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 02/06/2026 have been fully considered but are not persuasive or are moot in view of a new grounds of rejection.
Applicant argues, “Kabram's "beam location" refers to a location of the brain that an ultrasound beam is directed to from an externally worn array of ultrasound transducers. See Kabrams, paragraph [0166]….a change in operation of the implanted neurostimulation system comprising at least one of a change in the detection parameter set and a change in the stimulation parameter set. (claim 39)…”
In regards to claim 39, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated) further comprising triggering the determining of the CRE biomarker in response to a change in operation of the implanted neurostimulation system ([0168]: change in seizure strength results in a change in operation of the implanted neurostimulation system as beam location is changed to provides a second seizure strengthen used for comparison; [0149]) comprising at least one of
a change in the detection parameter set (optional) and
a change in the stimulation parameter set ([0157]: changes location to aim stimulating ultrasound beam which would change stimulation parameter set); [0166]: stimulation may be more effective when brain region to be treated is changed; [0160]).
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 11-13 and 25-41 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
In re claim 11, the limitation “the plurality of data types exclude electrical activity of a brain” is unclear, specifically because the claim also recites wherein the plurality of data types “comprise counts, rates, durations, distribution, classification, brain activity type, and/or power measures, each derived from records of electrical activity of the patient's brain”. Specifically, it is unclear how the plurality of data types exclude electrical activity of a brain while also comprising features derived from records of electrical activity of the patient’s brain. For examination purposes, the limitation “the plurality of data types exclude electrical activity of a brain” is interpreted as excluding records of electrical activity themselves (for instance, excluding EEG signals by themselves), and NOT excluding data derived from the records of electrical activity, such as counts, rates, durations, distribution, classification, brain activity type, and/or power measures.
The following is a quotation of the first paragraph 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.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
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 of carrying out his invention.
Claims 11-13 and 25-41 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, 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 applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
In re claim 11, the limitations, “the plurality of data types exclude electrical activity of a brain and comprise counts, rates, durations, distribution, classification, brain activity type, and/or power measures, each derived from records of electrical activity of the patient's brain that were sensed and stored by the implanted neurostimulation system” introduce new matter that is not supported by the specification.
Although Applicant’s specification disclose “the first training set includes: 1) data types derived from records of electrical activity of a brain sensed and stored by implanted neurostimulation systems, and classified as ictal records, while excluding data types of records of electrical activity of the brain sensed and stored by implanted neurostimulation systems and classified as interictal records, and 2) at least one patient feature” [0006], and “the second training set includes: 1) data types derived from records of electrical activity of the brain sensed and stored by implanted neurostimulation systems, and classified as interictal records, while excluding data types of records of electrical activity of the brain sensed and stored by implanted neurostimulation systems and classified as ictal records, and 2) at least one patient feature” [0006], Applicant’s specification fails to explicitly disclose electrical activity of the brain being excluded.
In other words, although the specification discloses data types of records of electrical activity being excluded, the specification fails to explicitly disclose records of electrical activity being excluded.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 11, 25-27, 29-31, 33-36, and 39 are rejected under 35 U.S.C. 103 as being unpatentable over Kabrams et al. (US 2020/0194120) in view of Arcot Desai et al. (US 2019/0117978) (hereinafter referred to as Arcot Desai (‘978)) in view of Crowder et al. (US 2016/0228705).
