DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first
inventor to file provisions of the AIA .
Response to Arguments
Applicant's arguments filed 2/23/2026 have been fully considered but they are not persuasive.
35 U.S.C. 101:
Regarding independent claims 1, 9, and 17, applicant argues that the actual technical
features/steps cannot be performed merely in the human mind as a mental process and cannot be considered a mere abstract idea. Specifically, the applicant argues that these limitation do not recite an abstract idea:
extract gait features from EEG signals acquired from a subject.
the first stage blocks and the second stage blocks being trained based on different gait data obtained solely from the subject at different time points.
at least one of the second stage blocks is a feature extractor block, each of the feature extractor block corresponding to respective ones of the first stage blocks.
After careful consideration, the examiner respectfully argues that the additional elements either
recite extra-solution activity or computer implementation to perform the abstract idea. Reasons are discussed below:
“Extract gait features from EEG signals acquired from a subject.” This limitation in combination with the feature extractor block is recite as pre-solution activity to the step of data gathering, specifically EEG signals.
“The first stage blocks and the second stage blocks being trained based on different gait data obtained solely from the subject at different time points.” The first and second stage blocks are recited as machine learning components for a generic processor. This is recited with no additional structure and is used as computer implementation to analyze different gait data at different time points, which is an abstract idea.
“At least one of the second stage blocks is a feature extractor block, each of the feature extractor block corresponding to respective ones of the first stage blocks.” The feature extractor block in combination with each first stage block is recited as pre-solution activity to the step of data gathering.
Applicant is reminded that abstract ideas cannot provide a practical application or significantly
more (e.g., an improvement). Both Step 2A Prong 2 and Step 2B require an additional element, not an abstract idea, to provide a practical application or significantly more (e.g., an improvement). See Genetic Technologies Limited v. Merial LLC (Fed Cir 2016). Here, the additional elements of claims 1 and 17 are merely generically recited computer elements used as tools for executing the abstract ideas or insignificant extra-solution activity.
35 U.S.C. 102 and 103:
Regarding applicant' s arguments that Chang teaches away from the claimed limitation, it is
noted that for a reference to be considered to teach away from a claimed limitation such reference must criticize, discredit, or otherwise discourage the proposed limitation. In re Fulton, 73 USPQ2d 1141 (Fed. Cir. 2004). The applicant is further advised that disclosed examples and/or preferred embodiments do not constitute a teaching away from a broader disclosure or nonpreferred embodiments, even if such nonpreferred embodiments are described as somewhat inferior. See In re Susi, 169 USPQ 423 (CCPA 1971), and In re Gurley, 31 USPQ2d 1130 (Fed. Cir. 1994).
In this case, the applicant argues that Chang teaches extracting gait data based on
movement/physical data instead of EEG signal, teaches away from the proposed combination. The examiner respectfully disagrees and argues that it is well known for EEG signals to determine gait features by capturing cortical activity patterns associated with locomotion. It is also disclosed in [70] that “the exosuit may include sensors (e.g., one or more electroencephalograph (EEG) sensors) that may be able to monitor brain activity that may be used to detect a user's desire to perform a particular movement. For example, if the user is sitting down, an EEG sensor may sense the user's desire to stand up and cause the exosuit to prime itself to assist the user in a sit-to-stand assistive movement.”
Furthermore, applicant argues that Chang does not teach “using a two-stage machine learning
model, extract gait features from EEG signals acquired from a subject; the first stage blocks and the second stage blocks being trained based on different gait data obtained solely from the subject at different time points; wherein at least one of the second stage blocks is a feature extractor block, each of the feature extractor block corresponding to respective ones of the first stage blocks.”
