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 .
This Office Action is made in response to applicant’s preliminary amendment filed on 01/16/2025. Claims 1-20 are currently pending in the application. An action follows below:
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(B) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
Claims 9 and 19 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention.
As per claims 9 and 19, these claims recite three expressions. Since the specification, including these claims, does not explicitly define all terms and/or variables in these expressions, it is considered that the invention is not clearly defined.
Further, note that the specification is not the measure of invention. Therefore, limitations contained therein can’t be read into the claims for the purpose of avoiding the prior art. See In re Sporck, 55 CCPA 743, 386 F.2d 924, 155 USPQ 687 (1968). In the instant case, all terms and/or variables in these expressions should be defined in the claims, in order to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
Claims 1-20 are rejected under 35 U.S.C. 112(a), as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for pre-AIA the inventor(s), at the time the application was filed, had possession of the claimed invention.
Note that, in order to satisfy its burden under the written description requirement, a patent application must disclose the full scope of the claim. Univ. of Rochester v. G.D. Searle & Co., 358 F.3d 916, 920 (Fed. Cir. 2004) (The purpose of the written description requirement is to “ensure that the scope of the right to exclude, as set forth in the claims, does not overreach the scope of the inventor’s contribution to the field of art as described in the patent specification.”.)
As per claim 1, this claim recites limitations, “a decoder application that configures the processor to: obtain a neural signal from the neural signal recorder; translate the neural signal into a command for an interface device communicatively coupled to the decoder, using the neural decoder model; infer an intended target of a user based on the command using the inference model; annotate the neural signal with the inferred intended target; and retrain the neural decoder model using the annotated neural signal as training data”, in last 10 lines. A person skilled in the computer or relevant art would have readily recognized that, the decoder application is mere coding software including instructions stored in the memory and does not have any physical structure capable of performing any function(s) itself. In the instant case, the original disclosure, including this claim, does not explicitly discuss in detail how the decoder application, which is a mere software including instructions stored in the memory and does not have any physical structure, itself performs a function, “configuring the processor to perform operations,” as recited in the above underlined limitations, so as to reasonably convey to one skilled in the relevant art that the inventor(s), at the time the application was filed, had possession of the claimed invention. Further, see the above bolded note.
Accordingly, the original disclosure does not contain such description and details regarding to the above underlined limitations of this claim, so as to reasonably convey to one skilled in the relevant art that the inventor(s), at the time the application was filed, had possession of the claimed invention.
As per claims 2-10, these claims are therefore rejected for at least the reason set forth in claim 1.
In addition to claim 2, this claim further recites limitations, “wherein the decoder application further directs the processor to: obtain an additional neural signal from the neural signal recorder; translate the additional neural signal into an additional command for the interface device using the retrained neural decoder model; and enact the additional command using the interface device.” A person skilled in the computer or relevant art would have readily recognized that, the decoder application is mere coding software including instructions stored in the memory and does not have any physical structure capable of performing any function(s) itself. In the instant case, the original disclosure, including this claim, does not explicitly discuss in detail how the decoder application, which is a mere software including instructions stored in the memory and does not have any physical structure, itself performs a function, “directing the processor to perform operations,” as recited in the above underlined limitations, so as to reasonably convey to one skilled in the relevant art that the inventor(s), at the time the application was filed, had possession of the claimed invention. Further, see the above bolded note. Accordingly, the original disclosure does not contain such description and details regarding to the above underlined limitations of this claim, so as to reasonably convey to one skilled in the relevant art that the inventor(s), at the time the application was filed, had possession of the claimed invention.
In addition to claim 6, this claim further recites limitations, “wherein the decoder application further configures the processor to obtain a confidence value from the inference model indicating a predicted accuracy of the inferred intended target.” See the discussion in the rejection of claim 1. Accordingly, the original disclosure does not contain such description and details regarding to the above underlined limitations of this claim, so as to reasonably convey to one skilled in the relevant art that the inventor(s), at the time the application was filed, had possession of the claimed invention.
In addition to claim 10, this claim further recites limitations, “wherein the decoder application further directs the processor to: obtain a plurality of neural signals from the neural signal recorder recorded during a predefined time window; translate each neural signal from the plurality of neural signals into a respective command for the interface device, using the neural decoder model; infer an intended target of a user based on each respective command; annotate each neural signal from the plurality of neural signals with the respective inferred intended target; and retrain the neural decoder model using the annotated plurality of neural signals.” See the discussion in the rejection of claim 2. Accordingly, the original disclosure does not contain such description and details regarding to the above underlined limitations of this claim, so as to reasonably convey to one skilled in the relevant art that the inventor(s), at the time the application was filed, had possession of the claimed invention.
