DETAILED ACTION
Applicant’s arguments, filed on 01/05/2026, have been fully considered. The following rejections and/or objections are either reiterated or newly applied. They constitute the complete set presently being applied to the instant application.
Applicants have amended their claims, filed on 01/05/2026, and therefore rejections newly made in the instant office action have been necessitated by amendment with the exception of claim 1 which simply incorporated the subject matter of claim 13.
Claims 1-12 and 14-20 are the current claims hereby under examination.
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 .
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 1-12 and 14-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more. Under the two-step 101 analysis, the claims fail to satisfy the criteria for subject matter eligibility.
Regarding Step 1, claims 1-20 are all within at least one of the four statutory categories.
Claim 1 and its dependent claims disclose a method (process).
Claim 19 discloses a device (machine).
Claim 20 discloses a storage medium (machine).
Regarding Step 2A, Prong One, the independent claims 1, 19, and 20 recite an abstract idea. In particular, the claims generally recite the following:
extracting, for each of a plurality of epochs in a session, a first plurality of features from first motion data from a multi-channel motion sensor, wherein the first plurality of features comprises: one or more first motion features; one or more time-domain respiration features extracted from a first channel of a first stream of motion data derived from the first motion data, the first channel corresponding to a selected channel of the multi-channel motion sensor; and one or more frequency-domain respiration features extracted from a second channel of a second stream of motion data derived from the first motion data, the second channel corresponding to the selected channel of the multi-channel motion sensor;
in accordance with a determination that one or more first criteria are satisfied, classifying, using the first plurality of features for the plurality of epochs, a state for each of the plurality of epochs as one of a plurality of sleep states, the plurality of sleep states including a first sleep state corresponding to a wake state, a second sleep state corresponding to a rapid eye movement sleep state, and a third sleep state corresponding to one or more non-rapid eye movement sleep states;
identifying, using the classification of each of the plurality of epochs, a first sleep interval of consecutive epochs classified as a respective sleep state of the plurality of sleep states preceded by a second sleep interval of consecutive epochs classified as a different respective sleep state and followed by a third sleep interval of consecutive epochs classified as the different respective sleep state: and in accordance with the first sleep interval being shorter than a threshold number of consecutive epochs, reclassifying the consecutive epochs of the first sleep interval from the respective sleep state to the different respective sleep state.
These elements recited in claims 1, 19, and 20 are drawn to abstract ideas since they involve a mental process that can be practically performed in the human mind including observation, evaluation, judgement, and opinion and using pen and paper.
Extracting, for each of a plurality of epochs in a session, a first plurality of features from first motion data from a multi-channel motion sensor, wherein the first plurality of features comprises: one or more first motion features; one or more time-domain respiration features extracted from a first channel of a first stream of motion data derived from the first motion data, the first channel corresponding to a selected channel of the multi-channel motion sensor; and one or more frequency-domain respiration features extracted from a second channel of a second stream of motion data derived from the first motion data, the second channel corresponding to the selected channel of the multi-channel motion sensor is drawn to an abstract idea since it is a mental process that can be practically performed in the human mind, or with the aid of pen and paper. A person of ordinary skill in the art could reasonably receive the first motion data and extract the different plurality of features mentally or with the aid of pen and paper. These techniques are based on algorithms and calculations and mathematical principles, which can be performed by hand. The mathematics are not overly complicated to perform using pen and paper given enough time, therefore these are defined as abstract ideas. There is nothing to suggest an undue level of complexity in extracting, for each of a plurality of epochs in a session, a first plurality of features from first motion data from a multi-channel motion sensor, wherein the first plurality of features comprises: one or more first motion features; one or more time-domain respiration features extracted from a first channel of a first stream of motion data derived from the first motion data, the first channel corresponding to a selected channel of the multi-channel motion sensor; and one or more frequency-domain respiration features extracted from a second channel of a second stream of motion data derived from the first motion data, the second channel corresponding to the selected channel of the multi-channel motion sensor.
In accordance with a determination that one or more first criteria are satisfied, classifying, using the first plurality of features for the plurality of epochs, a state for each of the plurality of epochs as one of a plurality of sleep states, the plurality of sleep states including a first sleep state corresponding to a wake state, a second sleep state corresponding to a rapid eye movement sleep state, and a third sleep state corresponding to one or more non-rapid eye movement sleep states sensor is drawn to an abstract idea since it is a mental process that can be practically performed in the human mind, or with the aid of pen and paper. A person of ordinary skill in the art could reasonably receive the first motion data and extract the different plurality of features mentally or with the aid of pen and paper. These techniques are based on algorithms, calculations, mathematical principles, evaluation, and judgement, which can be performed by hand. The mathematics are not overly complicated to perform using pen and paper given enough time, therefore these are defined as abstract ideas. There is nothing to suggest an undue level of complexity in in accordance with a determination that one or more first criteria are satisfied, classifying, using the first plurality of features for the plurality of epochs, a state for each of the plurality of epochs as one of a plurality of sleep states, the plurality of sleep states including a first sleep state corresponding to a wake state, a second sleep state corresponding to a rapid eye movement sleep state, and a third sleep state corresponding to one or more non-rapid eye movement sleep states.
Identifying, using the classification of each of the plurality of epochs, a first sleep interval of consecutive epochs classified as a respective sleep state of the plurality of sleep states preceded by a second sleep interval of consecutive epochs classified as a different respective sleep state and followed by a third sleep interval of consecutive epochs classified as the different respective sleep state: and in accordance with the first sleep interval being shorter than a threshold number of consecutive epochs, reclassifying the consecutive epochs of the first sleep interval from the respective sleep state to the different respective sleep state is drawn to an abstract idea since it is a mental process that can be practically performed in the human mind, or with the aid of pen and paper. A person of ordinary skill in the art could reasonably examine the plurality of epochs and reclassify the sleep interval if it is smaller than a predetermined threshold mentally. These techniques are based on evaluation and judgement, which can be performed in the mind through simple observation of data. There is nothing to suggest an undue level of complexity in identifying, using the classification of each of the plurality of epochs, a first sleep interval of consecutive epochs classified as a respective sleep state of the plurality of sleep states preceded by a second sleep interval of consecutive epochs classified as a different respective sleep state and followed by a third sleep interval of consecutive epochs classified as the different respective sleep state: and in accordance with the first sleep interval being shorter than a threshold number of consecutive epochs, reclassifying the consecutive epochs of the first sleep interval from the respective sleep state to the different respective sleep state.
Regarding Step 2A, Prong Two, claims 1, 19, and 20 do not recite elements that integrate the exception into a practical application. Therefore, the claims are directed to the abstract idea. The additional elements merely:
Recite the words “apply it” or an equivalent with the judicial exception, or include instructions to implement the abstract idea on a computer, or merely use the computer as a tool to perform the abstract idea (e.g., “processing circuitry”, and “non-transitory computer readable storage medium”), and
Add insignificant extra-solution activity (the pre-solution activity of: using generic data-gathering components (e.g., “a multi-channel motion sensor”)).
As a whole, the additional elements merely serve to gather information to be used by the abstract idea, while generically implementing it on a computer. There is no practical application because the abstract idea is not applied, relied on, or used in a meaningful way. The processing performed remains in the abstract realm, i.e., the result is not used for a treatment. No improvement to the technology is evident. Therefore, the additional elements, alone or in combination, do not integrate the abstract idea into a practical application.
Regarding Step 2B, claims 1, 19, and 20 do not include additional elements, alone or in combination, that are sufficient to amount to significantly more than the judicial exception (i.e., an inventive concept) for the same reasons as described above.
Claims 1, 19, and 20 do not recite additional elements that amount to significantly more than the judicial exception itself. In particular, “a multi-channel motion sensor” does not qualify as significantly more because the limitation merely describes a well-known data gathering sensor.
The data gathering step of “a multi-channel motion sensor” is nothing more than a conventional sensor. Such sensors are evidenced by:
US Patent Application No. 20230388765 (Chen) discloses motion sensors such as accelerometers being well-known sensors (Chen, [0076]);
US Patent Application No. 20230147448 (Yanase) discloses a conventional sleep sensor comprised of a body motion sensor that is an accelerometer (Yanase, [0080]);
US Patent Application No. 20180317852 (MacDonald) discloses motion sensors such as accelerometers as being conventional (MacDonald, [0028]).
Further, the element of processing circuitry and a non-transitory computer readable storage medium in claims 1, 19, and 20 does not qualify as significantly more because this limitation is simply appending well-understood, routine and conventional activities previously known in 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 in the industry (see Electric Power Group, 830 F.3d 1350 (Fed. Cir. 2016); Alice Corp. v. CLS Bank Int’l, 110 USPQ2d 1976 (2014)) and/or a claim to an abstract idea requiring no more than being stored on a computer readable medium which is a well-understood, routine and conventional activity previously known in the industry (see Electric Power Group, 830 F.3d 1350 (Fed. Cir. 2016); Alice Corp. v. CLS Bank Int’l, 110 USPQ2d 1976 (2014); SAP Am. v. InvestPic, 890 F.3d 1016 (Fed. Circ. 2018)).
