DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Information Disclosure Statement
There are references listed in the specification on [Page 1, Lines 14 – 32], [Page 5, Lines 27 – 28], [Page 6, Lines 26 – 28 and Lines 37 – 38], [Page 7, lines 1 – 2, lines 10 – 11] , [Page 10, lines 10 – 14], [Page 19, Lines 30 – 31], and [Page 20, Lines 21 – 24]. The listing of references in the specification is not a proper information disclosure statement. 37 CFR 1.98(b) requires a list of all patents, publications, or other information submitted for consideration by the Office, and MPEP § 609.04(a) states, "the list may not be incorporated into the specification but must be submitted in a separate paper." Therefore, unless the references have been cited by the examiner on form PTO-892 or already cited in an IDS, they have not been considered.
Specification
The disclosure is objected to because it contains an embedded hyperlink and/or other form of browser-executable code at [Page 1, Line 19]. Applicant is required to delete the embedded hyperlink and/or other form of browser-executable code; references to websites should be limited to the top-level domain name without any prefix such as http:// or other browser-executable code. See MPEP § 608.01.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are:
“first feature extraction unit” to “generate…time-domain features” in Claim 6 (and its dependent claims)
The claim limitation is interpreted according to [Page 11, Lines 20 – 25] “Fig. 5 shows schematically by way of example a preferred architecture of the (trained) machine learning model…a first feature extraction unit FE1.”, [Page 23, Lines 39 – 40], and “machine learning model, the trained machine learning model and the segmentation unit may be 40 stored in the memory (50).” The “first feature extraction unit” is shown as the generic block element “FE1” of the block diagram Fig. 5.
“second feature extraction unit” to “generate…spectrogram features” in Claim 6 (and its dependent claims)
The claim limitation is interpreted according to paragraph [Page 11, Lines 20 – 25] “Fig. 5 shows schematically by way of example a preferred architecture of the (trained) machine learning model…a second feature extraction unit FE2,” and “machine learning model, the trained machine learning model and the segmentation unit may be 40 stored in the memory (50).” The “second feature extraction unit” is shown as the generic block element “FE2” of the block diagram Fig. 5.
“feature combination units” to “generate…the joint representation” in Claim 6 (and its dependent claims)
The claim limitation is interpreted according to [Page 11, Lines 20 – 25] “Fig. 5 shows schematically by way of example a preferred architecture of the (trained) machine learning model…a feature combination unit FC” and “machine learning model, the trained machine learning model and the segmentation unit may be 40 stored in the memory (50).” The “feature combination unit” is shown as the generic block element “FC” of the block diagram Fig. 5.
“body action detection, identification and/or characterization unit” to “generate…event information” in Claim 6 (and its dependent claims)
The claim limitation is interpreted according to paragraph [Page 11, Lines 20 – 25] “Fig. 5 shows schematically by way of example a preferred architecture of the (trained) machine learning model…a body action detection, identification and/or characterization unit BU,” and “machine learning model, the trained machine learning model and the segmentation unit may be 40 stored in the memory (50).” The “body action detection, identification and/or characterization unit” is shown as the generic block element “BU” of the block diagram Fig. 5.
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1 - 16 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 (line 2), Claim 15 (line 5), and Claim 16 (line 4) each recite the term “the signal(s)”. There is insufficient antecedent basis for this limitation in the claim. It is unclear if this is intended to be the same or different than the previously-recited one or more signals of one or more modalities. For the purposes of examination, the term “the signal(s)” is deemed to claim “the one or more signals”. Claims 2 – 14 are similarly rejected due to their dependence on Claim 1.
Claim 1 (line 5 and line 6), Claim 15 (line 8 and line 9) and Claim 16 (line 7 and line 8) each recite the term “for each signal”. It is unclear if this is intended to be the same or different than the previously-recited “signal(s)” and “one or more signals”. For the purposes of examination, the term “for each signal” is deemed to claim “for each signal of the one or more signals”. Claims 2 – 14 are similarly rejected due to their dependence on Claim 1.
Claim 1 (line 10), Claim 15 (lines 13 - 14), and Claim 16 (lines 12 - 13) each recite the term “the second input of the machine learning model”. There is insufficient antecedent basis for this limitation in the claim. It is unclear if this is intended to be the same or different than the previously-recited trained machine learning model. For the purposes of examination, the term “the second input of the machine learning model” is deemed to claim “the second input of the trained machine learning model.” Claims 2 – 14 are similarly rejected due to their dependence on Claim 1.
Claim 1 (lines 13 - 14), Claim 15 (lines 16-17), and Claim 16 (lines 15 - 16) each recite the term “on the basis of the time-domain representation”. There is insufficient antecedent basis for this limitation in the claim. There is no previously-recited basis as such. For the purposes of examination, the term “on the basis of the time-domain representation” is deemed to claim “based on the time-domain representation”. Claims 2 – 14 are similarly rejected due to their dependence on Claim 1.
Claim 1 (lines 14- 15), Claim 15 (lines 17 - 18), and Claim 16 (lines 16 - 17) each recite the term “on the basis of the spectrogram representation”. There is insufficient antecedent basis for this limitation in the claim. There is no previously-recited basis as such. For the purposes of examination, the term “on the basis of the spectrogram representation” is deemed to claim “based on the spectrogram representation”. Claims 2 – 14 are similarly rejected due to their dependence on Claim 1.
Claim 1 (lines 16 - 17), Claim 15 (line 19), and Claim 16 (lines 18 - 19) each recite the term “on the basis of the time-domain features”. There is insufficient antecedent basis for this limitation in the claim. There is no previously-recited basis as such. For the purposes of examination, the term “on the basis of the time-domain features” is deemed to claim “based on the time-domain features”. Claims 2 – 14 are similarly rejected due to their dependence on Claim 1.
Claim 1 (line 18), Claim 15 (line 20), and Claim 16 (line 20) each recite the term “features of all signals”. It is unclear if these are intended to be the same or different than the previously-recited one or more signals. For the purposes of examination, the term “features of all signals” is deemed to claim “features of the one or more signals”. Claims 2 – 14 are similarly rejected due to their dependence on Claim 1.
