not apparent, and the
Examiner does not cogently explain, how such a modification would arrive
at a preeclampsia recognizer that determines whether preeclampsia exists in
a patient using a multidimensional feature vector comprising a feature set
comprising derivative features of PPG and ECG signal values.
The Examiner maintains that Sogin’s obesity factor, with its PPG and
ECG features, meets the feature requirement of claim 1 and that “Sogin’s
additional reliance on a blood assay is not excluded by Appellant’s claim
language.” Ans. 9. Although the Examiner is correct that claim 1 does not
exclude the additional reliance on a blood assay, claim 1 does require that
the preeclampsia determine, using a multidimensional feature vector created
using a feature set comprising derivative features of the signal values of the
PPG and ECG signals, whether preeclampsia exists in the patient. However,
as discussed above, although Sogin’s multivariate analysis method uses a
plurality of factors including the level of sFlt-1 or other biomarker
associated with preeclampsia measured from a sample of blood and a
physical parameter, which could include measurements from PPG (pulse
oximetry) and/or ECG signals (see Sogin ¶¶ 58, 60, 62, 69), to determine the
hypertensive state (i.e., PE, CHTN, PE+CHTN, or normal) of the patient,
Sogin’s method uses a feature set comprising only the level(s) of one or
more biomarkers, measured from a sample of blood, not a feature set
comprising derivative features of the PPG and/or ECG signals, in
determining whether preeclampsia exists in the patient. Thus, even if Gill’s
technique for processing PPG and ECG signals to assess cardiac
functionality, as an obesity factor/physical parameter, were incorporated into
Sogin, it is not apparent, and the Examiner does not cogently explain, how
Appeal 2021-001883
Application 14/401,657

7
the signal values taught by Gill would be used to determine whether
preeclampsia exists using Sogin’s system.
For the above reasons, we do not sustain the rejection of claim 1, or of
claims 4 and 10, which depend from claim 1, as being unpatentable over
Sogin and Gill.
Obviousness—Martin, Gill, and Najarian
The Examiner finds that Martin discloses a preeclampsia detection
system comprising one or more PPG and ECG sensors; a memory storing a
preeclampsia recognizer; and a processor configured to execute the
preeclampsia recognizer to generate, using PPG and ECG signals received
from the PPG and ECG sensors, a feature set comprising derivative features
of signal values and determine whether preeclampsia exists in a patient.
Non-Final Act. 13–14. The Examiner finds that Martin refers generally to:
(1) calculating various corresponding pulses between the received PPG
signals and the received ECG signal to determine variability of R-R interval,
and the R-wave peak and its timing with respect to the PPG wave; and
(2) monitoring the amplitude of the PPG signal with respect to shape and
features of the dicrotic notch. Non-Final Act. 14. The Examiner finds,
however, that Martin lacks details as to how various timing features should
be calculated. Non-Final Act. 14. The Examiner also finds that Martin
discloses use of a classifier to draw diagnostic conclusions, but lacks details
on the use of the classifier regarding endothelial dysfunction, including
preeclampsia; more particularly, Martin does not disclose creating a
multidimensional feature vector from a feature set comprising parameters
from two more different physiologic classes and using the vector to
determine whether preeclampsia exists in the patient. Non-Final Act. 15.
Appeal 2021-001883
Application 14/401,657

8
The Examiner finds that Gill teaches monitoring both PPG and ECG
signals for evaluating patient health to identify corresponding pulses
between the PPG and ECG signals comprising the features set forth in claim
1 and determining, based at least in part on the identified corresponding
pulses, signal values as recited in claim 1. Non-Final Act. 15–16. The
Examiner determines it would have been obvious to modify Martin’s
endothelial function monitoring and cardiovascular timing features to
include specific timing features derived by a combination of ECG and PPG
because Martin discloses calculating these features generally “without
sufficient detail, and because Gill teaches that these specifically calculated
features allow for accurate and low-cost calculation of important parameters
for assessment of cardiac functionality.” Non-Final Act. 16.
The Examiner finds that Najarian teaches creating a multidimensional
feature vector using a generated feature set comprising parameters from two
or more different physiologic classes and using the vector to derive a
diagnosis, and teaches a preeclampsia recognizer comprising a classifier
trained using patient data. Non-Final Act. 16. The Examiner determines it
would have been obvious to modify Martin’s preeclampsia diagnostic
technique “with the signal processing and machine learning algorithms of
Najarian, because Najarian teaches that its processing allows for early
diagnosis of heart related conditions, in addition to improving previous
methods of cardiac diagnosis.” Non-Final Act. 16.
Appellant challenges the Examiner’s rationale in combining Gill with
Martin. Appeal Br. 14–16. In particular, Appellant submits that Martin
discloses monitoring vasoconstrictive response from a PPG signal, which
may indicate a change in endothelial function, while Gill determines and
Appeal 2021-001883
Application 14/401,657