In view of the above, the additional elements individually do not integrate the exception into a practical application and do not amount to significantly more than the above judicial exception. Looking at the limitations as an ordered combination (that is, as a whole) adds nothing that is not already present when looking at the elements individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely provide conventional computer implementation, i.e., the computer is simply a tool to perform the process.
Regarding the dependent claims, claims 2-18 depend on claim 1. The dependent claims merely further define the abstract idea, routine data gathering using conventional sensors, or are additional data output that is well-understood, routine, and previously known to the industry.
For example, the following are dependent claims reciting abstract ideas and can be performed in the human mind or insignificant pre- or post-solution activity:
(Claim 2): “wherein the third sleep state corresponds to first-stage non-rapid eye movement sleep state and wherein the plurality of sleep states includes a fourth sleep state corresponding to a second-stage non-rapid eye movement sleep state and a third-stage non-rapid eye movement sleep state” further defines the abstract idea since it further defines the types of sleep states involved in the mental process;
(Claim 3): “wherein the third sleep state corresponds to a first-stage non-rapid eye movement sleep state, wherein the plurality of sleep states includes a fourth sleep state corresponding to a second-stage non-rapid eye movement sleep state, and wherein the plurality of sleep states includes a fifth sleep state corresponding to a third-stage non-rapid eye movement sleep state” further defines the abstract idea since it further defines the types of sleep states involved in the mental process;
(Claim 4): “further comprising: in accordance with a determination that the one or more first criteria are not satisfied, forgoing classifying the state for each of the plurality of epochs” is based in evaluation and judgement, therefore further defines the mental process;
(Claim 5): “wherein the one or more first criteria include a criterion that is satisfied when the session is longer than a threshold duration” is based in evaluation and judgement, therefore further defines the mental process;
(Claim 6): “wherein the one or more first criteria include a criterion that is satisfied when an electronic device including the multi-channel motion sensor is detected in contact with a body part during the session” is based in evaluation and judgement, therefore further defines the mental process;
(Claim 7): “further comprising: in accordance with a determination that one or more second criteria are satisfied, storing or displaying sleep intervals based on the classification of each of the plurality of epochs, wherein the sleep intervals include a sleep interval corresponding to the first sleep state, a sleep interval corresponding to the second sleep state, and a sleep interval corresponding to the third sleep state” is based in evaluation and judgement, therefore further defines the mental process;
(Claim 8): “wherein the one or more second criteria include a criterion that is satisfied when a total duration of the epochs classified different than the first sleep state is greater than a threshold duration” is based in evaluation and judgement, therefore further defines the mental process;
(Claim 9): “further comprising: in accordance with a determination that one or more third criteria are satisfied, storing or displaying the sleep intervals based on the classification of each of the plurality of epochs, wherein the sleep intervals corresponding to the second sleep state and the sleep interval corresponding to the third sleep state are merged” is based in evaluation and judgement, therefore further defines the mental process;
(Claim 10): “wherein the classifying is performed by a bidirectional long-short-term-memory machine learning model” is based in mathematical concept that can be performed mentally or with the aid of pen and paper. These techniques are based on algorithms and calculations and mathematical principles, which can be performed by hand. The mathematics are not overly complicated to perform using pen and paper given enough time, therefore these are defined as abstract ideas;
(Claim 11): “further comprising: scaling the first plurality of features to a common range of values for use by the bidirectional long-short-term-memory machine learning model” is based in mathematical concept that can be performed mentally or with the aid of pen and paper. These techniques are based on algorithms and calculations and mathematical principles, which can be performed by hand. The mathematics are not overly complicated to perform using pen and paper given enough time, therefore these are defined as abstract ideas;
(Claim 12): “further comprising: estimating a probability for each of the plurality of sleep states for each of the plurality of epochs, and classifying the state for each of the plurality of epochs using a maximum among the probability for each of the plurality of sleep states for each of the plurality of epochs” is based in mathematical concept that can be performed mentally or with the aid of pen and paper. These techniques are based on algorithms and calculations and mathematical principles, which can be performed by hand. The mathematics are not overly complicated to perform using pen and paper given enough time, therefore these are defined as abstract ideas;
(Claim 14): “wherein the multi-channel motion sensor comprises a three-axis accelerometer” is generic data-gathering components (see (Chen, [0076]); Yanase, [0080]; MacDonald, [0028]).;
(Claim 15): “further comprising: filtering the first motion data using a high-pass filter, wherein the one or more first motion features are extracted from the first motion data after filtering using the high-pass filter” is based in mathematical concept that can be performed mentally or with the aid of pen and paper. These techniques are based on algorithms and calculations and mathematical principles, which can be performed by hand. The mathematics are not overly complicated to perform using pen and paper given enough time, therefore these are defined as abstract ideas;
(Claim 16): “further comprising: filtering the first motion data using a band-pass filter to generate the first stream of motion data” is based in mathematical concept that can be performed mentally or with the aid of pen and paper. These techniques are based on algorithms and calculations and mathematical principles, which can be performed by hand. The mathematics are not overly complicated to perform using pen and paper given enough time, therefore these are defined as abstract ideas;
(Claim 17): “further comprising: filtering the first motion data using a low-pass filter; and down-sampling the first motion data from a first sampling rate to a second sampling rate lower than the first sampling rate” is based in mathematical concept that can be performed mentally or with the aid of pen and paper. These techniques are based on algorithms and calculations and mathematical principles, which can be performed by hand. The mathematics are not overly complicated to perform using pen and paper given enough time, therefore these are defined as abstract ideas;
(Claim 18): “further comprising: for each epoch: converting the first motion data into the first frequency domain representation for a first channel of the multi-channel motion sensor, a second frequency domain representation for the second channel of the multi-channel motion sensor, and a third frequency domain representation for a third channel of the multi-channel motion sensor; and computing a first signal-to-noise ratio using the first frequency domain representation, a second signal-to-noise ratio using the second frequency domain representation, and a third signal-to-noise ratio using the third frequency domain representation; wherein the selected channel corresponds to a respective channel of the first channel, the second channel, or the third channel with a maximum signal-to-noise ratio among the first signal-to-noise ratio, second signal-to-noise ratio and third signal-to-noise ratio” is based in mathematical concept that can be performed mentally or with the aid of pen and paper. These techniques are based on algorithms and calculations and mathematical principles, which can be performed by hand. The mathematics are not overly complicated to perform using pen and paper given enough time, therefore these are defined as abstract ideas.
The dependent claims do not recite significantly more than the abstract ideas. Therefore, claims 1-12 and 14-20 are rejected as being directed to non-statutory subject matter.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
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 1-9, 14, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Arnold (US 20160007934) in view of McDarby (US 20200337634) and Liu (CN 106333691). Citations to CN 106333691 will refer to the English Machine Translation that accompanies this Office Action.
Regarding independent claim 1, Arnold teaches a method ([0048]: “There are at least two base versions of the sleep detection method. The first is a motion sensor (e.g., an accelerometer) solution, which uses data from a three-axis motion sensor”) comprising:
extracting, for each of a plurality of epochs in a session, a first plurality of features from first motion data from a multi-channel motion sensor ([0045]: “a motion sensor (e.g., a multi-axis accelerometer) 112 generates motion data samples that represent motion for a plurality of time intervals (e.g., the motion sensor 112 may be in a wearable electronic device and the motion data represent motion of that device).”), wherein the first plurality of features comprises:
one or more first motion features ([0040]: “Such an embodiment that detects blocks of time in which the wearer is asleep may obtain a set of features for one or more periods of time from motion data obtained from a set of one or more motion sensors or data derived therefrom”);
one or more time-domain respiration features ([0111]: “Human sleep patterns can be described in terms of discrete stages of sleep. At the highest descriptive level, these break down into REM (rapid eye movement), and non-REM sleep stages, the latter including various light and deep sleep stages. Patterns of movement, heart-rate, heart-rate variability (HRV), and respiration change according to sleep stage. These patterns can be detected using data from a motion sensor (such as a three axis accelerometer) along with a photoplethysmographic sensor that allows measurement of heart-rate, HRV, and respiration”; [0158]: “the PPG data from a time window that includes the moment of interest is utilized to calculate at least one or more of the user's heart rate data, heart rate variability data, and/or respiration data”. The respiration data is determined over a time interval, therefore it is time-domain respiration features.).
