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 .
Status of Claims
Claims 1-9 and 21-22 are rejected. Claims 10-20 are canceled.
Response to Arguments
Drawing Objections
The previous drawing objections have been withdrawn in view of the Replacement Sheet.
Claim Objections
The previous claim objections have been withdrawn in view of the amendment.
Claim Rejections - 35 USC § 112
The previous 112(b) rejections have been withdrawn in view of the amendment.
Claim Rejections - 35 USC § 101
Applicant's arguments filed 12/2/25 have been fully considered but they are not persuasive.
Applicant asserts that the claims cannot practically be performed in the human mind, and thus does not recite a mental process. However, the Examiner disagrees. The limitations of “for each of the plurality of subjects, defining…a set of features for each of a plurality of time intervals of the triaxial accelerometer training data; clustering…the sets of features of the triaxial accelerometer training data into a number of clusters to obtain a cluster assignment for each of the sets of features of each of the plurality of subjects; fitting…a hidden Markov model to the triaxial accelerometer training data cluster assignments to generate a plurality of hidden states; and identifying…which of the plurality of hidden states are sleep states based on a frequency of occurrence of each of the hidden states during all sleep periods from the annotated timeline of the triaxial accelerometer training data for the plurality of subjects, thereby…to classify a sleep or wake status of a new subject based on triaxial accelerometer data from the new subject” are a process, as drafted, covers performance of the limitation that can be performed by a human mind (including an observation, evaluation, judgment, opinion) under the broadest reasonable standard. For example, these limitations are nothing more than a medical professional defining features on a paper based on the triaxial accelerometer data, drawing clusters of the data based on the features, determining a cluster assigned for the features, fitting the cluster assignments to generate a plurality of hidden states, and analyzing print outs of clusters of data relates to sleep states to classify a sleep or wake status of a subject.
Applicant states that claim 1 as amended recites many feature that cannot practically be performed in the human mind and are “tied to a specific machine.” However, the Examiner disagrees. The computing device is recited at a high level of generality and amounts to a part of a generic computer. Additionally, a triaxial accelerometer device is not recited in the claims, only triaxial accelerometer training data. However, under Step 2B of the rejection below it has been shown that a triaxial accelerometer is conventional in the art.
Moreover, claim 1 recites “fitting, using the computing device, a hidden Markov model to the triaxial accelerometer training data cluster assignments to generate a plurality of hidden states.” Fitting an HMM using the Baum-Welch algorithm, as explained by the specifications, involves complex mathematical operations including iterative expectation-maximization to estimate model parameters (transition probabilities, emission probabilities, and initial state probabilities). However, claim 1 does not recite the Baum Welch algorithm. While claim 4 does recite the Baum Welch algorithm, the specifics of it are not claimed. Claim 1 recites “fitting, using the computing device, a hidden Markov model to the to the triaxial accelerometer training data cluster assignments to generate a plurality of hidden states.” The fitting step is related to the hidden Markov model, which is nothing more than the computer implementation/automation of an abstract mental process of screening a patient, which is what a physician typically does with a patient in a diagnostic setting. Under the broadest reasonable interpretation, this limitation amounts to nothing more than a medical professional using pen and paper to fit cluster assignments to generate a plurality of hidden states.
Additionally, the computing device is recited at a high level of generality and amounts to a part of a generic computer.
Applicant states that similar to the claims in Thales Visionix Inc. v. United States, 850 F.3d 1343, 1348-49 (Fed. Cir. 2017), claim 1 uses specific triaxial accelerometer sensors and applies the Baum-Welch algorithm to fit an HMM to accurately determine when a subject is sleeping or awake. However, the Examiner disagrees. A triaxial accelerometer device is not recited in the claims, only triaxial accelerometer training data. Under Step 2B of the rejection below it has been shown that a triaxial accelerometer is conventional in the art.
Applicant states that the claimed invention represents a technological solution through allowing longitudinal actigraphy data to be helpful even for people with “erratic sleeping patterns or changing schedules, or in cases of shorter duration studies,” and can thus “broaden the usability of actigraphy for clinical, research, and consumer device purposes.” However, this is not recited in the claims. In claim 1, the last step of classifying a sleep or wake status of a new subject based on triaxial accelerometer data from the new subject amounts to the abstract idea itself. An improvement to the abstract idea is still an abstract idea.
