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 Arguments
Applicant's arguments filed 4/9/2026 have been fully considered but they are not persuasive.
Regarding 35 USC § 112(a) and 112(b), Applicant argues on pg. 6 of the Remarks that “rejections of claims 1-14 are moot in light of the amendments.” However, the 112(b) rejections of claims 10 and 11, as previously detailed on pg. 7-8 in the non-final rejection filed 1/9/2026, are maintained since no amendments have been made to claim 10 or 11. The 112(a) and 112(b) rejections have been withdrawn for claims 1-9 and 12-14 in light of the amendments.
Regarding 35 USC § 101, Applicant argues on pg. 7 and 8 of the Remarks that “the list of features…of independent claim 1 are not mental processes, nor can the claimed subject matter be practically performed in the human mind.” The Examiner respectfully disagrees. The Examiner notes that the limitations recited on pg. 7 and 8 of remarks include additional limitations that are not in question for consideration of reciting a judicial exception. The Examiner reminds Applicant that the limitations in question for evaluation under Step 2A, prong 1 are the following: “…the multi-output neural network is trained to simultaneously compute one or more values associated with a second physiological metric…generate…the one or more values associated with the first physiological metric and the one or more values associated with the second physiological metric, wherein the first physiological metric is associated with the first physiological phenomenon and the second physiological metric is associated with the second physiological phenomenon that is distinct from the first physiological phenomenon.” As noted in the non-final rejection filed 1/9/2026, the limitations are directed to an abstract idea mathematical concept and mental process since the action of computing metrics can be performed with equations and mathematical calculations. Further, generating values from the metric is another form calculations. Citing MPEP 2106.04(a)(2)(I): “It is important to note that a mathematical concept need not be expressed in mathematical symbols, because "[w]ords used in a claim operating on data to solve a problem can serve the same purpose as a formula." In re Grams, 888 F.2d 835, 837 and n.1, 12 USPQ2d 1824, 1826 and n.1 (Fed. Cir. 1989).” As detailed in the non-final office action filed on 1/9/2026, metrics are based on functions, which are mathematical concepts. Further, citing MPEP 2106.04(a)(2)(III): “The courts consider a mental process (thinking) that "can be performed in the human mind, or by a human using a pen and paper" to be an abstract idea.” The limitations recited above can be performed with a pen and paper in the human mind since the calculations for physiological metrics are based on functions. It is reasonable that one of ordinary skill could make calculations in their head with using a metric formula.
Applicant argues on page 9 of Remarks that “independent claim 1 as currently presented includes an additional element that reflects an improvement in the functioning of a computer or an improvement to other technology or technical field.” The Examiner respectfully disagrees. On pg. 9 of Remarks, the Applicant recites paragraphs [0012] and [0150] of the instant specification as evidence for an improvement to a wearable device. The Examiner notes that the Applicant is relying on details of an algorithm not presented in the independent claim. Further, on pg. 9 and 10 of Remarks, the Applicant argues that the recited features “are a technological improvement and have a practical application of providing improved wearable device technology for detection of illnesses and conditions of a user.” The recited features merely link the judicial exception to a field of use (See MPEP 2106.05(h)). The additional limitations merely link the judicial exception to the field of wearable technology. Additionally, the judicial exception abstract idea is relied upon solely as the improvement, and does not include additional limitations that integrate the judicial exception into a practical application. Citing MPEP 2106.05(a): “It is important to note, the judicial exception alone cannot provide the improvement. The improvement can be provided by one or more additional elements. See the discussion of Diamond v. Diehr, 450 U.S. 175, 187 and 191-92, 209 USPQ 1, 10 (1981)) in subsection II, below.” The improvement to the field of wearable technology is relied upon entirely by the abstract idea mental process and mathematical concept of calculating physiological metrics. The Examiner notes that the improvements recited on pg. 10 of Remarks, in reference to para. [0171] of the instant specification are not presented in the independent claim of claim 1. Similarly, the Applicant recites additional paragraphs of the specification that are not presented in independent claim 1. The Applicant argues that the additional limitations are an improvement to the processing time on pg. 12 of remarks. However, claim 1 merely recites generic computer structure (processor) for improving processing time. The claim is not directed towards improvements to the computer structure, nor do the claims or instant specification recite specific structure or algorithms for improving the processing of input physiological data. The Examiner maintains that the claim does not integrate the judicial exception into a practical application.