However, Arnold does not teach the respiration features being extracted from the motion data.
McDarby discloses a system and method for determining sleep stage. Specifically, McDarby teaches one or more time-domain respiration features extracted from a first channel of a first stream of motion data derived from the first motion data ([0052]: “the processor may be used to extract information about breathing and motion, and higher order information such as the sleep stage … A motion sensor (for detection of general bodily movement and respiration)”. Arnold discloses the motion data coming from acceleration data along a single axis, therefore the single axis the first channel. The first stream of motion data is the raw detected motion data from a time interval, and the first motion data is the data from the accelerometer from Arnold). Arnold and McDarby are analogous arts as they are both related to systems used to monitor parameters of a user to determine sleep stage.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the respiration features being derived from the motion data from McDarby into the method from Arnold as it allows all the information to be derived from one type of sensor, which can reduce the amount of sensors required for monitoring.
The Arnold/McDarby combination teaches the first channel corresponding to a selected channel of the multi-channel motion sensor (Arnold, [0045]: “the motion data samples may characterize a measurement of acceleration along an axis of movement”; [0229]: “the motion data can be accelerometer data (e.g., acceleration data along a single axis)”. The motion data is determined from accelerometer data along a single axis, which is the selected channel of the multi-channel motion sensor. Since the respiration data is extracted from the motion data, then the respiration data is extracted from a first channel of motion data.).
However, the Arnold/McDarby combination does not teach one or more frequency-domain respiration features extracted from a second channel of a second stream of motion data derived from the first motion data.
McDarby teaches one or more frequency-domain respiration features extracted from a second channel of a second stream of motion data derived from the first motion data ([0059]: “A processing means is provided to take the original movement signal (the entire or raw detected movement signal) and to split it into “respiration” and “non-respiration” signals, by using frequency domain filtering”; [0052]: “the processor may be used to extract information about breathing and motion, and higher order information such as the sleep stage … A motion sensor (for detection of general bodily movement and respiration)”. The second channel is the same channel as the first channel, since first motion data from Arnold is only from a single axis, therefore the frequency domain features are extracted from the second channel. The second stream of motion data is the raw detected motion signal in a time interval, and the first motion data is the data from the motion sensor from Arnold. In this combination, McDarby uses the motion data from Arnold and performs frequency domain filtering to split the data into respiration and non-respiration signals, which are the frequency-domain respiration features.).
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the frequency-domain respiration features extracted from motion data from McDarby into the Arnold/McDarby combination as it allows the method to only use the data related to the respiration of the user, which ensures that only the correct data is being processed and used for analysis.
The Arnold/McDarby combination teaches the second channel corresponding to the selected channel of the multi-channel motion sensor (Arnold, [0045]: “the motion data samples may characterize a measurement of acceleration along an axis of movement”; [0229]: “the motion data can be accelerometer data (e.g., acceleration data along a single axis)”. The motion data is determined from accelerometer data along a single axis, which is the selected channel of the multi-channel motion sensor. Since the respiration data is extracted from the motion data, then the respiration data is extracted from a second channel of motion data. Additionally, since the first channel and the second channel are the same selected channel, they are both the single axis that the acceleration data is taken from.);
in accordance with a determination that one or more first criteria are satisfied (Arnold, [0236]: “while in one embodiment of the invention the periods of time are of a fixed length, alternative embodiments may support variable length. While in one embodiment of the invention the periods of time are 30 seconds, alternative embodiments may select the period of time from within a range, for example, of 5-120 seconds. The one or more samples from the accelerometer along the axis during a given one of the periods of time are collectively represented by recording a single state (the worn state or the not—worn state) for that time period. This reduces data volume for processing, storage, and/or transmission. As such, the length of the periods of time are selected to sufficiently reduce data volume while preserving sufficiently meaningful data for subsequent operations described herein; [0140]: “FIG. 3E illustrates the deriving of user statuses (e.g., active/inactive/not-worn status) using data during different time spans according to one embodiment of the invention where not-worn status is included. The not-worn status is at reference 260 (representing moments of interest (not shown) classified as not-worn), where the embodiment determines that the user is not wearing the electronic wearable device around 9:00 am.”; Fig. 3E. The first criteria includes determining if the periods of time are at a sufficient length, as well as determining if the user is wearing the device.), classifying, using the first plurality of features for the plurality of epochs, a state for each of the plurality of epochs as one of a plurality of sleep states (Arnold, [0052]: “At task box 4B, the time block classifier 126 uses the activity levels to assign blocks of time a sleep state selected from one of a plurality of sleep states”), the plurality of sleep states including a first sleep state corresponding to a wake state, a second sleep state corresponding to a rapid eye movement sleep state, and a third sleep state corresponding to one or more non-rapid eye movement sleep states (Arnold, [0052]: “The sleep states can include, among other things, an awake state and an asleep state. As described in greater detail below, some embodiments may include sleep states that represent different stages of sleep of a user”; [0111]: “Human sleep patterns can be described in terms of discrete stages of sleep. At the highest descriptive level, these break down into REM (rapid eye movement), and non-REM sleep stages, the latter including various light and deep sleep stages”; [0127]: “the statistical features derived may be used to train a multi-class classifier (via supervised machine learning) 128 to classify periods of time as a particular sleep stage (e.g., REM, various stage of non-REM (NREM) such as light or deep sleep).”);
identifying, using the classification of each of the plurality of epochs, a first sleep interval of consecutive epochs classified as a respective sleep state of the plurality of sleep states preceded by a second sleep interval of consecutive epochs classified as a different respective sleep state and followed by a third sleep interval of consecutive epochs classified as the different respective sleep state; and reclassifying the consecutive epochs of the first sleep interval from the respective sleep state to the different respective sleep state (Arnold, [0141]: “in the middle of a user's sleep, they somehow satisfy the not-worn criteria. In this case, the “not worn” time span is preceded and succeeded by “asleep” time. In this case, embodiments may classify the not-worn period as asleep time since it's unlikely that the user actually took off the device in the middle of a sleep period.”.).
However, the Arnold/McDarby combination does not teach reclassifying the sleep interval based on the first sleep interval being shorter than a threshold number of consecutive epochs.
Liu discloses an apparatus and method for judging sleep state. Specifically, Liu teaches classifying the sleep interval in accordance with the first sleep interval being shorter than a threshold number of consecutive epochs ([0058]: “once the cumulative sum within 5 minutes reaches a maximum threshold, the system enters an active state and automatically exits the sleep state”. Liu discloses only changing the state of the user if it is over a predetermined time, therefore this threshold of time can be incorporated into the Arnold/McDarby combination by only correcting the first sleep interval when it is less than a predetermined time threshold.). Arnold, McDarby, and Liu are analogous art as they are all related to systems and methods that monitor a user’s motion to determine sleep state.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the threshold from Liu into the Arnold/McDarby combination as the threshold allows the method to reclassify sleep intervals only if they are shorter than a certain length. Since a quick change in sleep intervals is most likely to be incorrectly assigned, the threshold allows the method to correct this incorrect reading and provide a more accurate result and classification to the user.
Regarding claim 2, the Arnold/McDarby/Liu combination teaches the method of claim 1, wherein the third sleep state corresponds to first-stage non-rapid eye movement sleep state and wherein the plurality of sleep states includes a fourth sleep state corresponding to a second-stage non-rapid eye movement sleep state and a third-stage non-rapid eye movement sleep state (Arnold, [0052]: “The sleep states can include, among other things, an awake state and an asleep state. As described in greater detail below, some embodiments may include sleep states that represent different stages of sleep of a user”; [0111]: “Human sleep patterns can be described in terms of discrete stages of sleep. At the highest descriptive level, these break down into REM (rapid eye movement), and non-REM sleep stages, the latter including various light and deep sleep stages”; [0144]: “FIG. 4B illustrates the result of the determination of sleep stages for the moments of interest at different time windows according one embodiment of the invention. In this embodiment, there are three NREM stages. Other embodiments may have different NREM stages.”; McDarby, [0034]-[0036]: “Stage N1—this is the lightest stage of sleep, and is characterized by the appearance of some low amplitude waves at multiple frequencies interspersed with the alpha waves for >50% of an epoch. There may also be sharp vertex waves, some slow eye movements on the EOG and/or an overall lowering of the frequency of EEG. Stage N2—this is a slightly deeper stage of sleep, and is marked by the appearance of sleep spindles and K-complexes, on a background of mixed frequency signals. Sleep spindles are bursts of higher frequency activity (e.g. >12 Hz). K-complexes are distinct isolated bipolar waves lasting about 1-2 seconds. Stage N3 is the deepest stage of sleep (in the original R&K classification, there were two distinct stages called Stage 3 and Stage 4). This is characterised by the appearance of slow waves (e.g., 1-2 Hz frequency) for at least 20% of an epoch.”. The first sleep state is the awake state, the second sleep state is the REM sleep state, the third sleep state is Stage N1, the fourth sleep state is Stage N2 and Stage N3. As stated in Arnold, sleep stages can be combined into one sleep stage, therefore Stage N2 and Stage N3 can be combined into one sleep state, which is the fourth sleep state.).