9
monitors PEI based on correspondences between features of an ECG signal
and features of a PPG signal, and contends that the Examiner “fails to
explain how the signal values determined from ECG and PPG signals in Gill
would be used in Martin’s system in any particular way, much less for the
purpose of determining ‘whether preeclampsia exists’ in a patient.” Appeal
Br. 14–15. According to Appellant, “[n]either Martin nor Gill provide[s]
any teaching or suggestion regarding how PEI may be used to specifically
monitor endothelial function.” Appeal Br. 16.
Appellant’s argument that Martin and Gill do not teach or suggest
how PEI may be used to monitor endothelial function is not persuasive.
Martin teaches that “[s]imultaneous access to an ECG signal in addition to
the oximeter pulse timing can offer an indication of vascular tone, which can
augment the sensitivity or specificity of any conclusions regarding arousal,
respiratory effort, or sympathetic tone.” Martin ¶ 60. More particularly,
timing information gleaned from the ECG can define the overall duration of
PEI plus the transit time of the pulse through the systemic arteries to the
peripheral probe site; this overall duration is known as “pulse-transit time, or
PTT” and “is affected by vascular tone (also referred to as arterial
stiffness).” Martin ¶ 60. Thus, monitoring the PEI, as taught by Gill, is
informative as to the degree or progression of vasoconstriction, which, as
Martin discloses, may indicate change in endothelial function. Martin ¶ 168.
The Examiner’s rationale in combining Gill with Martin therefore has
rational underpinnings. See Ans. 7 (explaining that “Martin teaches general
use of cardiac timing features but no detail as to how they are calculated or
found” and that “Gill teaches detailed calculations and that the specific
Appeal 2021-001883
Application 14/401,657

10
cardiac timing features of Gill are critical for assessing cardiac
functionality”).
Appellant argues that Martin fails to disclose a preeclampsia
recognizer with “the ability [to] ‘determine whether preeclampsia exists in
the patient.” Appeal Br. 10–11. Appellant submits that Martin merely
discloses the ability to use a PPG signal to monitor a degree of
vasoconstrictive response in a patient” and that, although “this
vasoconstrictive response ‘may indicate change in endothelial function,’ it
cannot be used to actually determine whether or not a patient has
preeclampsia.” Appeal Br. 10 (citing Martin ¶ 168).
Appellant characterizes the disclosure of Martin as follows:
Martin is generally directed to a system that observes
trends in patient PPG data, primarily focusing on patient
breathing efforts during sleep (Martin, Abstract and Paragraph
[0033]). In particular, Martin discusses its system may observe
a degree of vasoconstrictive response in a patient through
analysis of collected PPG data (id. at Paragraph [0168]).
Martin notes that “[t]he degree of vasoconstrictive response
seen in the PPG acquired from a finger pulse oximeter, trended
over a given period (e.g., days, weeks, and/or months), may
indicate change in endothelial function” (id. (emphasis added)).
In other words, Martin discloses that PPG signals can be used
to assess the degree of vasoconstrictive response, which might
indicate a change in endothelial function that should be
investigated further by a physician.
Appeal Br. 11. According to Appellant, “Martin describes change in
endothelial function as a marker only of whether a patient status is
improving or worsening,” but “nothing in Martin suggests that the
vasoconstrictive response assessed from trended PPG signals could be used
to determine what condition is causing the change in patient status.”
Appeal 2021-001883
Application 14/401,657