After further search and consideration, the examiner respectfully disagrees and argues:
using a two-stage machine learning model, extract gait features from EEG signals acquired from a subject (paragraph 70-72, 210 and 284-288); EEG sensors may be used to detect user’s desire to perform gait movements. A combination of algorithmic and machine learning models is disclosed. Furthermore, the machine learning engine is disclosed to operate in a second period of time indicating a two-stage machine learning model, as disclosed in [35] of the applicant’s specification. It is disclosed that the machine learning engine operates in at least two period of time. Therefore, a two-stage machine learning model is disclosed.
and determine a predicted gait based on the gait features, wherein the two-stage machine learning model includes multiple first stage blocks and multiple second stage blocks, the first stage blocks and the second stage blocks being trained based on different gait data obtained solely from the subject at different time points (paragraph 204-210 and 284-288); A combination of machine learning models may use one or more sets of matrices. The matrices may utilize multiple learning algorithms and functions. Therefore, multiple first stage blocks and multiple second stage blocks are disclosed. The learning model may be used for any given particular time or planned activity. Therefore, a first and second stage may be used to train different gait data (from algorithms and functions) at different time points.
wherein at least one of the second stage blocks is a feature extractor block, each of the feature extractor block corresponding to respective ones of the first stage blocks (paragraph 240). “Machine learning models can be built using any suitable features, such as the biomechanical signals discussed above or additional features that a model may identify and extract.” It is already disclosed that a combination of machine learning models may use one or more sets of matrices. The matrices may utilize multiple learning algorithms and functions. Therefore, the combination of machine learning models may have a second stage block to be a feature extractor block corresponding to each first stage block and used to extract features.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Claims 1, 9 and 17 recite an apparatus, method, and a non-transitory computer with instructions for performing operations of the device comprising:
extracting gait features from the subject;
determining a predicted gait based on the gait features;
To determine whether a claim satisfies the criteria for subject matter eligibility, the claim is
evaluated according to a stepwise process as described in MPEP 2106(III) and 2106.03-2106.05. The instant claims are evaluated according to such analysis.
Step 1: Is the claim to a process, machine, manufacture or composition of matter?
Claim 1 is directed to an apparatus claim 17 is directed to a method to perform the steps of the
method and thus meet the requirements for step 1.
Step 2A (Prong 1): Does the claim recite an abstract idea, law of nature, or natural
phenomenon?
Claims 1 and 17 recite an apparatus and method for performing operations of the device comprising:
extracting gait features from the subject;
determining a predicted gait based on the gait features;
If a claim limitation, under its broadest reasonable interpretation, covers performance of the
limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Therefore, claims 1 and 17 recite an abstract idea of a mental process.
Claims 1 and 17 recite the abstract idea of a mental process. The limitations as drafted in the
claims, under its broadest reasonable interpretation, covers performance of the claimed steps in the mind, but for the recitation of a generic processor. Other than reciting a generic processing system and memory, nothing in the elements of the claims precludes the step from practically being performed in the mind or manually by a clinician. For example:
“Extracting gait features from the subject;” A physician may look at a patient and obtain gait data from observation.
“Determining a predicted gait based on the gait features;” Based on observed data, a physician may predict an outcome of the patients’ movements.
Step 2A (Prong 2): Does the claim recite additional elements that integrate the judicial
exception into a practical application?
Claims 1 and 17 recite the additional elements of a “processor”, “a memory”, “EEG”, “a two-stage machine learning model” and a “multiple first stage blocks and multiple second stage blocks”, “a feature extractor block”, which are being interpreted as a processor of a data gathering device.
EEG is recited as a pre-solution activity to the step of data gathering.
“A two-stage machine learning model”, “multiple first stage blocks and multiple second stage blocks”, and “a feature extractor block”, are mere computer implementation to analyze the gait data to provide a predicted outcome. These additional elements are well-known and the technological improvements is towards the abstract idea of analyzing.
However, these elements are recited at a high level of generality performing the function of generic data processing such that they amount to no more than mere instructions to simply implement the abstract idea using generic computer components. See MPEP 2106.05(b) and (f).
Accordingly, the additional elements do not integrate the abstract idea into a practical
application.
Step 2B: Does the claim recite additional elements that amount to significantly more than the
judicial exception?
The additional elements when considered individually and in combination are not enough to
qualify as significantly more than the abstract idea. As discussed above with respect to integration of the abstract idea into a practical application, “processor”, “a memory”, “EEG”, “a two-stage machine learning model” and a “multiple first stage blocks and multiple second stage blocks”, “a feature extractor block”, which are being interpreted as a processor of a data gathering device
EEG is recited as a pre-solution activity to the step of data gathering.