Regarding claim 3, the Arnold/McDarby/Liu combination teaches the method of claim 1, wherein the third sleep state corresponds to a first-stage non-rapid eye movement sleep state, wherein the plurality of sleep states includes a fourth sleep state corresponding to a second-stage non-rapid eye movement sleep state, and wherein the plurality of sleep states includes a fifth sleep state corresponding to a third-stage non-rapid eye movement sleep state (Arnold, [0052]: “The sleep states can include, among other things, an awake state and an asleep state. As described in greater detail below, some embodiments may include sleep states that represent different stages of sleep of a user”; [0111]: “Human sleep patterns can be described in terms of discrete stages of sleep. At the highest descriptive level, these break down into REM (rapid eye movement), and non-REM sleep stages, the latter including various light and deep sleep stages”; [0144]: “FIG. 4B illustrates the result of the determination of sleep stages for the moments of interest at different time windows according one embodiment of the invention. In this embodiment, there are three NREM stages. Other embodiments may have different NREM stages.”; McDarby, [0034]-[0036]: “Stage N1—this is the lightest stage of sleep, and is characterized by the appearance of some low amplitude waves at multiple frequencies interspersed with the alpha waves for >50% of an epoch. There may also be sharp vertex waves, some slow eye movements on the EOG and/or an overall lowering of the frequency of EEG. Stage N2—this is a slightly deeper stage of sleep, and is marked by the appearance of sleep spindles and K-complexes, on a background of mixed frequency signals. Sleep spindles are bursts of higher frequency activity (e.g. >12 Hz). K-complexes are distinct isolated bipolar waves lasting about 1-2 seconds. Stage N3 is the deepest stage of sleep (in the original R&K classification, there were two distinct stages called Stage 3 and Stage 4). This is characterised by the appearance of slow waves (e.g., 1-2 Hz frequency) for at least 20% of an epoch.”. The first sleep state is the awake state, the second sleep state is the REM sleep state, the third sleep state is Stage N1, the fourth sleep state is Stage N2, and the fifth sleep state is Stage N3.).
Regarding claim 4, the Arnold/McDarby/Liu combination teaches the method of claim 1, further comprising: in accordance with a determination that the one or more first criteria are not satisfied, forgoing classifying the state for each of the plurality of epochs (Arnold, [0236]: “while in one embodiment of the invention the periods of time are of a fixed length, alternative embodiments may support variable length. While in one embodiment of the invention the periods of time are 30 seconds, alternative embodiments may select the period of time from within a range, for example, of 5-120 seconds. The one or more samples from the accelerometer along the axis during a given one of the periods of time are collectively represented by recording a single state (the worn state or the not—worn state) for that time period. This reduces data volume for processing, storage, and/or transmission. As such, the length of the periods of time are selected to sufficiently reduce data volume while preserving sufficiently meaningful data for subsequent operations described herein”; [0140]: “FIG. 3E illustrates the deriving of user statuses (e.g., active/inactive/not-worn status) using data during different time spans according to one embodiment of the invention where not-worn status is included. The not-worn status is at reference 260 (representing moments of interest (not shown) classified as not-worn), where the embodiment determines that the user is not wearing the electronic wearable device around 9:00 am.”; Fig. 3E. Arnold discloses only classifying periods of time if they reach a specific length or if the user is wearing the device, therefore satisfying the first criteria. Therefore, the opposite is true, where if the period of time does not reach the threshold length or it is determined that the user is not wearing the device, it is not classified as a sleep state.).
Regarding claim 5, the Arnold/McDarby/Liu combination teaches the method of claim 4, wherein the one or more first criteria include a criterion that is satisfied when the session is longer than a threshold duration (Arnold, [0236]: “while in one embodiment of the invention the periods of time are of a fixed length, alternative embodiments may support variable length. While in one embodiment of the invention the periods of time are 30 seconds, alternative embodiments may select the period of time from within a range, for example, of 5-120 seconds. The one or more samples from the accelerometer along the axis during a given one of the periods of time are collectively represented by recording a single state (the worn state or the not—worn state) for that time period. This reduces data volume for processing, storage, and/or transmission. As such, the length of the periods of time are selected to sufficiently reduce data volume while preserving sufficiently meaningful data for subsequent operations described herein”. The first criteria includes criterion that the period of time reaches a certain length, which is the threshold duration. ).
Regarding claim 6, the Arnold/McDarby/Liu combination teaches the method of claim 4, wherein the one or more first criteria include a criterion that is satisfied when an electronic device including the multi-channel motion sensor is detected in contact with a body part during the session (Arnold, [0140]: “FIG. 3E illustrates the deriving of user statuses (e.g., active/inactive/not-worn status) using data during different time spans according to one embodiment of the invention where not-worn status is included. The not-worn status is at reference 260 (representing moments of interest (not shown) classified as not-worn), where the embodiment determines that the user is not wearing the electronic wearable device around 9:00 am.”; Fig. 3E. The first criteria includes criterion that the device needs to be in contact with the user’s body (determining whether the device is worn or not), therefore teaching on this limitation.).
Regarding claim 7, the Arnold/McDarby/Liu combination teaches the method of claim 1, further comprising: in accordance with a determination that one or more second criteria are satisfied, storing or displaying sleep intervals based on classification of each of the plurality of epochs (Arnold, [0152]: “The movement measurements and/or the determined blocks of time with the states and/or the sleep stages may be presented to the user (e.g., on a display of the device or another electronic device (e.g., a tablet/smartphone/computer which receives the data from the wearable electronic device, generates the data, or receives the data from another electronic device (e.g., a server))) or stored in the wearable electronic device for a period in time sufficient to present or communicate such data to a secondary device”), wherein the sleep intervals include a sleep interval corresponding to the first sleep state, a sleep interval corresponding to the second sleep state, and a sleep interval corresponding to the third sleep state (Arnold, [0144]: “FIG. 4B illustrates the result of the determination of sleep stages for the moments of interest at different time windows according one embodiment of the invention. In this embodiment, there are three NREM stages. Other embodiments may have different NREM stages.”; McDarby, [0034]-[0036]: “Stage N1—this is the lightest stage of sleep, and is characterized by the appearance of some low amplitude waves at multiple frequencies interspersed with the alpha waves for >50% of an epoch. There may also be sharp vertex waves, some slow eye movements on the EOG and/or an overall lowering of the frequency of EEG. Stage N2—this is a slightly deeper stage of sleep, and is marked by the appearance of sleep spindles and K-complexes, on a background of mixed frequency signals. Sleep spindles are bursts of higher frequency activity (e.g. >12 Hz). K-complexes are distinct isolated bipolar waves lasting about 1-2 seconds. Stage N3 is the deepest stage of sleep (in the original R&K classification, there were two distinct stages called Stage 3 and Stage 4). This is characterised by the appearance of slow waves (e.g., 1-2 Hz frequency) for at least 20% of an epoch.”).
Regarding claim 8, the Arnold/McDarby/Liu combination teaches the method of claim 7, wherein the one or more second criteria include a criterion that is satisfied when a total duration of the epochs classified different than the first sleep state is greater than a threshold duration (Arnold, [0052]: “At task box 4B, the time block classifier 126 uses the activity levels to assign blocks of time a sleep state selected from one of a plurality of sleep states. The sleep states can include, among other things, an awake state and an asleep state. As described in greater detail below, some embodiments may include sleep states that represent different stages of sleep of a user. Further, some embodiments may include sleep states that represent different types of non-sleeping activities (e.g., a restless state). Each block of time can span one or more of the periods of time characterized by the activity levels. In some cases, the activity levels covered by the blocks of time may be homogeneous activity levels, while other cases may allow for heterogeneous activity levels”; Fig. 4B; Liu, [0058]: “once the cumulative sum within 5 minutes reaches a maximum threshold, the system enters an active state and automatically exits the sleep state”. Liu discloses only changing the state of the user if it is over a predetermined time, therefore this threshold of time can be incorporated into the Arnold/McDarby combination by only correcting the first sleep interval when it is less than a predetermined time threshold. Since the Arnold/McDarby/Liu combination only reclassifies the epochs classified different than the first sleep state if the duration is less than a threshold, it is obvious that the opposite is true, wherein the classification is not changed when the duration is longer than a threshold duration, and therefore the different sleep state is displayed.).
