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
The information disclosure statement (IDS) submitted on April 09, 2025 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1 of the subject matter eligibility test (see MPEP 2106.03).
Claims 1-8 is directed to a “method” which describes one of the four statutory categories of patentable subject matter, i.e., a process.
Claims 9-16 is directed to a “device” which describes one of the four statutory categories of patentable subject matter, i.e., a machine.
Claims 17-20 are drawn to a “non-transitory computer-readable medium” which describes one of the four statutory categories, i.e., a manufacture.
Step 2A of the subject matter eligibility test (see MPEP 2106.04).
Prong One:
Claims 1, 9 and 17 recite (“sets forth” or “describes”) the abstract idea of “a mental process” (MPEP 2106.04(a)(2).III.), substantially as follows: “providing the first physiological signal as an input to a first transformer model; providing the second physiological signal as an input to a second transformer model; providing an output of the first transformer mode land an output of the second transformer model as inputs to a third transformer model; generating an estimated BP value corresponding to the first physiological signal and the second physiological signal based on an output of the at least one BP estimation model”.
In claims 1, 9 and 17, the above recited steps can be practically performed in the human mind, with the aid of a pen and paper or with a generic computer, in a computer environment, or merely using the generic computer as a tool to perform the steps. The transformer model may be as simple as a mathematical equation or a conversion function. A function has an input and an output. Hence, a person would be able to use the first and the second physiological signals as inputs for the conversion function and receive output via mathematical operation, which, depending on the complexity, may be done mentally, with a pen and paper, or with a generic computational device. The output from the first and the second models may be used as an input of the third model, and the output of the third model may be used as an input of the BP estimation model. And at the end, an estimation of the blood pressure may be generated. There is nothing recited in the claim to suggest an undue level of complexity in how the transformer models are constructed and operated and how the blood pressure may be estimated or generated. Therefore, a person would be able to perform the above steps mentally or with a generic computer.
Prong Two: Claims 1, 9 and 17 do not include additional elements that integrate the mental process into a practical application.
This judicial exception is not integrated into a practical application. In particular, the claims recites additional steps of (1) a first sensor for obtaining a first physiological signal; (2) a second sensor for obtaining a second physiological signal; (3) a processor.
The steps in (1) and (2) represent merely data gathering or pre-solution activities that are necessary for use of the recited judicial exception and are recited at a high level of generality with conventionally used tools (see below Step IIB for further details).
The step in (3) represents a conventional tool for implementing the recited steps, and is recited at a high level of generality.
As a whole, the additional elements merely serve to gather and feed information to the abstract idea and to output a notification based on the abstract idea, while generically implementing it on conventionally used tools. There is no practical application because the abstract idea is not applied, relied on, or used in a meaningful way. No improvement to the technology is evident, and the estimated blood pressure is not outputted in any way such that a practical benefit is realized. Therefore, the additional elements, alone or in combination, do not integrate the abstract idea into a practical application.
Step 2B of the subject matter eligibility test (see MPEP 2106.05).
Claims 1, 9 and 17 do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above, the claims recite additional steps of obtaining the first and the second physiological signals and a processor. These steps represents mere data gathering, and a conventional tool that are necessary for use of the recited judicial exception and are recited at a high level of generality.
A signal must be sensed, i.e., “obtained” as recited, by a sensor in order for it to be further processed and used meaningfully. Claims 1, 9 and 17 do not recite any further details in regard to the type of physiological signals and the type of sensors. For the purpose of providing evidence to support that such additional elements are routine, well-known and conventional, Examiner refers to claims 2, 10 and 18 and cites Liu et al., “Continuous blood pressure estimation from electrocardiogram and photoplethysmogram during arrhythmias”. Frontiers in Physiology. 2020 Sep 09, herein after Liu, to evidence a PPG sensor configured for acquiring PPG signals, and an ECG sensor configured for acquiring ECG signals (FIG.1(A) and (B)), and a processor for implementing the data analysis elements (p.4, Col. Left. Model Construction and Validation). Hence, these additional limitations merely represent insignificant, conventional pre-solution activities well-understood in the industry of blood pressure estimation
Accordingly, these additional steps amount to no more than insignificant conventional extra-solution activity. Mere insignificant conventional extra-solution activity cannot provide an inventive concept. The claims hence are not patent eligible.