“A two-stage machine learning model”, “multiple first stage blocks and multiple second stage blocks”, and “a feature extractor block”, are mere computer implementation to analyze the gait data to provide a predicted outcome. These additional elements are well-known and the technological improvements is towards the abstract idea of analyzing.
which are being interpreted as a processor of a data gathering device as recited to perform the steps of:
extracting gait features from the subject;
determining a predicted gait based on the gait features;
amount to no more than mere instructions to apply the exception using generic computer
components. Mere instructions to apply an exception using generic components cannot provide an inventive concept. These additional elements are well‐understood, routine (For example Chang et al. US Pub.: US 20190283247 A1, hereinafter Chang) teaches a data gathering device with a processor and memory, and conventional limitations that amount to mere instructions or elements to implement the abstract idea. In addition, the end result of the system/method, the essence of the whole, is a patent-ineligible concept. Therefore, the claims are not patent eligible.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis
for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1-9, 15, and 17-20 are rejected under 35 U.S.C. 102(a)(1) as being unpatentable by
Chang et al. US Pub.: US 20190283247 A1, hereinafter Chang.
Regarding claim 1, Chang teaches a system for gait training, comprising:
memory storing instructions (fig. 11, 1104; paragraph 188-189);
and a processor coupled to the memory and configured to process the stored instructions to implement (fig. 11, 1102; paragraph 188-189);:
a gait prediction module configured to: using a two-stage machine learning model, extract gait features from EEG signals acquired from a subject (paragraph 70-72, 210 and 284-288); EEG sensors may be used to detect user’s desire to perform gait movements. A combination of algorithmic and machine learning models is disclosed. Furthermore, the machine learning engine is disclosed to operate in a second period of time indicating a two-stage machine learning model, as disclosed in [35] of the applicant’s specification.
and determine a predicted gait based on the gait features, wherein the two-stage machine learning model includes multiple first stage blocks and multiple second stage blocks, the first stage blocks and the second stage blocks being trained based on different gait data obtained solely from the subject at different time points (paragraph 204-210 and 284-288); A combination of machine learning models may use one or more sets of matrices. The matrices may utilize multiple learning algorithms and functions. Therefore, multiple first stage blocks and multiple second stage blocks are disclosed. The learning model may be used for any given particular time or planned activity.
wherein at least one of the second stage blocks is a feature extractor block, each of the feature extractor block corresponding to respective ones of the first stage blocks (paragraph 240). “Machine learning models can be built using any suitable features, such as the biomechanical signals discussed above or additional features that a model may identify and extract.”
Regarding claim 2, Chang teaches wherein the second stage blocks include a self-attention block
(SAB) (paragraph 210). It is disclosed in [63] of the applicant’s specification that the SAB is “configured to consider the information of all the time points and assign weights to different time points based on importance.” The art discloses a learning network that has “numeric weights (e.g., initially configured with initial weight values) that can be tuned based on experience, making the neural network adaptive to inputs and capable of learning (e.g., learning pattern recognition).” Therefore, an SAB is disclosed.
Regarding claim 3, Chang teaches wherein the processor is further configured to implement: a
feature fusion module configured to: align the predicted gait and an acquired gait of the subject to obtain a gait data; and determine a gait recovery assessment based on features extracted from the gait data using a convolutional neural network (paragraph 28, 210, and 240). The art discloses extracting acquired gait features and predict a recovery rate. A convolutional neural network is disclosed because learning models are configured to also analyze recognition patterns from images provided from the LEDs of the exosuit.
Regarding claim 4, Chang teaches wherein the convolutional neural network is trained from gait
data acquired from multiple subjects (paragraph 183 and 210). The data can be communicated to a central resource, where data from multiple participants can be processed.
Regarding claim 5, Chang teaches wherein each of the feature extractor block is identical to the
respective first stage block (paragraph 240 and 284-288). The feature extractor block is disclosed as a machine learning model. This model may be identical to any one of the first stage blocks in both the first and second stage machine learning models.
Regarding claim 6, Chang teaches wherein the first stage blocks and the second stage blocks are
trained sequentially in each gait training session (paragraph 98 and 284-288). Real time alignment guidance is disclosed. Furthermore, the machine learning model is operating in at least a two stage machine learning model, wherein the first and second stage is a first and second time period, respectively. Therefore, the blocks are trained sequentially.