Applicant recites on pg. 13 of Remarks MPEP 2106.05(d). However, in the non-final Office Action filed on 01/09/2026, no argument was made by the Examiner that the additional limitations (i.e., the limitations that are not recited in step 2A, prong 1 as reciting judicial exceptions) are well-understood, routine, or conventional. Therefore, the argument is moot. Applicant argues on pg. 13 of Remarks that “[a]s a result of these additional elements, these additional elements provide ‘improved communication reliability, reduced latency, improved user experience related to reduced processing,…”. The Examiner notes that the recited improvements are not presented in claim 1. Claim 1 does not recite structure or any other additional matter besides the judicial exception abstract idea that results in these improvements. Instead, the improvements rely upon the judicial exception. Merely reciting the generic computer implementation of the judicial exception into a processor does amount to significantly more than the abstract idea itself. The Examiner maintains that the claim does not recite significantly more than the abstract idea.
In summary, the Examiner maintains the rejection under U.S.C. 101.
Applicant argues on pg. 15 of remarks that Baker does not disclose where the multi-output neural network of Zhang “comprises a single neural network architecture….”. The Examiner notes that the recitation of “comprises” in claim 1 is open-ended (Citing MPEP 2111.03: The transitional term "comprising", which is synonymous with "including," "containing," or "characterized by," is inclusive or open-ended and does not exclude additional, unrecited elements or method steps). Therefore, Baker meets the limitation of a single neural network architecture. Further, the Examiner respectfully disagrees that Baker does not remedy the deficiencies of claim 1. Baker suggests distinct and separate metrics in para. [0020]. Baker suggests that the metrics estimated could be a number of different, unrelated parameters. Figure 3 suggests multiple outputs that are separate (i.e., not a single output). In summary, Zhang, Baker, and Pho, in combination, address the original and amended limitations of claims 1-14. See 35 USC § 103 rejections below for more details and claim mapping for the amended claims.
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 1-14 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1
Claim 1 is recites a machine (i.e., a system). Therefore, claim 1 is directed to a statutory category of invention.
Step 2A, prong 1
Claim 1 recites: “wherein the multi-output neural network comprises a single neural network architecture trained to simultaneously compute a plurality of different, separate outputs including one or more values associated with a first physiological metric and one or more values associated with a second physiological metric.” Calculating values for a metric is a mathematical concept abstract idea in that the values are generated using mathematical formulas and functions. Computing one or more values associated with a physiological metric is so broadly claimed as to encompass a mathematical concept abstract idea. The specification recites calculating average temperature, respiratory rates, acceleration and angular values, sleep metrics, activity metrics, and breathing quality metrics (para. [0213]: “the functions may be stored or transferred over as one or more instructions or code…”). Likewise, the calculations are so broadly claimed that a human could perform the calculations in their head.
Additionally, claim 1 recites, “generate, via a single pass of the multi-output neural network, the plurality of different, separate outputs including the one or more values associated with the first physiological metric and the one or more values associated with the second physiological metric…“. The one or more values are being output are mathematical concepts in that they are essentially results of the previous calculations. The recitation is also a mental process abstract idea in that on could generate values in their head based on readings from the accelerometer or the PPG sensor. Similar to the calculations, a human could perform mathematical calculations and obtain results, thus generating one or more values. The Examiner notes that the additional limitations from the amendments to claim 1 still recite a judicial exception.
Step 2A, prong 2
Claim 1 does not include any additional limitations that amount to integration of the abstract idea into a practical application. Claim 1 recites the additional elements of “a wearable device configured to acquire physiological data from a user, the physiological data comprising at least one of heartbeat data collected via photoplethysmogram (PPG) measurements from one or more light-emitting components and one or more light-receiving components of the wearable device or motion data collected via one or more accelerometers of the wearable device; and one or more processors communicatively coupled with the wearable device, wherein the one or more processors are configured to: receive the physiological data acquired via the wearable device via one or more electronic signals; input one or more input streams into the multi-output neural network, wherein one or more input streams comprises at least one of a first input stream corresponding to the heartbeat data or a second input stream corresponding to the motion data; wherein the first physiological metric and the second physiological are distinct and based on a shared set of features from at least one of the first input stream or the second input stream; wherein the multi-output neural network is configured to transfer knowledge between different computations of the plurality of different, separate outputs through the single neural network architecture using the shared set of features; and transmit an instruction to a graphical user interface (GUI) of a user device associated with the wearable device, the instruction configured to cause the GUI to display one or more messages based at least in part on generating the one or more values associated with the first physiological metric and the one or more values associated with the second physiological metric.” Acquiring physiological data from a user, collecting heartbeat and motion data, receiving the data, inputting the data into a neural network, and transmitting the data to a GUI amounts to insignificant, extra-solution activity of data gathering. The common computer elements and data gathering sensors are recited at a high level of generality to simply amount to generic computer implementation of an abstract idea. Displaying one or more messages on a GUI amounts to insignificant post-solution activity (data displaying and/or reporting). Lastly, transferring knowledge of computations using a shared set of features (which the instant specification discloses as “input streams” in para. [0127]) is merely generic computer implementation of an abstract idea. The Examiner notes that “transferring knowledge” is equivalent to transferring or transmitting data, which is a generic computer function. Therefore, the additional limitations of claim 1 do not amount to integration of the abstract idea into a practical application. The Examiner notes that the additional limitations from the amendments to claim 1 still do not integrate the abstract idea into a practical application.