11
Appeal Br. 12. Appellant submits that “Martin lists preeclampsia as one
example of a condition that [may] lead to a change in endothelial function,”
but that detection of a vasoconstrictive response that may indicate a change
in endothelial function does not necessarily mean that the patient has
preeclampsia. Appeal Br. 12.
Appellant’s characterization of Martin is consistent with our reading
of Martin’s disclosure regarding trending vasoconstrictive response over
time to indicate a change in endothelial function. Martin discloses a
controller configured to trend the vasoconstrictive response of a patient by
processing signals from a pulse oximeter to assess endothelial dysfunction.
Martin ¶¶ 33–34. Martin further discloses that the degree of
vasoconstrictive response seen in the PPG from the pulse oximeter, trended
over a given period, “may indicate change in endothelial function, which
may be a marker of improving or worsening patient status (e.g., the onset of
[preeclampsia], etc.).” Martin ¶ 168 (emphasis added). The reasonable
inference from Martin’s use of “may,” “e.g.,” and “etc.” is that the detection
of a degree of vasoconstrictive response from the PPG signal may indicate
endothelial dysfunction, which may be a marker of a medical condition, one
of which could be the onset of preeclampsia. In other words, Martin
discloses that preeclampsia is one possible implication of the detection of a
degree of vasoconstrictive response in the PPG signal indicating endothelial
dysfunction, but implies that there may be other causes for such a response
as well. Martin does not provide any specific guidance as to how to
distinguish endothelial dysfunction resulting from preeclampsia from
endothelial dysfunction resulting from other conditions. Martin does
disclose that “[s]imultaneous access to and ECG signal in addition to the
Appeal 2021-001883
Application 14/401,657

12
pulse oximeter timing can offer an indication of vascular tone, which can
augment the sensitivity or specificity of any conclusions regarding arousal,
respiratory effort, or sympathetic tone” and that “[t]he ECG can indicate the
moment of electrical systole, so the time delay between this central cardiac
timing and the arrival of the pulse at the periphery can define the
left-ventricle’s pre-ejection period plus the transit time of the pulse through
the systemic arteries to the peripheral probe site” (i.e., pulse-transit time or
PTT). Martin ¶ 60. However, Martin does not suggest that this additional
PTT information from the ECG is sufficient to distinguish endothelial
dysfunction resulting from preeclampsia from endothelial dysfunction
resulting from other conditions.4
The Declaration of Tammy Euliano (“Declarant”), dated March 17,
2020 (hereinafter “Euliano Decl.”), a Professor of Anesthesiology and
Obstetrics and Gynecology who avers that she has specialized in Obstetric
Anesthesia for over 20 years and published peer reviewed manuscripts
related to obstetric anesthesia and preeclampsia,5 supports our interpretation
of Martin’s disclosure with regard to endothelial dysfunction and

4 We make no determination herein as to whether the Specification and
drawings of the present application provide adequate description of an
algorithm (preeclampsia recognizer) that effectively distinguishes signal
values derived from corresponding features of ECG and PPG signals in a
patient having preeclampsia from signal values derived from corresponding
features of ECG and PPG in patients having other conditions, such as heart
failure, hypovolemic shock, or chronic hypertension, characterized by
vasoconstriction and/or endothelial dysfunction. See Sharrock 5:38–61;
13:13–14 (teaching that increased vasoconstriction and/or endothelial
dysfunction are seen in all these conditions).
5 Declarant is named as a co-inventor on the present application. See
Euliano Decl. ¶ 1.
Appeal 2021-001883
Application 14/401,657

13
preeclampsia. See Euliano Decl. ¶¶ 3, 13–14. Declarant states that “Martin
is clear that the trend in vasoconstrictive response over a period of time
indicates ‘a change in endothelial function.’” Euliano Decl. ¶ 14 (citing
Martin ¶ 34). However, Declarant avers that “nothing in Martin’s system
enables a determination to be made as to whether preeclampsia exists in a
patient. Indeed, it is possible for a patient to have a change in endothelial
function without having preeclampsia.”6 Euliano Decl. ¶ 14. Thus,
according to Declarant, “Martin does not provide guidance to one of
ordinary skill in the art as to how ECG and PPG signals may be used to
make a determination as to whether preeclampsia exists in a patient.”
Euliano Decl. ¶ 14.
The Examiner attempts to dismiss the Euliano Declaration as
“provid[ing] opinion on legal conclusions of obviousness, and reasoning
which amounts to a denial that the art of record teaches determination of
[preeclampsia] in a patient based on a difference of opinion in the teachings
of the art” and characterizes “Appellant’s reasoning that Martin is not
efficacious or enabling” as “just a denial of the teachings of Martin.”
Ans. 6. The Examiner also notes “the interest of [Declarant’s] opinion in
this case [as] that of a co-inventor, which is considered less persuasive than
that of a disinterested person.” Ans. 6.
The Examiner’s characterization of Declarant’s averments about
whether Martin discloses a recognizer that determines whether preeclampsia

