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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 12/2/2025 has been entered.
Response to Arguments
Applicant's arguments filed 11/7/2025 regarding the 35 USC 112(a), 101 and 102 rejections have been fully considered but they are not persuasive.
Regarding the 35 USC 112(a) rejection, Applicant argues on p. 4 under the Remarks that “Specifically, Fig. 3 (at right) discloses the specific steps performed by the claims, rather than describing a “black-box algorithm”. There are no particular steps to train the Cycle-GAN to learn ECG restoration”. The Examiner respectfully disagrees.
Figure 3 states at step 302 “Adapt cycle-GAN to transform ECG signals” with no details how the cycle-GAN is particularly adapted. Furthermore, the term “adapt” indicates that the cycle-GAN must be transformed or trained some particular way to be able to transform ECG signals. Claim 1 further states “transforming, by the device, at least one of a one-dimensional or two-dimensional version cycle-consistent adversarial networks trained to transform electrocardiogram signals…”. The positive recitation of “trained to transform electrocardiogram signals” indicates there are steps needed to adapt the cycle-CAN to learn ECG restoration and the term “transforming” also indicates the cycle-CAN is somehow transformed to provide the ECG restoration. Par. [0027] of the specification simply repeats the intended function of element 302 in Fig. 3 and does not lay out any particular steps indicating how applicant transforms a cycle-CAN to provide the claimed function.
Applicant further argues on pp.4-5, “One of ordinary skill in the art, for instance a MSc student who is familiar with machine learning, could understand from the specification how to train an 1D Cycle-GAN model, as recited by the claims”. However, as note din MPEP §2161.01, “It is not enough that one skilled in the art could write a program to achieve the claimed function because the specification must explain how the inventor intends to achieve the claimed function to satisfy the written description requirement. See, e.g., Vasudevan Software, Inc. v. MicroStrategy, Inc., 782 F.3d 671, 681-683, 114 USPQ2d 1349, 1356, 1357 (Fed. Cir. 2015).”
Regarding the 35 USC 101 rejection, Applicant again argues on p. 6 under the Remarks, “Such functionalities are not merely mathematical concepts and relationships since they require transformation and restoration of electrical signals, which are not manual nor mental processes.” It appears applicant is conflating two separate abstract ideas into one. The Examiner did not assert the abstract idea as being a mental process so any arguments regarding what can or cannot be performed mentally are moot. The Examiner instead asserted the limitations of ”transforming… at least one of a one-dimensional or two-dimensional version cycle-consistent adversarial networks trained to transform electrocardiogram signals form at least one different dataset”; and “restoring…the at least one set of corrupted electrocardiogram segments based upon a one- or two-dimensional operational cycle-generative adversarial network trained over the batches without any tuning or pre-processing, wherein restoring the at least one set of corrupted electrocardiogram segments is according to a random blend of at least one of a diminished Q, R and S (QRS) wave amplitude, noise, signal cuts, baseline wandering, and bias” are mathematical concepts since, at their core, machine learning networks, neural networks, etc. are a series of mathematical correlations between data inputs and nodes to predict a data output. Therefore, the “transforming” and “restoring steps” above amount to mathematical correlations of ECG data points, wherein mathematical concepts and relationships are not patent eligible.
Applicant’s arguments on p. 7 regarding the process being “fully automatic” are not persuasive in that the claim only requires generic computer implementation of the abstract idea. Applicant’s arguments regarding diagnosing hidden/undetected arrhythmia events from the restored ECG are not persuasive in that the claims do not require diagnosing any arrythmias and instead state “configured for at least one of arrhythmia classification or peak detection”, This is a passive claim recitation and does not require the active steps of diagnosis as argued. Even if the output of a classification where claimed in generic terms, this would most likely result in the insignificant, extra-solution activity of data reporting.
Regarding step 2B, Applicant argues an improvement to computer capabilities. However, any purported improvement is in the abstract idea itself and not with regards to any additional elements outside of the abstract idea. The rejection is maintained.