Step 2B
Claim 1 does not include any additional limitations that amount to significantly more than the abstract idea. Claim 1 recites the additional elements of “a wearable device configured to acquire physiological data from a user, the physiological data comprising at least one of heartbeat data collected via photoplethysmogram (PPG) measurements from one or more light-emitting components and one or more light-receiving components of the wearable device or motion data collected via one or more accelerometers of the wearable device; and one or more processors communicatively coupled with the wearable device, wherein the one or more processors are configured to: receive the physiological data acquired via the wearable device via one or more electronic signals; input one or more input streams into the multi-output neural network, wherein one or more input streams comprises at least one of a first input stream corresponding to the heartbeat data or a second input stream corresponding to the motion data; wherein the first physiological metric and the second physiological are distinct and based on a shared set of features from at least one of the first input stream or the second input stream; wherein the multi-output neural network is configured to transfer knowledge between different computations of the plurality of different, separate outputs through the single neural network architecture using the shared set of features; and transmit an instruction to a graphical user interface (GUI) of a user device associated with the wearable device, the instruction configured to cause the GUI to display one or more messages based at least in part on generating the one or more values associated with the first physiological metric and the one or more values associated with the second physiological metric.” Acquiring physiological data from a user, collecting heartbeat and motion data, receiving the data, inputting the data into a neural network, and transmitting the data to a GUI amounts to insignificant, extra-solution activity of data gathering. The common computer elements and data gathering sensors are recited at a high level of generality to simply amount to generic computer implementation of an abstract idea. Displaying one or more messages on a GUI amounts to insignificant post-solution activity (data displaying and/or reporting). Lastly, transferring knowledge of computations using a shared set of features (which the instant specification discloses as “input streams” in para. [0127]) is merely generic computer implementation of an abstract idea. The Examiner notes that “transferring knowledge” is equivalent to transferring or transmitting data, which is a generic computer function. Therefore, the additional limitations of claim 1 do not amount to significantly more than the abstract idea. The Examiner notes that the additional limitations from the amendments to claim 1 still do not amount to significantly more than the abstract idea.
Dependent claims
Claims 2, 3, and 12 further define extra-solution activity for data gathering.
Claims 4-7, 9-11, and 13 further define abstract idea mathematical concept abstract ideas.
Claim 8 further defines generic computer structure.
Claim 14 further defines the mathematical concept and mental process abstract ideas by generally linking the abstract idea to a field of use.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 10 and 11 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.
Regarding claim 10, the claim recites “wherein the one or more values associated with the first physiological metric comprises a first time scale and the one or more values associated with the second physiological metric comprise a second time scale same as the first time scale.” .” Due to the claim’s distinction between a “first time scale” and “a second time scale,” and that the second time scale is the “same as the first time,” it is unclear what the difference is between the time scales.
Regarding claim 11, the claim recites “wherein the one or more values associated with the first physiological metric comprises a first resolution and the one or more values associated with the second physiological metric comprise a second resolution same as the first resolution.” Due to the claim’s distinction between a “first resolution” and “a second resolution,” and that they are the “same resolution,” it is unclear what the difference is between the resolutions.
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.
Claim(s) 1-8 and 10-14 are rejected under 35 U.S.C. 103 as being unpatentable over Zhang et al. (US 20230114833 A1, "Zhang") in view of Baker et al. (US 20090287070 A1, “Baker”).