6 Declarant does not identify any other specific conditions (other than
preeclampsia) for which endothelial dysfunction may be a marker. The
Euliano Declaration would have greater probative value if it had identified
other such conditions.
Appeal 2021-001883
Application 14/401,657

14
exists in a patient from pulse oximeter and/or ECG sensor data as expressing
a difference of opinion with, or denial of, Martin’s disclosure is inapposite.
As noted above, we find Declarant’s averments about Martin in this regard
to be a reasonable representation/explanation of what Martin discloses.
Further, although a declaration by a co-inventor may be less persuasive than
one made by a disinterested person, it is not to be disregarded for that reason
alone and may be relied on when sufficiently convincing. Cf. In re
McKenna, 203 F.2d 717, 720 (stating that an affidavit by an applicant or
co-applicant as to the advantages of the invention in the application is less
persuasive than one made by a disinterested person, but that such an
affidavit “is not to be disregarded for that reason alone and may be relied on
when sufficiently convincing”).
Although still maintaining that Martin discloses detecting the onset of
preeclampsia, the Examiner also finds that “this feature is taught specifically
by Najarian.” Ans. 4–5 (citing Najarian ¶ 286). However, for the reasons
that follow, Najarian does not make up for the deficiency in Martin in this
regard.
The Examiner finds that Najarian teaches creating a multidimensional
feature vector using a generated feature set comprising parameters from two
or more different physiologic classes and using the feature set to derive a
diagnosis, and that Najarian teaches a preeclampsia recognizer comprising a
classifier trained using patient data. Non-Final Act. 16 (citing Najarian
¶¶ 136, 140–143, 234, 286). Appellant submits that Najarian does not
disclose a system capable of determining whether preeclampsia exists in a
patient. Appeal Br. 17. According to Appellant, Najarian discusses using
heart rate variability, measured through signal processing methods described
Appeal 2021-001883
Application 14/401,657

15
therein, for studying critical care parameters, and “then recites a long and
exhaustive list of over thirty example critical care parameters, one of which
being preeclampsia.” Appeal Br. 16 (citing Najarian ¶ 286). As Appellant
correctly points out, “[t]his is the only instance where Najarian mentions
preeclampsia.” Appeal Br. 16.
Appellant argues:
Furthermore, Najarian does not describe or suggest how
the measure of heart rate variability through the described
signal processing methods would be different in different
critical care parameter cases. That is, Najarian includes a
plethora of largely distinct and different critical care parameters
(e.g., pneumonia or a viral infection or a stroke) that may be
studied based on the described heart rate variability. One of
ordinary skill in the art would clearly recognize that each
critical care parameter exhibits different characteristics and
symptoms, and that the possibility of a single method of
diagnosis being equally applicable in the same manner for
every critical care parameter is very unlikely, if not
non-existent. Indeed, there is no reasonable expectation of
success for generating a single method of diagnosis for every
critical care parameter listed by Najarian. Najarian would
therefore not enable one of ordinary skill in the art to create a
multidimensional feature vector that may be used by a classifier
to determine whether preeclampsia exists in a patient, as
required by [c]laim 1.
Appeal Br. 18.
Najarian discloses an apparatus that “can help care workers estimate
the extent of blood volume loss, distinguish blood volume loss from
physiological activities associated with exercise[,] and predict the presence
[and] extent of cardiovascular disease.” Najarian ¶ 22. Najarian’s apparatus
can “collect electronic heart-related signals from an individual and relate this
data to an ECG waveform” by, in one embodiment, analyzing the signals
Appeal 2021-001883
Application 14/401,657

