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 .
Response to Amendment
The amendment filed on 11/04/2025 has been entered. Claims 1-6 have been amended. Claims 1-6 remain pending.
The previously raised objections are withdrawn because the issues have been properly corrected.
On Pages 6-7 of Remarks, the Applicant states that amendments for Claims 1-6 resolve the issues previously raised under 35 U.S.C. 112(f) and 35 U.S.C. 112(a).
As the Applicant states, “one or more processors and a memory” has been added and provides sufficient structure to perform the claimed functions, which as the Applicant stats should resolve the issues previously raised under 35 U.S.C. 112(f). The Examiner respectful disagrees. The above recited “one or more processors and a memory” is not disclosed anywhere in Specification. Even if “one or more processors and a memory” is disclosed in Specification, generic processor and memory as recited is not sufficient structure for performing the functions. However, as shown in the section of 35 U.S.C. 112(f), some previously raised issues are resolved by some other amendments.
As the Applicant states, the data receiving unit is disclosed as receiving signals and values from a PPG sensor and a non-invasive blood pressure measurement device, which are newly added in the amended Claim 1 and as the Applicant states should resolve the issue previously raised under 35 U.S.C. 112(a). The Examiner respectful disagrees, because the recited PPG sensor and blood pressure measurement device are modules or units that acquire or send the data, but not module that receives the data.
As the Applicant states, the derived variable extraction unit is described as generating a time-related variable based on the most recent non-invasive blood pressure reading (e.g. the time elapsed since the most recent blood pressure measurement), which as the Applicant states is a concrete algorithmic basis for the derived variable extraction unit. The Examiner respectful disagrees. In the amended Claim 1, Lines 18-19, the newly added limitation of “the derived variable represents an elapsed time” is not disclosed in Specification. Instead, Specification discloses that the derived variable is generated or calculated based on or according to the elapsed time. Therefore, it is still unclear what structure (either specific algorithm or method) is used for obtaining the derived variable.
On Page 8 of Remarks, the Applicant states that amendments for Claims 1-5 resolve the issue previously raised under 35 U.S.C. 112(b). The Examiner agrees and withdraws the rejection.
Response to Arguments
35 U.S.C. 101 rejections
On Pages 8-12 of Remarks, the Applicant argues that when considered as a whole, Claims 1-6 are directed to a specific improved technological process, but not a mere mental process or mathematical algorithm. In the following, the Examiner respectfully responds to the specific arguments, and more details can be found in the section of 35 U.S.C. 101 rejection.
As the Applicant argues (Pages 8-10), the PPG signal and blood measurement are specific types of medical sensor data, a person could not mentally take the data and calculate a continuous blood pressure waveform, and the deep learning models could not be implemented mentally and are recited with specific processing features. The Examiner respectfully disagrees. PPG and blood pressure measurements are routinely performed in clinical practice, and are common function for many wearable devices such as watch, so including such measurement data would not provide an inventive concept. For one of ordinary skill in the art, e.g. a clinician or anyone who regularly read PPG and blood pressure diagrams, it is not difficult to mentally estimate a blood pressure waveform based on a PPG waveform and one or more cuff-measured blood pressure values. The Examiner agrees that deep learning models could not be implemented mentally; however, the models are recited in the claims and even in Specification in a very generic way. In the amended Claims 1 and 3, the limitation of cross-attention component is newly added. First, as the concept of cross attention was invented multiple years ago and has been advanced rapidly, the concept has evolved into multiple variations. Second, in both the claims and Specification, the term of “cross-attention” is merely mentioned, without any detail or specification on how such mechanism is applied to PPG signals.
As the Applicant argues (Pages 10-11), the claims are integrated into a practical application, and certain steps (such as receiving data and converting and transmitting results) are not insignificant extra-solution activities. The Examiner respectfully disagrees. The application is about estimating continuous arterial blood pressure. First, according to MPEP 2106.05(h), generally linking the use of a judicial exception to a particular technological environment or field of use does not integrate a judicial exception into a practical application. Second, to estimate continuous blood pressure, receiving data and transmitting result are indeed necessary steps, but it is obvious that how these steps are performed would not possibly affect the estimation of blood pressure waveform. Therefore, the abovementioned steps are regarded as insignificant extra-solution activities.
As the Applicant argues (Pages 11), the claims are not directed to a judicial exception, because they recite an ordered combination of elements, use specific deep learning techniques, and enable enhanced accuracy and early disease prediction. The Examiner respectfully disagrees. As discussed above and in the section of 35 U.S.C. 101, the claims recite an abstract idea, which involves limitations that are performed by either mental process and/or mathematical calculations.