Claim 1 (line 19), Claim 15 (line 21), and Claim 16 (line 20) each recite the term “on the basis of the joint representation”. There is insufficient antecedent basis for this limitation in the claim. There is no previously-recited basis as such. For the purposes of examination, the term “on the basis of the joint representation” is deemed to claim “based on the joint representation”. Claims 2 – 14 are similarly rejected due to their dependence on Claim 1. Similarly, the term “on the basis of the joint representation” recited in Claim 6 (line 22) is similarly deemed to claim “based on the joint representation”.
Claim 6 (line 10) recites the term “the number of signals”. There is insufficient antecedent basis for this limitation in the claim. It is unclear if these are intended to be the same or different than the previously-recited “signal(s)” and “one or more signals”. For the purposes of examination, the term “the number of signals” is deemed to claim “a number of the one or more signals.”
Claim 6 (lines 3 – 6 and 10) recites the term “a number n of first inputs, a number n of second inputs, a number n of first feature extraction units, a number n of second feature extraction units” and “wherein n is the number of signals, wherein n an integer equal to or greater than 1”. In the context of the whole claim, it appears that there could be as many inputs, first feature extraction units, second feature extraction units as there are signals, and it is unclear if there are actually as many “units” as there are signals processed. Does this mean that an entirely different unit is created for each individual signal that is processed, such that each signal is processed with a different framework? As such, is this referring to software that passing data through a unit for each individual signal? For the purposes of examination, the terms “a number n of first inputs, a number n of second inputs, a number n of first feature extraction units, a number n of second feature extraction units” and “wherein n is the number of signals, wherein n an integer equal to or greater than 1” are deemed to claim “processing n of the signal(s) using one or more second inputs, one or more first feature extraction units, one or more second feature extraction units” and “wherein n is the number of signal(s), wherein n an integer equal to or greater than 1”. Claims 7 – 10 are similarly rejected due to their dependence on Claim 6.
Claim 6 (line 11) recites the term “configured to receive a time-domain representation of a signal”. It is unclear if this is intended to be a the same or different time-domain representation of the same or different signal(s) that were previously-recited. For the purposes of examination, the term “configured to receive a time-domain representation of a signal” is deemed to claim “configured to receive the time-domain representation of the signal(s)”.
Claim 6 (line 12 - 13) recites the term “configured to receive a spectrogram representation of a signal”. It is unclear if this is intended to be a the same or different spectrogram representation of the same or different signal(s) that were previously-recited. For the purposes of examination, the term “configured to receive spectrogram representation of a signal” is deemed to claim “configured to receive the spectrogram representation of the signal(s)”.
Claim 6 (lines 14 - 15) recites the term “on the basis a time-domain representation, time-domain features”. It is unclear if this is intended to be a the same or different time-domain representation and time-domain features that were previously-recited. For the purposes of examination, the term “on the basis a time-domain representation, time-domain features” is deemed to claim “based on the time-domain representation, the time-domain features”.
Claim 6 (lines 15 - 16) recites the term “on the basis a spectrogram representation, spectrogram features”. It is unclear if this is intended to be a the same or different spectrogram representation and spectrogram features that were previously-recited. For the purposes of examination, the term “on the basis a spectrogram representation, spectrogram features” is deemed to claim “based on the spectrogram representation, the spectrogram features”.
Claim 11 (lines 2 - 3) recites the term “the basis of one-or more time-domain feature vectors”. There is insufficient antecedent basis for this limitation in the claim. There is no previously-recited basis as such. For the purposes of examination the term “the basis of one-or more time-domain feature vectors” is deemed to claim “a basis of one-or more time-domain feature vectors”.
Claim 11 (line 3) recites the limitation “representing one or more time-domain features”. It is unclear if these are intended to be the same or different than the previously-recited time-domain features. For the purposes of examination, the term “representing one or more time-domain features” is deemed to claim “representing one or more of the time-domain features”.
Claim 11 (line 4) recites the limitation “representing one or more spectrogram features”. It is unclear if these are intended to be the same or different than the previously-recited spectrogram features. For the purposes of examination, the term “representing one or more spectrogram features” is deemed to claim “representing one or more of the spectrogram features”.
Claim 12 (line 2) recites the limitation "the property of differentiability". There is insufficient antecedent basis for this limitation in the claim. The is no such previously-recited property. For the purposes of examination, the term “the property of differentiability” is deemed to claim “a property of differentiability”.
Claim 13 (lines 4 – 5) recites the term “the probability that the one or more signals at the respective timestep are caused by a cough event”. There is insufficient antecedent basis for this limitation in the claim. For the purposes of examination, the term “the probability that the one or more signals at the respective timestep are caused by a cough event” is deemed to claim “a probability that the one or more signals at the respective timestep are caused by a cough event”.
Claim 14 (line 3) recites the term “the number of cough events captured in the one or more signals”. There is insufficient antecedent basis for this term in the claim. For the purposes of examination, the term “the number of cough events captured in the one or more signals” is deemed to claim “a number of cough events captured in the one or more signals”.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1 - 16 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Regarding Claim 1, the claim recites "an act or step, or series of acts or steps" and is therefore a process, which is a statutory category of invention (Step 1). The claim is then analyzed to determine whether it is directed to any judicial exception (Step 2A, Prong 1).
Regarding Claims 15 and 16, the claims each recite an apparatus, which is one of the statutory categories of invention (Step 1). The claim is then analyzed to determine whether it is directed to any judicial exception (Step 2A, Prong 1).
Each of Claims 1 - 16 has been analyzed to determine whether it is directed to any judicial exceptions.
Step 2A, Prong 1
Each of Claims 1 - 16 recites at least one step or instruction for observations, evaluations, judgments, and opinions, which are grouped as a mental process under the 2019 PEG. The claimed invention involves making observations, evaluations, judgments, and opinions, which are concepts performed in the human mind under the 2019 PEG.
Accordingly, each of Claims 1 - 16 recites an abstract idea.