Applicant asserts that claim 1 as amended provides for the use of a specific technique to improve the functioning of a computer and to solve a technological problem arising in the medical field of actigraphy and sleep monitoring. However, the Examiner disagrees. The computing device is recited at a high level of generality and amounts to a part of a generic computer.
Applicant states that similar to the claims in SRI Int’l, Inc. v. Cisco Sys., 930 F.3d at 1304 (2019), claim 1 uses a specific technique to solve a technological problem in the field of automated sleep monitoring: enabling accurate sleep classification using widely available triaxial accelerometer data rather than proprietary activity measures. However, the steps related to the sleep classification are directed to the abstract idea. An improvement to the abstract idea is still an abstract idea.
Applicant states that similar to the claims in Enfish, LLC v. Microsoft Corp., 822 F.3d 1327, 1336 (Fed. Cir. 2016), claim 1 is similarly directed to a specific improvement in how computers perform sleep monitoring by enabling computers to perform sleep classification using widely available sensor data rather than requiring proprietary measures. However, MPE 2106.05(f) discloses:
Claims that amount to nothing more than an instruction to apply the abstract idea using a generic computer do not render an abstract idea eligible.
Applicant states that similar to the claims in McRO, Inc. v. Bandai Namco Games Am. Inc., 837 F.3d 1299, 1314 (Fed. Cir. 2016), claim 1 uses specific rules (identifying sleep states based on frequency of occurrence from the annotated timeline) applied in a specific order (clustering, HMM fitting, frequency analysis, classifier training) to create a trained sleep classifier. However, each case turns on its own facts. In the case of McRO, the approach employed the computer to perform a distinct process to automate a task previously performed by humans. McRO, 837 F.3d at 1314 (Fed. Cir. 2016). This is not the case for the instant application, because sleep and wake analysis has been performed by a computer before.
Applicant states that similar to Thales Visionix, 850 F.3d at 1348-49, claim 1 requires obtaining triaxial accelerometer training data from specific accelerometer devices, and the entire training process depends on processing this device-specific sensor data. However, the Examiner disagrees. A triaxial accelerometer device is not recited in the claims, only triaxial accelerometer training data. Under Step 2B of the rejection below it has been shown that a triaxial accelerometer is conventional in the art.
Applicant respectfully submits that claim 1 provides non-routine or unconventional features that are not well understood in the art. However, the Examiner disagrees. Under Step 2B of the rejection below it has been shown that a triaxial accelerometer is conventional in the art.
Applicant states that the claimed invention is similar to Bascom Global Internet Servs., Inc. v. AT&T Mobility LLC, 827 F.3d 1341, 1350 (Fed. Cir. 2016), where the Federal Circuit held that "an inventive concept can be found in the non-conventional and non-generic arrangement of known, conventional pieces." However, MPEP 2106.05 discloses:
As made clear by the courts, the "‘novelty’ of any element or steps in a process, or even of the process itself, is of no relevance in determining whether the subject matter of a claim falls within the § 101 categories of possibly patentable subject matter." Intellectual Ventures I v. Symantec Corp., 838 F.3d 1307, 1315, 120 USPQ2d 1353, 1358 (Fed. Cir. 2016) (quoting Diamond v. Diehr, 450 U.S. at 188–89, 209 USPQ at 9). See also Synopsys, Inc. v. Mentor Graphics Corp., 839 F.3d 1138, 1151, 120 USPQ2d 1473, 1483 (Fed. Cir. 2016) ("a claim for a new abstract idea is still an abstract idea. The search for a § 101 inventive concept is thus distinct from demonstrating § 102 novelty.").
Applicant states that the identifying step is not merely data gathering or insignificant post-solution activity. However, the identifying step is directed to the abstract idea.
Applicant asserts that the Office action has asserted that obtaining triaxial accelerometer data amounts to well-understood, routine, conventional activity. However, the Office action's assertion addresses only the obtaining step, not the identifying step. The obtaining step is directed to well-understood, routine, and conventional activity. The identifying step is directed to the abstract idea.