As per claims 11-20, these claims are substantially similar to claims 1-10 except that claims 11-20 are method claims and claims 1-10 are apparatus claims. See the discussion in the rejection of claims 1-10 for similar limitations. Accordingly, the original disclosure does not contain such description and details regarding to the above underlined limitations of these claims, so as to reasonably convey to one skilled in the relevant art that the inventor(s), at the time the application was filed, had possession of the claimed invention.
Notice to Applicant(s)
Examiner notes that the specification is not the measure of invention. Therefore, limitations contained therein can’t be read into the claims for the purpose of avoiding the prior art. See In re Sporck, 55 CCPA 743, 386 F.2d 924, 155 USPQ 687 (1968).
Further, the names/ terms of the features/elements used in the pending application or pending claims may be different from the names/terms of the matching features/ elements of the prior arts; however, the matching features/ elements of the prior arts contain all characteristics/ functions of the features/elements DEFINED by the pending claims.
Note that in order to avoid confusion, the below citations in the below rejection(s) are mere one or more places in the reference to disclose the "claimed" limitation(s) and/or are directed to one or more of embodiments disclosed by the cited reference(s). In other words, the “claimed” features/limitations may be read in other places in the reference or other embodiments of the reference. In order to better understand how the claimed limitations are taught by the reference(s), a review of the entire reference(s) is suggested by the examiner. Applicant is reminded a prior art reference must be considered in its entirety, i.e., as a whole, including portions that would lead away from the claimed invention as not all relevant paragraphs may have been cited in the rejection. W.L. Gore & Associates, Inc. v. Garlock, Inc., 721 F.2d 1540, 220 USPQ 303 (Fed. Cir. 1983), cert. denied, 469 U.S. 851 (1984).
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 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.
Due to the above rejections under 35 U.S.C. 112(a) and/or 35 U.S.C. 112(b) and, for the sake of applying the prior art(s) in order to compact prosecution, the following art rejections are based as best understood by Examiner in view of the originally filed specification and drawings.
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.
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 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-4, 6, 8, 11-14, 16 and 18 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Francis et al. (US 2019/0025917 A1; hereinafter Francis.)
As per claim 1, Francis discloses a closed-loop recalibrating brain-computer interface (BCI) (see at least Figs. 1A, 1B; Abstract; ¶ [0224]; ¶ [0253], disclosing a reinforcement learning brain-machine interface (RL-BMI) can have a policy that governs how detected signals, emanating from a motor cortex of a subject's brain, are translated into action; closed loop RL-BMI) comprising:
a neural signal recorder configured to record brain activity (see at least ¶ [0012], disclosing: detecting a motor signal having a characteristic and emanating from a motor cortex of a subject’s brain;) and
a decoder (see at least Figs. 1A, 1B; ¶ [0111], disclosing the SAC-BMI agent (actor) 102 strives to decode the motor signal to result in an intended action of the external actuator (end effector)) comprising:
a processor (see at least Fig. 17; ¶ [0265], disclosing a processing system 1302 including a processor for executing instructions;) and
a memory (see at least Fig. 17; ¶ [0265], disclosing instructions stored in a machine-readable medium 1310,) where the memory contains:
a neural decoder model (see at least Figs. 1A, 1B; ¶¶ [0008], [0110]-[0111], disclosing the actor [102] being the portion of the system that decodes neural activity into actions;)
an inference model (see at least Figs. 1A, 1B; ¶¶ [0008], [0110]-[0111], disclosing the critic decoding the neural activity from the same region or another into an evaluative signal used to update the actor;) and
a decoder application that configures the processor to (see at least ¶¶ [0270]-[0271], disclosing the server 1406 is configured to run applications that may be accessed and controlled at the client devices 1402):
obtain a neural signal from the neural signal recorder (see at least Figs. 1A, 1B; ¶¶ [0109], [0111], disclosing decoded neural signals from the motor cortex as an actor; The actor 102 receives motor signal from the brain 106;)
translate the neural signal into a command for an interface device communicatively coupled to the decoder, using the neural decoder model (see at least Figs. 1A, 1B; ¶¶ [0110]-[0111], disclosing the SAC-BMI agent (actor) 102 striving to decode the motor signal to result in an intended action of the external actuator (end effector);)
infer an intended target of a user based on the command using the inference model (see at least ¶¶ [0110], [0134], [0185], disclosing the critic 104 receives reward signal from the brain 106; BMI user would like to correct his or her movements on the way to reaching a target and that, in real world situations, there is no model of the environment; TD learning is a logical RL algorithm to use. Actor-critic methods are TD methods that have the actor (policy) and the critic (estimated value function or the evaluative feedback signal provider) exist as two independent entities; the evaluative feedback was simply based on whether the “hand feedback” cursor was moving toward or away from the rewarding target;)
annotate the neural signal with the inferred intended target (see at least ¶ [0110], disclosing the multisensory feedback to the brain 106 with respect to the action performed results in a critic signal, which is labeled as rewarding or non-rewarding by a classifier;) and
retrain the neural decoder model using the annotated neural signal as training data (see at least ¶ [0110], disclosing such an evaluative scalar feedback is used to adapt the RL-BMI agent.)