In re claim 11, Kabrams discloses a method of modifying an operation of an implanted [0003, 0154] neurostimulation system of a patient [0125, 0127, 0154] based on information included in a subject-patient dataset ([0155-0157]: processor receives real-time sensor readings i.e. subject-patient dataset from sensors; [0052-0053]) comprising
a plurality of patient features ([0148]: one or more signals detected from the brain may be inputted; [0167]: includes input from sensors as well as an output for an acoustic signal; [0156]) and
a plurality of data types ([0125]: EEG measurements; [0141]: electrodes 104 record EEG measurements and would be a plurality of data types due to the plurality of electrodes);
wherein the plurality of patient features
are non-physiological (combination of detection parameters ([0156]: detection parameter set is a program used to determine that the brain is experiencing a seizure) and a stimulation parameter set ([0033-0034]: acoustic signal may be an ultrasound signal which will have frequency, spatial resolution, and a power density)) and
comprise
a detection parameter set (see above),
a stimulation parameter set (see above),
a location of a lead of the implanted neurostimulation system (optional), and/or
a type of the lead (optional);
the plurality of data types are records of electrical activity of the patient's brain [0125, 0167] that were sensed [0042] and stored ([0222]: memory can be incorporated in any embodiment; [0225]) by the implanted neurostimulation system [0042, 0225]; and
the implanted neurostimulation system is configured
to detect an electrographic seizure event in sensed electrical activity [0156] based on the detection parameter set [0156], and
to stimulate the patient's brain based on the stimulation parameter set ([0156]: acoustic signal provided based on seizure detection; [0033-0034]),
the method comprising:
determining a clinical response estimate (CRE) biomarker ([0049]: EEG signal is used to detect an epileptic seizure; [0125]) indicative of the patient's electrographic seizure [0125] by:
applying a first machine-learned model (fig. 13: 1306; [0091]: first trained statistical model is a first machine-learned model; [0216-0217; 0096]) to the subject-patient dataset to identify a plurality of key inputs ([0216-0220]: first trained statistical model uses signal data to indicate a predicted strength of a symptom of a neurological disorder and determines if the predicted strength exceeds a threshold, which is identifying a key input i.e. identifying signal data that shows presence of a symptom; [0016]: signal may comprise of multiple signals) comprising
a first key input ([0016]: electrical signal portion of a signal; [0154]) corresponding to a first data type selected from EEG signals [0154], and
a second key input corresponding to a first patient feature ([0016]: signal may comprise of multiple signals, for instance a mechanical signal; [0145]: mechanical signal includes an audible signal that allows a seizure to be heard and is interpreted as a patient feature),
applying a second machine-learned model ([0220]: second trained statistical model) to only the plurality of key inputs ([0218-0220]: the second trained statistical model is only applied to the signal associated with the predicted strength exceeding the threshold) to derive an input dataset from the subject-patient dataset ([0220]: input dataset is data from the signal that is associated with predicted strength exceeding the threshold),
wherein
the second machine-learned model is different from the first machine-learned model [0221] and
the input dataset
is a subset of the subject-patient dataset ([0220]: data associated with the predicted strength exceeding the threshold is a subset of all the data) and
comprises
the first key input ([0016]: signal includes electrical activity),
the second key input ([0016]: signal includes mechanical activity)
at least one additional input corresponding to either of a second data type different from the first data type and selected from counts, rates, durations, distribution, classification, brain activity type, and power measures (optional), or a second patient feature different from the first patient feature (see above, where the first patient feature can be a detection parameter set) and selected from the detection parameter set, the stimulation parameter set (see above, where the second patient feature can be a stimulation parameter set), the location of a lead of the implanted neurostimulation system, and the type of the lead,
processing only the input dataset to obtain a plurality of model inputs ([0149]: machine model algorithms analyze data to predict a strength of a seizure; [0220]: the second trained statical model only uses the input dataset indicating a strength exceeding the threshold to conduct further analysis to support or contradict the predicted strength of the symptom; [0157]), and
applying a machine-learned CRE model to only the plurality of model inputs to determine the CRE biomarker ([0167]: machine learning takes input recordings from various sensors to detect biomarkers that detect the presence of a seizure; [0220]: additional analysis is done in response to the predicted strength exceeding the threshold therefore only the plurality of model inputs are used),
wherein the machine-learned CRE model is trained on datasets across a patient population [0148];
comparing the CRE biomarker to a seizure criterion ([0137]: EEG signals are above a predetermined threshold i.e. seizure criterion); and
responsive to the criterion not being met ([0137]: criterion not being met occurs when the EEG signals are above the threshold i.e. when the seizure risk level is too high), adjusting the detection parameter set by changing a detection parameter ([0150]: if algorithm is not certain within a confidence threshold that the patient is having a seizure, then the algorithm can ask the patient whether or not they had a seizure, i.e. adjust a detection parameter, and retrain the algorithm); and
responsive to a detection of an electrographic event based on the adjusted detection parameter set [0137], delivering a stimulation therapy to the patient based on the stimulation parameter set ([0137]: brain is stimulated in response to EEG signal not being within the threshold).