As the Applicant argues (Pages 11-12), the claims include “significantly more”, because the combination of receiving data, applying deep learning models and outputting results is not routine, and the recited deep learning implementation is not generic (e.g. use of a cross-attention neural network). The Examiner respectfully disagrees. The procedure of receiving data, applying analysis method and outputting results is routine and conventional, so that it does not provide an inventive concept. As discussed above, the deep learning implementation (even including the newly added limitation of “cross-attention”) is recited in a very generic way.
35 U.S.C. 103 rejections
On Pages 12-17 of Remarks, the Applicant argues that the 35 U.S.C. 103 rejections for Claims 1-6 should be traversed. In the following, the Examiner respectfully responds to the specific arguments, and more details can be found in the section of 35 U.S.C. 103 rejection.
On Page 13 of Remarks, the Applicant argues that a) the reference Addison does not teach a separate processing unit solely to compute a derived variable from the NIBP reading, and that b) the reference Wang uses the cuff measurement directly for calibration, and a person of ordinary skill in the art (POSITA) would not be motivated to use such derived variable (such as elapsed time since last cuff reading) for modification. The Examiner respectfully disagrees. a) In Addison, it is explicitly disclosed that "features 604 … the elapsed time at the current time since the time of the most recent calibration point" can be determined by the "CNIBP monitoring device 100" (Addison, Para 0097; "CNIBP monitoring device 100 may also determine features 604 as the value of the elapsed time at time t since the most recent calibration point. For example, CNIBP monitoring device 100 may derive feature 604 by subtracting time t from the time of the most recent calibration point."). b) In reference Wang, the cuff measurement (real SBP/DBP in Fig. 1) is used to obtain "calibrated-estimated SBP/DBP" (see Fig. 1). In other words, the cuff measurement is used for calibrating blood pressure at its systolic and diastolic time points only, but not any other time points. Further, the calibrated-estimated SBP/DBP combines with the estimated PVR waveform to estimate central artic blood pressure using a linear regression method. Obviously, in this method of Wang, the accuracy of estimated blood pressure at time points other than systolic and diastolic points would largely depend on the accuracy of the "PPG to PVR transformer 15", a model trained on other subjects' data. Now that the patient's real SBP/DBP measured by cuff is available, it is obviously beneficial to consider the option of using such measurement to assist estimating blood pressure at other time points for potentially higher accuracy, and for such estimation, it is obvious to first determine how far away the time point to be predicted is from the cuff measurement. This would motivate one of ordinary skill in the art to consider modification by the reference of Addison, in which both blood pressure measured at the most recent calibration point and an elapsed time since the most recent calibration point are used as input to estimate blood pressure from PPG signals (Addison, Para 0056).
On Pages 14 of Remarks, the Applicant argues that c) Addison does not explicitly teach a separate modeling stage that takes prior outputs and new inputs to generate a waveform, and that d) the structural arrangement of three sequential learning models in Claim 1 is not taught or suggested by Addison, and it would be a significant redesign by combining Wang's two model plus regression approach with Addison's neural network approach. The Examiner respectfully disagrees. c) The limitation in original Claim 1 recites "a blood pressure waveform prediction unit configured to input the primary arterial blood pressure waveform and the second feature values to a third learning model, and predict the continuous arterial blood pressure waveform". This can be interpreted to be the same as Addison's Fig. 8 for a conceptual diagram of an example continuous non-invasive blood pressure (CNIBP) monitoring device. Specifically, the entire network shown in Fig. 8 may correspond to "third learning model"; "features 600" may correspond to "primary arterial blood pressure waveform" (Addison, Para 0078, "Features 600 may be values of a set of metrics derived from a PPG signal over a specific time period", indicating "features 600" to be "prior outputs" from some prior processing of PPG signal); "features 604" may corresponds to "second feature value". More details can be found in Addison, Para 0087-0098. d) First, the three learning models in Claim 1 is not sequential as argued above: the first learning model need not to be implemented before the second learning model. Wang discloses a structural arrangement very similar to Claim 1, with a difference that instead of outputting a continuous arterial blood pressure waveform, Wang directly output two clinically important metrics (SBP-C and PP-C; see Para 0089) as an example. To output the entire blood pressure waveform, it would be natural to modify Wang by replacing the relatively simple "CBP estimator" by Addison's method. Both Wang and Addison aim to obtain a continuous measurement of blood pressure, and both the references achieve the goal by analyzing PPG signals. Wang discloses "A continuous measurement of BPs is necessary for medical monitoring and diagnosis by physicians" (Para 0004), and "The method or system is comfortable, continuous, and suitable for all-day use, and the accuracy of estimation of BPs and/or CBPs are sufficiently improved" (Para 0008). Addison discloses "Continuous non-invasive blood pressure (CNIBP) monitoring systems allow a blood pressure of a patient to be tracked continuously … The CNIBP monitoring system may periodically recalibrate the CNIBP monitoring system based on blood pressure measured by a non-invasive blood pressure monitoring system, such as an inflatable cuff-type blood pressure monitoring system." (Para 0001).