As per claim 2, Francis discloses the decoder application further directing the processor to: obtain an additional neural signal from the neural signal recorder (see at least ¶¶ [0008], [0009], [0015];) translate the additional neural signal into an additional command for the interface device using the retrained neural decoder model (see at least ¶¶ [0008], [0009], [0015];) and enact the additional command using the interface device (see at least ¶¶ [0008], [0009], [0015], disclosing that the system will automatically update itself, if for instance there are changes in the neural input to the system. In the above idealization we assumed a perfect critic that could decode the evaluative signal from the neural activity; the neural activity in M1 can be mapped to desired movements by a decoder (actor) and the corresponding reward expectation signal extracted from the same neural ensemble could be utilized as an evaluative signal (critic) of the performed action to allow subsequent autonomous BMI improvement; adjusting the policy based on the evaluation signal such that a subsequent motor signal, emanating from the motor cortex and having the characteristic, results in a second action, by the device, different from the first action.)
As per claim 3, Francis discloses the neural decoder model being a supervised machine learning model (see at least ¶ [0111], disclosing supervised actor-critic reinforcement learning brain machine interface (SAC-BMI) environment.)
As per claim 4, Francis discloses the neural signal recorder being an intracortical microelectrode array; an electrocorticography device; or an electroencephalography device (see at least ¶ [0156], disclosing chronically implanted bilaterally in the primary motor cortex with, for example, 96 channel platinum microelectrode arrays.)
As per claim 6, Francis discloses the decoder application further configuring the processor to obtain a confidence value from the inference model indicating a predicted accuracy of the inferred intended target (see at least ¶ [0151], disclosing to provide a confidence measure to the critic’s output.)
As per claim 8, Francis discloses the interface device being a computer providing a movable cursor in a 2-dimensional (2D) virtual environment (see at least ¶ [0110], disclosing an appropriate action of the external actuator (end effector) 112 including, without limitations, computer cursor.)
As per claims 11-14, 16 and 18, see the discussion in the rejections of claims 1-4, 6 and 8 for similar limitations.
Claims 5 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Francis in view of Shenoy et al. (US 2021/0064135 A1; hereinafter Shenoy.)
As per claims 5 and 15, Francis does not teach the inference model is a recurrent neural network. However, in the same field of endeavor, Shenoy discloses a related brain-computer interface (BCI) (see at least Fig. 1; Abstract; ¶ [0036],) comprising a decoder application for decoding neural signals, wherein the inference model is a recurrent neural network (see at least Abstract; ¶ [0011], disclosing decoding intended symbols from neural activity; the symbol model is selected from the group consisting of: recurrent neural networks (RNNs), long short-term memory (LSTM) networks, temporal convolutional networks, and hidden Markov models (HMMs),) thereby improving the efficiency of the BCI system (see at least ¶ [0012].) Thus, it would have been obvious to one of ordinary skill in the art at the time before the effective filing date of invention of the pending application to utilize the recurrent neural network, as the inference model, in the BCI and the associated method of the Francis reference, in view of the teaching in the Shenoy reference, to improve the above modified BCI of the Francis reference for the predictable result of improving the efficiency of the BCI system.
Claims 7, 10, 17 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Francis in view of Even-Chen et al. (US 2017/0042440 A1; hereinafter Even-Chen.)