Specifically, Claims 1 - 16 recite (underlined are observations, judgments, evaluations, or opinions, which are grouped as a mental process under the 2019 PEG) (additional elements bolded, see Step 2A, prong 2);
Claims 1, 15, and 16:
For Claim 1:
A computer-implemented method, the method comprising the steps:
For Claim 15:
A computer system comprising:
a processor; and
a memory storing an application program configured to perform, when executed by the processor, an operation, the operation comprising:
For Claim 16:
A non-transitory computer readable medium having stored thereon software instructions that, when executed by a processor of a computer system, cause the computer system to execute the following steps:
For each of Claims 1, 15, and 16:
receiving one or more signals of one or more modalities, the signal(s) being related to an event,
providing a trained machine learning model, wherein the trained machine learning model comprises an output, and, for each signal, a first input, and a second input,
for each signal:
generating a time-domain representation of the signal,
generating a spectrogram representation of the signal,
inputting the time-domain representation into the first input of the trained machine learning model and the spectrogram representation into the second input of the machine learning model, wherein the trained machine learning model is configured and trained to
- generate time-domain features on the basis of the time-domain representation,
- generate spectrogram features on the basis of the spectrogram representation,
- generate a joint representation on the basis of the time-domain features and the spectrogram features of all signals, and
- generate, on the basis of the joint representation, an event information, the event information indicating whether and/or to what extent the event is related to a body action,
receiving, from the trained machine learning model the event information,
outputting the event information.
(observation, judgment or evaluation, which is grouped as a mental process under the 2019 PEG);
These underlined limitations describe a mathematical calculation and/or a mental process, as a skilled practitioner is capable of performing the recited limitations and making a mental assessment thereafter. Examiner notes that nothing from the claims suggests that the limitations cannot be practically performed by a human with the aid of a pen and paper, or by using a generic computer as a tool to perform mathematical calculations and/or mental process steps in real time. Examiner additionally notes that nothing from the claims suggests and undue level of complexity that the mathematical calculations and/or the mental process steps cannot be practically performed by a human with the aid of a pen and paper, or using a generic computer as a tool to perform mathematical calculations and/or mental process steps. For example, in Independent Claims 1, 15, and 16, these limitations include:
Observation and judgment to generate a time-domain representation of the signal,
Observation and judgment to generate a spectrogram representation of the signal,
Observation and judgment to input the time-domain representation into the first input of the trained machine learning model and the spectrogram representation into the second input of the machine learning model,
- Observation and judgment to generate time-domain features on the basis of the time-domain representation,
Observation and judgment to generate spectrogram features on the basis of the spectrogram representation,
Observation and judgment to generate a joint representation on the basis of the time-domain features and the spectrogram features of all signals
Observation and judgment to generate, on the basis of the joint representation, an event information, the event information indicating whether and/or to what extent the event is related to a body action,
Observation and judgment to receive, from the trained machine learning model the event information,
all of which are grouped as mental processes under the 2019 PEG.
Similarly, the dependent claims include the following abstract limitations, in addition the aforementioned limitations in Independent Claims 1, 15, and 16 (underlined observation, judgment or evaluation, which is grouped as a mental process under the 2019 PEG):
generate, on the basis a time- domain representation, time-domain features
Observation and judgment to generate, on the basis a time- domain representation, time-domain features
generate, on the basis a spectrogram representation, spectrogram features
Observation and judgment to generate, on the basis a spectrogram representation, spectrogram feature
generate, at least partially on all time-domain features and all spectrogram features, the joint representation
Observation and judgment to generate, at least partially on all time-domain features and all spectrogram features, the joint representation,
generate, on the basis of the joint representation, the event information,
Observation and judgment to generate, on the basis of the joint representation, the event information,
the joint representation is generated, at least partially on the basis of one or more time-domain feature vectors representing one or more time-domain features and one or more spectrogram feature vectors representing one or more spectrogram features by one or more of the following operations: element-wise multiplication, element-wise addition, cross- product, stacking on top or combinations thereof.
Observation and judgment to generate the joint representation, at least partially on the basis of one or more time-domain feature vectors representing one or more time-domain features and one or more spectrogram feature vectors representing one or more spectrogram features by one or more of the following evaluation operations: element-wise multiplication, element-wise addition, cross- product, stacking on top or combinations thereof.
output a vector, the vector containing a number of timesteps and for each time step a probability value, the probability value indicating the probability that the one or more signals at the respective timestep are caused by a cough event.
Observation and judgment to evaluate and output a vector, the vector containing a number of timesteps and for each time step a probability value, the probability value indicating the probability that the one or more signals at the respective timestep are caused by a cough event.
output a number, the number being equal to the number of cough events captured in the one or more signals.
Observation and judgment to evaluate and output a number, the number being equal to the number of cough events captured in the one or more signals.
all of which are grouped as mental processes under the 2019 PEG.
Accordingly, as indicated above, each of the above-identified claims recite an abstract idea.
Step 2A, Prong 2
The above-identified abstract ideas in each of Independent Claims 1, 15, and 16 (and their respective Dependent Claims) are not integrated into a practical application under 2019 PEG because the additional elements (identified above in Independent Claims 1, 15, and 16), either alone or in combination, generally link the use of the above-identified abstract ideas to a particular technological environment or field of use. More specifically, the additional elements of:
“trained machine learning model”
“first inputs”, “second inputs”
“first feature extraction unit”, “second feature extraction unit”, “convolutional neural network”
“one or more feature combination units”
“body action detection, identification and/or characterization unit”, “recurrent neural network”
“output”
“processor”
“memory”
“application program”
“non-transitory computer readable medium”
Additional elements recited include an “trained machine learning model”, “first inputs”, “second inputs”, “first feature extraction unit”, “second feature extraction unit”, “convolutional neural network”, “one or more feature combination units”, “body action detection, identification and/or characterization unit”, “recurrent neural network”, “output”, “processor”, “memory”, “application program”, and “non-transitory computer readable medium” in the Independent Claims 1 and 11, their dependent claims. These component are recited at a high level of generality, i.e., as a generic computer processor performing a generic function of processing data (the executing). These generic hardware component limitations for “trained machine learning model”, “first inputs”, “second inputs”, “first feature extraction unit”, “second feature extraction unit”, “convolutional neural network”, “one or more feature combination units”, “body action detection, identification and/or characterization unit”, “recurrent neural network”, “output”, “processor”, “memory”, “application program”, and “non-transitory computer readable medium” are no more than mere instructions to apply the exception using generic computer and hardware components. As such, these additional elements do not impose any meaningful limits on practicing the abstract idea.