On Pages 15 of Remarks, the Applicant argues that the combination of Wang and Addison does not output "blood pressure state" as claimed. The Examiner respectfully disagrees. When a patient is diagnosed as normal state versus abnormal state based on a physiologic parameter, we typically do not consider the state and the parameter value being equal, particularly for those parameters that are not quantitative enough. However, blood pressure is such a well-quantified parameter that its diastolic and systolic values fully define whether a patient is normal or hypertensive. For example, a measurement of less than 120/80 mmHg is normal. In addition, under a broadest reasonable interpretation, a set of systolic and diastolic blood pressure values, which immediately indicates to one of ordinary skill in the art what a patient's blood pressure state is at, is one informative type of blood pressure state.
On Pages 15-16 of Remarks, the Applicant further argues that the reasoning for combining Wang and Addison is flawed or insufficient. The Examiner respectfully disagrees. As acknowledged by the Applicant, Wang indeed uses a "multi-model system", similar to the claimed method, with the differences that Wang does not determine a derived variable based on elapsed time and does not explicitly output a continuous blood pressure waveform. For the issue of elapse time, Addison realizes the importance of such elapse time. Fig. 4 of Addison demonstrates that calibration of estimated blood pressure using a recent measurement (at time t0, t1 or t2) can substantially increase accuracy of the estimated blood pressure. For the issue of outputting a continuous blood pressure waveform, as Addison proposes, with measured blood pressure and the elapse time from such measurement as input, blood pressure at non-measurement time points can be accurately estimated from PPG signals. Hence, it is obvious for one of ordinary skill in the art to modify Wang by Addison to accurately estimate a blood pressure waveform from PPG signals.
On Page 16 of Remarks, the Applicant argues that several technical obstacles would arise in attempting to combine Wang and Addison, and that specifically Addison's neural network would overlap in function with Wang's GPR model and linear regression. The Examiner respectfully disagrees. Addison's neural network would not necessarily overlap in function with Wang's GPR model and linear regression. Wang's GPR model is used to calibrate estimated SBP/DBP (see Fig. 1 of Wang), which is a calibration on the magnitude or intensity of the signal; on the other hand, the elapse time as determined and used in Addison measures how far a predicted time point is away from the recent cuff measurement, so would largely impact along the time axis. Hence, the two pieces of information (calibrated estimated SBP/DBP and accurate elapse time) can be naturally combined to achieve more accurate estimation of blood pressure. Wang's linear regression is a method for deriving characteristic blood pressure values from a continuous waveform, which is different from Addison's neural network of estimating a continuous blood pressure waveform. Therefore, Wang can be modified by Addison without significant technical obstacles.
On Page 17 of Remarks, the Applicant argues that e) Addison's mention of using an elapsed time feature does not amount to a suggestion that one must do so, nor how to integrate it with Wang's method, and that f) Addison does not specifically teach constructing the third learning model. The Examiner respectfully disagrees. e) In Addison, the phrase "elapsed time … since the most recent calibration time" is recited for about 40 times, and in abstract ("… determine, using the continuous non-invasive blood pressure model and based at least in part on inputting the calibration data determined at the calibration point, the values of the set of metrics, and an elapsed time at the particular time since the calibration point into the continuous non-invasive blood pressure model, a blood pressure of the patient at the particular time."). With such disclosure, one of ordinary skill in the art would consider using such elapsed time feature, and would not consider its integration with Wang's method to be challenging. f) Addison discloses training the model in Paras 0048-0059, and shows the training system in Fig. 2.
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.
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 limitations are:
“a data receiving module configured to receive …” in Claim 1.
A review of the Specification discloses indicates that no structural information of the “data receiving module” is provided in the Specification (also refer to the section of 35 U.S.C 112(a)).
“a photoplethysmography analysis module comprising a first neural network model configured to process the PPG signal …” in Claim 1.
A review of the Specification discloses that the corresponding structure for “a photoplethysmography analysis module comprising a first neural network model” is formed of “a deep learning-based algorithm such as Resnet algorithm, U-net algorithm, Attention algorithm, and Self-Attention algorithm” (Para 0042).
“a derived variable extraction module configured to compute a derived variable …” in Claim 1.