Regarding claim 9, the Arnold/McDarby/Liu combination teaches the method of claim 7, further comprising: in accordance with a determination that one or more third criteria are satisfied, storing or displaying the sleep intervals based on the classification of each of the plurality of epochs (Arnold, [0152]: “The movement measurements and/or the determined blocks of time with the states and/or the sleep stages may be presented to the user (e.g., on a display of the device or another electronic device (e.g., a tablet/smartphone/computer which receives the data from the wearable electronic device, generates the data, or receives the data from another electronic device (e.g., a server))) or stored in the wearable electronic device for a period in time sufficient to present or communicate such data to a secondary device”), wherein the sleep intervals corresponding to the second sleep state and the sleep interval corresponding to the third sleep state are merged (Arnold, [0144]: “FIG. 4B illustrates the result of the determination of sleep stages for the moments of interest at different time windows according one embodiment of the invention. In this embodiment, there are three NREM stages. Other embodiments may have different NREM stages.”; McDarby, [0034]-[0036]: “Stage N1—this is the lightest stage of sleep, and is characterized by the appearance of some low amplitude waves at multiple frequencies interspersed with the alpha waves for >50% of an epoch. There may also be sharp vertex waves, some slow eye movements on the EOG and/or an overall lowering of the frequency of EEG. Stage N2—this is a slightly deeper stage of sleep, and is marked by the appearance of sleep spindles and K-complexes, on a background of mixed frequency signals. Sleep spindles are bursts of higher frequency activity (e.g. >12 Hz). K-complexes are distinct isolated bipolar waves lasting about 1-2 seconds. Stage N3 is the deepest stage of sleep (in the original R&K classification, there were two distinct stages called Stage 3 and Stage 4). This is characterised by the appearance of slow waves (e.g., 1-2 Hz frequency) for at least 20% of an epoch.”).
Regarding claim 14, the Arnold/McDarby/Liu combination teaches the method of claim 1, wherein the multi-channel motion sensor comprises a three-axis accelerometer (Arnold, [0147]: “The motion data may in some cases be accelerometer data received from an accelerometer (e.g., a three axis accelerometer) that represents motion along three axes”).
Regarding independent claim 19, Arnold teaches an electronic device (Abstract, “Aspects of generating movement measures that quantify motion data samples generated by a wearable electronic device are discussed herein”) comprising:
a multi-channel motion sensor ([0045]: “a motion sensor (e.g., a multi-axis accelerometer) 112 generates motion data samples that represent motion for a plurality of time intervals (e.g., the motion sensor 112 may be in a wearable electronic device and the motion data represent motion of that device).”);
and processing circuitry coupled to the multi-channel motion sensor ([0039]: “The wearable electronic device may be configured to (e.g., by a set of one or more processors executing instructions) obtain the motion data samples generated by the set of motion sensors.”), the processing circuitry programmed to:
extract, for each of a plurality of epochs in a session, a first plurality of features from first motion data from a multi-channel motion sensor ([0045]: “a motion sensor (e.g., a multi-axis accelerometer) 112 generates motion data samples that represent motion for a plurality of time intervals (e.g., the motion sensor 112 may be in a wearable electronic device and the motion data represent motion of that device).”), wherein the first plurality of features comprises:
one or more first motion features ([0040]: “Such an embodiment that detects blocks of time in which the wearer is asleep may obtain a set of features for one or more periods of time from motion data obtained from a set of one or more motion sensors or data derived therefrom”);
one or more time-domain respiration features ([0111]: “Human sleep patterns can be described in terms of discrete stages of sleep. At the highest descriptive level, these break down into REM (rapid eye movement), and non-REM sleep stages, the latter including various light and deep sleep stages. Patterns of movement, heart-rate, heart-rate variability (HRV), and respiration change according to sleep stage. These patterns can be detected using data from a motion sensor (such as a three axis accelerometer) along with a photoplethysmographic sensor that allows measurement of heart-rate, HRV, and respiration”; [0158]: “the PPG data from a time window that includes the moment of interest is utilized to calculate at least one or more of the user's heart rate data, heart rate variability data, and/or respiration data”. The respiration data is determined over a time interval, therefore it is time-domain respiration features.).
However, Arnold does not teach the respiration features being extracted from the motion data.
McDarby discloses a system and method for determining sleep stage. Specifically, McDarby teaches one or more time-domain respiration features extracted from a first channel of a first stream of motion data derived from the first motion data ([0052]: “the processor may be used to extract information about breathing and motion, and higher order information such as the sleep stage … A motion sensor (for detection of general bodily movement and respiration)”. Arnold discloses the motion data coming from acceleration data along a single axis, therefore the single axis the first channel. The first stream of motion data is the raw detected motion data from a time interval, and the first motion data is the data from the accelerometer from Arnold). Arnold and McDarby are analogous arts as they are both related to systems used to monitor parameters of a user to determine sleep stage.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the respiration features being derived from the motion data from McDarby into the method from Arnold as it allows all the information to be derived from one type of sensor, which can reduce the amount of sensors required for monitoring.
The Arnold/McDarby combination teaches the first channel corresponding to a selected channel of the multi-channel motion sensor (Arnold, [0045]: “the motion data samples may characterize a measurement of acceleration along an axis of movement”; [0229]: “the motion data can be accelerometer data (e.g., acceleration data along a single axis)”. The motion data is determined from accelerometer data along a single axis, which is the selected channel of the multi-channel motion sensor. Since the respiration data is extracted from the motion data, then the respiration data is extracted from a first channel of motion data.).
However, the Arnold/McDarby combination does not teach one or more frequency-domain respiration features extracted from a second channel of a second stream of motion data derived from the first motion data.
McDarby teaches one or more frequency-domain respiration features extracted from a second channel of a second stream of motion data derived from the first motion data ([0059]: “A processing means is provided to take the original movement signal (the entire or raw detected movement signal) and to split it into “respiration” and “non-respiration” signals, by using frequency domain filtering”; [0052]: “the processor may be used to extract information about breathing and motion, and higher order information such as the sleep stage … A motion sensor (for detection of general bodily movement and respiration)”. The second channel is the same channel as the first channel, since first motion data from Arnold is only from a single axis, therefore the frequency domain features are extracted from the second channel. The second stream of motion data is the raw detected motion signal in a time interval, and the first motion data is the data from the motion sensor from Arnold. In this combination, McDarby uses the motion data from Arnold and performs frequency domain filtering to split the data into respiration and non-respiration signals, which are the frequency-domain respiration features.).
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the frequency-domain respiration features extracted from motion data from McDarby into the Arnold/McDarby combination as it allows the method to only use the data related to the respiration of the user, which ensures that only the correct data is being processed and used for analysis.
The Arnold/McDarby combination teaches the second channel corresponding to the selected channel of the multi-channel motion sensor (Arnold, [0045]: “the motion data samples may characterize a measurement of acceleration along an axis of movement”; [0229]: “the motion data can be accelerometer data (e.g., acceleration data along a single axis)”. The motion data is determined from accelerometer data along a single axis, which is the selected channel of the multi-channel motion sensor. Since the respiration data is extracted from the motion data, then the respiration data is extracted from a second channel of motion data. Additionally, since the first channel and the second channel are the same selected channel, they are both the single axis that the acceleration data is taken from.);
in accordance with a determination that one or more first criteria are satisfied (Arnold, [0236]: “while in one embodiment of the invention the periods of time are of a fixed length, alternative embodiments may support variable length. While in one embodiment of the invention the periods of time are 30 seconds, alternative embodiments may select the period of time from within a range, for example, of 5-120 seconds. The one or more samples from the accelerometer along the axis during a given one of the periods of time are collectively represented by recording a single state (the worn state or the not—worn state) for that time period. This reduces data volume for processing, storage, and/or transmission. As such, the length of the periods of time are selected to sufficiently reduce data volume while preserving sufficiently meaningful data for subsequent operations described herein; [0140]: “FIG. 3E illustrates the deriving of user statuses (e.g., active/inactive/not-worn status) using data during different time spans according to one embodiment of the invention where not-worn status is included. The not-worn status is at reference 260 (representing moments of interest (not shown) classified as not-worn), where the embodiment determines that the user is not wearing the electronic wearable device around 9:00 am.”; Fig. 3E. The first criteria includes determining if the periods of time are at a sufficient length, as well as determining if the user is wearing the device.), classifying, using the first plurality of features for the plurality of epochs, a state for each of the plurality of epochs as one of a plurality of sleep states (Arnold, [0052]: “At task box 4B, the time block classifier 126 uses the activity levels to assign blocks of time a sleep state selected from one of a plurality of sleep states”), the plurality of sleep states including a first sleep state corresponding to a wake state, a second sleep state corresponding to a rapid eye movement sleep state, and a third sleep state corresponding to one or more non-rapid eye movement sleep states (Arnold, [0052]: “The sleep states can include, among other things, an awake state and an asleep state. As described in greater detail below, some embodiments may include sleep states that represent different stages of sleep of a user”; [0111]: “Human sleep patterns can be described in terms of discrete stages of sleep. At the highest descriptive level, these break down into REM (rapid eye movement), and non-REM sleep stages, the latter including various light and deep sleep stages”; [0127]: “the statistical features derived may be used to train a multi-class classifier (via supervised machine learning) 128 to classify periods of time as a particular sleep stage (e.g., REM, various stage of non-REM (NREM) such as light or deep sleep).”);
identifying, using the classification of each of the plurality of epochs, a first sleep interval of consecutive epochs classified as a respective sleep state of the plurality of sleep states preceded by a second sleep interval of consecutive epochs classified as a different respective sleep state and followed by a third sleep interval of consecutive epochs classified as the different respective sleep state; and reclassifying the consecutive epochs of the first sleep interval from the respective sleep state to the different respective sleep state (Arnold, [0141]: “in the middle of a user's sleep, they somehow satisfy the not-worn criteria. In this case, the “not worn” time span is preceded and succeeded by “asleep” time. In this case, embodiments may classify the not-worn period as asleep time since it's unlikely that the user actually took off the device in the middle of a sleep period.”.).