Dependent Claims
The following dependent claims merely further define the abstract idea and are, therefore, directed to an abstract idea for similar reasons:
describing how the transformer models are trained (claims 3-6, 11-14 and 19-20);
describing the subcomponents of the BP estimation model (claims 7 and 15);
describing further input/output operations (claims 8 and 16).
The following dependent claims merely further describe the extra-solution activities and therefore, do not amount to significantly more than the judicial exception or integrate the abstract idea into a practical application for similar reasons:
describing the sensors and the physiological signals (claims 2, 10 and 18);
Taken alone and in combination, the additional elements do not integrate the judicial exception into a practical application at least because the abstract idea is not applied, relied on, or used in a meaningful way. They also do not add anything significantly more than the abstract idea. Their collective functions merely provide computer/electronic implementation and processing, and no additional elements beyond those of the abstract idea. Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements individually. There is no indication that the combination of elements improves the functioning of a computer, output device, improves technology other than the technical field of the claimed invention, etc. Therefore, the claims are rejected as being directed to non-statutory subject matter.
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 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 1-2, 7-10 and 15-18 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Liu et al., “Continuous blood pressure estimation from electrocardiogram and photoplethysmogram during arrhythmias”. Frontiers in Physiology. 2020 Sep 09, hereinafter Liu.
Claims 1, 9 and 17. Liu teaches in FIG.1 a method for performing cuffless blood pressure (BP) measurement, the method comprising:
obtaining a first physiological signal and a second physiological signal associated with a user (FIG.1(A): experimental scene of simultaneous acquisition of ECG, PPG and IBP signals; and FIG.1(B) typical ECG and PPG waveforms recorded);
providing the first physiological signal as an input to a first transformer model; providing the second physiological signal as an input to a second transformer model (p.3, Col. Right. Feature Extraction: based on the physiological background between BP and corresponding ECG and PPG signals, 15 crucial features (numbered from 1 to 15 and listed in Table 1) containing cardiovascular information were extracted from ECG and PPG signals in each cardiac cycle for BP estimation) – in Table 1, each calculation used for the corresponding feature extraction is considered a model that takes the first physiological signal (ECG signals) and the second physiological signal (PPG signals) as an input and transforms the input into the output (the corresponding extracted feature);
providing an output of the first transformer model and an output of the second transformer model as inputs to a third transformer model (p.4, Col. Left. Model Construction and Validation. Four machine learning algorithms were implemented to establish the relationship between the extracted features and the reference BP values; and p.5, Col. Right, ¶-2: The machine learning algorithms mentioned above were implemented to construct an individual BP estimation model for each patient) – the extracted features are the output of the first and the second transformer model. The “BP estimation model” is the third transformer model as claimed;
providing an output of the third transformer model to at least one BP estimation model (FIG.3: Best Algorithm); and generating an estimated BP value corresponding to the first physiological signal and the second physiological signal based on an output of the at least one BP estimation model (p.5, Col. Left, ¶-2: “the estimated BP (SBP or DBP) with the proposed regression algorithm”; and p.7, Col. Right. Machine Learning Algorithm Selection. Figure 4B: estimated BP values (SBP-RFR and DBP-RFR, marked in red)) – since the BP estimation model is constructed using the extracted features from the ECG and the PPG signals, the estimated BP values from the BP estimation model would corresponds to the ECG and the PPG signals, i.e., “the first physiological signal and the second physiological signal” as claimed.
In regard to claim 9, Liu teaches in FIG.1 the electronic device comprising the first sensor (the ECG electrode on the wrist), and second sensor (the finger-tip PPG sensor).