A review of the Specification discloses indicates that no structural information of the “derived variable extraction module” is provided in the Specification (also refer to the section of 35 U.S.C 112(a)).
“a blood pressure state prediction module configured to predict a blood pressure state of the patient using the continuous arterial blood pressure waveform” in Claim 1.
A review of the Specification discloses that the corresponding structure for a “blood pressure state prediction unit” is formed of determining systolic and diastolic blood pressure from a blood pressure waveform, and classifying blood pressure state using the determined systolic and diastolic blood pressure (Para 0055).
Because these claim limitations are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, they 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 these limitations interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitations to avoid 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 limitations recite sufficient structure to perform the claimed function so as to avoid 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 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 1-6 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 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.
Claim 1, Lines 2-9, recites “… monitoring system, comprising a photoplethysmography (PPG) sensor … a non-invasive blood pressure measurement device … and one or more processors and a memory …”. The Specification does not disclose any detail on the recited sensor, device, processors and memory or how these components are comprised in the system.
Claim 1, Line 10, recites “a data receiving module configured to receive …”. The Specification does not disclose any detail on this limitation.
Claim 1, Lines 18-19, recites “the derived variable represents an elapsed time …”. The Specification does not disclose any detail on this limitation.
Claim 6, Lines 10-11, recites “the derived variable represents an elapsed time …”. The Specification does not disclose any detail on this limitation.
Claims 2-5 are rejected under 35 U.S.C. 112(a) as they inherit the deficiency of Claim 1.
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 1-6 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
With regard to Claims 1-5:
Step 1: the claims are drawn to a system/apparatus, one of the four statutory categories.
Step 2A, Prong One:
The claims recite the limitations of “process the PPG signal”, “compute a derived variable … ”, “predict a blood pressure state …” in Claim 1, and “generate the derived variable based on a time elapsed …” in Claim 2,. All these limitations are, under their broadest reasonable interpretation, limitations that cover performance of the limitation in the mind and/or by mathematical calculations. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind and/or by mathematical calculations but for the recitation of computer components and deep learning models at a high level of generality, then it falls within the “Mental Processes” and/or “Mathematical Concepts” grouping of abstract ideas. Accordingly, the claims recite an abstract idea.
Step 2A, Prong Two:
This judicial exception is not integrated into a practical application. In particular, the claims recite the additional elements – PPG sensor, non-invasive blood pressure measurement device, receiving PPG and blood pressure values, a first neural network, a second neural network, a third neural network that includes an attention mechanism, and inputting data and outputting results from the networks in Claim 1, the third neural network comprising a cross-attention component and a feed-forward component in Claim 3, receiving the data through various ports of a commercial patient monitoring device in Claim 4, and converting the predicted waveform using D-A converter and transmitting the continuous waveform to either commercial patient monitoring device or central monitoring device or medical record system in Claim 5. The elements of a first, a second and a third neural network are recited at a very high level of generality, which is tantamount to implementing the processes on a generic computer. The other elements of data receiving, conversion and transmitting are insignificant extra-solution activities. Accordingly, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The claims are directed to an abstract idea.
Step 2B:
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of neural network model are recited at a very high level of generality, which is tantamount to implementing the processes on a generic computer, and the other elements of receiving, converting and transmitting data are insignificant extra-solution activities, which cannot provide an inventive concept.
For the reasons set forth above, Claims 1-5 are not patent eligible.
With regard to Claim 6:
Step 1: the claim is drawn to a method/process, one of the four statutory categories.
Step 2A, Prong One:
The claim recites the limitations of “extracting a derived variable …”, and “predicting a blood pressure state …”, which are, under their broadest reasonable interpretation, limitations that cover performance of the limitation in the mind and/or mathematical calculations. If a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind and/or by mathematical calculations, but for the recitation of computer components and deep learning methods in a high level of generality, then it falls within the “Mental Processes” and/or “Mathematical Concepts” grouping of abstract ideas. Accordingly, the claim recites an abstract idea.
Step 2A, Prong Two:
This judicial exception is not integrated into a practical application. In particular, the claim recites the additional elements of a first learning model, a second learning model and a third learning model, with a very high level of generality, which is tantamount to implementing the processes on a generic computer, and the other additional elements of receiving PPG and blood pressure values and transmitting the analysis result to an external device are insignificant extra-solution activities. Accordingly, the additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea.
Step 2B:
The claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of three learning models are recited with a very high level of generality, which is tantamount to implementing the processes on a generic computer, and the elements of receiving PPG and blood pressure values and transmitting results are insignificant extra-solution activities, which cannot provide an inventive concept.