Kabrams fails to disclose
the plurality of data types
exclude electrical activity of a brain and
comprise
counts,
rates,
durations,
distribution,
classification,
brain activity type, and/or
power measures,
each derived from records of electrical activity of the patient's brain that were sensed and stored by the implanted neurostimulation system;
determining a clinical response estimate (CRE) biomarker indicative of the patient's electrographic seizure rate…
…a first key input corresponding to a first data type selected from counts, rates, durations, distribution, classification, brain activity type, and power measures,
comparing the CRE biomarker to a seizure rate criterion.
Regarding the limitations,
“the plurality of data types
exclude electrical activity of a brain and
comprise
counts,
rates,
durations,
distribution,
classification,
brain activity type, and/or
power measures,
each derived from records of electrical activity of the patient's brain that were sensed and stored by the implanted neurostimulation system;
…a first key input corresponding to a first data type selected from counts, rates, durations, distribution, classification, brain activity type, and power measures,”
Arcot Desai (‘978) teaches an implanted neurostimulation system [0014] comprising a plurality of data types ([0008]: EEG records are stored to obtain input feature vectors extracted from an EEG record; [0061-0062])
the plurality of data types
exclude electrical activity of a brain ([0061-0062]: input records store data derived from EEG instead of just EEG data; [0008]) and
comprise
classification ([0061]: EEG classification such as seizure, seizure onset, baseline, etc.) derived from records of electrical activity of the patient's brain that were sensed [0024, 0061-0062] and stored by the implanted neurostimulation system [0014];
applying a first machine-learned model [0008] to a subject-patient dataset ([0008]: EEG records and clinical information [0049] are interpreted as a subject-patient dataset) to identify a plurality of key inputs comprising a first key input corresponding to a first data type selected from classification (see above), and
a second key input corresponding to a first patient feature selected from the detection parameter set ([0004]: detection parameter and simulation parameter are included in patient’s medical history for optimizing therapy) and the stimulation parameter set [0004].
Arcot Desai (‘978) further teaches that clinical information may be associated with each EEG record received [0052], which includes information related to a trigger event that caused the implanted neurostimulation system to record and store the EEG [0052], as well classification for the EEG [0052], which allows feature vectors to be selected based on a common characteristic [0061] to save computation time and power [0061].
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient taught by Kabrams, to provide the plurality of data types exclude electrical activity of a brain and comprise classification derived from records of electrical activity of the patient's brain that were sensed and stored by the implanted neurostimulation system, as taught by Arcot Desai (‘978) because clinical information may be associated with each EEG record received, which includes information related to a trigger event that caused the implanted neurostimulation system to record and store the EEG, as well classification for the EEG, which allows feature vectors to be selected based on a common characteristic to save computation time and power.
Regarding the limitations,
“determining a clinical response estimate (CRE) biomarker indicative of the patient's electrographic seizure rate…
comparing the CRE biomarker to a seizure rate criterion”,
Crowder teaches an analogous EEG detection system [0002], and teaches
determining a clinical response estimate (CRE) biomarker ([0265]: overall effectiveness value) indicative of a patient's electrographic seizure rate ([0265]: overall effectiveness value is used to determine electrographic seizures per week) and
comparing the CRE biomarker to a seizure rate criterion [0265].
Crowder further teaches that the CRE biomarker may be compared with a specified time period to determine is a stimulation parameter subspace is effective or not [0265].
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide determining a clinical response estimate (CRE) biomarker indicative of the patient's electrographic seizure rate and comparing the CRE biomarker to a seizure rate criterion, as taught by Crowder, because the CRE biomarker may be compared with a specified time period to determine is a stimulation parameter subspace is effective or not.
In re claim 25, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated) wherein the machine-learned model applied to the subject-patient dataset to identify the plurality of key inputs is one of a
decision tree model,
a neural network model [0167], and
a trained supervised machine learning model ([0167]: SVM).
In re claim 26, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated) wherein deriving the first key input from the record of electrical activity of the patient's brain comprises applying a machine-learned model [0202] to the record to determine a brain activity type that corresponds to the first key input ([0202]: CNN detects seizure).
In re claim 27, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated) wherein deriving the first key input from the record of electrical activity of the patient's brain comprises applying a deep learning model [0190] to the record of electrical activity to determine a numeric value that corresponds to the first key input ([0180]: data is annotated from 0 to 1 based on seizure detection; [0178]).
In re claim 29, the proposed combination fails to yield wherein
the implanted neurostimulation system comprises an implanted lead, and
the second key input corresponds to a type of implanted lead.
