DETAILED ACTION
Applicant' s arguments, filed 06/23/2026 have been fully considered. The following rejections and/or objections are either reiterated or newly applied. They constitute the complete set presently being applied to the instant application.
Applicants have amended their claims, filed 12/11/2025, and therefore rejections newly made in the instant office action have been necessitated by amendment.
Claims 1-16 are the current claims hereby under examination.
All references to Applicant’s specification are made using the paragraph numbers assigned in the US publication of the present application US 2022/0225926 A1.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-5 8-12, and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Higgins US Patent Application Publication Number US 2011/0201944 A1 hereinafter Higgins in view of Osorio US Patent Application Publication Number US 2005/0197590 A1 hereinafter Osorio in view of Echauz international Patent Application Publication Number WO 03/030734 A2 hereinafter Echauz, and further in view of Fried US Patent Application Publication Number US 2017/0113046 A1 hereinafter Fried.
Regarding claim 1 Higgins discloses an epilepsy monitoring device (Abstract) comprising:
a first unit (Paragraph 0152: the electrode array; Fig. 13 reference 12) comprising:
a first body portion configured to be placed on or implanted within a head of a user (Paragraph 0152; fig. 13 reference 12 depicts the electrode arrays including a surface and depth electrode array. These electrode arrays have individual connecting wires which join together into a single wire (reference 16) that leads back to the implantable assembly (reference 14). The point where the wires connect is considered to be “a first unit” and “a first body portion” because the connecting point of these two wires satisfies all requirements set for by the claim)
a sensor unit connected to the first body portion and including a first sensor configured to measure actual normal cranial nerve signals and actual convulsive cranial nerve signals (Paragraphs 0059-0060: the electrode arrays may include cranial nerve electrodes; Paragraphs 0053 and 0055: the systems monitors EEG signals during normal periods and seizure events), and
a stimulation electrode connected to the first body portion and configured to apply cranial nerve stimulation treatment to a brain of the user based on a brain stimulation signal (Paragraph 0063: the stimulation cuff may provide stimulation to cranial nerves; Paragraph 0066: the implanted assembly provides the electrical stimulation signal; Paragraph 0152: the electrodes implanted in the brain may be used for stimulation; Paragraphs 0161-0162, 0166, and 0170: the processing sub-assembly located within the implantable assembly may execute the detection algorithms and provide the stimulation signal); and
a second unit electrically connected to the first unit (Paragraphs 0063-0064: the implanted assembly is in wired or wireless communication with the electrode arrays) and comprising:
a second body portion configured to be placed on or implanted within a body of the user other than the head of the user (Paragraph 0064: the implanted assembly may be implanted in a sub-clavicular pocket or abdomen),
a battery housed within the second body portion and configured to supply power to the first unit (Paragraph 0071: the implanted assembly includes a power source; Paragraphs 0066: the implanted assembly may provide the electrical stimulation and thus supplied power to the first unit),
communication circuitry comprising a first transceiver, the communication circuitry configured for wireless communication with an external device (Paragraphs 0054 and 0071-0073: the implanted assembly wirelessly communicates with an external assembly; Paragraph 0089: the telemetry transmitter).
a processor and a memory storing a classification model (Paragraphs 0069 and 0072: the memory sub-assembly and processing sub-assembly; Paragraphs 0161-0162 and 0166: the processing sub-assembly located within the implantable assembly may store and execute the classifier)
wherein the processor is configured to:
receive the actual normal cranial nerve signals and the actual convulsive cranial nerve signals in time series from the first sensor of the first unit (Paragraphs 0059-0060: the electrode arrays may include cranial nerve electrodes; Paragraphs 0053 and 0055: the systems monitors EEG signals during normal periods and seizure events; Paragraph 0068-0069: the second unit receives signals from the electrodes of the first unit.)
wherein the classification model has been pre-trained using, as training data (1) the actual normal cranial nerve signals and the actual convulsive cranial nerve signals (Paragraphs 0157-0159: the feature extraction and classification models are customized to the user to characterize a subject’s condition; Paragraphs 0164-0165: the seizure detection algorithms may be trained on the annotated EEG data collected from the patient),
wherein the processor is further configured to execute the trained classification model on a cranial nerve signal received from the first sensor to determine whether the user has convulsions (Paragraphs 0161-0162 and 0166: the processing sub-assembly located within the implantable assembly may execute the detection algorithms; Paragraphs 0157-0159: the feature extraction and classification models are customized to the user to characterize a subject’s condition; Paragraphs 0153-0154 and 0166-0167: EEG data is received and processed in substantially real time to detect seizures)
Higgins fails to further disclose the device wherein the communication circuitry comprises a second transceiver, and the processor is configured to: analyze the power spectrum density to detect a specific frequency band that includes a frequency having the greatest separation width between the actual normal cranial nerve signals and the actual convulsive cranial nerve signals, and generate virtual normal cranial nerve signals and virtual convulsive cranial nerve signals having features similar to features extracted from the actual normal cranial nerve signals and the actual convulsive cranial nerve signals, the features including the specific frequency band detected from the actual normal cranial nerve signals and the actual convulsive cranial nerve signals and the classification model also being pre-trained using (2) the virtual normal cranial nerve signals and the virtual convulsive cranial nerve signals having the specific frequency band.