For the reasons set forth above, Claim 6 is not patent eligible.
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.
Claims 1-3 are rejected under 35 U.S.C. 103 as being unpatentable over Wang et al (US 20230293117 A1; hereafter Wang), in view of Addison et al (US 20210193311 A1; hereafter Addison) and Aguirre et al (Sensors 2021, 21, 2167; hereafter Aguirre).
With regard to Claim 1, Wang discloses a deep learning-based (Wang, Para 0087; “… after the deep learning processing of Approximation Network and Refinement Network”) continuous arterial blood pressure monitoring system (Wang, Para 0008; “The method or system is comfortable, continuous, and suitable for all-day use, and the accuracy of estimation of BPs and/or CBPs are sufficiently improved.”), comprising:
a photoplethysmography (PPG) sensor configured to measure a PPG signal of a patient (Wang, Para 0035; “an upper-arm wearable apparatus including a PPG sensor and sensing modeling-used PPG waveform signals …”);
a non-invasive blood pressure measurement device configured to obtain a blood pressure measurement value of the patient (Wang, Abstract; “A system for estimating BPs … comprises … a cuff-based BP measuring apparatus …”), and
one or more processors and a memory storing instructions that, when executed by the one or more processors (Wang, Para 0066; “… the PPG to BP subsystem 101 … may be a computer or a circuit system, or have partial functions performed at the upper-arm wearable apparatus 11”; Para 0067; “the PPG to CBP subsystem 102 may be a computer or a circuit system”), cause the one or more processors to implement:
a data receiving module (Wang, Fig. 1 shows that the PPG to BP subsystem (101) includes the PPG signal Receiver and analyzer (13) and also receives “Real SBP/DBP”) configured to receive the PPG signal and the blood pressure measurement value (Wang, Para 0065; “The cuff-based BP measuring apparatus 12 has been calibrated so as to provide the PPG to BP estimator and calibrator 14 with an accurate blood pressure. … The PPG waveform signals and/or other vital signs are wirelessly transmitted to and shown in a PPG signal receiver and analyzer 13.”);
a photoplethysmography analysis module (PPG to PVR transformer 15) comprising a first neural network model (Wang, Para 0067; “a PPG to PVR transformer 15 converts the PPG waveform signals to refined PVR waveforms using deep machine learning”) configured to process the PPG signal to generate a primary arterial blood pressure waveform (refined PVR waveforms);
a non-invasive blood pressure analysis module (PPG to BP estimator and calibrator 14) comprising a second neural network model (Wang, Para 0066; “PPG to BP estimator and calibrator 14 … it uses the age grouping method and further calibrates preliminary values by machine learning (ML) algorithms or other algorithms (e.g. regression algorithms, artificial neural networks (ANN), fuzzy logic, and support vector machine) …”.) configured to input the derived variable and the blood pressure measurement value and to output a second feature value (calibrated-estimated SBP/DBP);
a blood pressure waveform prediction module (CBP estimator) being configured to receive the primary arterial blood pressure waveform (refined PVR waveforms) and the second feature value (calibrated-estimated SBP/DBP) (Wang, Fig. 1 shows that both estimated PVR and estimated SBP/DBP are input into CBP estimator.); and
a blood pressure state prediction module (the CBP estimator 16) configured to predict a blood pressure state (CBPs) of the patient using the continuous arterial blood pressure waveform (Wang, Para 0088; “the CBP estimator 16 further uses linear regression equations to estimate CBPs in clinic use, and substitutes the calibrated-estimated BPs and waveform parameters of the user’s refined PVR waveform into the linear regression equation to have estimated CBPs.”).
Wang does not clearly and explicitly disclose implementing:
a derived variable extraction module configured to compute a derived variable from the blood pressure measurement value, wherein the derived variable represents an elapsed time since a most recent non-invasive blood measurement of the patient, and
the blood pressure waveform prediction module comprising a third neural network model that includes an attention mechanism and predicts a continuous arterial blood pressure waveform.