Further additional elements from Independent Claims 1, 15, and 16 include pre-solution activity limitations, such as:
receiving one or more signals of one or more modalities, the signal(s) being related to an event,
providing a trained machine learning model, wherein the trained machine learning model comprises an output, and, for each signal, a first input, and a second input,
wherein the trained machine learning model is configured and trained
outputting the event information.
a processor; and
a memory storing an application program configured to perform, when executed by the processor, an operation, the operation comprising:
A non-transitory computer readable medium having stored thereon software instructions that, when executed by a processor of a computer system, cause the computer system to execute the following steps:
In addition the aforementioned extra-solution activity limitations in Independent Claims 1, 15, and 16, additional extra-solution activity limitations recited in the dependent claims include:
wherein the body action is selected from: cough, snoring, sneezing, hiccups, vomiting, shouting, swallowing, wheezing, shortness of breath, chewing, teeth grinding, chills, convulsions, spasm.
wherein the body action is or comprises one or more cough events.
wherein at least one signal of the one or more signals is an audio signal.
wherein at least one signal of the one or more signals is selected from: an electromyographic signal, an electrocardiogram signal, an accelerometer signal, a chest impedance signal, a plethysmographic signal, a temperature signal, a heart rate signal, a blood pressure signal.
These pre-solution measurement elements are insignificant extra-solution activity, setting up the parameters of the system, and serve as data-gathering for the subsequent steps.
The “trained machine learning model”, “first inputs”, “second inputs”, “first feature extraction unit”, “second feature extraction unit”, “convolutional neural network”, “one or more feature combination units”, “body action detection, identification and/or characterization unit”, “recurrent neural network”, “output”, “processor”, “memory”, “application program”, and “non-transitory computer readable medium” as recited in Independent Claims 1, 15, and 16and their dependent claims are generically recited computer and hardware elements which do not improve the functioning of a computer, or any other technology or technical field. Nor do these above-identified additional elements serve to apply the above-identified abstract idea with, or by use of, a particular machine, effect a transformation or apply or use the above-identified abstract idea in some other meaningful way beyond generally linking the use thereof to a particular technological environment, such that the claim as a whole is more than a drafting effort designed to monopolize the exception. Furthermore, the above-identified additional elements do not add a meaningful limitation to the abstract idea because they amount to simply implementing the abstract idea on a computer. For at least these reasons, the abstract ideas identified above in Independent Claims 1, 15, and 16 (and their respective dependent claims) is not integrated into a practical application under 2019 PEG.
Moreover, the above-identified abstract idea is not integrated into a practical application under 2019 PEG because the claimed method and system merely implements the above-identified abstract idea (e.g., mental process and certain method of organizing human activity) using rules (e.g., computer instructions) executed by a computer processor as claimed. In other words, these claims are merely directed to an abstract idea with additional generic computer elements which do not add a meaningful limitation to the abstract idea because they amount to simply implementing the abstract idea on a computer. Additionally, Applicant’s specification does not include any discussion of how the claimed invention provides a technical improvement realized by these claims over the prior art or any explanation of a technical problem having an unconventional technical solution that is expressed in these claims. That is, like Affinity Labs of Tex. v. DirecTV, LLC, the specification fails to provide sufficient details regarding the manner in which the claimed invention accomplishes any technical improvement or solution. Thus, for these additional reasons, the abstract idea identified above in Independent Claims 1, 15, and 16 (and their respective dependent claims) is not integrated into a practical application under the 2019 PEG.
Accordingly, Independent Claims 1, 15, and 16 (and their respective dependent claims) are each directed to an abstract idea under 2019 PEG.
Step 2B –
None of Claims 1 – 16 include additional elements that are sufficient to amount to significantly more than the abstract idea for at least the following reasons.
These claims require the additional elements of: “trained machine learning model”, “first inputs”, “second inputs”, “first feature extraction unit”, “second feature extraction unit”, “convolutional neural network”, “one or more feature combination units”, “body action detection, identification and/or characterization unit”, “recurrent neural network”, “output”, “processor”, “memory”, “application program”, and “non-transitory computer readable medium” as recited in Independent Claims 1, 15, and 16 and their dependent claims.
The additional elements of the “trained machine learning model”, “first inputs”, “second inputs”, “first feature extraction unit”, “second feature extraction unit”, “convolutional neural network”, “one or more feature combination units”, “body action detection, identification and/or characterization unit”, “recurrent neural network”, “output”, “processor”, “memory”, “application program”, and “non-transitory computer readable medium” Claims 1 - 20, as discussed with respect to Step 2A Prong Two, amounts to no more than mere instructions to apply the exception using generic computer and hardware components. The same analysis applies here in 2B, i.e., mere instructions to apply an exception using a generic computer component cannot integrate a judicial exception into a practical application at Step 2A or provide an inventive concept in Step 2B.
The above-identified additional elements are generically claimed computer components which enable the above-identified abstract idea(s) to be conducted by performing the basic functions of automating mental tasks. The courts have recognized such computer functions as well understood, routine, and conventional functions when claimed in a merely generic manner (e.g., at a high level of generality) or as insignificant extra-solution activity. See, Versata Dev. Group, Inc. v. SAP Am., Inc. , 793 F.3d 1306, 1334, 115 USPQ2d 1681, 1701 (Fed. Cir. 2015); and OIP Techs., 788 F.3d at 1363, 115 USPQ2d at 1092-93.
Per Applicant’s specification, the “trained machine learning model” is generically described on [Page 18, Lines 21 – 22] with “the machine learning model is or comprises one or more convolutional neural networks (CNN)…”, and [Page 18, Line 40] “The machine learning model may be or comprise one or more recurrent neural networks.”
Per Applicant’s specification, “first inputs”, “second inputs” is defined as described in the 112f interpretation above as part of the trained machine learning model, with [Page 11, Line 26] with “The first input I l is provided by the first feature extraction unit FE 1” and [Page 11, Lines 20 – 25] “Fig. 5 shows schematically by way of example a preferred architecture of the (trained) machine learning model…first input I.” The “first feature input” is shown as the “I1” arrow of the block diagram Fig. 5.
Per Applicant’s specification, the first feature extraction unit”, “second feature extraction unit”, “convolutional neural network”, is defined as described in the 112f interpretation above as part of the trained machine learning model, with [Page 11, Lines 20 – 25] “Fig. 5 shows schematically by way of example a preferred architecture of the (trained) machine learning model…a first feature extraction unit FE1.”, [Page 23, Lines 39 – 40], “machine learning model, the trained machine learning model and the segmentation unit may be 40 stored in the memory (50).”, and [Page 18, Lines 23 – 26] “A CNN is a class of deep neural networks…commonly applied to analyzing visual imagery (such as spectrogram representations). A CNN comprises an input layer with input neurons, an output layer 25 with at least one output neuron, as well as multiple hidden layers between the input layer and the output layer.” The “first feature extraction unit” is shown as the generic block element “FE1” of the block diagram Fig. 5.