Regarding claim 7, Chang teaches wherein each of the feature extractor block of a present gait
training session corresponds to respective ones of the first stage blocks from a previous gait training session (paragraph 240 and 284-288). The feature extractor block is disclosed as a machine learning model. This model may be identical to any one of the first stage blocks in both the first and second stage machine learning models.
Regarding claim 8, Chang teaches wherein the first stage blocks are trained over multiple gait
training sessions (paragraph 98 and 284-288). Real time alignment guidance is disclosed. Furthermore, the first stage blocks are trained over gait sessions in real-time.
Regarding claim 9, Chang teaches wherein during training of the second stage blocks,
parameters of each of the feature extractor block are fixed (paragraph 240 and 284-288). The feature extractor block is disclosed as a machine learning model. This model may be identical to any one of the first stage blocks in both the first and second stage machine learning models.
Regarding claim 15, Chang teaches wherein the processor is further configured to implement: a visual module configured to: provide a visual representation of the subject, the visual representation comprising a visualization of one leg of the subject corresponding to the predicted gait; and a visualization of another leg of the subject corresponding to an acquired gait (fig. 11, 1112; paragraph 193-194). User subsystem 1100 may also include one or more output assemblies 1112 that may present information (e.g., graphical, audible, and/or tactile information) to a user of subsystem 1100.
Regarding claim 17, Chang teaches a method of gait training, comprising:
using a two-stage machine learning model, extracting gait features from EEG signals acquired from a subject (paragraph 70-72, 210 and 284-288); EEG sensors may be used to detect user’s desire to perform gait movements. A combination of algorithmic and machine learning models is disclosed. Furthermore, the machine learning engine is disclosed to operate in a second period of time indicating a two-stage machine learning model, as disclosed in [35] of the applicant’s specification.
and determining a predicted gait based on the gait features, wherein the two-stage machine learning model includes multiple first stage blocks and multiple second stage blocks, the first stage blocks and the second stage blocks being trained based on different gait data obtained solely from the subject at different time points (paragraph 204-210 and 284-288); A combination of machine learning models may use one or more sets of matrices. The matrices may utilize multiple learning algorithms and functions. Therefore, multiple first stage blocks and multiple second stage blocks are disclosed. The learning model may be used for any given particular time or planned activity.
wherein at least one of the second stage blocks is a feature extractor block, each of the feature extractor block corresponding to respective ones of the first stage blocks (paragraph 240). “Machine learning models can be built using any suitable features, such as the biomechanical signals discussed above or additional features that a model may identify and extract.”
Regarding claim 18, Chang teaches further comprising: aligning the predicted gait and an acquired gait of the subject to obtain a gait data; and determining a gait recovery assessment based on features extracted from the gait data using a convolutional neural network (paragraph 28, 210, and 240). The art discloses extracting acquired gait features and predict a recovery rate. A convolutional neural network is disclosed because learning models are configured to also analyze recognition patterns from images provided from the LEDs of the exosuit.
Regarding claim 19, Chang teaches further comprising: providing a visual representation of the
subject, the visual representation comprising a visualization of one leg of the subject corresponding to the predicted gait; and a visualization of another leg of the subject corresponding to an acquired gait (fig. 11, 1112; paragraph 73-75 and 193-194). Figure 1A-C shows LDM in two leg portions of the patient. User subsystem 1100 may also include one or more output assemblies 1112 that may present information (e.g., graphical, audible, and/or tactile information) to a user of subsystem 1100.
Regarding claim 19, Chang teaches further comprising training the first stage blocks over multiple gait training sessions (paragraph 98 and 284-288). Real time alignment guidance is disclosed. Furthermore, the first stage blocks are trained over gait sessions in real-time.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections
set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 10-11 are rejected under 35 U.S.C. 103 as being unpatentable over Chang.
Regarding claim 10-11, Chang discloses “wherein the EEG signals are input into the feature
extractor block and the respective variable block,” but does not disclose expressly the “wherein the second stage blocks comprise at least one pair of parallel blocks, each of the at least one pair of parallel blocks comprises one of the feature 31 extractor block in parallel with a respective variable block, and are identical.” It would have been an obvious matter of duplication of parts to a person of ordinary skill in the art to modify the system as taught by Chang with “the second stage blocks comprise at least one pair of parallel blocks, each of the at least one pair of parallel blocks comprises one of the feature 31 extractor block in parallel with a respective variable block,” because the court held that mere duplication of parts has no patentable significance unless a new and unexpected result is produced.) In re Harza, 274 F.2d 669, 124 USPQ 378 (CCPA 1960).