Crowder teaches wherein
an implanted neurostimulation system [0091] comprises an implanted lead (fig. 9B: any one of leads 904 or 906; [0091]), and
a second key input ([0137]: detection of neurological conditions) corresponds to a type of implanted lead ([0137]: EEG data is used to detect neurological conditions; [0095]: lead senses electrographic activity).
Crowder further teaches that in response to the detection of the neurological condition, therapeutic stimulation can be performed [0151, 0243].
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide wherein the implanted neurostimulation system comprises an implanted lead, and the second key input corresponds to a type of implanted lead, as taught by Crowder, because the implanted lead can be used to detect health conditions, which can be an indication that stimulation is needed.
In re claim 30, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated) wherein the second key input corresponds to a diagnosis of the patient [0182].
In re claim 31, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated) wherein processing the input dataset comprises combining a plurality of data types to obtain a corresponding model input [0167].
In re claim 33, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated) wherein processing the input dataset comprises filtering a record of electrical activity of the brain [0167].
In re claim 34, the proposed combination fails to yield wherein the filtering comprises extracting spectral power in specific frequency bands of the record.
Crowder teaches wherein a filtering ([0144]: Fourier transforms provide filtering) comprises extracting spectral power in specific frequency bands of the record ([0178]: power determined for ECoG in various frequency bands; see also “TABLE 2-continued” on page 17: “total spectral power”).
Crowder further teaches that ECoG ([0051]: electrographic activity) waveforms and other electrophysiological waveforms may be analyzed in a frequency domain [0111] to detect neurological states [0145] and assess effectiveness of stimulation [0145].
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide wherein the filtering comprises extracting spectral power in specific frequency bands of the record, as taught by Crowder, because doing so can be used to detect neurological states and assess effectiveness of stimulation.
In re claim 35, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated) wherein processing the input dataset comprises combining a plurality of patient features to obtain a corresponding model input [0167].
In re claim 36, the proposed combination fails to yield
wherein combining the plurality of patient features comprises inputting each of the plurality of patient features to a logic operation,
wherein an output of the logic operation is the corresponding model input.
Crowder teaches
wherein combining a plurality of patient features [0144-0145] comprises inputting each of the plurality of patient features to a logic operation [0144-0146],
wherein an output of the logic operation is the corresponding model input ([0144-0146]: Boolean combination used to analyze physiological data sensed from patient and combines results that will be used to detect neurological events).
Crowder further teaches that a pathological event may have been determined as being detected when a combination of conditions occurs, for instance a logical combination [0146], and that the results of the algorithms may be logically combined as necessary to detect neurological events or states [0145], as well as to assess effectiveness of a stimulation subspace [0145].
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide wherein combining the plurality of patient features comprises inputting each of the plurality of patient features to a logic operation, wherein an output of the logic operation is the corresponding model input, as taught by Crowder, because a pathological event may have been determined as being detected when a combination of conditions occurs, for instance a logical combination, and that the results of the algorithms may be logically combined as necessary to detect neurological events or states, as well as to assess effectiveness of a stimulation subspace.
In re claim 39, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated) further comprising triggering the determining of the CRE biomarker in response to a change in operation of the implanted neurostimulation system ([0168]: change in seizure strength results in a change in operation of the implanted neurostimulation system as beam location is changed to provides a second seizure strengthen used for comparison; [0149]) comprising at least one of
a change in the detection parameter set (optional) and
a change in the stimulation parameter set ([0157]: changes location to aim stimulating ultrasound beam which would change stimulation parameter set); [0166]: stimulation may be more effective when brain region to be treated is changed; [0160]).
Claims 12-13 are rejected under 35 U.S.C. 103 as being unpatentable over Kabrams et al. (US 2020/0194120) in view of Arcot Desai et al. (US 2019/0117978) (hereinafter referred to as Arcot Desai (‘978)) in view of Crowder et al. (US 2016/0228705) in view of Guttag et al. (US 2006/0111644).
In re claim 12, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated)
wherein processing the input dataset comprises applying a first machine-learned model ([0093]: first trained statistical model; [0215-0218]: processor provides data from signal as input to the first trained statistical model; [0224]: various embodiments can be combined) to a first subset of the input dataset to obtain one or more of the plurality of model inputs ([0218-0219]: signal detected from brain as input goes into the first trained statistical model to output predicted strength of symptom),
wherein the first machine-learned model is trained on datasets that
include data types derived from records of electrical activity of the brain classified as ictal records ([0091-0093]: first trained statistical model has high sensitivity which includes data types from ictal records i.e. instances of seizures; [0180]: times during a seizure i.e. ictal records are included in the algorithm for training; [0178-0179]).