Addison in the same field of endeavor discloses implementing:
a derived variable extraction module configured to compute a derived variable from the blood pressure measurement value (Addison, Para 0097; “CNIBP monitoring device 100 may derive feature 604 by subtracting time t from the time of the most recent calibration point.”), wherein the derived variable represents an elapsed time since a most recent non-invasive blood measurement of the patient (Addison, Para 0097; “… features 604 as the value of the elapsed time at time t since the most recent calibration point.”), and
the blood pressure waveform prediction module comprising a third neural network model (a neural network algorithm) and predicts a continuous arterial blood pressure waveform (Addison, Para 0043; “a neural network algorithm may enable processing circuitry 110 to more accurately determine the continuous blood pressure of patient 101 using features extracted from a PPG signal along with calibration data from a most recent calibration point.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Wang, as suggested by Addison, in order to determine the time elapsed as the derived variable and to predict a continuous arterial blood pressure waveform. One of ordinary skill in the art would have been motivated to make the modification of determining the derived variable for the benefit of calibrating the estimated blood pressure with the most recent calibration data to achieve highest estimation accuracy (Addison, Para 0043; “a neural network algorithm may enable processing circuitry 110 to more accurately determine the continuous blood pressure of patient 101 using features extracted from a PPG signal along with calibration data from a most recent calibration point …”), and to make the modification of predicting a continuous blood pressure waveform for the benefit of continuous monitoring of patients with hypertension (Wang, Para 0003-0004; “Hypertension is a major cardiovascular risk factor contributing to various medical conditions, diseases, and events such as heart attacks, heart failure, aneurisms, strokes, and kidney disease. … A continuous measurement of BPs is necessary for medical monitoring and diagnosis by physicians”).
Wang and Addison do not explicitly disclose a neural network model that includes an attention mechanism.
Aguirre in the same field of endeavor discloses a neural network model that includes an attention mechanism (Aguirre, Abstract; “this methodology is capable of transforming PPG into an ABP pulse …”; Page 6, Para 2; “The proposed deep learning architecture is inspired by seq2seq encoder-decoder [36] models with attention mechanism.”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Wang and Addison, as suggested by Aguirre, in order to include an attention mechanism in the network. One of ordinary skill in the art would have been motivated to make the modification for the benefit of improved efficiency and accuracy for estimating blood pressure by allowing decoder of the network to dynamically focus on more relevant features as show in encoder of the network.
With regard to Claim 2, Wang, Addison and Aguirre disclose all the limitations in Claim 1 as discussed above, and do not clearly and explicitly disclose wherein the derived variable extraction module is configured to generate the derived variable based on a time elapsed from a time point of the most recent non-invasive blood pressure measurement to a target time point for arterial blood pressure prediction.
Addison further discloses wherein the derived variable extraction module is configured to generate the derived variable based on a time elapsed from a time point of the most recent non-invasive blood pressure measurement to a target time point for arterial blood pressure prediction (Addison, Para 0097; “CNIBP monitoring device 100 may derive feature 604 by subtracting time t from the time of the most recent calibration point.”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Wang, Addison and Aguirre, as further suggested by Addison, in order to generate the derived variable based on the time elapse from time of calibration to time of prediction. One of ordinary skill in the art would have been motivated to make the modification for the benefit of calibrating the estimated blood pressure with the most recent calibration data to achieve highest estimation accuracy (Addison, Para 0043; “a neural network algorithm may enable processing circuitry 110 to more accurately determine the continuous blood pressure of patient 101 using features extracted from a PPG signal along with calibration data from a most recent calibration point …”).
With regard to Claim 3, Wang, Addison and Aguirre disclose all the limitations in Claim 1 as discussed above, which include disclosing the third neural network model (Addison, Para 0043; “a neural network algorithm”) to integrate the primary arterial blood pressure waveform and the second feature value (Wang, Fig. 1 shows that both estimated PVR and estimated SBP/DBP are input into CBP estimator) when generating a continuous final blood pressure waveform (Addison, Para 0043; “a neural network algorithm may enable processing circuitry 110 to more accurately determine the continuous blood pressure of patient 101 using features extracted from a PPG signal along with calibration data from a most recent calibration point.”).
Wang, Addison and Aguirre as discussed above do not clearly and explicitly disclose wherein the neural network model comprises a cross-attention component and a feed-forward neural network component.
Addison further discloses wherein the neural network model comprises a feed-forward neural network component (Addison, Fig. 5 shows a neural network, in which at least sequence input layer, fully connected layer 512 and regression output layer 514 are feed-forward components), and Aguirre further discloses wherein the neural network model comprises a cross-attention component (Aguirre, Figure 6 shows a model architecture including a cross-attention component. The attention module (green box) combines information from both the encoder (the lower-left part) and the decoder (the lower-right part), indicating the attention to be cross-attention. Also note that in the shown architecture, multiple feed-forward (toward the direction of output) components are also used). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Wang, Addison and Aguirre, as further suggested by Addison and Aguirre, in order to comprise a cross-attention component and a feed-forward component in the network. One of ordinary skill in the art would have been motivated to make the modification for the benefit of constructing a neural network with cross-attention mechanism to achieve higher efficiency and accuracy as compared to regular neural networks.