Claim Rejections - 35 USC § 102 & 103
Applicant’s arguments with respect to claims 1-9 and 21-22 have been considered but are moot in view of the new ground of rejection.
Specification
The specification is objected to as failing to provide proper antecedent basis for the claimed subject matter. See 37 CFR 1.75(d)(1) and MPEP § 608.01(o). Correction of the following is required:
“hidden states” in claims 1, 21, and 22;
“a new subject” in claim 1; and
“a threshold ratio” in claims 21-22.
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 1-9 and 21-22 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception, specifically an abstract idea without significantly more.
Step 1
The claimed invention in claims 1-9 and 21-22 are directed to statutory subject matter as the claims recite a process for training a classifier to determine sleep and or wake status of a subject.
Step 2A, Prong One
Regarding claim 1, the recited steps are directed to a mental process of performing concepts in a human mind or by a human using a pen and paper (see MPEP 2106.04(a)(2) subsection (III)).
Regarding claim 1, the limitations of “for each of the plurality of subjects, defining…a set of features for each of a plurality of time intervals of the triaxial accelerometer training data; clustering…the sets of features of the triaxial accelerometer training data into a number of clusters to obtain a cluster assignment for each of the sets of features of each of the plurality of subjects; fitting…a hidden Markov model to the triaxial accelerometer training data cluster assignments to generate a plurality of hidden states; and identifying…which of the plurality of hidden states are sleep states based on a frequency of occurrence of each of the hidden states during all sleep periods from the annotated timeline of the triaxial accelerometer training data for the plurality of subjects, thereby…to classify a sleep or wake status of a new subject based on triaxial accelerometer data from the new subject” are a process, as drafted, covers performance of the limitation that can be performed by a human mind (including an observation, evaluation, judgment, opinion) under the broadest reasonable standard. For example, these limitations are nothing more than a medical professional defining features on a paper based on the triaxial accelerometer data, drawing clusters of the data based on the features, determining a cluster assigned for the features, fitting the cluster assignments to generate a plurality of hidden states, and analyzing print outs of clusters of data relates to sleep states to classify a sleep or wake status of a subject.
Step 2A, Prong Two
For claim 1, the judicial exception is not integrated into a practical application. In particular, claim 1 recites “obtaining, at a computing device, triaxial accelerometer training data and an annotated timeline of the triaxial accelerometer training data for a plurality of subjects.” The step of obtaining triaxial accelerometer data amounts to pre-solution activity of data gathering. The computing device is recited at a high level of generality and amounts to a part of a generic computer. The steps of defining, clustering, fitting, and identifying are related to the hidden Markov model, which are nothing more than the computer implementation/automation of an abstract mental process of screening a patient, which is what a physician typically does with a patient in a diagnostic setting. Merely including instructions to implement an abstract idea on a computer does not integrate a judicial exception into practical application.
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 element of obtaining triaxial accelerometer data amounts to nothing more than mere pre-solution activity of data gathering, which does not amount to an inventive concept. Moreover, obtaining triaxial accelerometer data is well-understood, routine, and conventional activity as evidenced by US 20100131227 (¶2-a conventional tri-axial accelerometer measures acceleration in a three-dimensional space), US 20120176309 (¶5-a conventional three dimensional (3D) pointing device 1, such as a 3D mouse, that includes a three-axis accelerometer 11), and US 20140117059 (¶8-a conventional triaxial accelerometer). Further, simply appending well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception, e.g., a claim to an abstract idea requiring no more than a generic computer to perform generic computer functions that are well-understood, routine and conventional activities previously known to the industry, as discussed in Alice Corp., 573 U.S. at 225, 110 USPQ2d at 1984 (see MPEP § 2106.05(d)).
Regarding dependent claims 2-9 and 21-22, the limitations of claim 1 further define the limitations already indicated as being directed to the abstract idea.
Claims 2-4 and 7-9 further define the details of the machine learning model, which is nothing more than the computer implementation/automation of an abstract mental process of screening a patient, which is what a physician typically does with a patient in a diagnostic setting.
Claims 5-6 are further directed to the abstract idea.