The proposed combination fails to yield wherein the first machine-learned model is trained on datasets that exclude data types of records of electrical activity of the brain classified as interictal records.
Guttag teaches an analogous EEG detection system [0002] and teaches
wherein a first machine-learned model ([0035]: first classifier is a first machine-learned model) is trained on datasets that exclude data types of records of electrical activity of a brain classified as interictal records ([0280]: longer epochs can be used to avoid detecting inter-ictal discharges; [0054-0055]: detector comprises of the classifier )
Guttag further teaches that sometimes inter-ictal discharges are better avoided to minimize false-positives [0280, 0346-0347].
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide wherein the first machine-learned model is trained on datasets that exclude data types of records of electrical activity of the brain classified as interictal records, as taught by Guttag, because sometimes it’s preferred to exclude interictal records so that false-positive detections are avoided.
In re claim 13, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated)
wherein processing the input dataset comprises applying a second machine-learned model [0094] to a second subset of the input dataset ([[0094]: second subset of the input data set is the signal that goes to the second trained statistical model) to obtain one or more of the plurality of model inputs ([0094]: input from signal is entered into the second trained statistical model to output either an agreement or contradiction of the predicted strength; [0091-0093]),
wherein the second machine-learned model is trained on datasets that…
exclude data types of records of electrical activity of the brain classified as ictal records ([0095]: second trained statistical model is trained to have high sensitivity and high specificity which means that false positive seizures i.e. ictal records will be excluded).
The proposed combination fails to yield wherein the second machine-learned model is trained on datasets that include data types derived from records of electrical activity of the brain classified as interictal records.
Guttag teaches wherein a second machine-learned model ([0035]: second classifier is a second machine-learned model is trained on datasets that include data types derived from records of electrical activity of the brain classified as interictal records ([0237]: non-seizure i.e. interictal records are included as part of the training).
Guttag further teaches that reference EEG waveforms can include inter-ictal discharges so that false-positives can be determined [0070, 0346] or so stimulation can be applied in response to inter-ictal discharge detections [0070].
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide wherein the second machine-learned model is trained on datasets that include data types derived from records of electrical activity of the brain classified as interictal records, as taught by Guttag, because sometimes it’s preferred to include interictal records such as when determining false-positives or determining when stimulation should be applied.
Claims 28 and 32 are rejected under 35 U.S.C. 103 as being unpatentable over Kabrams et al. (US 2020/0194120) in view of Arcot Desai et al. (US 2019/0117978) (hereinafter referred to as Arcot Desai (‘978)) in view of Crowder et al. (US 2016/0228705) in view of Ford et al. (US 2017/0281071).
In re claim 28, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated) wherein the deep learning model is configured to assign a numeric value between 0 and 1 to the record of electrical activity (see in re claim 27 above).
The proposed combination fails to yield wherein the deep learning model is a regression model trained on data from other patients.
Ford teaches an analogous method of detecting electrical activity in a brain [0002], and teaches wherein a deep learning model [0119] is a regression model [0119, 0152] trained on data from other patients [0171].
Ford further teaches that data science techniques can use regressions and deep learning algorithms to train models to recognize and categorize patients outside of a test set [0011, 0119].
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide wherein the deep learning model is a regression model trained on data from other patients, as taught by Ford, because data science techniques can use regressions and deep learning algorithms to train models to recognize and categorize patients outside of a test set.
In re claim 32, the proposed combination fails to yield
wherein combining the plurality of data types comprises inputting
a first data type corresponding to long train events and
a second data type corresponding to delta power to a decision regression model,
wherein an output of the decision regression model is the corresponding model input.
Crowder teaches
wherein combining the plurality of data types ([007]: neurostimulator may analyze extracted features in combination) comprises inputting
a first data type corresponding to long train events ([0065]: occurs when ECG comprises high voltage spiking of more than 5 seconds) and
a second data type corresponding to delta power ([0087]: rounded theta-delta waveforms are observed in ECG activity).