Per Applicant’s specification, the “one or more feature combination units” is defined as described in the 112f interpretation above as part of the trained machine learning model, with [Page 11, Lines 20 – 25] “Fig. 5 shows schematically by way of example a preferred architecture of the (trained) machine learning model…a feature combination unit FC” and “machine learning model, the trained machine learning model and the segmentation unit may be 40 stored in the memory (50).” The “feature combination unit” is shown as the generic block element “FC” of the block diagram Fig. 5.
Per Applicant’s specification, the “body action detection, identification and/or characterization unit”, “recurrent neural network” is defined as described in the 112f interpretation above as part of the trained machine learning model, with [Page 11, Lines 20 – 25] “Fig. 5 shows schematically by way of example a preferred architecture of the (trained) machine learning model…a body action detection, identification and/or characterization unit BU,” and “machine learning model, the trained machine learning model and the segmentation unit may be 40 stored in the memory (50).” The “body action detection, identification and/or characterization unit” is shown as the generic block element “BU” of the block diagram Fig. 5.
Per Applicant’s specification, the “output” is defined as described in the 112f interpretation above as part of the trained machine learning model, with [Page 8, Lines 3 - 4] with “The output of the machine learning model can be a classification result, a regression result, a segmentation result and/or another result as described herein” and [Page 11, Lines 20 – 25] “Fig. 5 shows schematically by way of example a preferred architecture of the (trained) machine learning model…an output O.” The output O is shown as the arrow element O of the block diagram Fig. 5.
Per Applicant’s specification, the “processor” is defined generically at [Page 22, Lines 31 – 33] with “…a "computer", that unit which comprises a processor for carrying out logical operations”, [Page 23, lines 5 – 12], and [Page 22, Lines 25 – 28] “The operations…may be performed by…at least one general-purpose computer system...” The “processor “ is shown as generic block element “processing unit 20” in Fig. 17.
Per Applicant’s specification, the “memory”, is defined generically at [Page 23, lines 3 – 4] with “The computer may include one or more of each of a number of components such as, for example, a processing unit (20) connected to a memory (50) (e.g., storage device).” The “memory” is shown as generic block element “memory (50)” in Fig. 17.
Per Applicant’s specification, the “application program” is defined generically at [Page 22, Lines 27 – 28] with “at least one general purpose computer system…at least one computer program stored in a typically nontransitory computer readable storage medium.”
Per Applicant’s specification, the “non-transitory computer readable medium” is defined generically at [Page 22, Lines 29 – 30] with “"non-transitory" is used herein to exclude transitory, propagating signals or waves, but to otherwise include any volatile or non-volatile computer memory technology suitable to the application.”
Accordingly, in light of Applicant’s specification, the claimed terms “trained machine learning model”, “first inputs”, “second inputs”, “first feature extraction unit”, “second feature extraction unit”, “convolutional neural network”, “one or more feature combination units”, “body action detection, identification and/or characterization unit”, “recurrent neural network”, “output”, “processor”, “memory”, “application program”, and “non-transitory computer readable medium” are reasonably construed as a generic computing and hardware devices. Like SAP America vs Investpic, LLC (Federal Circuit 2018), it is clear, from the claims themselves and the specification, that these limitations require no improved computer resources, just already available computers, with their already available basic functions, to use as tools in executing the claimed process.
Furthermore, Applicant’s specification does not describe any special programming or algorithms required for the “trained machine learning model”, “first inputs”, “second inputs”, “first feature extraction unit”, “second feature extraction unit”, “convolutional neural network”, “one or more feature combination units”, “body action detection, identification and/or characterization unit”, “recurrent neural network”, “output”, “processor”, “memory”, “application program”, and “non-transitory computer readable medium”. This lack of disclosure is acceptable under 35 U.S.C. §112(a) since this hardware performs non-specialized functions known by those of ordinary skill in the computer arts. By omitting any specialized programming or algorithms, Applicant's specification essentially admits that this hardware is conventional and performs well understood, routine and conventional activities in the computer industry or arts. In other words, Applicant’s specification demonstrates the well-understood, routine, conventional nature of the above-identified additional elements because it describes these additional elements in a manner that indicates that the additional elements are sufficiently well-known that the specification does not need to describe the particulars of such additional elements to satisfy 35 U.S.C. § 112(a) (see Berkheimer memo from April 19, 2018, (III)(A)(1) on page 3). Adding hardware that performs “‘well understood, routine, conventional activit[ies]’ previously known to the industry” will not make claims patent-eligible (TLI Communications).
The recitation of the above-identified additional limitations in Claims 1 – 16 amounts to mere instructions to implement the abstract idea on a computer. Simply using a computer or other machinery in its ordinary capacity for economic or other tasks (e.g., to receive, store, or transmit data) or simply adding a general-purpose computer or computer components after the fact to an abstract idea (e.g., a fundamental economic practice or mathematical equation) does not provide significantly more. See Affinity Labs v. DirecTV, 838 F.3d 1253, 1262, 120 USPQ2d 1201, 1207 (Fed. Cir. 2016) (cellular telephone); and TLI Communications LLC v. AV Auto, LLC, 823 F.3d 607, 613, 118 USPQ2d 1744, 1748 (Fed. Cir. 2016) (computer server and telephone unit). Moreover, implementing an abstract idea on a generic computer, does not add significantly more, similar to how the recitation of the computer in the claim in Alice amounted to mere instructions to apply the abstract idea of intermediated settlement on a generic computer.
A claim that purports to improve computer capabilities or to improve an existing technology may provide significantly more. McRO, Inc. v. Bandai Namco Games Am. Inc., 837 F.3d 1299, 1314-15, 120 USPQ2d 1091, 1101-02 (Fed. Cir. 2016); and Enfish, LLC v. Microsoft Corp., 822 F.3d 1327, 1335-36, 118 USPQ2d 1684, 1688-89 (Fed. Cir. 2016). However, a technical explanation as to how to implement the invention should be present in the specification for any assertion that the invention improves upon conventional functioning of a computer, or upon conventional technology or technological processes. That is, the disclosure must provide sufficient details such that one of ordinary skill in the art would recognize the claimed invention as providing an improvement. Here, Applicant’s specification does not include any discussion of how the claimed invention provides a technical improvement realized by these claims over the prior art or any explanation of a technical problem having an unconventional technical solution that is expressed in these claims. Instead, as in Affinity Labs of Tex. v. DirecTV, LLC 838 F.3d 1253, 1263-64, 120 USPQ2d 1201, 1207-08 (Fed. Cir. 2016), the specification fails to provide sufficient details regarding the manner in which the claimed invention accomplishes any technical improvement or solution.