Claim 21 is directed to the abstract idea. The limitation of “wherein identifying which of the plurality of hidden states are sleep states comprises identifying hidden states having a frequency of occurrence during the sleep periods from the annotated timeline of the triaxial accelerometer training data based on a threshold ratio” is a process, as drafted, covers performance of the limitation that can be performed by a human mind (including an observation, evaluation, judgment, opinion) under the broadest reasonable standard. For example, this limitation is nothing more than a medical professional analyzing print outs of sleep states to identify hidden states as sleep states based on a threshold ratio.
Claim 22 is directed to the abstract idea. The limitation of “identifying which of the plurality of hidden states are awake states, wherein identifying which of the plurality of hidden states are awake states comprises identifying hidden states having a frequency of occurrence during all awake periods from the annotated timeline of the triaxial accelerometer training data based on a threshold ratio” is a process, as drafted, covers performance of the limitation that can be performed by a human mind (including an observation, evaluation, judgment, opinion) under the broadest reasonable standard. For example, this limitation is nothing more than a medical professional analyzing print outs of awake states to identify hidden states as awake states based on a threshold ratio.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1 and 8 are rejected under 35 U.S.C. 103 as being unpatentable over Nguyen (NPL “Unsupervised Clustering of Free-Living Human Activities using Ambulatory Accelerometry” published in 2007) in view of Li (NPL ” A Hidden Markov Model Based Unsupervised Algorithm for Sleep/Wake Identification Using Actigraphy” published in 2020).
Regarding claim 1, Nguyen teaches a process for training a classifier to determine sleep and or wake status of a subject based on triaxial accelerometer data, the process comprising: obtaining, at a computing device (page 4896, left col., Data Collection section, ¶2-computer), triaxial accelerometer training data (page 4896, left col., Data Collection section, ¶1-a small and lightweight triaxial accelerometer device…was worn; page 4896, left col., Data Collection section, ¶2-all data signals were recorded to a SD memory card on the device, and were transferred to a computer for analysis when monitoring was finished) and an annotated timeline of the triaxial accelerometer training data (page 4896, left col., last ¶- the subject was instructed to concurrently record the nature and time of each activity (where possible). The diary was used to register the actual activities with the accelerometer signals, which can be used as reference when evaluating algorithms; Table I; Fig. 1); defining, using the computing device, a set of features for each of a plurality of time intervals of the triaxial accelerometer training data (Fig. 2-Coherent clusters (using ‘raw’ acceleration features) obtained from a subject’s activity as shown in Fig. 1; page 4897, left col., ¶1-four acceleration-based feature vectors were used in experiments, namely, raw accelerometer data (‘raw’ – 3 features), statistics of the raw data describing the mean, standard deviation, frequency-domain energy, and the correlation between each pair of axis (‘raw-stats’ –12 features) [7], tilt angle and time-domain energy (‘theta energy’ – 2 features) [3] and signal magnitude area (‘sma’–1 feature)); clustering, using the computing device, the sets of features of the triaxial accelerometer training data into a number of clusters to obtain a cluster assignment for each of the sets of features (Fig. 2-Coherent clusters (using ‘raw’ acceleration features) obtained from a subject’s activity as shown in Fig. 1; Table II); fitting, using the computing device, a hidden Markov model to the triaxial accelerometer training data cluster assignments to generate a plurality of hidden states (page 4895, right col., Coherent Event Clustering section, ¶2-these segments are used to initialise the HMM…hyperparameters for each of the clusters (i.e. states); page 4896, left col., Unusual Event Clustering section, ¶1-the unusual event clustering algorithm adopted uses a semi-supervised adapted HMM topology; Fig. 3-Unusual event clustering of the same subject’s activity as shown in Fig. 1. Vertical axis represents the usual, and the first and second detected unusual event); and identifying, using the computing device, which of the plurality of hidden states are sleep states based on a frequency of occurrence of each of the hidden states during all sleep periods from the annotated timeline of the triaxial accelerometer training data (page 4898, left col., ¶1-separates the rest/sleep (usual event) from movements and activity (unusual event); Fig. 3-Unusual event clustering of the same subject’s activity as shown in Fig. 1. Vertical axis represents the usual, and the first and second detected unusual event; based on Fig. 3-the sleep states are determined based on the frequency of the detected usual event and wake states would be determined based on the frequency of the detected unusual events; Fig. 1), thereby enabling the hidden Markov model to classify a sleep or wake status of a new subject based on triaxial accelerometer data from the new subject (Abstract-unsupervised pattern recognition; page 4895, right col., Coherent Event Clustering section, ¶1-a hidden Markov model (HMM); page 4898, left col., ¶1-separates the rest/sleep (usual event) from movements and activity (unusual event); Fig. 3-Unusual event clustering of the same subject’s activity as shown in Fig. 1. Vertical axis represents the usual, and the first and second detected unusual event; based on Fig. 3-the sleep states are determined based on the frequency of the detected usual event and wake states would be determined based on the frequency of the detected unusual events; Fig. 1). However, Nguyen does not teach being for a plurality of subjects.