Crowder further teaches that data corresponding to long train events provides evidence of a hypersynchronous seizure onset [0064], and delta power being observed provides evidence of propagated activity [0087], which can be used to determine effective stimulation parameters [0087]. Crowder further teaches that features extracted from sensed signals may be analyzed in combination [0007] to detect a neurological event [0007].
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide wherein combining the plurality of data types comprises inputting a first data type corresponding to long train events and a second data type corresponding to delta power, as taught by Crowder, because features such as a hypersynchronous seizure onset, and evidence of propagated activity, may be analyzed in combination to detect a neurological event, and the propagated activity can also be used to determine effectiveness of stimulation paraments.
Regarding the limitation,
“wherein combining the plurality of data types comprises inputting …the second data type corresponding to delta power to a decision regression model,
wherein an output of the decision regression model is the corresponding model input”,
see the proposed combination yielded in re claim 28 above, where it would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide wherein combining the plurality of data types comprises inputting the second data type corresponding to delta power to a decision regression model, wherein an output of the decision regression model is the corresponding model input, as taught by Ford, for substantially the same reasons as discussed in re claim 28 above.
Claim 37 is are rejected under 35 U.S.C. 103 as being unpatentable over Kabrams et al. (US 2020/0194120) in view of Arcot Desai et al. (US 2019/0117978) (hereinafter referred to as Arcot Desai (‘978)) in view of Crowder et al. (US 2016/0228705) in view of Arcot Desai et al. (US 2020/0272857) (hereinafter referred to as Arcot Desai (‘857)).
In re claim 37, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated)
wherein the plurality of patient features comprise
one or more of a patient monitoring component [0181] and
a patient diagnoses [0181-0182]/treatment [0137],
wherein
the patient diagnoses/treatment comprises at least one of
seizure frequency during pre-implant period,
lobe of epilepsy onset [0137],
presence of sclerosis,
presence of dysplasia.
The proposed combination fails to yield
wherein the plurality of patient features comprise one or more of a patient demographic,
wherein the patient demographic comprises at least one of
age,
duration of epilepsy,
sex,
type of job, and
geographical location.
Arcot Desai (‘857) teaches system (fig. 1B: 100) comprising an implanted neurostimulation system (102) and a records classification processor (104) which classifies a plurality of patient features ([0014]: EEG records; [0056]: record may include combination of criteria such as power spectral density of EEG; [0048]: may also include non-physiological information)
wherein the plurality of patient features comprise one or more of a patient demographic ([0048]: non-physiological information may be included, such as a patient’s demographics),
wherein the patient demographic comprises
age [0048] and
sex [0048].
Arcot Desai (‘857) further teaches that although EEG records are used, other physiological information and non-physiological information may also be included in a dataset [0048].
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide wherein the plurality of patient features comprise one or more of a patient demographic, wherein the patient demographic comprises age and sex, as taught by Arcot Desai (‘857), because other physiological information and non-physiological information may also be included in a dataset, such as patient demographic.
Claim 38 is rejected under 35 U.S.C. 103 as being unpatentable over Kabrams et al. (US 2020/0194120) in view of Arcot Desai et al. (US 2019/0117978) (hereinafter referred to as Arcot Desai (‘978)) in view of Crowder et al. (US 2016/0228705) in view of Alkawwas (US 2002/0045836).
In re claim 38, the proposed combination fails to yield further comprising triggering the determining of the CRE in response to a replacement of a component of the implanted neurostimulation system.
Alkawwas teaches measuring biopotential signals generated by a body [0003], and teaches triggering a determining of a desired lead biopotential measurement [0074, 0069] in response to a replacement of a component ([0074]: replacing electrode patches) of a biopotential monitoring system [0074, 0014].
Alkawwas further teaches that recalibration is needed during replacement to correct any errors [0074].
The proposed combination would be for a component of the implanted neurostimulation system (for instance, the sensor of Kabrams: [0050]: sensor which measures EEG signal and is part of the implanted device) yielded by the proposed combination to trigger determining of the CRE in response to being replaced.
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide triggering the determining of the CRE in response to a replacement of a component of the implanted neurostimulation system, as taught by the replacement of the electrode patches of Alkawwas triggering a recalibration, because recalibration is needed after a component is replaced to correct errors.