Regarding claim 1, Zhang discloses a system for utilizing a multi-output neural network (Abstract; Fig. 1 and 2), comprising: a wearable device (wearable device 104; Fig. 1) configured to acquire physiological data from a user, the physiological data comprising at least one of a heartbeat data collected via photoplethysmogram (PPG) (para. [0043]; "a PPG sensor assembly") measurements from one or more light-emitting components (para. [0071]; “optical transmitter”) and one or more light-receiving components (para. [0071]; “optical receiver”) of the wearable device (wearable device 104; Fig. 1) or motion data collected via one or more accelerometers (para. [0028]; "accelerometer data (e.g., movement/motion data; para. [0075]; "The ring 104 may include one or more motion sensors 245, such as one or more accelerometers); of the wearable device (wearable device 104; Fig. 1); and one or more processors communicatively coupled with the wearable device (para. [0053]; " The processing module 230-a of the ring 104 may include one or more processors (e.g., processing units)"), wherein the one or more processors (Fig. 2; processing module 230-a) are configured to: receive the physiological data acquired via the wearable device via one or more electronic signals (Fig. 2; para. [0053]; "For example, the processing module 230-a may transmit/receive data to/from the modules and other components of the ring 104, such as the sensors"; input one or more input streams (para. [0116]; "In particular, the motion data collected/processed at 305 through 320, and the PPG data collected and processed at 325 through 335, may be inputted into a machine learning model" into the multi-output neural network (para. [0116]; “machine learning model”), wherein the one or more input streams comprises at least one of a first input stream corresponding to the heartbeat data (para. [0116]; Fig. 3; the input stream going from 325 to 335) or a second input stream corresponding to the motion data (para. [0116]; Fig. 3; the input stream going from 305 to 320), wherein the multi-output neural network comprises a single neural network architecture (para. [0116]; “machine learning model”; The use of “comprises” is open-ended, meaning the disclosure of a machine learning model reads on a single neural network since at least a single neural network is disclosed.),trained to simultaneously compute a plurality of different, separate outputs including one or more values associated with a first physiological metric (para. [0078]; “For example, the processing module 230 may calculate and store various metrics, such as sleep metrics”) and one or more values associated with a second physiological metric (para. [0078; “The ring 104, or other computing/wearable device, may calculate a variety of values/metrics with respect to motion) wherein the first physiological metric and the second physiological metric are distinct and based on a shared set of features from at least one of the first input stream or the second input stream (Fig. 3; The input stream from the PPG data (325) and the motion data (305) are input into 340 to output heart rate (345)). generate, via a single pass of the multi-output neural network, the plurality of different, separate outputs including the one or more values associated with the first physiological metric (para. [0118]; " As such, the machine learning model may be trained to receive PPG data and motion data as inputs, and generate heart rate”) and the one or more values associated with the second physiological metric (para. [0078; “The ring 104, or other computing/wearable device, may calculate a variety of values/metrics with respect to motion), wherein the multi-output neural network is configured to transmit an instruction to a graphical user interface (GUI) (para. [0141]; GUI 500) of a user device associated with the wearable device, the instruction configured to cause the GUI to display one or more messages (The user device may display recommendations and/or information associated with the heart rate data via a message) based at least in part on generating the one or more values associated with the first physiological metric and the one or more values associated with the second physiological metric (para. [0025]; " Some electronic devices (e.g., wearable devices 104, user devices 106) may measure physiological parameters of respective users 102, such as photoplethysmography waveforms, continuous skin temperature, a pulse waveform, respiration rate, heart rate, heart rate variability (HRV), actigraphy, galvanic skin response, pulse oximetry, and/or other physiological parameters"). However, Zhang doesn’t disclose where the outputs are distinct and separate, based on a shared set of features (inputs), or where the neural network is configured to transfer knowledge between different computations of the plurality of different, separate outputs.
Baker, in the same field of endeavor of using neural networks to predict physiological parameters, discloses a neural network used to process physiological data. Baker discloses wherein the outputs are distinct and separate (para. [0020]: “The reader will understand that the systems and methods described herein may be equally adapted to the estimation and display of any physiological parameter of any patient (human or non-human) generated by any monitoring device, including but not limited to pulse rate, blood pressure, temperature, cardiac output, respiration parameters, and measures of blood constituents other than oxygenation.”), based on a shared set of inputs (para. [0036]: “Rather, the neural network, during the process of training, creates a non-linear and complex model that generates an output based on all the inputs.”), and where the neural network is configured to transfer knowledge between different computations of the plurality of different, separate outputs (Fig. 3; The output of the first layer becomes an input of each separate node in the next layer. This is a transfer of data (knowledge between separate computations at each node; para. [0037]: “FIG. 3 illustrates an embodiment of a neural network. The neural network 300 is represented diagrammatically as a series of interconnected nodes. The references 302, 304, 308 designate various network layers, including an input layer 302, an output layer 308 and a hidden layer 304. In FIG. 3 only one hidden layer is represented, but in alternative embodiments the network may comprise two or more hidden layers, each having any number of hidden nodes. In the embodiment shown, the output layer 308 delivers the estimated physiological parameter, in this case SpO.sub.2, in an output 312-1. A second output 312-2 is also illustrated, such as would deliver an estimated accuracy. Depending on the embodiment, the outputs 312 and inputs 310 may be single values or may represent ongoing streams of data.”).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to further modify the system of Zhang to further include where the outputs are distinct and separate, based on a shared set of features, and where the network transfers knowledge between computations of outputs, as disclosed by Baker. One of ordinary skill would recognize that it would have been obvious to include different output types than just heart rate. Further, Baker suggests a plethora of modifications for output types for a multi-output neural network. One of ordinary skill would also recognize that sharing features would improve efficiency, as multiple calculations could be made with the same input. Lastly, one of ordinary skill would recognize that sharing knowledge between nodes would be advantageous in that it would improve the efficiency of the computation time for the device. Therefore, it would have been obvious to modify the device of Zhang to further include the features of Baker.