16
using “a mathematical operation to determine the presence of a heart-related
injury or illness.” Najarian ¶ 22. Najarian’s apparatus monitors certain
heart-related measures and transforms them “into values of the measure of
[an] ECG waveform, such as P, Q, R, S, and T components using
mathematical techniques which then have predictive value in regards to
outcome in response to injury and illness.” Najarian ¶ 24. One of
Najarian’s primary concerns in developing this apparatus, or model, is to
study heart rate variability as a potential means to detect central volume
blood loss and injury-illness severity in critically ill and injured subjects.
Najarian ¶ 21.
Najarian discloses a processor adapted to generate derived data from
ECG waveforms by applying a mathematical operation using wavelet
transformation analysis, which identifies at least one time interval between a
series of repeating wave components. Najarian ¶ 29. Najarian also discloses
using a sensor device that generates data indicative of various physiological
and contextual parameters ranging from heart rate, pulse rate, ECG or EKG,
or blood pressure, to respiration rate, skin temperature, core body
temperature, body fat, hydration level, glucose or blood sugar level, activity
level, body position, pressure on muscles or bones, and UV exposure and
absorption. Najarian ¶ 103. Najarian discloses that the microprocessor of
the sensor device “may be programmed to derive information relating to an
individual’s physiological state based on the data indicative of one or more
physiological parameters.” Najarian ¶ 126. In other words, the sensor
device must be programmed to adapt it to the physiological parameters that
the user wishes to target or study.
Appeal 2021-001883
Application 14/401,657

17
Najarian discloses that one aspect of Najarian’s invention “relates to a
sophisticated algorithm development process for creating a wide range of
algorithms for generating information relating to a variety of variables from
the data received from the plurality of physiological and/or contextual
sensors.” Najarian ¶ 133. Najarian enumerates a myriad of different
variables including, but not limited to, energy expenditure, calorie intake,
sleep states, and activity states. Najarian ¶ 133. “According to [Najarian’s]
algorithm development process, linear or non-linear mathematical models or
algorithms are constructed that map the data from the plurality of sensors to
a desired variable. The first step of this process is to collect data from
subjects placed into situations simulating real-world conditions to develop a
database of raw data (measured using sensors that will be used in practice)
and data consisting of verifiably accurate data measurements and
extrapolated or derived data from more accurate lab equipment; the database
is split into training and test sets. Najarian ¶ 136. The next step requires
building a mathematical model that relates the raw data to the corresponding
verifiable standard data using well known machine learning techniques
including artificial neural nets, decision trees, memory-based methods, and
the like. Najarian ¶ 137. Najarian discusses the variety of derivatives, or
functions, that can be used to develop an overall model. Najarian
¶¶ 140–143. The algorithm includes at least one context detector that
produces a weight expressing the probability that a given portion of
collected data indicates the subject was in each of several possible contexts
(e.g., at rest or active), and a regression algorithm for each context that
computes a continuous prediction taking raw or derived data as input.
Najarian ¶ 143. The algorithm also includes a post-processor which outputs
Appeal 2021-001883
Application 14/401,657

18
the parameter of interest that is being measured or predicted by the
algorithm. Najarian ¶ 143.
Najarian illustrates, in Figure 15, an example algorithm developed for
measuring an individual’s energy expenditure. See Najarian ¶ 144.
However, Najarian teaches that the disclosed algorithm development process
may be used to create algorithms that enable the sensor device to detect and
measure various other parameters, such as duress, states of unconsciousness,
fatigue, shock, drowsiness, heat stress, dehydration, state of readiness, health
and/or metabolic status, for example. Najarian ¶ 145. Najarian also
emphasizes that an algorithm developed with one sensor device will not
work well on sensor devices that are not substantially structurally identical
to the sensor device used to create the algorithm. Najarian ¶ 145.
Najarian teaches that the disclosed system is applicable in diagnostic
settings, such as drug therapy calibrations, drug delivery monitoring,
post-surgical or rehabilitation environments. Najarian ¶ 234. Najarian
discloses using heart rate variability information regarding cardiovascular
activities “to evaluate the degree of hemorrhagic shock, a critical care
parameter, and assist in assessing the effects of treatment before
cardiovascular collapse occurs.” Najarian ¶ 286. Najarian also offers a
laundry list of various “[o]ther critical care parameters which may be studied
with the present invention.” Najarian ¶ 286. The critical care parameters
included in Najarian’s list cover a wide range of parameters, such as
bleeding (traumatic and nontraumatic hemorrhage, coagulopathies), various
wounds or injuries (wounds, burns, traumatic brain injury, spinal cord
injury, acute poisonings, drownings, hyperthermia, and hypothermia),
infections (bacterial infection, viral infection, fungal infection, pneumonia,
Appeal 2021-001883
Application 14/401,657