Claims 4-5 are rejected under 35 U.S.C. 103 as being unpatentable over Wang, Addison and Aguirre, in view of Barak et al (US 20200253564 A1; hereafter Barak).
With regard to Claim 4, Wang, Addison and Aguirre disclose all the limitations in Claim 1 as discussed above. Wang, Addison and Aguirre do not explicitly and clearly disclose wherein the data receiving module is configured to receive the photoplethysmography signal and the non-invasive blood pressure measurement value through a serial port or a LAN port of a commercial patient monitoring device.
Barak in the same field of endeavor discloses wherein the data receiving module is configured to receive the photoplethysmography signal and the non-invasive blood pressure measurement value (Barak, Para 0268; “PCM 103 of FIG. 2 … comprises a port connected to line 202, for communicating data to hardware assembly 201 from PCM 103. This port serves the purpose of reading the calibration sensor values and optionally also tracking sensor values from the PCM.”) through a serial port or a LAN port of a commercial patient monitoring device (Barak, Para 0279; “Interface circuit 304 may format data for transmission serial RS232 connection, USB, Ethernet (LAN) …”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Wang, Addison and Aguirre, as suggested by Barak, in order to receive the measurements through the port(s) of a commercial monitoring device. One of ordinary skill in the art would have been motivated to make the modification for the benefit of making use of the capability of a commercial monitoring device in handling the various sensors and their measurements (Barak, Para 0259; “… enable a user to program PCM 103 to graphically display different biometric data versus time (such as ECG and/or pressure wave) …”) (Barak, Para 0260; “PCM 103 is programmed to determine from the cuff based measurements, a value for MAP, a value for Ps, and a value for Pd.”).
With regard to Claim 5, Wang, Addison and Aguirre disclose all the limitations in Claim 1 as discussed above. Wang, Addison and Aguirre do not explicitly and clearly disclose comprising: a data transmitting module configured to convert the predicted continuous arterial blood pressure waveform into an analog voltage via a digital-to-analog (D/A) converter and transmit the continuous arterial blood pressure waveform to a blood pressure measurement module of a commercial patient monitoring device, or transmit the continuous arterial blood pressure waveform in digital form to a central monitoring device or an electronic medical record system.
Barak in the same field of endeavor discloses comprising a data transmitting module configured to convert the predicted continuous arterial blood pressure waveform into an analog voltage via a digital-to-analog (D/A) converter and transmit the continuous arterial blood pressure waveform to a blood pressure measurement module of a commercial patient monitoring device, or transmit the continuous arterial blood pressure waveform in digital form to a central monitoring device or an electronic medical record system (Barak, Para 0279; “In one embodiment, this data is sent to PCM 103 in analog form using D/A converter 505.”; Para 0023; “… PCMs (Patient Care Monitors)… ”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Wang, Addison and Aguirre, as suggested by Barak, in order to convert the predicted blood pressure waveform using D-A converter and transmit it to a commercial monitoring device. One of ordinary skill in the art would have been motivated to make the modification for the benefit of transmitting the result in a compatible form to a commercial monitoring device where the result can be further processed and/or displayed (Barak, Para 0279; “In one embodiment, this data is sent to PCM 103 in analog form using D/A converter 505. In embodiments, interface circuit 504 converts BPV and/or SVV data to a form compatible with PCM 103, and transmits the formally compatible data via wire 202 to PCM 103.”) (Barak, Para 0257-0258; “PCM 103 comprises display 104 and user controls 108. Display 104 can preferably display graphic representation of biometric data versus time 104.”).
Claim 6 is rejected under 35 U.S.C. 103 as being unpatentable over Wang and Addison, in view of Barak et al (US 20200253564 A1; hereafter Barak).