Regarding claim 2, Baker and Zhang, in combination, disclose the system of claim 1 (see above). Further, Zhang discloses wherein the physiological data comprises temperature data (para. [0077]; “The ring 104 may store a variety of data described herein. For example, the ring 104 may store temperature data, such as raw sampled temperature data and calculated temperature data.”), wherein the one or more input streams comprises at least one of the first input stream, the second input stream, or a third input stream correspond to the temperature data (para. [0077]; “The ring 104 may store a variety of data described herein. For example, the ring 104 may store temperature data, such as raw sampled temperature data and calculated temperature data.”). However, Zhang doesn’t expressly disclose wherein the first physiological metric and the second physiological metric are based on the shared set of features from at least one of the first input stream, the second input stream, or the third input stream, or a combination thereof.
Baker discloses wherein the first physiological metric (Fig. 3; 312-1) and the second physiological metric (312-2) are based on the shared set of features from at least one of the first input stream, the second input stream, or the third input stream, or a combination thereof (Fig. 3; para. [0037]: “FIG. 3 illustrates an embodiment of a neural network. The neural network 300 is represented diagrammatically as a series of interconnected nodes. The references 302, 304, 308 designate various network layers, including an input layer 302, an output layer 308 and a hidden layer 304. In FIG. 3 only one hidden layer is represented, but in alternative embodiments the network may comprise two or more hidden layers, each having any number of hidden nodes. In the embodiment shown, the output layer 308 delivers the estimated physiological parameter, in this case SpO.sub.2, in an output 312-1. A second output 312-2 is also illustrated, such as would deliver an estimated accuracy. Depending on the embodiment, the outputs 312 and inputs 310 may be single values or may represent ongoing streams of data.”)
It would have been obvious for one of ordinary skill in the art to further modify the device of Zhang to include multiple input streams that share that share features, as disclosed by Baker, with the system of claim 1, as disclosed by Zhang and Baker in combination. One of ordinary skill would recognize that doing so would improve the efficiency of the system to by sharing input data. This would improve the computation efficiency of the system.
Regarding claim 3, Zhang, in combination with Baker, disclose the system of claim 2 (see above). Zhang discloses wherein the physiological data comprises blood oxygen data (para. [0077]; “The ring 104 may store a variety of data described herein. For example, the ring 104 may store temperature data, such as raw sampled temperature data and calculated temperature data.”), wherein the one or more input streams comprises at least one of the first input stream, the second input stream, a third input stream, or a fourth input stream corresponding to the blood oxygen data (para. [0077]; “The ring 104 may store a variety of data described herein. For example, the ring 104 may store temperature data, such as raw sampled temperature data and calculated temperature data.”). However, Zhang doesn’t expressly disclose wherein the first physiological metric and the second physiological metric are based on the shared set of features from at least one of the first input stream, the second input stream, or the third input stream, or a combination thereof.
Baker discloses wherein the first physiological metric (Fig. 3; 312-1) and the second physiological metric (312-2) are based on the shared set of features from at least one of the first input stream, the second input stream, or the third input stream, or the fourth input stream, or a combination thereof (Fig. 3; para. [0037]: “FIG. 3 illustrates an embodiment of a neural network. The neural network 300 is represented diagrammatically as a series of interconnected nodes. The references 302, 304, 308 designate various network layers, including an input layer 302, an output layer 308 and a hidden layer 304. In FIG. 3 only one hidden layer is represented, but in alternative embodiments the network may comprise two or more hidden layers, each having any number of hidden nodes. In the embodiment shown, the output layer 308 delivers the estimated physiological parameter, in this case SpO.sub.2, in an output 312-1. A second output 312-2 is also illustrated, such as would deliver an estimated accuracy. Depending on the embodiment, the outputs 312 and inputs 310 may be single values or may represent ongoing streams of data.”)