Li relates generally to a Hidden Markov Model (HMM) based unsupervised algorithm that can automatically and effectively infer sleep/wake states (page 2, Abstract). Li further teaches the invention using the following step:
being for a plurality of subjects (page 5, ¶2-we propose a sleep/wake identification method based on Hidden Markov Model (HMM) that has several advantages over existing methods. First, it is an unsupervised algorithm that does not require training data such as PSG and sleep logs to train the model. Second, it can be directly applied to datasets from different devices and populations, as it is data-driven that makes full use of the information contained in the dataset to learn and separate sleep and wake states).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Nguyen to include being for a plurality of subjects of Li in order to characterize individual activity patterns (Li, page 2, Abstract), since different
populations can have very different sleep and activity patterns (Li, page 3, last ¶-page 4, ¶1).
Regarding claim 8, the combination of Nguyen and Li teaches the process of claim 1, wherein the features include at least one item selected from a list including: standard deviation of each axis, a coefficient of variation of each axis, a range along each axis, an interquartile range of each axis, a number of small peak (local maxima) in each axis, a number of medium peaks in each axis, and a number of large peaks in each axis (Nguyen, page 4897, left col., ¶1-four acceleration-based feature vectors were used in experiments, namely, raw accelerometer data (‘raw’ – 3 features), statistics of the raw data describing the mean, standard deviation, frequency-domain energy, and the correlation between each pair of axis (‘raw-stats’ –12 features)).
Claims 2 and 21-22 are rejected under 35 U.S.C. 103 as being unpatentable over Nguyen in view of Li as applied to claim 1 above, and further in view of Nathan (US 20170188895 filed on 3/21/17).
Regarding claim 2, the combination of Nguyen and Li teaches the process of claim 1. However, the combination of Nguyen and Li does not teach wherein each of the plurality of time intervals is around 1 minute.
Nathan teaches wherein each of the plurality of time intervals is around 1 minute (¶94-the time segment can be 30 seconds to several minutes long. There are several feature sets that we use for training the classifier).
Nathan relates to the field of personal monitoring devices, systems, and methods (¶3).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Nguyen to include wherein each of the plurality of time intervals is around 1 minute of Nathan in order to analyze human body part motion (Nathan, ¶8).
Regarding claim 21, the combination of Nguyen and Li teaches the process of claim 1. However, the combination of Nguyen and Li teaches wherein identifying which of the plurality of hidden states are sleep states comprises identifying hidden states having a frequency of occurrence during the sleep periods from the annotated timeline of the triaxial accelerometer training data based on a threshold ratio.
Nathan teaches wherein identifying which of the plurality of hidden states are sleep states comprises identifying hidden states having a frequency of occurrence during the sleep periods from the annotated timeline of the triaxial accelerometer training data based on a threshold ratio (¶173-binning of motion types to find which motion type occurred most frequently. The confidence of this motion type is calculated as the ratio of frequency of the most frequently occurring motion and frequency of all other motions. A threshold is also used on the minimum possible value of frequency of most frequently occurring motion; ¶50- these algorithms aim to determine and recognize the following normal or typical human motions or activities: 1. Sleeping, 2. Walking, 3. Running, 4. Exercising).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Nguyen to include wherein identifying which of the plurality of hidden states are sleep states comprises identifying hidden states having a frequency of occurrence during the sleep periods from the annotated timeline of the triaxial accelerometer training data based on a threshold ratio of Nathan in order to make a decision if a motion type has occurred or not (Nathan, ¶64).