19
sepsis, septic shock), cardiovascular conditions (acute and chronic heart
failure, preeclampsia, eclampsia, acute respiratory failure, pulmonary
embolism, stroke, cerebral aneurysm), neuromuscular disease, metabolic
disorders (hyperthyroid, hypothyroid, adrenal insufficiency), and diabetic
ketoacidosis. Najarian ¶ 286. Appellant is correct that this laundry list is the
only instance of Najarian mentioning preeclampsia.
Najarian’s list of “critical care parameters which may be studied” with
Najarian’s method is very wide ranging and includes parameters that differ
significantly from one another in terms of the body systems involved, the
sensor data that would be pertinent, and the derivatives, or functions, that
would be required to appropriately analyze the sensor data to correlate
features of the data to the particular critical care parameter of interest. Thus,
absent some understanding of the features, or derivations of features, of the
signals that correlate to a determination of preeclampsia, as distinguished
from other conditions that are characterized by endothelial dysfunction or
vasoconstrictive response, a person having ordinary skill in the art would not
have a reasonable expectation of success in using an algorithm created to
evaluate the degree of hemorrhagic shock to determine whether a subject has
preeclampsia. As pointed out in the Euliano Declaration, “Najarian does not
provide any guidance as to how one of skill in the art would expect heart rate
variability to be different in a case where a patient has a viral infection or a
stroke, for example, compared to when a patient has preeclampsia.” Euliano
Decl. ¶ 18. Further, as discussed above, Najarian discusses a myriad of
linear and non-linear functions and machine learning techniques and
regressions that may be used to generate algorithms. Najarian ¶¶ 140–143.
Appeal 2021-001883
Application 14/401,657

20
The Examiner finds that “Najarian teaches a machine learning
approach which is trained using patient data per condition.” Ans. 8.
However, Najarian does not disclose what combination of sensors, sensor
placement, and functions/machine learning techniques/regressions would be
required to create an algorithm having the ability to determine whether a
particular subject has preeclampsia, as distinguished from other conditions
that may be characterized by vasoconstrictive response or endothelial
dysfunction.
At best, Najarian discloses a general approach to developing an
algorithm, or series of algorithms, and post-processor, using machine
learning techniques and linear and non-linear functions and regressions, to
evaluate critical care parameters, with particular application to evaluating
the degree of severity of hemorrhage in a subject, and invites the industry to
apply this general technique to a myriad of other critical care parameters of
interest without providing any guidance as to how to collect and analyze
sensor data using such a technique to identify preeclampsia. This is not
enough to give a person of ordinary skill in the art a reasonable expectation
of success in combining Najarian’s teachings with Martin, which, as
discussed above, fails to provide any specific guidance as to how to
distinguish endothelial dysfunction resulting from preeclampsia from
endothelial dysfunction resulting from other conditions, to create a
preeclampsia recognizer that determines whether preeclampsia exists in a
patient, as called for in claim 1. See Medichem, S.A. v. Rolabo, S.L., 437
F.3d 1157, 1165 (Fed.Cir.2006) (stating that an artisan does not have a
reasonable expectation of success where the prior art does nothing more than
invite the artisan “merely to vary all parameters or try each of numerous
Appeal 2021-001883
Application 14/401,657