As per claims 7 and 17, Francis does not teach “the closed-loop recalibrating BCI does not require manual recalibration when used at least every other day,” as claimed. However, Even-Chen teaches a closed-loop recalibrating BCI that does not require manual recalibration when used at least every other day (see at least ¶ [0034] disclosing “… conducted six (J) and four (L) days of closed-loop BMI experiments. Each day we calibrated the task difficulty by changing grid size and hold time to keep the monkey's success rate at "80% (actual experimental session success rates ranged from 76% to 82%)…”; ¶ [0040] disclosing “… We predicted that an automatic error detector that can spare the user from manual correction would improve BMI performance. The ability to predict putative errors before target selection and to decode errors with high accuracy around 400 ms after target selection enabled us to implement two modes of error detection and mitigation for online BMI use. The first mode is error prevention (FIG. 3B, green), which estimates whether the upcoming target selection is erroneous based on the neural activity a few milliseconds before the end of the hold period; The second mode is error deletion (FIG. 3B, purple), which detects errors after they occur (400 ms after target selection) and automatically "undoes" the presumably incorrect target selection …”; ¶ [0042] disclosing “… recordings to be quite stable from day to day, we pretrained the error detector on data collected on previous days. Each mode was compared separately to a standard BMI control without error detection in an A-B-A block format. We found that both modes improved the monkeys' typing rate across all days (t-test, p<0.01) by alleviating the need to correct mistakes (i.e., either by preventing errors or by auto-deleting them)…”.) Thus, it would have been obvious to one of ordinary skill in the art at the time before the effective filing date of invention of the pending application to utilize the recalibration taught by Even-Chen, specifically “a closed-loop recalibrating BCI that does not require manual recalibration when used at least every other day,” discussed above, in the BCI and the associated method of the Francis reference, in view of the teaching in the Even-Chen reference, to improve the above modified BCI of the Francis reference for the predictable result of improving the efficiency of the BCI system.
As per claims 10 and 20, Francis teaches obtain a plurality of neural signals from the neural signal recorder and translate each neural signal from the plurality of neural signals into a respective command for the interface device, using the neural decoder model (see at least ¶¶ [0110], [0111], [0134], [0185]); infer an intended target of a user based on each respective command, annotate each neural signal from the plurality of neural signals with the respective inferred intended target and retrain the neural decoder model using the annotated plurality of neural signals (see at least ¶¶ [0110], [0111], [0134], [0185].) Francis does not teach obtain a plurality of neural signals from the neural signal recorder recorded during a predefined time window. However, Even-Chen teaches to obtain a plurality of neural signals from the neural signal recorder recorded during a predefined time window (see at least ¶ [0034]: “… conducted six (J) and four (L) days of closed-loop BMI experiments …”.) Thus, it would have been obvious to one of ordinary skill in the art at the time before the effective filing date of invention of the pending application to utilize the recalibration taught by Even-Chen, specifically “obtain a plurality of neural signals from the neural signal recorder recorded during a predefined time window,” discussed above, in the BCI and the associated method of the Francis reference, in view of the teaching in the Even-Chen reference, to improve the above modified BCI of the Francis reference for the predictable result of improving the efficiency of the BCI system.
Claims 9 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Francis.
As per claims 9 and 19, these claims further recite a set of mathematic equations/ expressions, which were derived from the well-known mathematic equations/expressions/laws of the well-known hidden Markov model to be used as the interference model. Moreover, in the present case, since the well-known mathematic equations/expressions/laws cannot be patented, it follows that the addition of old and necessary antecedent steps of establishing values for the variables in the equations/expressions cannot convert the unpatentable subject matter to patentable subject matter. See In re Christensen, 178 USPQ 35 (CCPA 1973). Accordingly, these claims are unpatentable over Francis.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. Wettersten et al. (US 10,955,918 B1; see at least Figs. 14A-14F and the corresponding description; Abstract,) Mineault (US 11,314,329 B1; see at least Figs. 14A-14F and the corresponding description; Abstract,) and Provenza et al. (US 2021/0106830 A1;see at least Figs. 1-6 and the corresponding description; Abstract,) each discloses a related recalibrating brain-computer interface (BCI) and an associate recalibration method.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Jimmy H Nguyen whose telephone number is (571) 272-7675. The examiner can normally be reached on Monday-Friday 8:30AM-6PM.
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, Temesghen Ghebretinsae, can be reached at (571) 272-3017. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/Jimmy H Nguyen/
Primary Examiner, Art Unit 2626
Prosecution Insights