However, the Arnold/McDarby combination does not teach reclassifying the sleep interval based on the first sleep interval being shorter than a threshold number of consecutive epochs.
Liu discloses an apparatus and method for judging sleep state. Specifically, Liu teaches classifying the sleep interval in accordance with the first sleep interval being shorter than a threshold number of consecutive epochs ([0058]: “once the cumulative sum within 5 minutes reaches a maximum threshold, the system enters an active state and automatically exits the sleep state”. Liu discloses only changing the state of the user if it is over a predetermined time, therefore this threshold of time can be incorporated into the Arnold/McDarby combination by only correcting the first sleep interval when it is less than a predetermined time threshold.). Arnold, McDarby, and Liu are analogous arts as they are all related to systems and methods that monitor a user’s motion to determine sleep state.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the threshold from Liu into the Arnold/McDarby combination as the threshold allows the method to reclassify sleep intervals only if they are shorter than a certain length. Since a quick change in sleep intervals is most likely to be incorrectly assigned, the threshold allows the method to correct this incorrect reading and provide a more accurate result and classification to the user.
Regarding independent claim 20, Arnold teaches a non-transitory computer readable storage medium storing instructions ([0042]: “The embodiment may then store, in non-transitory machine readable storage medium, data associating the period of time with a not-worn state.”), which when executed by an electronic device including processing circuitry ([0039]: “The wearable electronic device may be configured to (e.g., by a set of one or more processors executing instructions) obtain the motion data samples generated by the set of motion sensors.”) to:
extract, for each of a plurality of epochs in a session, a first plurality of features from first motion data from a multi-channel motion sensor ([0045]: “a motion sensor (e.g., a multi-axis accelerometer) 112 generates motion data samples that represent motion for a plurality of time intervals (e.g., the motion sensor 112 may be in a wearable electronic device and the motion data represent motion of that device).”), wherein the first plurality of features comprises:
one or more first motion features ([0040]: “Such an embodiment that detects blocks of time in which the wearer is asleep may obtain a set of features for one or more periods of time from motion data obtained from a set of one or more motion sensors or data derived therefrom”);
one or more time-domain respiration features ([0111]: “Human sleep patterns can be described in terms of discrete stages of sleep. At the highest descriptive level, these break down into REM (rapid eye movement), and non-REM sleep stages, the latter including various light and deep sleep stages. Patterns of movement, heart-rate, heart-rate variability (HRV), and respiration change according to sleep stage. These patterns can be detected using data from a motion sensor (such as a three axis accelerometer) along with a photoplethysmographic sensor that allows measurement of heart-rate, HRV, and respiration”; [0158]: “the PPG data from a time window that includes the moment of interest is utilized to calculate at least one or more of the user's heart rate data, heart rate variability data, and/or respiration data”. The respiration data is determined over a time interval, therefore it is time-domain respiration features.).
However, Arnold does not teach the respiration features being extracted from the motion data.
McDarby discloses a system and method for determining sleep stage. Specifically, McDarby teaches one or more time-domain respiration features extracted from a first channel of a first stream of motion data derived from the first motion data ([0052]: “the processor may be used to extract information about breathing and motion, and higher order information such as the sleep stage … A motion sensor (for detection of general bodily movement and respiration)”. Arnold discloses the motion data coming from acceleration data along a single axis, therefore the single axis the first channel. The first stream of motion data is the raw detected motion data from a time interval, and the first motion data is the data from the accelerometer from Arnold). Arnold and McDarby are analogous arts as they are both related to systems used to monitor parameters of a user to determine sleep stage.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the respiration features being derived from the motion data from McDarby into the method from Arnold as it allows all the information to be derived from one type of sensor, which can reduce the amount of sensors required for monitoring.
The Arnold/McDarby combination teaches the first channel corresponding to a selected channel of the multi-channel motion sensor (Arnold, [0045]: “the motion data samples may characterize a measurement of acceleration along an axis of movement”; [0229]: “the motion data can be accelerometer data (e.g., acceleration data along a single axis)”. The motion data is determined from accelerometer data along a single axis, which is the selected channel of the multi-channel motion sensor. Since the respiration data is extracted from the motion data, then the respiration data is extracted from a first channel of motion data.).
However, the Arnold/McDarby combination does not teach one or more frequency-domain respiration features extracted from a second channel of a second stream of motion data derived from the first motion data.
McDarby teaches one or more frequency-domain respiration features extracted from a second channel of a second stream of motion data derived from the first motion data ([0059]: “A processing means is provided to take the original movement signal (the entire or raw detected movement signal) and to split it into “respiration” and “non-respiration” signals, by using frequency domain filtering”; [0052]: “the processor may be used to extract information about breathing and motion, and higher order information such as the sleep stage … A motion sensor (for detection of general bodily movement and respiration)”. The second channel is the same channel as the first channel, since first motion data from Arnold is only from a single axis, therefore the frequency domain features are extracted from the second channel. The second stream of motion data is the raw detected motion signal in a time interval, and the first motion data is the data from the motion sensor from Arnold. In this combination, McDarby uses the motion data from Arnold and performs frequency domain filtering to split the data into respiration and non-respiration signals, which are the frequency-domain respiration features.).
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the frequency-domain respiration features extracted from motion data from McDarby into the Arnold/McDarby combination as it allows the method to only use the data related to the respiration of the user, which ensures that only the correct data is being processed and used for analysis.
The Arnold/McDarby combination teaches the second channel corresponding to the selected channel of the multi-channel motion sensor (Arnold, [0045]: “the motion data samples may characterize a measurement of acceleration along an axis of movement”; [0229]: “the motion data can be accelerometer data (e.g., acceleration data along a single axis)”. The motion data is determined from accelerometer data along a single axis, which is the selected channel of the multi-channel motion sensor. Since the respiration data is extracted from the motion data, then the respiration data is extracted from a second channel of motion data. Additionally, since the first channel and the second channel are the same selected channel, they are both the single axis that the acceleration data is taken from.);
in accordance with a determination that one or more first criteria are satisfied (Arnold, [0236]: “while in one embodiment of the invention the periods of time are of a fixed length, alternative embodiments may support variable length. While in one embodiment of the invention the periods of time are 30 seconds, alternative embodiments may select the period of time from within a range, for example, of 5-120 seconds. The one or more samples from the accelerometer along the axis during a given one of the periods of time are collectively represented by recording a single state (the worn state or the not—worn state) for that time period. This reduces data volume for processing, storage, and/or transmission. As such, the length of the periods of time are selected to sufficiently reduce data volume while preserving sufficiently meaningful data for subsequent operations described herein; [0140]: “FIG. 3E illustrates the deriving of user statuses (e.g., active/inactive/not-worn status) using data during different time spans according to one embodiment of the invention where not-worn status is included. The not-worn status is at reference 260 (representing moments of interest (not shown) classified as not-worn), where the embodiment determines that the user is not wearing the electronic wearable device around 9:00 am.”; Fig. 3E. The first criteria includes determining if the periods of time are at a sufficient length, as well as determining if the user is wearing the device.), classifying, using the first plurality of features for the plurality of epochs, a state for each of the plurality of epochs as one of a plurality of sleep states (Arnold, [0052]: “At task box 4B, the time block classifier 126 uses the activity levels to assign blocks of time a sleep state selected from one of a plurality of sleep states”), the plurality of sleep states including a first sleep state corresponding to a wake state, a second sleep state corresponding to a rapid eye movement sleep state, and a third sleep state corresponding to one or more non-rapid eye movement sleep states (Arnold, [0052]: “The sleep states can include, among other things, an awake state and an asleep state. As described in greater detail below, some embodiments may include sleep states that represent different stages of sleep of a user”; [0111]: “Human sleep patterns can be described in terms of discrete stages of sleep. At the highest descriptive level, these break down into REM (rapid eye movement), and non-REM sleep stages, the latter including various light and deep sleep stages”; [0127]: “the statistical features derived may be used to train a multi-class classifier (via supervised machine learning) 128 to classify periods of time as a particular sleep stage (e.g., REM, various stage of non-REM (NREM) such as light or deep sleep).”);
identifying, using the classification of each of the plurality of epochs, a first sleep interval of consecutive epochs classified as a respective sleep state of the plurality of sleep states preceded by a second sleep interval of consecutive epochs classified as a different respective sleep state and followed by a third sleep interval of consecutive epochs classified as the different respective sleep state; and reclassifying the consecutive epochs of the first sleep interval from the respective sleep state to the different respective sleep state (Arnold, [0141]: “in the middle of a user's sleep, they somehow satisfy the not-worn criteria. In this case, the “not worn” time span is preceded and succeeded by “asleep” time. In this case, embodiments may classify the not-worn period as asleep time since it's unlikely that the user actually took off the device in the middle of a sleep period.”.).