Regarding the 35 USC 102 rejection with respect to Mohebbanaaz et al. (“Removal of Noise from ECG Signals using Residual Generative Adversarial Network”), herein “Moh”, Applicant argues “Applicant submits that additive noise is only one type of artifact that can corrupt the ECG signals, and the noise type is not necessarily Gaussian nor independent from the signal. The noise severity (variance) cannot be known in advance either. In contrast, the claims restore the ECG signal corrupted by a random blend of several artifacts such as noise (with any type and severity), cuts, baseline wandering, bias, etc., and therefore are a blind ECG restoration method, not a simple ECG denoising, as disclosed by Mohebbanaaz.” One issue this claim and argument presents is, if the noise in the signal is truly random, it is unclear how the claim can assert the noise is a particular combination of diminished QRS wave amplitude, noise, signal cuts, baseline wandering, and bias as claimed. Something cannot be random and ordered at the same time. His language has introduced a 35 USC 112b rejection for indefiniteness. To the Examiner’s bets understanding of the claimed invention, it is understood that noise is prevalent in any electrical signal, such as an ECG signal and the noise can have various unknown sources. Moh likewise acknowledges this and list some various types of noise that can be present: baseline wander, powerline interference, artifacts, motion of electrodes, etc. (I. Introduction). Moh then states other channel noises like additive white gaussian noise can be present (I. Introduction). Moh then discloses utilizing GANs since they “are able to remove various noise at various recording conditions at the same time (I. Introduction). Therefore, the Examiner maintains Moh discloses utilizing GANs to better remove the various, random noises that can be present in an ECG signal, which is what Applicant is currently claiming to the best understanding of the claims scope in light of Applicant’s specification. Regarding the language “without any tuning or pre-processing” the Examiner notes this is performed by the GAN after it is trained. This does not indicate the training is performed with random noise. However, the white gaussian noise used by Moh to train the GAN is still random but again, the claim language has no bearing on the training process. Moh does disclose utilizing the trained GAN to remove random noise from ECG signals after it is trained. Moh utilizes a known injected noise to calibrate the GAN but envisions applying the GAN as a denoising algorithm after it is trained (“This motivated us to propose a network based on GAN to remove noise from ECG signal”). This is the purpose of the entire study. Moh tested the hypothesis that it could be done; confirmed it and thus, the next logical step would be to apply it.
Regarding the 35 USC 102 rejection with respect to Antczak (“A Generative Adversarial Approach To ECG Synthesis And Denoising”), Applicant argues “Antczak discloses that generative adversarial networks (GAN) are known to produce synthetic data that are difficult to discern from real ones by humans. Antczak discloses ECG data augmentation, using ECG synthesis to produce realistically looking ECG signals, but cannot be applied for blind ECG restoration, as recited in the claims.” However, the Examiner notes that the claims do not recite any limitations requiring blind source separation as argued. Instead, the claims only require transforming a corrupted electrocardiogram segment to a “restored” state using a GAN. Antczak accomplishes the claimed function by taking in a nosy (corrupted signal) and producing a denoised signal (see Fig. 4). Applicant has not defined in the claim any particular source of the ECG data so the claim covers the synthetic data relied on by Antczak. There is nothing in the claim precluding synthetic data. Furthermore, Antczak discloses the noisy ECG signal includes multiple sources of noise and therefore a blend of nose as claimed (p. 3, Noise Model).
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1 and 2 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
When dealing with computer-implemented inventions, MPEP §2161.01 states:
“Similarly, original claims may lack written description when the claims define the invention in functional language specifying a desired result but the specification does not sufficiently describe how the function is performed or the result is achieved. For software, this can occur when the algorithm or steps/procedure for performing the computer function are not explained at all or are not explained in sufficient detail (simply restating the function recited in the claim is not necessarily sufficient). In other words, the algorithm or steps/procedure taken to perform the function must be described with sufficient detail so that one of ordinary skill in the art would understand how the inventor intended the function to be performed. See MPEP §§ 2163.02 and 2181, subsection IV.”
The present claim set requires an algorithm, namely a one or two-dimensional operational cycle-generative adversarial network, to restore a set of corrupted electrocardiogram segments. The specification of the application sets forth examples of the types of signals the algorithm can produce (Fig. 2a-2f) but fails to disclose any particular process or details of process the algorithm actually performs to obtain these results. The claim also states that the adversarial network is “trained over the batches” but the specification does not set forth any particular process by which this algorithm is trained. Par. [0029] of the specification states, “At step 303, once a one- or two-dimensional operational cycle-GAN is trained over the batches, the generator self-ONN trained for the “corrupted” to “clean” ECG segment transformation can then be used for the ECG restoration”. The Examiner notes this is an intended/functional result but does not amount to a sufficient description of how the algorithm actually works to achieve those results. Applicant has essentially claimed and described a black-box algorithm that lacks a sufficient description regarding how this algorithm actually works to produce the desired result.
Additionally, the Examiner notes, “It is not enough that one skilled in the art could write a program to achieve the claimed function because the specification must explain how the inventor intends to achieve the claimed function to satisfy the written description requirement. See, e.g., Vasudevan Software, Inc. v. MicroStrategy, Inc., 782 F.3d 671, 681-683, 114 USPQ2d 1349, 1356, 1357 (Fed. Cir. 2015).”, See MPEP §2161.01.
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 and 2 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 recites the limitation "the batches" in line 11 of the claim. There is insufficient antecedent basis for this limitation in the claim.
Claim 2 has the same issue.