With regard to Claim 6, Wang discloses a continuous arterial blood pressure monitoring method (Wang, Para 0008; “The method or system is comfortable, continuous, and suitable for all-day use, and the accuracy of estimation of BPs and/or CBPs are sufficiently improved.”), comprising:
receiving, from a photoplethysmography sensor and a non-invasive blood pressure device, a photoplethysmography (PPG) signal and a non-invasive blood pressure measurement value of a patient (Wang, Para 0065; “The cuff-based BP measuring apparatus 12 has been calibrated so as to provide the PPG to BP estimator and calibrator 14 with an accurate blood pressure. … The PPG waveform signals and/or other vital signs are wirelessly transmitted to and shown in a PPG signal receiver and analyzer 13.”),
acquiring a primary arterial blood pressure waveform from the PPG signal using a first learning model (Wang, Para 0067; “a PPG to PVR transformer 15 converts the PPG waveform signals to refined PVR waveforms using deep machine learning”) executed by a processor (Wang, Para 0067; “the PPG to CBP subsystem 102 may be a computer or a circuit system”);
inputting the derived variable and the non-invasive blood pressure measurement value to a second learning model (Wang, Para 0066; “PPG to BP estimator and calibrator 14 … it uses the age grouping method and further calibrates preliminary values by machine learning (ML) algorithms or other algorithms (e.g. regression algorithms, artificial neural networks (ANN), fuzzy logic, and support vector machine) …”) executed by the processor, and extracting a second feature value (calibrated-estimated SBP/DBP);
inputting the primary arterial blood pressure waveform (refined PVR waveforms) and the second feature value (calibrated-estimated SBP/DBP) to an analysis module (CBP estimator); and
predicting a blood pressure state of the patient using the continuous arterial blood pressure waveform (Wang, Para 0088; “the CBP estimator 16 further uses linear regression equations to estimate CBPs in clinic use, and substitutes the calibrated-estimated BPs and waveform parameters of the user’s refined PVR waveform into the linear regression equation to have estimated CBPs.”);
Wang does not explicitly and clearly disclose:
extracting a derived variable using the non-invasive blood pressure measurement value, wherein the derived variable represents an elapsed time since a most recent non-invasive blood pressure measurement of the patient,
using a learning model to predict a continuous arterial blood pressure waveform, and
transmitting the continuous arterial blood pressure waveform and the blood pressure state of the patient to an external patient monitoring device or to an electronic medical record system.
Addison in the same field of endeavor discloses
extracting a derived variable using the non-invasive blood pressure measurement value (Addison, Para 0097; “CNIBP monitoring device 100 may derive feature 604 by subtracting time t from the time of the most recent calibration point.”), wherein the derived variable represents an elapsed time since a most recent non-invasive blood pressure measurement of the patient (Addison, Para 0097; “… features 604 as the value of the elapsed time at time t since the most recent calibration point.”), and
using a learning model (a neural network algorithm) to predict a continuous arterial blood pressure waveform (Addison, Para 0043; “a neural network algorithm may enable processing circuitry 110 to more accurately determine the continuous blood pressure of patient 101 using features extracted from a PPG signal along with calibration data from a most recent calibration point.”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Wang, as suggested by Addison, in order to determine the time elapsed as the derived variable and to predict a continuous arterial blood pressure waveform. One of ordinary skill in the art would have been motivated to make the modification of determining the derived variable for the benefit of calibrating the estimated blood pressure with the most recent calibration data to achieve highest estimation accuracy (Addison, Para 0043; “a neural network algorithm may enable processing circuitry 110 to more accurately determine the continuous blood pressure of patient 101 using features extracted from a PPG signal along with calibration data from a most recent calibration point …”), and to make the modification of predicting a continuous blood pressure waveform for the benefit of continuous monitoring of patients with hypertension (Wang, Para 0003-0004; “Hypertension is a major cardiovascular risk factor contributing to various medical conditions, diseases, and events such as heart attacks, heart failure, aneurisms, strokes, and kidney disease. … A continuous measurement of BPs is necessary for medical monitoring and diagnosis by physicians”).
Wang and Addison do not clearly and explicitly disclose transmitting the continuous arterial blood pressure waveform and the blood pressure state of the patient to an external patient monitoring device or to an electronic medical record system.
Barak in the same field of endeavor discloses transmitting the continuous arterial blood pressure waveform and the blood pressure state of the patient to an external patient monitoring device or to an electronic medical record system (Barak, Para 0279; “In one embodiment, this data is sent to PCM 103 in analog form using D/A converter 505.” Para 0023; “… PCMs (Patient Care Monitors)… ”). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Wang and Addison, as suggested by Barak, in order to transmit the estimated blood pressure waveform and state to an external device. One of ordinary skill in the art would have been motivated to make the modification for the benefit of transmitting the result in a compatible form to a commercial monitoring device where the result can be further processed and/or displayed (Barak, Para 0279; “In one embodiment, this data is sent to PCM 103 in analog form using D/A converter 505. In embodiments, interface circuit 504 converts BPV and/or SVV data to a form compatible with PCM 103, and transmits the formally compatible data via wire 202 to PCM 103.”) (Barak, Para 0257-0258; “PCM 103 comprises display 104 and user controls 108. Display 104 can preferably display graphic representation of biometric data versus time 104.”).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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 LEI ZHANG whose telephone number is (571)272-7172. The examiner can normally be reached Monday-Friday 8am-5pm E.T..
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 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.
/L.Z./Examiner, Art Unit 3798
/PASCAL M BUI PHO/Supervisory Patent Examiner, Art Unit 3798