Claim 40 is are rejected under 35 U.S.C. 103 as being unpatentable over Kabrams et al. (US 2020/0194120) in view of Arcot Desai et al. (US 2019/0117978) (hereinafter referred to as Arcot Desai (‘978)) in view of Crowder et al. (US 2016/0228705) in view of Kuperman et al. (US 2022/0180993).
In re claim 40, the proposed combination yields (all mapping directed to Kabrams unless otherwise stated) further comprising triggering the determining of the CRE biomarker in response to a change in therapy ([0168]: measuring changes seizure strength requires change in therapy by moving beam position; [0149]) comprising a change in dose of a current anti-seizure medication or a change of anti-seizure medication ([0166]: change in brain region for stimulation to treat seizure; [0039]: symptom includes seizure; [0053]).
The proposed combination fails to yield further comprising triggering the determining of the CRE biomarker in response to a change in drug therapy comprising a change in dose of a current anti-seizure medication or a change of anti-seizure medication.
Kuperman teaches monitoring and managing neurological diseases [0017] by determining seizure data [0017] to generate a personalized dose-response profile [0017], and teaches triggering determining of a CRE biomarker ([0040]: machine learning algorithm use used to analyze stored data to manage a neurological disease such as by identifying seizure events and predicting patient responses to treatment to generate treatment recommendations) in response to a change in drug therapy comprising a change in dose of a current anti-seizure medication ([0026]: therapeutic efficacy can be measured as changes in doses due to tolerance).
Kuperman further teaches that analysis can provide a report [0055] that includes a personalized dose response profile [0055] indicating seizure burden with respect to medication dosage [0055] and efficacy can be better understood and correlated based on changes to medications [0020].
Kuperman also teaches that seizure variability may be complicated to understand [0028] due to an improvement In seizure counts after a medication change actually being caused by a different reason than the medication change [0028], therefore, a baseline is needed [0028] for efficacy calculations.
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide triggering the determining of the CRE biomarker in response to a change in drug therapy comprising a change in dose of a current anti-seizure medication or a change of anti-seizure medication, as taught by Kuperman, because a personalized dose response profile indicating seizure burden with respect to medication dosage and efficacy can be better understood and correlated based on changes to medications, and also because seizure variability may be complicated to understand, therefore, a baseline is needed related to therapeutic changes for efficacy calculations.
Claim 41 is rejected under 35 U.S.C. 103 as being unpatentable over Kabrams et al. (US 2020/0194120) in view of Arcot Desai et al. (US 2019/0117978) (hereinafter referred to as Arcot Desai (‘978)) in view of Crowder et al. (US 2016/0228705) in view of Bonnet et al. (US 2016/0375249).
In re claim 41, the proposed combination fails to yield comprising triggering the determining of the CRE biomarker in response to a change in patient habits comprising a change in patient diet, a change in patient sleep, or a change in patient exercise..
Bonnet teaches an electrical stimulation device [0003] comprising triggering a determining of a CRE biomarker ([0190]: each time patient is stimulated, implantable device obtains actual effect on patient i.e. a CRE biomarker) in response to a change in patient habits ([0150-0153]: therapy adapted based on changes in patient activity which can be measured via an accelerometer; [0190]: once patient is stimulated with a new set of parameters, actual effect on the patient will be calculated; [0171]) comprising
a change in patient sleep [0189] or
a change in patient exercise [0189].
Bonnet further teaches that neurostimulation devices should be able to adapt therapy quickly [0150-0153], especially due to changes in patient activity [0150-0153]. Bonnet further teaches that control device settings can be updates based on patient activity such as exercise and sleep [0189].
It would have been obvious to someone of ordinary skill in the art at the time the instant invention was filed to modify the method of modifying an operation of an implanted neurostimulation system of a patient yielded by the proposed combination, to provide comprising triggering the determining of the CRE biomarker in response to a change in patient habits comprising a change in patient diet, a change in patient sleep, or a change in patient exercise, as taught by Bonnet, because neurostimulation devices should be able to adapt therapy quickly, especially due to changes in patient activity, such as exercise and sleep.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Contact
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RUMAISA R BAIG whose telephone number is (571)270-0175. The examiner can normally be reached Mon-Fri: 8am- 5pm.
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, David Hamaoui can be reached on (571) 270-5625. 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.
/RUMAISA RASHID BAIG/Examiner, Art Unit 3796 /DAVID HAMAOUI/SPE, Art Unit 3796
Prosecution Insights