Claim 1 recites, “…according to a random blend of at least one of diminished Q, R, and S (QRS) wave amplitude, noise, signal cuts, baseline wandering, and bias.” It is unclear if the phrase “at least one” only applies to the diminished Q, R, and S (QRS) wave amplitude or if the phrase “at least one” extends to noise, signal cuts, baseline wandering, and bias. The combination of features claimed is presently unclear.
Additionally, Applicant claims a “random blend” of “diminished Q, R, and S (QRS) wave amplitude, noise, signal cuts, baseline wandering, and bias.” Applicant is defining a “random” set as a set of knowns, which would appear to contradict the blend being random. Something cannot be both random and categorized/known. Also, the specification fails to support the phrase “random blend” and instead uses the term “blend”, see par. [0021]. The Examiner suggest amending the claim to maintain consistent language with the originally filed disclosure.
Claim 2 contains the same issues.
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 and 2 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1
Claims 1 and 2 recite a method and apparatus
Step 2A, Prong 1
Claims 1 and 2 recite the abstract idea of ”transforming… at least one of a one-dimensional or two-dimensional version cycle-consistent adversarial networks trained to transform electrocardiogram signals form at least one different dataset”; and “restoring…the at least one set of corrupted electrocardiogram segments based upon a one- or two-dimensional operational cycle-generative adversarial network trained over the batches without any tuning or pre-processing, wherein restoring the at least one set of corrupted electrocardiogram segments is according to a random blend of at least one of a diminished Q, R and S (QRS) wave amplitude, noise, signal cuts, baseline wandering, and bias”. These fall under the mathematical concept abstract idea grouping since, at their core, machine learning networks, neural networks, etc. are a series of mathematical correlations between data inputs and nodes to predict a data output. Therefore, the “transforming” and “restoring steps” above amount to mathematical correlations of ECG data points, wherein mathematical concepts and relationships are not patent eligible.
Step 2A, Prong 2
Claims 1 and 2 do not include any additional elements that integrate the abstract idea into a practical application. Claims 1 and 2 include the step of “selecting…at least one set of clean electrocardiogram segments, and at least one set of corrupted electrocardiogram segments. This is insignificant, extra-solution activity (mere data gathering) Claims 1 and 2 also recite “wherein the transformed electrocardiogram signal are used as a baseline and configured for at least one of arrhythmia classification or peak detection.” This is not an active method step and reads like the signal is made available for these functions, which would fall under data gathering/reporting as well, which amounts to insignificant, extra-solution activity. Additionally, Claim 2 includes a processor and memory, which are recited at such a high level of generality that it amounts only to generic computer implementation of the abstract idea.
Step 2B
Claims 1 and 2 do not include any additional elements that amount to significantly more than the abstract idea itself. Claims 1 and 2 include the step of “selecting…at least one set of clean electrocardiogram segments, and at least one set of corrupted electrocardiogram segments”. This is insignificant, extra-solution activity (mere data gathering). Claims 1 and 2 also recite “wherein the transformed electrocardiogram signal are used as a baseline and configured for at least one of arrhythmia classification or peak detection.” This is not an active method step and reads like the signal is made available for these functions, which would fall under data gathering/reporting as well, which amounts to insignificant, extra-solution activity. Additionally, Claim 2 includes a processor and memory, which are recited at such a high level of generality that it amounts only to generic computer implementation of the abstract idea.
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.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 1 and 2 ae rejected under 35 U.S.C. 102(a)(1) and 102(a)(2) as being anticipated by Mohebbanaaz et al. (“Removal of Noise from ECG Signals using Residual Generative Adversarial Network”), herein “Moh”. Moh discloses utilizing a computerized system (processor and memory) to obtain clean ECG signals from a database, such as the MIT-BIH arrhythmia database (p.3, “III. Results and Discussion”); training the GAN on these signals (p. 3, “The proposed model is trained until the required minimum loss is obtained”); and restoring a corrupted ECG signal using the GAN (the corrupted ECG is fed to G in Fig. 1; see also p.2-3 discussion in general).
Claims 1 and 2 ae rejected under 35 U.S.C. 102(a)(1) and 102(a)(2) as being anticipated by Antczak (“A Generative Adversarial Approach To ECG Synthesis And Denoising”). Antczak discloses utilizing a computerized system (processor and memory) to denoise ECG data (denoising is equivalent to removing corruption and thus restoring ECG data) by relying on a GAN that is trained on clean ECG data and then implemented, after training, to restore ECG data to a more accurate state (see pp. 1-2, see also restored result in Fig. 4 under “ECG-GAN Denoiser” in which the signal is “restored” from the noised signal).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALLEN PORTER whose telephone number is (571)270-5419. The examiner can normally be reached Mon - Fri 9:00-6:00 EST.
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, Carl Layno can be reached at 571-272-4949. 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.
/ALLEN PORTER/Primary Examiner, Art Unit 3796