21
possible choices until one possibly arrived at a successful result,” without
either giving an indication of which parameters are critical or some
“direction as to which of many possible choices is likely to be successful”).
As discussed above, Gill discloses assessing cardiac functionality by
determining an interval indicative of pre-injection interval (PEI) from the
time from a detected predetermined feature of an ECG signal to detected
predetermined features of a signal, such as a PPG signal, indicative of
changes in arterial volume. Gill ¶¶ 28, 46. However, Gill does not mention
preeclampsia, much less teach how to correlate timing feature(s) of the ECG
and PPG signals with the presence or absence of preeclampsia in a patient.
Thus, Gill does not make up for the deficiency in Martin and Najarian.
For the above reasons, we do not sustain the rejection of claim 1, or of
claims 4, 5, 16, 17, 20, and 22, which depend from claim 1, as being
unpatentable over Martin, Gill, and Najarian.
Obviousness—Martin, Gill, Najarian, and Sharrock
In rejecting claim 21, which depends from claim 1, the Examiner
relies on Sharrock’s teaching of using “augmentation index and other
cardiovascular timing ratios and features to determine the existence of
preeclampsia” and determines it would have been obvious to employ this
technique, using pulse pressure wave timing obtained from PPG and ECG as
taught by Martin (in combination with Gill and Najarian), because “Sharrock
discloses that augmentation is altered in patients with early [preeclampsia]
. . . and assists in distinguishing [preeclampsia] from other forms of
endothelial dysfunction” and because Sharrock teaches the need to more
clearly diagnose preeclampsia because it is a major cause of maternal and
fetal mortality. Non-Final Act. 19–20 (citing Sharrock 5:22–26, 56–61;
Appeal 2021-001883
Application 14/401,657

22
8:34–64; 13:54–61; Abstract). The teachings of Sharrock in regard to using
its technique in analyzing the pulse waveforms to better distinguish
preeclampsia from other conditions characterized by increased
vasoconstriction and/or endothelial dysfunction (see Sharrock 5:17–61;
13:54–14:4) appears to overcome the aforementioned deficiency in the
combination of Martin, Gill, and Najarian, in that it teaches a technique for
analyzing the pulse features that would help distinguish preeclampsia from
other conditions, thereby providing a reasonable expectation of success in
combining the teachings of Martin, Gill, Najarian, and Sharrock to create a
preeclampsia recognizer that can determine whether preeclampsia exists in a
patient.
Appellant does not present any arguments specifically contesting the
rejection of claim 21, aside from requesting its reversal on the basis that
claim 21 depends from independent claim 1. See Appeal Br. 25. In
particular, Appellant does not set forth any substantive argument contesting
the Examiner’s determination that the combination of Martin, Gill, Najarian,
and Sharrock renders obvious the subject matter of claim 21 and thus has
waived any argument of error. Accordingly, we sustain the rejection of
claim 21 as being unpatentable over Martin, Gill, Najarian, and Sharrock.
Obviousness—Martin, Gill, Najarian, and Farringdon or Almog
In rejecting claims 2, 3, 9–11, 13, and 14, the Examiner’s additional
reliance on Farringdon and Almog does not make up for the deficiency in
the combination of Martin, Gill, and Najarian discussed above. See Non-
Final Act. 20–24. Accordingly, we do not sustain the rejection of claims 2,
3, 13, and 14 as being unpatentable over Martin, Gill, Najarian, and
Appeal 2021-001883
Application 14/401,657

23
Farringdon; or the rejection of claims 9–11 as being unpatentable over
Martin, Gill, Najarian, and Almog.
New Grounds of Rejection—Obviousness
The Examiner’s determination that the subject matter of claim 21
would have been obvious in view of the combination of Martin, Gill,
Najarian, and Sharrock inherently subsumes within it a determination that
the subject matter of independent claim 1, from which claim 21 depends,
would have been obvious in view of Martin, Gill, Najarian, and Sharrock.
See Ormco Corp. v. Align Tech.Inc., 498 F.3d 1307, 1319 (Fed. Cir. 2007)
(when a dependent claim is “found to have been obvious, the broader claims
. . . must also have been obvious”). Thus, claim 1 is rejected under
35 U.S.C. § 103(a) as being unpatentable over Martin, Gill, Najarian, and
Sharrock. We designate this a new ground of rejection pursuant to our
authority under 37 C.F.R. § 41.50(b).
We also adopt the Examiner’s additional findings and reasoning set
forth on pages 16–18 and 20–24 in rejecting dependent claims 2–5, 9–11,
13, 14, 16, 17, 20, and 22, which Appellant does not contest. See Appeal
Br. 25 (requesting reversal of the rejections of the dependent claims only on
the basis that they depend from independent claim 1). Accordingly, claims
4, 5, 16, 17, 20, and 22 are rejected under 35 U.S.C. § 103(a) as being
unpatentable over Martin, Gill, Najarian, and Sharrock (applied in the same
manner as for claims 1 and 21); claims 2, 3, 13, and 14 are rejected under
35 U.S.C. § 103(a) as being unpatentable over Martin, Gill, Najarian,
Sharrock (applied in the same manner as for claims 1 and 21), and
Farringdon; and claims 9–11 are rejected under 35 U.S.C. § 103(a) as being
unpatentable over Martin, Gill, Najarian, Sharrock (applied in the same
Appeal 2021-001883
Application 14/401,657