For at least the above reasons, the apparatuses and method of Claims 1 - 16 are directed to applying an abstract idea as identified above on a general-purpose computer without (i) improving the performance of the computer itself, or (ii) providing a technical solution to a problem in a technical field. None of Claims 1 - 16 provides meaningful limitations to transform the abstract idea into a patent eligible application of the abstract idea such that these claims amount to significantly more than the abstract idea itself.
Taking the additional elements individually and in combination, the additional elements do not provide significantly more. Specifically, when viewed individually, the above-identified additional elements for Step 2A Prong 2 in Independent Claims 1, 15, and 16 (and their dependent claims) do not add significantly more because they are simply an attempt to limit the abstract idea to a particular technological environment. That is, neither the general computer elements nor any other additional element adds meaningful limitations to the abstract idea because these additional elements represent insignificant extra-solution activity. When viewed as a combination, these above-identified additional elements simply instruct the practitioner to implement the claimed functions with well-understood, routine and conventional activity specified at a high level of generality in a particular technological environment. As such, there is no inventive concept sufficient to transform the claimed subject matter into a patent-eligible application. When viewed as whole, the above-identified additional elements do not provide meaningful limitations to transform the abstract idea into a patent eligible application of the abstract idea such that the claims amount to significantly more than the abstract idea itself. Thus, Claims 1 - 16 merely apply an abstract idea to a computer and do not (i) improve the performance of the computer itself (as in Bascom and Enfish), or (ii) provide a technical solution to a problem in a technical field (as in DDR).
Therefore, none of the Claims 1 - 16 amounts to significantly more than the abstract idea itself. Accordingly, Claims 1 - 16 are not patent eligible and are rejected under 35 U.S.C. 101.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1 – 7, 9 – 11 and 13 - 16 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Anushiravani et. al, (US 2020/0151516 A1).
Regarding Claims 1, 15, and 16, Anushiravani discloses
For Claim 1: A computer implemented method, the method comprising the steps ([Abstract]; [0153] - [0154]):
For Claim 15: A computer system ([Abstract]; [0153] - [0154]) comprising: a processor ([0154] “…processor…”); and a memory storing an application program ([0153] “ a machine readable medium may be any tangible medium that may contain, or store, a program for use…machine readable storage medium would include…random access memory (RAM)…”) configured to perform, when executed by the processor, an operation ([0154] “program code, when executed by the processor of the computer… cause the functions/operations…to be implemented."), the operation comprising:
For Claim 16: A non-transitory computer readable medium ([0008] “non-transitory, computer-readable mediums.”; [0153]) having stored thereon software instructions that, when executed by a processor of a computer system, cause the computer system to execute the following steps ([0153] “machine readable medium… store, a program for use…”; [0154] “program code, when executed by the processor of the computer… cause the functions/operations…to be implemented."),:
For each of Claims 1, 15, and 16:
receiving one or more signals of one or more modalities ([0046] “sensory data from different sensors…audio…”; [0047]), the signal(s) being related to an event ([0046] “…audio of the coughing…”; [0047]),
providing a trained machine learning model ([0004] “artificial intelligence (Al) system”; [0007] “AI model…”; [0007] “…training a neural network…”), wherein the trained machine learning model comprises an output ([0007] “obtaining a first set of raw output predictions from the model…”), and, for each signal, a first input ([0063] “…passed to a neural network. “), and a second input ([0122] “…convolutional neural network…map the spectrogram”),
for each signal ([0046] “sensory data from different sensors…audio…”; [0047]; Figs 3A and 3B; Fig. 14),:
generating a time-domain representation of the signal (Fig 3B; [0063] “…time domain audio signal…”),
generating a spectrogram representation of the signal (Fig 26; [0121] “ a time-frequency representation of an audio signal such as a spectrogram”),
inputting the time-domain representation into the first input of the trained machine learning model ([0063] “time domain audio signal is passed to a neural network. “) and the spectrogram representation into the second input of the machine learning model ([0122] “a convolutional neural network can be trained to map the spectrogram to a lower resolution spectrogram…”), wherein the trained machine learning model is configured and trained ([0004] “artificial intelligence (Al) system”; [0007] “AI model…”; [0007] “…training a neural network…”) to
- generate time-domain features on the basis of the time-domain representation ([0064] “…the output feature vector…transforming the time domain signal to a more well-known time-frequency representation…”; [0065])
- generate spectrogram features on the basis of the spectrogram representation ([0122] “a convolutional neural network can be trained to map the spectrogram to a lower resolution spectrogram, wherein each element in the target corresponds to one grid box in the input sound (2603)…features of the…”; [0059] “Such features can include… Mel Frequency Cepstral Coefficients (MFCC)…spectral flatness…”),
- generate a joint representation on the basis of the time-domain features and the spectrogram features of all signals (Fig. 14, “Pooling Layer” and “Fully-Connected Layer”; Fig. 3A, Fig. 3B, Fig. 4; [0064] “…determine a final feature vector (312)…”; ([0065] “manually hand-designed features, such as Mel-frequency Cepstral Coefficients (MFCC)”; [0109]) and
- generate, on the basis of the joint representation an event information (Fig. 14, “LABELS” from “pooling layer” and “Fully-connected layer”; [0069]; [0122]), the event information indicating whether and/or to what extent the event is related to a body action ([0069] “a classification algorithm (401) that predicts user's symptoms (e.g., coughing, wheezing, type of the cough if any, snoring, teeth grinding, shortness of breath, agonal breathing,”; Fig 4; [0122] “class label will then belong to the maximum class confidence value over all windows “),
receiving, from the trained machine learning model the event information (Fig. 4, arrow with “symptom labels”)(Fig. 5, arrow to information from post processing, in box 505 including “dry cough, wet cough…”),
outputting the event information (Fig. 4, arrow with “symptom labels”)(Fig. 5, arrow to information from post processing, in box 505 including “dry cough, wet cough…”)(Examiner notes that an arrow in a flow-chart type diagram indicates broadly, an output),
Regarding Claim 2, Anushiravani discloses as described above, The computer-implemented method according to claim 1. For the remainder of Claim 2, Anushiravani discloses wherein the body action is selected from: cough ([0069] “…coughing… “), snoring ([0069] “…snoring…”), sneezing, hiccups, vomiting, shouting, swallowing, wheezing ([0069] “…wheezing…”), shortness of breath ([0069] “…shortness of breath…”), chewing ([0069] “…teeth grinding…”)(Examiner notes that teeth grinding is broadly chewing on one’s own teeth.), teeth grinding ([0069] “…teeth grinding…”), chills, convulsions, spasm ([0069] “…coughing… “)(Examiner notes that a cough is broadly a spasm.)