It would have been obvious for one of ordinary skill in the art to further modify the device of Zhang to include multiple input streams that share that share features, as disclosed by Baker, with the system of claim 1, as disclosed by Zhang and Baker in combination. One of ordinary skill would recognize that doing so would improve the efficiency of the system to by sharing input data. This would improve the computation efficiency of the system.
Regarding claim 4, the combination of Zhang and Baker disclose the system of claim 1 (see above). Further, Zhang discloses wherein the multi-output neural network includes weights (para. [0034]: “0034] In some aspects, the system 100 may utilize other biological rhythms to further improve physiological data collection, analysis, and processing by phase of these other rhythms. For example, if a weekly rhythm is detected within an individual's baseline data, then the model may be configured to adjust “weights” of data by day of the week.”). However, Zhang never expressly discloses wherein a training accuracy associated with training the multi-output neural is increased during training of the weights.
Baker discloses wherein a training accuracy associated with training the multi-output neural is increased during training of the weights (para. [0035]: “In embodiments of monitoring systems described in this disclosure, a neural network is used to create a more accurate estimate of a physiological parameter. In addition, the neural network may be used to generate an estimate of the accuracy of said physiologic parameter estimate. For example, in an embodiment the neural network is used to combine one or more estimates of a physiologic parameter with one or more associated signal quality metrics, creating a more accurate estimate of said physiologic parameter.” ; para. [0036]: “It should be noted that, when a neural network is used to combine two or more input estimates of a physiologic parameter, a neural network is not performing a simple selection or interpolation between the two input parameters. Rather, the neural network, during the process of training, creates a non-linear and complex model that generates an output based on all the inputs.”; para. [0039]: “To each node of the neural network 300 there corresponds a function F, which may be a nonlinear activation function such as a hyperbolic tangent function or a sigmoid function, as well as an activation level. Moreover, in the embodiment shown each node i of each layer is linked to the nodes j of the next layer; and a weighting Pij weights each link between a node i and a node j. This weighting weights the influence of the result of each node i in the calculation of the result delivered by each node j to which it is linked.”; Baker discloses narratively that the neural network is used to increase accuracy, and training weights are used at the nodes).
It would have been obvious for one of ordinary skill in the art to include the weights specifically for increasing accuracy during training, as disclosed by Baker, with the device of claim 1. One of ordinary skill in the art would recognize that using training weights to make the neural network more accurate would be an improvement to the device of Zhang, which discloses a similar device that already uses weights. Using the weights to increase accuracy would have been obvious for one of ordinary skill in the art, as Baker uses the weights for a similar purpose.
Regarding claim 5, Zhang, in combination with Baker, discloses the system of claim 1 (see above). Zhang further discloses wherein the multi-output neural network is trained to simultaneously compute one or more values associated with a third physiological metric (para. [0017]; “For instance, the machine learning model may be configured to differentiate between candidate heart rate measurements that are attributable to motion artifacts from candidate heart rate measurements that are indicative of the user's actual heart rate…By way of another example, the machine learning model may be configured to identify time-domain and/or frequency-domain features within the received PPG data and motion data (which may be used to identify candidate heart rate measurements), and may be configured to determine or estimate a heart rate for the user based on the identified features”), wherein the one or more processors are further configured to: generate, via the single pass of the multi-output neural network, the one or more values associated with the third physiological metric, wherein the third physiological metric is distinct from the first physiological metric and the second physiological metric (para. [0193]; The code may include instructions executable by a processor to receive physiological data associated with the user, the physiological data comprising PPG data and motion data collected throughout a first time interval via a wearable device associated with the user, determine a set of candidate heart rate measurements within the first time interval based at least in part on the PPG data, select a first heart rate measurement from the set of candidate heart rate measurements based at least in part on the received motion data, and determine a first heart rate for the user within the first time interval based at least in part on the selected first heart rate measurement). However, Zhang does not expressly disclose wherein the third physiological metric is based on a shared set of features.
Baker, in the same field of endeavor, discloses where the third physiological metric (para. [0017]: “For instance, the machine learning model may be configured to differentiate between candidate heart rate measurements that are attributable to motion artifacts from candidate heart rate measurements that are indicative of the user's actual heart rate…By way of another example, the machine learning model may be configured to identify time-domain and/or frequency-domain features within the received PPG data and motion data (which may be used to identify candidate heart rate measurements), and may be configured to determine or estimate a heart rate for the user based on the identified features”) is based on a shared set of features. (Fig. 3; multiple inputs and multiple outputs are shown for the neural network.).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to include a third physiological metric based on shared features or inputs, as disclosed by Baker, with the system of claim 1, as disclosed by Zhang in combination with Baker. One of ordinary skill would recognize that doing so would improve the efficiency of the system to by sharing input data. This would improve the computation efficiency of the system.