24
manner as for claims 1 and 21), and Almog. We designate these as new
rejections pursuant to 37 C.F.R. § 41.50(b).
CONCLUSION
The Examiner’s rejections of claims 1–5, 9–11, 13, 14, 16, 17, 20, 22
are REVERSED.
The Examiner’s rejection of claim 21 is AFFIRMED.
New grounds of rejection of claims 1–5, 9–11, 13, 14, 16, 17, 20, 22
are entered pursuant to 37 C.F.R. § 41.50(b).
DECISION SUMMARY
In summary:
Claims
Rejected
35
U.S.C.
§
Reference(s)/Basis Affirmed Reversed New
Ground
1, 4, 10 103(a) Sogin, Gill 1, 4, 10
1, 4, 5,
16, 17,
20, 22
103(a) Martin, Gill,
Najarian
1, 4, 5,
16, 17,
20, 22

1, 4, 5,
16, 17,
20–22
103(a) Martin, Gill,
Najarian, Sharrock
21 1, 4, 5,
16, 17,
20, 22
2, 3, 13,
14
103(a) Martin, Gill,
Najarian,
Farringdon
2, 3, 13,
14

9–11 103(a) Martin, Gill,
Najarian, Almog
9–11
2, 3, 13,
14
103(a) Martin, Gill,
Najarian, Sharrock,
Farringdon
2, 3, 13,
14
9–11 103(a) Martin, Gill,
Najarian, Sharrock,
Farringdon
9–11
Appeal 2021-001883
Application 14/401,657

25
Overall
Outcome
21 1–5, 9–
11, 13,
14, 16,
17, 20, 22
1–5, 9–
11, 13,
14, 16,
17, 20,
22

FINALITY OF DECISION
This decision contains a new ground of rejection pursuant to
37 C.F.R. § 41.50(b). 37 C.F.R. § 41.50(b) provides “[a] new ground of
rejection pursuant to this paragraph shall not be considered final for judicial
review.” 37 C.F.R. § 41.50(b) also provides:
When the Board enters such a non-final decision, the appellant,
within two months from the date of the decision, must exercise
one of the following two options with respect to the new ground
of rejection to avoid termination of the appeal as to the rejected
claims:
(1) Reopen prosecution. Submit an appropriate
amendment of the claims so rejected or new Evidence relating
to the claims so rejected, or both, and have the matter
reconsidered by the examiner, in which event the prosecution
will be remanded to the examiner. The new ground of rejection
is binding upon the examiner unless an amendment or new
Evidence not previously of Record is made which, in the
opinion of the examiner, overcomes the new ground of rejection
designated in the decision. Should the examiner reject the
claims, appellant may again appeal to the Board pursuant to this
subpart.
(2) Request rehearing. Request that the proceeding be
reheard under § 41.52 by the Board upon the same Record. The
request for rehearing must address any new ground of rejection
and state with particularity the points believed to have been
misapprehended or overlooked in entering the new ground of
rejection and also state all other grounds upon which rehearing
is sought.
Appeal 2021-001883
Application 14/401,657

26
Further guidance on responding to a new ground of rejection can be
found in the Manual of Patent Examining Procedure § 1214.01.
TIME PERIOD FOR RESPONSE
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R.
§ 1.136(a)(1)(iv).

AFFIRMED IN PART; 37 C.F.R. § 41.50(b)