In regard to claims 9 and 17, Liu teaches the processor and the non-transitory computer-readable medium storing instructions (p.4, Col. Left. Model Construction and Validation. Four machine learning algorithms were implemented…by using the Scikit-learn library in a Python programming environment) – the Python programming environment is considered the “processor” and the “non-transitory computer-readable medium storing instructions executed by at least one processor” as claimed.
Claims 2, 10 and 18. As applied to claim 1, Liu teaches that
the first physiological signal comprises an electrocardiogram (ECG) signal associated with the user, and wherein the second physiological signal comprises a photoplethysmogram (PPG) signal associated with the user (FIG.1(A): experimental scene of simultaneous acquisition of ECG, PPG and IBP signals; and FIG.1(B) typical ECG and PPG waveforms recorded).
Claims 7 and 15. Liu further teaches in p.6, Col. Right. Machine Learning Algorithm Selection
the at least one BP estimation model comprises a systolic BP (SBP) estimation model and a diastolic BP (DBP) estimation model (¶-1: based on its higher performance (i.e., lower RESMs) in BP estimation and faster training capacity compared with the AdaboostR model, the RFR model was selected as the best estimator algorithm; and ¶-2: FIG.4B presents…estimated BP values (SBP-RFR and DBP-RFR, marked in read) by using the RFR model).
Claims 8 and 16. Liu further teaches in p.6, Col. Right. Machine Learning Algorithm Selection
providing the output of the third transformer model to the SBP estimation model; generating an estimated SBP value based on an output of the SBP estimation model (¶-1: The RFR model…in SBP estimation);
providing the output of the third transformer model to the DBP estimation model; and generating an estimated DBP value based on an output of the DBP estimation model (¶-1: The RFR model in DBP estimation), wherein
the estimated BP value comprises the estimated SBP value and the estimated DBP value (¶-1: based on its higher performance (i.e., lower RESMs) in BP estimation and faster training capacity compared with the AdaboostR model, the RFR model was selected as the best estimator algorithm; and ¶-2: FIG.4B presents…estimated BP values (SBP-RFR and DBP-RFR, marked in read) by using the RFR model).
Claim Rejections - 35 USC § 103
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 set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied 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.
Claims 3, 11 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Liu in view of Velo et al., US 2020/0100693 A1, hereinafter Velo.
Claims 3, 11 and 19. Liu further teaches that
the first transformer model, the second transformer model, the third transformer model, and the at least one BP estimation model are trained using a pre-training process and a user-specific training process, wherein the pre-training process is performed using a first training dataset corresponding to a plurality of users, and wherein the user-specific training process is performed based on a second training dataset corresponding to the user (FIG.3: A: the training set and B: the training is performed for the data set from patients 1 to 35; and p.5, Col. Left, the four machine learning algorithms implemented – in each of the four models (1)-(4), the pre-training and the training process are described).
Each of the models are built upon based on the extracted features from the PPG and the ECG signals, i.e., the output of the first, the second, and the third transformer models. The PPG and the ECGs signals are pre-processed, i.e., calibrated based on the IBP reference, as taught in p.3, Col. Right, ¶-1. This is considered the pre-training process of the transformer models.
In regard to the feature of the first transformer model, the second transformer model, the third transformer model being trained using a user-specific training process, in an analogous PPG/ECG based BP estimation field of endeavor, Velo teaches such a feature in claim 1: using at least one of a PPG based statistical or machine learning model to analyze the PPG signal obtained using the optical sensor…using at least one of an ECG based statistical or machine learning model to analyze the ECG signal obtained using the electrodes of the ECT sensor.
Therefore, it would have been obvious to one of the ordinary skilled in the art before the effective filing date of the claimed invention to have the device, the method or the program of Liu employ such features associated with the training of the transformer models as taught in Velo for the conventionally known advantage of allowing a more accurate analysis to be performed.
Claims 4-5, 12-13 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Liu in view of Velo, further in view of Murthy et al., US 2018/0247107 A1, hereinafter Murthy.