However, the Arnold/McDarby combination does not teach reclassifying the sleep interval based on the first sleep interval being shorter than a threshold number of consecutive epochs.
Liu discloses an apparatus and method for judging sleep state. Specifically, Liu teaches classifying the sleep interval in accordance with the first sleep interval being shorter than a threshold number of consecutive epochs ([0058]: “once the cumulative sum within 5 minutes reaches a maximum threshold, the system enters an active state and automatically exits the sleep state”. Liu discloses only changing the state of the user if it is over a predetermined time, therefore this threshold of time can be incorporated into the Arnold/McDarby combination by only correcting the first sleep interval when it is less than a predetermined time threshold.). Arnold, McDarby, and Liu are analogous arts as they are all related to systems and methods that monitor a user’s motion to determine sleep state.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the threshold from Liu into the Arnold/McDarby combination as the threshold allows the method to reclassify sleep intervals only if they are shorter than a certain length. Since a quick change in sleep intervals is most likely to be incorrectly assigned, the threshold allows the method to correct this incorrect reading and provide a more accurate result and classification to the user.
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over the Arnold/McDarby/Liu combination as applied to claim 1 above, and further in view of Statan (US 20240090827).
Regarding claim 10, the Arnold/McDarby/Liu combination teaches the method of claim 1, wherein classifying is performed by a machine learning model (Arnold, [0087]: “the user activity classifier 124 may include a classification model generated based on running a supervised machine learning algorithm”).
However, the Arnold/McDarby/Liu combination is silent on the specific type of machine learning algorithm used.
Statan discloses a method for monitoring and improving sleep data. Specifically, Statan teaches wherein the classifying is performed by a bidirectional long-short-term-memory machine learning model ([0095]: “the machine-learned data analysis model 1810 can otherwise include various machine-learned models such as neural networks (e.g., deep neural networks), other types of machine-learned models, including non-linear models, and/or linear models, or binary classifiers. Neural networks can include feed-forward neural networks, recurrent neural networks (e.g., long short-term memory recurrent neural networks), convolutional neural networks, or other forms of neural networks.”). Arnold, McDarby, and Statan are analogous arts as they are all related to systems used to monitor parameters of a user to monitor sleep.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to use the machine learning type from Statan into the method from the Arnold/McDarby/Liu combination as the combination is silent on the type of machine learning algorithm used, and Statan discloses a suitable machine learning type in an analogous device.
Claims 11-12 are rejected under 35 U.S.C. 103 as being unpatentable over the Arnold/McDarby/Liu/Statan combination as applied to claim 10 above, and further in view of Chan (US 20150190086).
Regarding claim 11, the Arnold/McDarby/Liu/Statan combination teaches the method of claim 10.
However, the Arnold/McDarby/Liu/Statan combination does not teach further comprising: scaling the first plurality of features to a common range of values for use by the bidirectional long-short-term-memory machine learning model.
Chan discloses an automated sleep stage analysis device. Specifically, Chan teaches further comprising: scaling the first plurality of features to a common range of values for use by the bidirectional long-short-term-memory machine learning model ([0034]: “After feature extraction, the plurality of features can be scaled so that a subsequent machine learning classifier is able to deal with features of varying magnitudes”). Arnold, McDarby, Statan, and Chan are analogous arts as they are all related to systems used to monitor parameters of a user to monitor sleep.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the scaling from Chan into the Arnold/McDarby/Liu/Statan combination as it allows the data to be scaled, which can allow it to be analyzed easier and compared to each other in a simpler way, which can provide a more comprehensive analysis.
Regarding claim 12, the Arnold/McDarby/Liu/Statan combination teaches the method of claim 10.
However, the Arnold/McDarby/Liu/Statan combination does not teach further comprising: estimating a probability for each of the plurality of sleep states for each of the plurality of epochs, and classifying the state for each of the plurality of epochs using a maximum among the probability for each of the plurality of sleep states for each of the plurality of epochs.
Chan teaches further comprising: estimating a probability for each of the plurality of sleep states for each of the plurality of epochs, and classifying the state for each of the plurality of epochs using a maximum among the probability for each of the plurality of sleep states for each of the plurality of epochs ([0035]: “After feature extraction, each epoch has a set of features associated with it (the feature vector X). The set of features and the feature vector X are used as an input to a machine learning classifier unit of the wireless sensor device. The machine learning classifier unit outputs a set of probabilities or "confidence" values that each epoch is associated with one of the 5 possible sleep stages. The machine learning classifier unit may be any classifier in which probabilistic outputs can be obtained or estimated”; [0040]: “temporal dynamics adjustment is utilized by the wireless sensor device after the machine learning classifier initially classifies each epoch based upon posterior probabilities. Considering the temporal dynamics of sleep can improve the automated sleep staging. For example, it is rare to go straight from being awake to stage N3 or REM sleep and deeper sleep stages are more likely early in the night and REM is more likely later in the night. Therefore, the wireless sensor device continually considers temporal dynamics and learns about what previous epochs were classified as to help with the classification of the current epoch that is being analyzed”; [0048]: “The wireless sensor device utilizes the outputs of the machine learning classifier to estimate posterior probabilities via step 442 classifying each epoch into one of the 5 stages of sleep. The posterior probabilities are converted into likelihoods via step 446 based upon learned prior probabilities via step 444. The wireless sensor device performs temporal dynamics adjustment on the outputted likelihood results via step 450 based upon learned transition probabilities”).
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the probability analysis from Chan into the Arnold/McDarby/Liu/Statan combination as it allows for additional analysis to be performed on the data, which can provide a more comprehensive and detailed analysis.
Claim 15 is rejected under 35 U.S.C. 103 as being unpatentable over the Arnold/McDarby/Liu combination as applied to claim 1 above, and further in view of Park (US 20070191742).
Regarding claim 15, the Arnold/McDarby/Liu combination teaches the method of claim 1.
However, the Arnold/McDarby/Liu combination does not teach further comprising: filtering the first motion data using a high-pass filter, wherein the one or more first motion features are extracted from the first motion data after filtering using the high-pass filter.
Park discloses a system to analyze a sleep structure. Specifically, Park teaches further comprising: filtering the first motion data using a high-pass filter, wherein the one or more first motion features are extracted from the first motion data after filtering using the high-pass filter ([0045]: “Low-Pass Filters (LPFs) 30a-30d and High-Pass Filters (HPFs) for outputting low-pass filtered signals and high-pass filtered signals 30e-30h from the outputs of the differential amplifiers 20a-20d, respectively, a multiplexer (MUX) 40 for selecting any one from among the outputs of the LPFs 30a-30d and the HPFs 30e-30h, a controller 50 for analyzing the sleeping person's sleeping pattern based on the output of the multiplexer”). Arnold, McDarby, and Park are analogous arts as they are all related to systems used to monitor parameters of a user to monitor sleep.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the filtering from Park into the Arnold/McDarby/Liu combination as it allows the method to filter data to ensure only the correct data is being used for analysis, which will provide the most accurate result.
Claim 16 is rejected under 35 U.S.C. 103 as being unpatentable over the Arnold/McDarby/Liu combination as applied to claim 1 above, and further in view of Gozani (US 20140309709).
Regarding claim 16, the Arnold/McDarby/Liu combination teaches the method of claim 1.
However the Arnold/McDarby/Liu combination does not teach further comprising: filtering the first motion data using a band-pass filter to generate the first stream of motion data.