Regarding Claim 3, Anushiravani discloses as described above, The computer-implemented method according to claim 1. For the remainder of Claim 3, Anushiravani discloses wherein the body action is or comprises one or more cough events ([0069] “a classification algorithm (401) that predicts user's symptoms (e.g., coughing, wheezing, type of the cough…”)
Regarding Claim 4, Anushiravani discloses as described above, The computer-implemented method according to claim 1. For the remainder of Claim 4, Anushiravani discloses wherein at least one signal of the one or more signals is an audio signal ([0058 ] “auditory data from each user…”; [0059] each audio signal”).
Regarding Claim 5, Anushiravani discloses as described above, The computer-implemented method according to claim 1. For the remainder of Claim 5, Anushiravani discloses wherein at least one signal of the one or more signals ([0046] “sensory data from different sensors…audio…”; [0047]) is selected from: an electromyographic signal ([0093] “PSG sensors…Electromyography (EMG) can be used to monitor…feature vector is extracted from all sensors…detect more vitals…”; Fig 12A), an electrocardiogram signal ([0093] “…Electrocardiography (ECG)”), an accelerometer signal ([0091] “…an accelerometer and RF to measure and analyzes user's movement…”), a chest impedance signal, a plethysmographic signal, a temperature signal ([0047] “ fourth sensor 105…thermometer…body temperature of the user 101”), a heart rate signal ([0093] “track a user's…heart rhythm…”), a blood pressure signal ([0047] “…fifth sensor 106 is blood pressure monitor…”).
Regarding Claim 6, Anushiravani discloses as described above, The computer-implemented method according to claim 1. For the remainder of Claim 6, Anushiravani discloses wherein the trained machine learning model ([0004] “artificial intelligence (Al) system”; [0007] “AI model…”; [0007] “…training a neural network…”), comprises
- a number n of first inputs ([0063] “time domain audio signal is passed to a neural network.“; Fig 3A and Fig 3B),
- a number n of second inputs ([0122] “a convolutional neural network can be trained to map the spectrogram to a lower resolution spectrogram…”; Fig 3A and Fig 3B),
- a number n of first feature extraction units (Fig 3A; “Extract features and concatenate” 304; (Fig. 3B, “convolutional layer” for processing the “time domain data” “A_1” and “A_2”)
- a number n of second feature extraction units((Fig 3A; “Extract features and concatenate” 304; Fig. 3B) Examiner notes that the units extract first and second features…),
- one or more feature combination units (Fig. 14, “pooling layer”)
- a body action detection, identification and/or characterization unit (Fig. 4, “classification algorithm” 401”), and
- an output (Fig. 4, arrow with “symptom labels”)(Fig. 5, arrow to information from post processing, in box 505 including “dry cough, wet cough…”)(Examiner notes that an arrow in a flow-chart type diagram indicates broadly, an output),
wherein n is the number of signals ([0046] “sensory data from different sensors…audio…”; [0047]; [0058] “…each audio signal…”; Figs 3A and Fig. 3B, Fig. 14)(Examiner notes the 112(b) interpretation above)
wherein n is an integer equal to or greater than 1 (Examiner notes that each of the mapped components above have at least one),
wherein each first input is configured to receive a time-domain representation of a signal ([0063] “time domain audio signal is passed to a neural network.” Fig 3A and Fig 3B),
wherein each second input is configured to receive a spectrogram representation of a signal ([0122] “a convolutional neural network can be trained to map the spectrogram to a lower resolution spectrogram…”; Fig 3A and 3B),
wherein each first feature extraction unit (Fig 3A; “Extract features and concatenate” 304; (Fig. 3B, “convolutional layer” for processing the “time domain data” “A_1” and “A_2”) is configured to generate, on the basis a time-domain representation, time-domain features ([0064] “…the output feature vector…transforming the time domain signal to a more well-known time-frequency representation…”; [0065])
wherein each second feature extraction unit ((Fig 3A; “Extract features and concatenate” 304; Fig. 3B) is configured to generate, on the basis a spectrogram representation, spectrogram features representation ([0122] “a convolutional neural network can be trained to map the spectrogram to a lower resolution spectrogram, wherein each element in the target corresponds to one grid box in the input sound (2603)…features of the…”;[0059] “Such features can include… Mel Frequency Cepstral Coefficients (MFCC)…spectral flatness…”),
wherein the one or more feature combination units (Fig. 14, “pooling layer”),are configured to generate, at least partially on all time-domain features and all spectrogram features (Fig. 14, “Pooling Layer”; Fig. 3A, Fig. 3B, Fig. 4; [0064] “…determine a final feature vector (312)…”; ([0065] “manually hand-designed features, such as Mel-frequency Cepstral Coefficients (MFCC)”; [0109]), the joint representation (Fig. 14, “Pooling Layer” and “Fully-Connected Layer”; Fig. 3A, Fig. 3B, Fig. 4)
wherein the body action detection, identification and/or characterization unit (Fig. 4, “classification algorithm” 401”) is configured to generate, on the basis of the joint representation, the event information (Fig. 14, arrow to “labels”; ([0069] “a classification algorithm (401) that predicts user's symptoms (e.g., coughing, wheezing, type of the cough if any, snoring, teeth grinding, shortness of breath, agonal breathing,”; Fig 4; [0122] “class label will then belong to the maximum class confidence value over all windows “),
wherein the output is configured to output the event information (Fig. 4, arrow with “symptom labels”)(Fig. 5, arrow to information from post processing, in box 505 including “dry cough, wet cough…”)(Examiner notes that an arrow in a flow-chart type diagram indicates broadly, an output),
Regarding Claim 7, Anushiravani discloses as described above, The computer-implemented method according to claim 6. For the remainder of Claim 7, Anushiravani discloses wherein one or more of the first feature extraction units and/or the second feature extraction units (Fig 3A; “Extract features and concatenate” 304; (Fig. 3B, “convolutional layer” for processing the “time domain data” “A_1” and “A_2”) are or comprise a convolutional neural network ([0063] “extracting features from raw data from the sensors using a bottle-neck convolutional neural network (CNN)”; Fig 3B).