Claims 4, 12 and 20. Liu and Velo combined teaches all the limitations of claim 3, yet neither teaches the claimed feature.
However, in an analogous machine learning model based medical data analysis field of endeavor, Murthy teaches that
at least one of the pre-training process and the user-specific training process is performed using a weighted contrastive loss function comprising a similarity metric ([0060]: to minimize such miss-classification errors during data-splitting, the softmax-loss is augmented with an error-driven, weighted contrastive loss function that helps block diagonalization of the confusion matrix; and [0061]: minimizing the weighted contractive loss results in a similarity metric of samples belonging to the same cluster to be small).
Therefore, it would have been obvious to one of the ordinary skilled in the art before the effective filing date of the claimed invention to have the device, the method or the program of Liu and Velo combined employ such features associated with the training of the models as taught in Murthy for the advantage of “minimize miss-classification errors”, as suggested in Murthy, [0060].
Claims 5 and 13. Murthy further teaches that
the similarity metric indicates a similarity between a first ground truth BP value corresponding to a first training sample and a second ground truth BP value corresponding to a second training sample ([0030]: the classification results for the training samples in the validation dataset are compared with the ground truth classifications for the training samples in the validation dataset to determine which training samples in the validation dataset have been incorrectly classified; and claim 13: the initial trained deep-network classifier is fine-tuned using a weighted contrastive loss function to penalize mix-classifications of training images into classes not within the same cluster as ground truth classes for the training images).
Claims 6 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Liu in view of Velo and Murthy, further in view of Vogler et al., US 2024/0403625 A1, hereinafter Vogler.
Claims 6 and 14. Liu, Velo and Murthy combined teaches all the limitations of claim 4.
Neither Liu nor Velo teaches the claimed feature.
However, in an analogous machine learning model based medical data analysis field of endeavor, Murthy teaches that
the similarity metric is used to cluster training samples included in at least one of the first training dataset and the second training dataset ([0030]: the classification results for the training samples in the validation dataset are compared with the ground truth classifications for the training samples in the validation dataset to determine which training samples in the validation dataset have been incorrectly classified; and claim 13: the initial trained deep-network classifier is fine-tuned using a weighted contrastive loss function to penalize mix-classifications of training images into classes not within the same cluster as ground truth classes for the training images).
Therefore, it would have been obvious to one of the ordinary skilled in the art before the effective filing date of the claimed invention to have the device, the method or the program of Liu and Velo combined employ such features associated with the similarity metrics as taught in Murthy for the advantage of “minimize miss-classification errors”, as suggested in Murthy, [0060].
Neither of Liu, Velo nor Murthy teaches that the training occurred in an embedding space.
However, in an analogous machine learning model training field of endeavor, Vogler teaches that the model training is in an embedded space ([0206]: the architecture of a machine learning model can be used to learn multimodal representation of data of two different modalities. If representations of data or three or more different modalities are to be generated by a machine learning model, such a machine learing model can comprise a third (and fourth, fifth,…) encoder and a third (and forth, fifth,…) decoder. All encoders are merged into one embedding: the joint representation of the different input data of different modality).
Therefore, it would have been obvious to one of the ordinary skilled in the art before the effective filing date of the claimed invention to have the device, the method or the program of Liu and Velo combined employ such features associated with the model training being in an embedded space as taught in Vogler for the advantage of “automatically generating an overall picture of a patient that is as complete as possible based on the available data across multiple modalities”, as suggested in Vogler, [0005].
Response to Arguments
Applicant’s arguments in regard to the rejections under 35 U.S.C. 101, 102(a)(1) and 103 have been fully considered but they are not persuasive. Applicant’s assertion and Examiner’s considerations are stated below.