Regarding claim 6, Zhang, in combination with Baker, discloses the system of claim, 1 (see above). Further, Baker discloses wherein the one or more processors are further configured to: train the multi-output neural network based at least in part on inputting the one or more input streams (Fig. 3; Inputs 310-1 – 310-n) into the multi-output neural network and generating the one or more values associated with the first physiological metric and the one or more values associated with the second physiological metric (para. [0036]; “Rather, the neural network, during the process of training, creates a non-linear and complex model that generates an output based on all the inputs.”; Para. [0007]; “The pulse oximeter includes: a microprocessor capable of calculating an estimated value of oxygen saturation of a patient's blood at least in part upon information received from a sensor; a neural network module capable of receiving the estimated value as an input and calculating a revised value of oxygen saturation of a patient's blood using a neural network”).
It would have been obvious to one of ordinary skill in the art to combine the neural network configured to be trained with the system of Zhang since doing so would improve the processing speed and accuracy of Zhang’s physiological data. Further, it would have been obvious to use multiple physiological outputs with Baker’s neural network since Baker discloses adapting the neural network for other types of outputs [para. 0020]). Doing this would enhance Zhang’s system by enabling it to predict multiple physiological values and metrics based on multiple physiological data types.
Regarding claim 7, Zhang, in combination with Baker, discloses the system of claim 1 (see above). Zhang further discloses wherein the one or more processors are further configured to: train the multi-output neural network based on a plurality of features within a training physiological dataset associated with a plurality of users (para. [0119]; the system 200 may train multiple versions of a machine learning model, such as for different demographics of users (e.g., different age groups, varying levels of activity/performance, different skin tones, etc.), for users with varying medical conditions, and the like. In this regard, different models which are tailored to different demographics of users may be used to further fine-tune the ability of the respective models to perform heart rate detection. For example, the system 200 may acquire physiological data from a user who is an avid runner, and may utilize a machine learning model trained on data from other runners to perform heart rate detection for the user.”), wherein the plurality of features includes at least one of a blood oxygen saturation, body temperature, pulse wave amplitude, heart rate, intensity of motion, or ambient noise level (para. [0040]: “Example measurements and determinations may include, but are not limited to, user skin temperature, pulse waveforms, respiratory rate, heart rate, HRV blood oxygen levels, and the like.”).
Regarding claim 8, Zhang, in combination with Baker, discloses the system of claim 1 (see above). Further, Zhang discloses wherein the one or more processors are further configured to: simultaneously pre-process the one or more input streams prior to inputting the one or more input streams into the multi-output neural network. (para. [0099]; “At 310, the wearable device, the user device, the servers, or any combination thereof, may preprocess the motion data.”; para. [0105]; “At 330, the wearable device, the user device, the servers, or any combination thereof, may preprocess the PPG data.”; Both 310 and 330 process physiological data simultaneously prior to merging the data and calculating the heart rate output).
Regarding claim 10, Zhang, in combination with Baker, discloses the system of claim 1 (see above). Further, Zhang discloses wherein the one or more values associated with the first physiological metric (para. [0093]; “PPG data”) comprises a first time scale and the one or more values associated with the second physiological metric (para. [0093]; “motion data.”) comprise a second time scale same as the first time scale. (para. [0093]; “Additionally, or alternatively, the machine learning model may be configured to identify time-domain and/or frequency-domain features within the received PPG data and motion data).
Regarding claim 11, Zhang, in combination with Baker, discloses the system of claim 1 (see above). Further, Zhang discloses wherein the one or more values associated with the first physiological metric comprises a first resolution and the one or more values associated with the second physiological metric comprise a second resolution same as the first resolution. (para. [0093]; “Additionally, or alternatively, the machine learning model may be configured to identify time-domain and/or frequency-domain features within the received PPG data and motion data).
Due to the lack of specificity in the claim as to what is considered a resolution, the claim is interpreted that a resolution can be any type of resolution. Additionally, the specification recites a “thirty second resolution (e.g., time scale) (para. [0131]).” Therefore, using the broadest reasonable interpretation of a resolution to mean any type of resolution, Zhang’s disclosure of time scales read as resolutions.