Regarding Claim 9, Anushiravani discloses as described above, The computer-implemented method according to claim 6. For the remainder of Claim 9, Anushiravani discloses wherein the body action detection, identification and/or characterization unit (Fig. 4, “classification algorithm” 401”; [0102] “a recurrent network for predicting symptoms from a sequence of data according to an embodiment…”), is or comprises a recurrent neural network ([0082] “models…are modeled using…recurrent neural networks…Fig. 18”; [0102]; Fig. 17).
Regarding Claim 10, Anushiravani discloses as described above, The computer-implemented method according to claim 9. For the remainder of Claim 10, Anushiravani discloses wherein the recurrent neural network comprises gated recurrent units ([0102] “…layers can be set to…memory units such as…Gated Recurrent Units (GRUs)”).
Regarding Claim 11, Anushiravani discloses as described above, The computer-implemented method according to claim 1. For the remainder of Claim 11, Anushiravani discloses wherein the joint representation is generated (Fig. 14, “Fully-Connected Layer”; Fig. 3A, Fig. 3B, Fig. 4; [0064] “…determine a final feature vector (312)…”;), at least partially on the basis of one or more time-domain feature vectors representing one or more time-domain features ([0064] “…the output feature vector has already been determined by transforming the time domain signal to a more well-known time-frequency representation…”) and one or more spectrogram feature vectors representing one or more spectrogram features ([0065] “manually hand-designed features, such as Mel-frequency Cepstral Coefficients (MFCC)”) by one or more of the following operations: element-wise multiplication, element-wise addition ([0109] “trained on a loss function which is the addition of the difference between norm 2 distance between the reference and positive example from the norm 2 distance between the reference and the negative example as shown in Equation 7…”), cross- product, stacking on top or combinations thereof.
Regarding Claim 13, Anushiravani discloses as described above, The computer-implemented method according to claim 1. For the remainder of Claim 13, Anushiravani discloses wherein the trained machine learning model is configured to output a vector ([0106] “ a feature vector is produced (2000)”; Fig 20), the vector containing a number of timesteps and for each time step a probability value ([0073] “…output layer…posterior probabilities”; Fig. 5), the probability value indicating the probability that the one or more signals at the respective timestep are caused by a cough event ([0073] “ Y is the output layer containing the posterior probabilities of each possible condition occurring…”).
Regarding Claim 14, Anushiravani discloses as described above, The computer-implemented method according to claim 1. For the remainder of Claim 14, Anushiravani discloses wherein the trained machine learning model is configured to output a number ([0100] “number of respiratory events from a certain location…”)(Examiner notes that the number of respiratory events is broadly an output of the classifier), the number being equal to the number of cough events captured in the one or more signals ([0100] “number of respiratory events from a certain location…”; Fig. 14. “LABEL”).
Claim 12 is rejected under 35 U.S.C. 102(a)(1) as being anticipated by Anushiravani et. al, (US 2020/0151516 A1), as evidenced by Chatterjee (“A Basic Introduction to Convolutional Neural Network”, Ref U on PTO-892)
Regarding Claim 12, Anushiravani discloses as described above, The computer-implemented method according to claim 1. For the remainder of Claim 12, Anushiravani discloses wherein the joint representation(Fig. 14, “Pooling Layer” and “Fully-Connected Layer”; Fig. 3A, Fig. 3B, Fig. 4) preserves the property of differentiability (Fig. 14, “Pooling Layer” and “Fully-Connected Layer”; [0097] “FIG. 14 depicts a CNN as a classification algorithm…”)(Examiner notes that using a CNN preserves the property of differentiability, as evidenced by Chatterjee in “A Basic Introduction to Convolutional Neural Network” at (Page 4, Figure -- “The basic layers of CNN”; [Page 4, 1st Full Paragraph]) ”A simple CNN is a sequence of layers, and every layer of a CNN transforms one volume of activations to another through a differentiable function.”))
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Anushiravani et. al, (US 2020/0151516 A1) in view of Peltonen et. al., (US 2023/0015028 A1).
Regarding Claim 8, Anushiravani discloses as described above, The computer-implemented method according to claim 7. For the remainder of Claim 8, Anushiravani discloses wherein the convolutional neural network of one or more of the first feature extraction units (Fig 3A; “Extract features and concatenate” 304; (Fig. 3B, “convolutional layer” for processing the “time domain data” “A_1” and “A_2”)
Anushiravani does not specifically disclose comprise shortcut connections.
Peltonen teaches a machine learning method for predicting the presence of a malady of the respiratory system using a spectrogram at least one pattern classifier CNN with shortcut connections to classify cough sound features [Abstract]; [0004]. Specifically for Claim 8, Peltonen teaches wherein the convolutional neural network comprise shortcut connections ([0126] “…process used to produce a CNN is to take a pretrained ResNet model, which is a residual network containing shortcut connections, such as ResNet-18, and use the convolutional layers of the model as a backbone…“)
Peltonen provides a motivation to combine at [0127] with “ResNet-18 is a convolutional neural network that is trained on more than a million images from the ImageNet database…has learned rich feature representations for a wide range of images”. A person having ordinary skill in the art before the effective filing date of the claimed invention would recognize that incorporating a CNN with shortcut connections, such as implementing ResNet-18 with the CNN, would be useful for attaining classification of spectrogram features to readily classify cough sounds from audio data visual representations.
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Anushiravani’s disclosed machine learning model with CNN for classifying respiratory event sounds, such as cough, at least in part using spectrogram with the ResNet-18 shortcut connections with CNN taught by Peltonen, creating a single machine learning model for classifying respiratory event sounds, such as cough, with effective feature recognition using shortcut connections with the CNN.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MELISSA J MONTGOMERY whose telephone number is (571)272-2305. The examiner can normally be reached Monday - Friday 7:30 - 5:00 ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Alexander Valvis can be reached at (571) 272 - 4233. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/MELISSA JO MONTGOMERY/Examiner, Art Unit 3791
/PATRICK FERNANDES/Primary Examiner, Art Unit 3791
Prosecution Insights