Regarding claim 22, the combination of Nguyen and Li teaches the process of claim 1. However, the combination of Nguyen and Li does not teach identifying which of the plurality of hidden states are awake states, wherein identifying which of the plurality of hidden states are awake states comprises identifying hidden states having a frequency of occurrence during all awake periods from the annotated timeline of the triaxial accelerometer training data based on a threshold ratio.
Nathan teaches identifying which of the plurality of hidden states are awake states, wherein identifying which of the plurality of hidden states are awake states comprises identifying hidden states having a frequency of occurrence during all awake periods from the annotated timeline of the triaxial accelerometer training data based on a threshold ratio (¶173-binning of motion types to find which motion type occurred most frequently. The confidence of this motion type is calculated as the ratio of frequency of the most frequently occurring motion and frequency of all other motions. A threshold is also used on the minimum possible value of frequency of most frequently occurring motion; ¶50- these algorithms aim to determine and recognize the following normal or typical human motions or activities: 1. Sleeping, 2. Walking, 3. Running, 4. Exercising; the activities that aren’t sleeping would be classified as an awake state).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Nguyen to include identifying which of the plurality of hidden states are awake states, wherein identifying which of the plurality of hidden states are awake states comprises identifying hidden states having a frequency of occurrence during all awake periods from the annotated timeline of the triaxial accelerometer training data based on a threshold ratio of Nathan in order to make a decision if a motion type has occurred or not (Nathan, ¶64).
Claim 3 is rejected under 35 U.S.C. 103 as being unpatentable over Nguyen in view of Li as applied to claim 1 above, and further in view of Srivastava (US 20200075167 filed on 8/30/19), hereinafter referred to as Sriv.
Regarding claim 3, the combination of Nguyen and Li teaches the process of claim 1. However, the combination of Nguyen and Li does not teach wherein the number of clusters is around 10 clusters.
Sriv teaches wherein the number of clusters is around 10 clusters (¶209-the maximum number of clusters is set to ten).
Sriv relates to greatly reducing the number of undiagnosed sleep conditions, by alerting users of serious problems (¶144).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Nguyen to include wherein the number of clusters is around 10 clusters of Sriv in order to determine good-to-bad sleep quality results, make up the possible recommendations, or the behavioural recipes to achieve good sleep (Sriv, ¶188).
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Nguyen in view of Li as applied to claim 1 above, and further in view of Xu (NPL “A Semi-supervised Hidden Markov Model-based Activity Monitoring System” published in 2011).
Regarding claim 4, the combination of Nguyen and Li teaches the process of claim 1. However, the combination of Nguyen and Li does not teach wherein fitting the hidden Markov model to the to the triaxial accelerometer training data cluster assignments comprise using a Baum Welch algorithm.
Xu teaches wherein fitting the hidden Markov model to the to the triaxial accelerometer training data cluster assignments comprise using a Baum Welch algorithm (Xu, page 1795, right col., HMM modeling section-use the Baum-Welch method [14] to estimate the model parameters; page 1795, left col. Data acquisition section-two Freescale triaxial accelerometers are required, with one attached to the waist, and the other one attached to the left thigh; page 1796, right col., Experimental Results section-for each subject, three universal HMMs corresponding to three activities using 5-minute data from the other five subjects are trained).
Xu relates to a semi-supervised HMM based activity monitoring system, that adapts the HMM for a specific subject from a general model in order to alleviate the requirement of a large training data set (Abstract).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Nguyen to include wherein fitting the hidden Markov model to the to the triaxial accelerometer training data cluster assignments comprise using a Baum Welch algorithm of Li in order to estimate the model parameters (Xu, page 1795, right col., HMM modeling section).
Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Nguyen in view of Li as applied to claim 1 above, and further in view of Todros (WO 2006054306 filed on 11/22/05).
Regarding claim 5, the combination of Nguyen and Li teaches the process of claim 1. However, the combination of Nguyen and Li does not teach wherein at least one sleeping state indicates a sleeping condition of at least one of the plurality of subjects.