Gozani discloses a system for detecting user sleep-wake state. Specifically, Gozani teaches further comprising: filtering the first motion data using a band-pass filter to generate the first stream of motion data ([0077]: “the single-axis acceleration components A.sub.x(t), A.sub.y(t), and A.sub.z(t) of accelerometer 152 are sampled at 50 Hz (although other sampling rates may also be used). Sampled acceleration components are band-pass filtered between 0.25 Hz and 12.0 Hz. The high-pass aspect of this band-pass filter reduces the effect of acceleration components which are not directly linked to body movement”). Arnold, McDarby, and Gozani are analogous arts as they are all related to systems used to monitor parameters of a user to monitor sleep.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the filtering from Gozani into the Arnold/McDarby/Liu combination as it allows the method to filter data to ensure only the correct data is being used for analysis, which will provide the most accurate result.
Claim 17 is rejected under 35 U.S.C. 103 as being unpatentable over the Arnold/McDarby/Liu combination as applied to claim 1 above, and further in view of Park and Heneghan (US 20210038087).
Regarding claim 17, the Arnold/McDarby/Liu combination teaches the method of claim 1.
However, the Arnold/McDarby/Liu combination does not teach further comprising: filtering the first motion data using a low-pass filter.
Park teaches further comprising: filtering the first motion data using a low-pass filter ([0045]: “Low-Pass Filters (LPFs) 30a-30d and High-Pass Filters (HPFs) for outputting low-pass filtered signals and high-pass filtered signals 30e-30h from the outputs of the differential amplifiers 20a-20d, respectively, a multiplexer (MUX) 40 for selecting any one from among the outputs of the LPFs 30a-30d and the HPFs 30e-30h, a controller 50 for analyzing the sleeping person's sleeping pattern based on the output of the multiplexer”).
However, the Arnold/McDarby/Liu/Park combination does not teach down-sampling the first motion data from a first sampling rate to a second sampling rate lower than the first sampling rate.
Heneghan discloses a method for monitoring physiological signs. Specifically, Heneghan teaches down-sampling the first motion data from a first sampling rate to a second sampling rate lower than the first sampling rate ([0109]: “the signal is thresholded and summed into non-overlapping two second bins to give an actimetry count. The two second epochs can then be downsampled to the appropriate epoch”). Arnold, McDarby, Park, and Heneghan are analogous arts as they are all related to systems used to monitor parameters of a user.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the down-sampling from Heneghan into the Arnold/McDarby/Liu/Park combination as it allows for more processing that can provide more accurate data that can be used for the analysis, which can provide a more accurate result.
Claim 18 is rejected under 35 U.S.C. 103 as being unpatentable over the Arnold/McDarby/Liu combination as applied to claim 1 above, and further in view of Nicolae (US 20220117558).
Regarding claim 18, the Arnold/McDarby/Liu combination teaches the method of claim 1, further comprising: for each epoch: converting the first motion data into a first frequency domain representation for the first channel of the multi-channel motion sensor, a second frequency domain representation for the second channel of the multi-channel motion sensor, and a third frequency domain representation for a third channel of the multi-channel motion sensor (McDarby, Fig. 11; Abstract: “At least a portion of the detected signals may be analyzed to calculate respiration variability. The respiration variability may include one or more of variability of respiration rate and variability of respiration amplitude”; [0059]: “A processing means is provided to take the original movement signal (the entire or raw detected movement signal) and to split it into “respiration” and “non-respiration” signals, by using frequency domain filtering”; [0052]: “the processor may be used to extract information about breathing and motion, and higher order information such as the sleep stage … A motion sensor (for detection of general bodily movement and respiration)”. The channels can be the same channel, since first motion data from Arnold is only from a single axis, therefore the frequency domain features are extracted from the channels. The processing of the original movement signal by splitting it into the respiration and non-respiration signals is the step of converting the first motion data (the original movement signal) into the frequency domain representations.).
However, the Arnold/McDarby/Liu combination does not teach computing a first signal-to-noise ratio using the first frequency domain representation, a second signal-to-noise ratio using the second frequency domain representation, and a third signal-to-noise ratio using the third frequency domain representation; wherein the selected channel corresponds to a respective channel of the first channel, the second channel, or the third channel with a maximum signal-to-noise ratio among the first signal-to-noise ratio, second signal-to-noise ratio and third signal-to-noise ratio.
Nicolae discloses a wearable system for monitoring a user. Specifically, Nicolae teaches computing a first signal-to-noise ratio using the first frequency domain representation, a second signal-to-noise ratio using the second frequency domain representation, and a third signal-to-noise ratio using the third frequency domain representation; wherein the selected channel corresponds to a respective channel of the first channel, the second channel, or the third channel with a maximum signal-to-noise ratio among the first signal-to-noise ratio, second signal-to-noise ratio and third signal-to-noise ratio ([02479]: “the system 2600 may evaluate the quality of a signal, e.g., using any conventional metrics such as signal-to-noise ratio”). Arnold, McDarby, and Nicolae are analogous arts as they are all related to systems used to monitor parameters of a user to monitor sleep.
Therefore, it would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to include the signal-to-noise ratio from Nicolae into the Arnold/McDarby/Liu combination as it allows the combination to perform more analysis on the data, ensuring a more accurate and comprehensive result.
Response to Arguments
All of applicant’s argument regarding the rejections and objections previously set forth have been fully considered and are persuasive unless directly addressed subsequently.
Applicant's arguments with respect to the 101 rejection have been fully considered but they are not persuasive. Applicant argues that the subject matter of the claims is integrated into a practical application, however all the limitations of the claims are either an abstract idea that can be performed mentally or are insignificant pre-solution activity such as using a generic sensor to gather data (i.e., an accelerometer) and then performing calculations and analysis on the data, which does not integrate the limitations into a practical idea and therefore are rejected under 35 U.S.C. 101. Applicant argues that the processing provides significant improvements to the real-life operations of the devices. This is not persuasive since the claimed processing merely provides a better algorithm, not an improved computer. An improved mental process is still a mental process even if such a mental process results in more accurate results.1,2 Also, having the claims focus on determining the sleep state of the human body is not itself limiting the claims to improving the technology because cases that involve practical, technological improvements extend beyond simply improving the accuracy of a prediction.3 See, e.g., McRO, Inc. v. Bandai Namco Games America Inc., 837 F.3d 1299, 1315 (Fed. Cir. 2016) (“The claimed process uses a combined order of specific rules that renders information into a specific format that is then used and applied to create desired results: a sequence of synchronized, animated characters.”); Finjan, Inc. v. Blue Coat Sys., Inc., 879 F.3d 1299, 1304 (Fed. Cir. 2018) (finding patent eligible a claim drawn to a behavior-based virus scan that protects against viruses that have been “cosmetically modified to avoid detection by code-matching virus scans”); Enfish, LLC v. Microsoft Corp., 822 F.3d 1327, 1330, 1333 (Fed. Cir. 2016) (discussing patent eligible claims directed to “an innovative logical model for a computer database” that included a self-referential table allowing for greater flexibility in configuring databases, faster searching, and more effective storage); CardioNet, LLC v. InfoBionic, Inc., 955 F.3d 1358, 1368 (Fed. Cir. 2020) (explaining that the claims at issue focus on a specific means for improving cardiac monitoring technology; they are not “directed to a result or effect that itself is the abstract idea and merely invoke generic processes and machinery” (quoting McRO, 837 F.3d at 1314)). The claims do not add additional limitations to remove the claim from the abstract idea, since all the claimed limitations are either generic data gathering or are calculation and evaluation steps of a mental process, therefore the 101 rejection is maintained.
Applicant’s arguments with respect to the 103 rejections have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ERIN K MCCORMACK whose telephone number is (703)756-1886. The examiner can normally be reached Mon-Fri 7:30-5.
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, Jason Sims can be reached at 5712727540. 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.
/E.K.M./Examiner, Art Unit 3791
/MATTHEW KREMER/Primary Examiner, Art Unit 3791
1 “[T]he improvement in computational accuracy alleged here does not qualify as an improvement to a technological
process; rather, it is merely an enhancement to the abstract mathematical calculation of haplotype phase itself...The
different use of a mathematical calculation, even one that yields different or better results, does not render patent
eligible subject matter.” In re Board of Trustees of Leland Stanford Junior University, 991 F.3d 1245 (Fed. Cir.
2021).
2 “[A] claim for a new abstract idea is still an abstract idea.” Synopsys, Inc. v. Mentor Graphics Corp, 839 F.3d 1138
(Fed. Cir. 2016).
3 See In re Board of Trustees of Leland Stanford Junior University, 991 F.3d 1245 (Fed. Cir. 2021).
Prosecution Insights