In regard to the rejection to claims 1, 9 and 17 under 35 U.S.C. 101, Applicant cited MPEP 2106.04(d) and asserted that the specification clearly articulates an improvement to the technologies, and the improvement is properly reflected in the claims, for example in the claimed arrangement of transformer model (emphasis added). Examiner respectfully disagrees and notes the following:
MPEP2106/04(d), as cited by the Applicant in the Remarks p.9, states (emphasis added):
A claim reciting a judicial exception is not directed to the judicial
exception if it also recites additional elements demonstrating that
the claim as a whole integrates the exception into a practical
application. One way to demonstrate such integration is when the
claimed invention improves the functioning of a computer or
improves another technology or technical field.
Examiner acknowledges that the claim should be considered as a whole to determine if it is patent eligible. Examiner would like to remind the Applicant that when considering whether the claim as a whole integrates the exception into a practical application, MPEP refers to “additional elements” that should be considered whether it integrates the judicial exception into a practical application, not the abstract idea itself (see the above underlined content). In the rejection, Step 2A, Prong Two analysis indicates that the additional elements are identified to be (1) a first sensor for obtaining a first physiological signal; (2) a second sensor for obtaining a second physiological signal; (3) a processor, while the asserted feature. However, Applicant asserted that the improvement is properly reflected in the claimed arrangement of transformer model (see the above bolded content). In the rejection, the arrangement of the transformer models is identified as part of the mental process (see Step 2A, Prong One analysis), hence would not be properly considered whether it would integrate the judicial exception (i.e., itself) into a practice application.
Applicant is suggested to address either the identified abstract idea is not a mental process for the Step 2A Prong One analysis, or the identified additional elements are sufficient to integrate the judicial exception for the Step 2A Prong Two analysis.
In regard to the rejection to claims 1, 9 and 17 under 35 U.S.C. 102(a)(1), Applicant asserted on p.13, ¶-1 that “as can be seen in FIG. 7A, a model 710 may include…a merged transformer model 712…Then, an output of the merged transformer model may be provided as an input to the SBP estimation model 713A and the DBP estimation model 713B”. Applicant further asserted on pp.14-15 that “Liu fails to disclose that the outputs of any BP estimation model are provided as inputs to the best algorithm. Instead, as discussed above, the four algorithms were trained using one dataset as input, and validated using another dataset as input, and then one of the four algorithms was further evaluated using a third dataset as input. Therefore, the cited references fail to disclose or suggest "providing the first physiological signal as an input to a first transformer model; providing the second physiological signal as an input to a second transformer model; providing an output of the first transformer model and an output of the second transformer model as inputs to a third transformer model; [ and] providing an output of the third transformer model to at least one BP estimation model," as claimed in claim 1.” Examiner respectfully disagrees and notes the following:
A “model”, by definition found in Merriam-Webster Dictionary, is a system of data presented as a mathematical description of an entity. A mathematical equation, a mathematical function, or a simulation are examples of a model. A “transformer model” hence is broadly interpreted as a function that transforms an input to an output.
The ECG and the PPG signals are the first input and the second input. In Table 1, each of the calculations to extract its corresponding feature is the first transformer model and the second transformer model, and th corresponding extracted features are the first output and the second output.
The selection of the 4 regression algorithms are the third model. The extracted features are entered as the third input to evaluate the 4 algorithms. The selection outcome of which algorithm is the best algorithm is the output. The selected best algorithm is provided as the BP estimation model for SBP and DBP estimation.
Examiner notes that (1) though the specification discloses that the first model and the second model are merged to the third model, this feature is not yet reflected in the claim. In the claim, the 4 transformer models are recited as 4 independent models; and (2) the first and the second output are provided as inputs to the third model, however, the third output is recited to be merely provided to the at least one BP estimation model (without being specified that it is provided as an input).
Based on the above considerations, the arguments are not sufficient to overcome the rejections, and claims 1-20 remain rejected.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to YI-SHAN YANG whose telephone number is (408) 918-7628. The examiner can normally be reached Monday-Friday 8am-4pm PST.
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, Pascal M Bui-Pho can be reached at 571-272-2714. 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.
/YI-SHAN YANG/Primary Examiner, Art Unit 3798