Todros teaches wherein at least one sleeping state indicates a sleeping condition of at least one of the plurality of subjects (page 3, lines 4-5-diagnosis of a sleep-related condition of a patient; page 5, lines 25-30-analyzing the physiological signals includes constructing a hidden Markov model (HMM) having model states corresponding to the sleep stages, and identifying a state sequence in the model that accords with the physiological signals).
Todros relates generally to physiological monitoring and diagnosis, and specifically to sleep recording and analysis (page 1, lines 10-11).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Nguyen to include wherein at least one sleeping state indicates a sleeping condition of at least one of the plurality of subjects of Todros in order for diagnosing certain pathological conditions affecting the quality of sleep of patient 22 (Todros, page 30, lines 15-16).
Claim 6 is rejected under 35 U.S.C. 103 as being unpatentable over Nguyen in view of Li as applied to claim 1 above, and further in view of Richter (US 20220406450 filed on 11/23/20).
Regarding claim 6, the combination of Nguyen and Li teaches the process of claim 1. However, the combination of Nguyen and Li does not teach wherein each hidden state is characterized by a frequency distribution of the clustering assignments occurring during that hidden state.
Richter teaches wherein each hidden state is characterized by a frequency distribution of the clustering assignments occurring during that hidden state (¶75-the sleep clusters are positioned in relation to the frequency contribution; ¶144-classifying sleep stages).
Richter relates generally to data processing, and more particularly to methods of artificial intelligence and automatic computer scoring of sleep stages from data acquired during PSG and applications of the same (¶3).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Nguyen to include wherein each hidden state is characterized by a frequency distribution of the clustering assignments occurring during that hidden state of Richter in order to classifying sleep stages (Richter, ¶144).
Claim 7 is rejected under 35 U.S.C. 103 as being unpatentable over Nguyen in view of Li as applied to claim 1 above, and further in view of Liu (NPL “Wearable Device Heart Rate and Activity Data in an Unsupervised Approach to Personalized Sleep Monitoring: Algorithm Validation” published in 2020).
Regarding claim 7, the combination of Nguyen and Li teaches the process of claim 1. However, the combination of Nguyen and Li does not teach wherein clustering assignments are captured as emission probabilities of different states in the hidden Markov model.
Liu teaches wherein clustering assignments are captured as emission probabilities of different states in the hidden Markov model (page 7, left col., ¶2-the estimated parameters for emission distribution in different states, we can generally classify the 2 hidden states as sleep and wake; Table 3).
Liu relates to develop an unsupervised personalized sleep/wake identification algorithm using multifaceted data to explore the benefits of incorporating both heart rate and activity level in these types of algorithms and to compare this approach’s output with that of an existing commercial wearable device’s algorithms (page 1, Objective section).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Nguyen to include wherein clustering assignments are captured as emission probabilities of different states in the hidden Markov model of Liu in order to develop a personalized and unsupervised sleep/wake identification approach (Liu, page 2, Objectives section, ¶1).
Claim 9 is rejected under 35 U.S.C. 103 as being unpatentable over Nguyen in view of Li as applied to claim 1 above, and further in view of Bach (US 20220133194 filed on 12/7/21).
Regarding claim 9, the combination of Nguyen and Li teaches the process of claim 1. However, the combination of Nguyen and Li does not teach wherein the clustering the features of the triaxial training data is performed using K-means clustering.
Bach teaches wherein the clustering the features of the triaxial training data is performed using K-means clustering (¶529-clustering data of the correlation matrices using k-means; ¶280-3-axis accelerometer; ¶590).
Bach relates to cluster analysis…to group the data of the correlation matrices into clusters, each of which can be characterized (¶342).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Nguyen to include wherein the clustering the features of the triaxial training data is performed using K-means clustering of Bach because it is particularly well-adapted to large data sets (Bach, ¶342).
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 LAURA HODGE whose telephone number is (571) 272-7101. The examiner can normally be reached M-F: 8:00 am-5:00 pm.
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, UNSU JUNG can be reached at (571) 272-8506. 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.N.H./Examiner, Art Unit 3792
/AMANDA L STEINBERG/Examiner, Art Unit 3792