Regarding claim 12, Zhang, in combination with Baker, discloses the system of claim 1 (see above). Further, Zhang discloses wherein the physiological data is acquired throughout a time interval that includes one or more sleep intervals of the user. (para. [0032]; “In this example, the ring 104-a may collect physiological data associated with the user 102-a, including temperature, heart rate, HRV, respiratory rate, and the like. In some aspects, data collected by the ring 104-a may be input to a machine learning classifier, where the machine learning classifier is configured to determine periods of time that the user 102-a is (or was) asleep. Moreover, the machine learning classifier may be configured to classify periods of time into different sleep stages…”.
Regarding claim 13, Zhang, in combination with Baker, discloses the system of claim 1 (see above). Further, Zhang discloses wherein at least one of the first physiological metric and the second physiological metric comprises at least one of a breathing disturbance event, a sleep staging metric, a blood oxygen metric, a blood pressure metric, a heartbeat metric, or a respiratory rate metric. (para. [0040]; “Example measurements and determinations may include, but are not limited to, user skin temperature, pulse waveforms, respiratory rate, heart rate, HRV, blood oxygen levels, and the like).
Regarding claim 14, Zhang, in combination with Baker, discloses the system of claim 1 (see above). Further, Zhang discloses wherein the wearable device comprises a wearable ring device. (Fig. 1; para. [0040]; “ring 104”).
Claim(s) 9 is rejected under 35 U.S.C. 103 as being unpatentable over Zhang et al. (US 20230114833 A1, “Zhang”), of Baker et al. (US 20090287070 A1, “Baker”), and Pho et al. (US 20220409187 A1, “Pho”).
Regarding claim 9, Zhang, in combination with Baker discloses the system of claim 1 (see above). However, Zhang and Baker do not disclose wherein each of the first physiological metric and the second physiological metric is associated with a probability that the user experienced an associated physiological event during a time period, and the associated physiological event is one of a breathing disturbance, deep sleep, or light sleep.
Pho, in the same field of endeavor of wearable sensor used to detect physiological events, discloses a wearable device used to detect illness and sleep patterns. Pho discloses wherein each of the first physiological metric and the second physiological metric (para. [0055]: “In some aspects, data collected by the ring 104-a may be input to a classifier, where the classifier is configured to determine illness risk metrics (or other metrics) associated with a likelihood or probability…”) is associated with a probability that the user experienced an associated physiological event (para. [0054]: “illness prediction metrics…”; These are used to predict an associated illness related to the metric.) during a time period (para. [0054]; “In some implementations, techniques described herein may compare physiological data (and rhythm parameters thereof) collected over different time intervals (e.g., first/reference time interval, second/prediction time interval) to identify a satisfaction of deviation criteria, where the satisfaction of one or more deviation criteria may be used to predict illness risk metrics (e.g., “risk scores”), illness prediction metrics, illness severity metrics, illness recovery metrics, and the like.), and the associated physiological event is one of a breathing disturbance, deep sleep, or light sleep. (para. [0047]: “In some aspects, the system 100 may detect periods of time during which a user 102 is asleep, and classify periods of time during which the user 102 is asleep into one or more sleep stages (e.g., sleep stage classification… where the machine learning classifier is configured to determine periods of time during which the user 102-a is (or was) asleep. Moreover, the machine learning classifier may be configured to classify periods of time into different sleep stages, including an awake sleep stage, a rapid eye movement (REM) sleep stage, a light sleep stage (non-REM (NREM)), and a deep sleep stage (NREM)”).
It would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine the system and physiological metrics of Zhang with the association between physiological metrics and probability metrics, as disclosed by Pho, since doing so would enable Zhang’s device to not only calculate metrics, but also predict if the user is experiencing a phenomenon. Further, it would have been obvious to one of ordinary skill to try and predict multiple types of physiological phenomenon such as those disclosed in Zhang to increase the range of possible physiological phenomenon the device is capable of detecting. It also would have been obvious to include specific events, such as a breathing disturbance, light sleep, or deep sleep since Pho discloses predicting these events with a similar device. One would recognize that these events could be predicted with a reasonable expectation of success. Therefore, it would have been obvious to include these events in the system of claim 1.
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 OWEN LEWIS MARSH whose telephone number is (571)272-8584. The examiner can normally be reached 7:30am – 5pm (M-Th), 8am – noon (F).
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, Jennifer McDonald can be reached at (571) 270-3061. 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.
/O.L.M./Examiner, Art Unit 3796
/Jennifer Pitrak McDonald/Supervisory Patent Examiner, Art Unit 3796
Prosecution Insights