Therefore, it would have been an obvious matter of design choice to modify Chang to obtain the invention as specified in the claim(s).
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over Chang in view of Kollada
et al. US Pub.: US 20220392637 A1, hereinafter Kollada.
Regarding claim 14, Chang does not explicitly teach wherein the feature extractor block is
configured as at least one of: a Temporal Convolution Block (TCB) and a Spatial Convolution Block (SCB).
Kollada, in the same field of endeavor, teaches wherein the feature extractor block is configured
as at least one of: a Temporal Convolution Block (TCB) and a Spatial Convolution Block (SCB) (paragraph 46-48 and 64). It is disclosed that “the dynamic encoding subnetwork is a convolutional-based subnetwork, which preserves temporal information of the embeddings from the different modalities.” This equates to the a Temporal Convolution Block. Furthermore, “the dynamic embeddings allows integration of information across modalities within a temporal context (e.g. facial expression change and vocal inflection while a salient word is uttered).”
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the processor of Chang to add the temporal Convolution Block from Kollada for the benefit of allowing integration of information across modalities within a temporal context.
Claims 16 is rejected under 35 U.S.C. 103 as being unpatentable over Chang in view of Connor
et al. US Pub.: US 20150309563 A1, hereinafter Connor.
Regarding claim 16, Chang teaches further comprising: a gait acquisition module for obtaining in
acquired gait of the subject (paragraph 124). Module 540 can leverage data obtained from sensors placed on the equipment in conjunction with or to the exclusion of sensors in the exosuit.
However, Chang does not teach wherein the gait acquisition module includes goniometers for measuring hip angles, knee angles, and ankle joint angles of the subject.
Conner, in the same field of endeavor, teaches wherein the gait acquisition module includes
goniometers for measuring hip angles, knee angles, and ankle joint angles of the subject (fig. 1; paragraph 55, 159, 163, 166, 210, and 213). The goniometer may be placed at the joints of a patient, shown in figure 1. It is disclosed in [159] “this device can span the surface of the portion of the body containing one or more body joints selected from the group consisting of: hip, knee and ankle.
Therefore, it would have been obvious to one of ordinary skill in the art before the effective
filing date of the claimed invention to modify the exosuit of Chang to add the goniometer at each of the hip, knee, and ankle joint from Conner for the benefit of measuring hip angles, knee angles, and ankle joint angles of the subject.
Claims 12-13 are indicated as having no prior art rejection. However, the claims are not
considered allowable subject matter due to the 35 U.S.C. 101 rejection disclosed above.
The following is a statement of reasons for the indication of no prior art rejection: The
closest prior art teaches a first and second stage block, disclosed as a combination of learning models, and a feature extractor model (paragraph 240 and 284-288). The feature extractor block is disclosed as a machine learning model. This model may be identical to any one of the first stage blocks in both the first and second stage machine learning models. Furthermore, the art teaches the EEG signals are input into each of the feature extractor block (paragraph 70-72, 210 and 284-288); EEG sensors may be used to detect user’s desire to perform gait movements. A combination of algorithmic and machine learning models is disclosed. Furthermore, the machine learning engine is disclosed to operate in a second period of time indicating a two-stage machine learning model, as disclosed in [35] of the applicant’s specification. However, no prior art was found to teach every limitation of dependent claim 12, specifically, “wherein during training of the second stage blocks, parameters of each of the feature extractor block are fixed and parameters of each of the variable block are updateable,” in combination with dependent claim 10.”
Furthermore, no prior art was found to teach every limitation of dependent claim 13, specifically, “wherein the second stage blocks include a concatenation block for receiving respective outputs from one of the at least one pair of parallel blocks,” in combination with dependent claim 10.”
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to THIEN J TRAN whose telephone number is (571)272-0486. The examiner can normally be reached M-F. 8:30 am - 5:30 pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Benjamin Klein can be reached at 571-270-5213. 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.
/T.J.T./Examiner, Art Unit 3792
/Benjamin J Klein/Supervisory Patent Examiner, Art Unit 3792
Prosecution Insights