Osorio teaches a system that analyzes signals representative of a subject's brain activity in a signal processor for information indicating the subject's current activity state and for predicting a change in the activity state. The system uses a combination of nonlinear filtering methods to perform real-time analysis of the electro-encephalogram (EEG) or electro-corticogram (ECoG) signals from a subject patient for information indicative of or predictive of a seizure, and to complete the needed analysis at least before clinical seizure onset. The preferred system then performs an output task for prevention or abatement of the seizure, or for recording pertinent data (Abstract). Thus, Osorio falls within the same field of endeavor as Applicant’s invention.
Osorio teaches the analysis of brainwave signals by performing power spectral density (PSD) analysis and matching the PSD signal to patterns representative of interictal, or normal, and ictal, or seizure, states. Osorio teaches that the system focuses on the PSD of frequency bands of past seizures that are maximally different from their respective normal segments. These bands of greatest difference are weighted more heavily in the determination of ictal states. (Paragraphs 0107 and 0144). Thus, Osorio teaches the calculation of PSD values for the normal and convulsive signal and the analysis of the PSD values to determine the frequency band with the greatest difference between normal and convulsive signals.
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to implement the PSD analysis and focus on the frequency bands with the greatest difference as taught by Osorio into the device of Higgins because weighting these frequency band more heavily than frequency bands with smaller changes as taught by Osorio (Paragraphs 0107 and 0144: the focus on maximally different frequency bands) may help improve seizure detection speed and accuracy since the most discriminative frequency bands are being more heavily considered than the less discriminative frequency bands.
Higgins in view of Osorio fails to further teach the device wherein the communication circuitry comprises a second transceiver, and the processor is configured to: generate virtual normal cranial nerve signals and virtual convulsive cranial nerve signals having features similar to features extracted from the actual normal cranial nerve signals and the actual convulsive cranial nerve signals, the features including the specific frequency band detected from the actual normal cranial nerve signals and the actual convulsive cranial nerve signals and the classification model also being pre-trained using (2) the virtual normal cranial nerve signals and the virtual convulsive cranial nerve signals having the specific frequency band.
Echauz teaches an epileptiform activity patient-specific template creation system which permits a user to efficiently develop an optimized set of patient-specific parameters for epileptiform activity detection algorithms. The epileptiform activity patient template creation system is primarily directed for use with an implantable neurostimulator system having EEG storage capability, in conjunction with a computer software program operating within a computer workstation having a processor, disk storage and input/output facilities for storing, processing and displaying patient EEG signals. The implantable neurostimulator is operative to store records of EEG data when neurological events are detected, when it receives external commands to record, or at preset or arbitrary times. The computer workstation operates on stored and uploaded records of EEG data to derive the patient-specific templates via a single local minimum variant of a multidimensional greedy line search process and a feature overlay process (Abstract), Thus Echauz falls within the same field of endeavor as Applicant’s invention.
Echauz teaches that additional data may be generated from the EEG data uploaded from an implanted neurostimulator by slightly modifying the collected data to create new datasets. The modifications may include adjustments to amplitude, noise, playback speed, and similar modifications. The modifications result in a much larger dataset which includes actual recorded data and the generated data surrogates which facilitates the generation of a set of detection parameters that maintains high sensitivity but is more specific than a set of detection parameters generated from only real recorded data. The generated data may be added to the data sets and used for generating a patient specific template (Page 11 lines 11-29; Page 31 line 32 – Page 33 line 6). Echauz further teaches that only signals that do not contain artifacts outliers or too much noise should be used in training (Page 53 lines 8-31: checking received signals for artifacts, noise, or post ictal activity; page 54 lines 26-31: detecting and discarding outlier signals; page 55 lines 20-25: marking signals as invalid for training if they contain substantial levels of artifacts)
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to implement the generation of virtual training data to aid in training a seizure detection algorithm as taught by Echauz into the device of Higgins in view of Osorio such that the generated data maintains the changes in PSD of the respective frequency bands because Echauz teaches that generating virtual data produces a larger dataset that allows the model to be trained to be more specific than a model trained only using real data while still maintaining high sensitivity (Echauz: Page 11 lines 11-29; Page 31 line 32 – Page 33 line 6) and ensuring that the generated data maintains the same PSD relationships in the most discriminative frequency bands would ensure that the virtual data maintains similar characteristics to true data and thus be more useful in training an accurate model by not introducing false relationships into the PSD analysis since Osorio teaches that the changes in the PSD are an important feature for seizure detection (Osorio: Paragraphs 0107 and 0144), and Echauz teaches that signals with too much noise, artifacts, or outliers should not be used for training (Echauz: Page 53 lines 8-31: checking received signals for artifacts, noise, or post ictal activity; page 54 lines 26-31: detecting and discarding outlier signals; page 55 lines 20-25: marking signals as invalid for training if they contain substantial levels of artifacts).
Higgins in view of Osorio further in view of Echauz fails to further teach the device wherein the communication circuitry comprises a second transceiver
Fried teaches systems and methods for restoring cognitive function are disclosed. In some implementations, a method includes, at a computing device, separately stimulating one or more of lateral and medial entorhinal afferents and other structures connecting to a hippocampus of an animal subject in accordance with a plurality of predefined stimulation patterns, thereby attempting to restore object-specific memories and location-specific memories; collecting a plurality of one or more of macro- and micro-recordings of the stimulation of hippocampal entorhinal cortical (HEC) system; and refining the computational model for restoring individual memories in accordance with a portion of the plurality of one or more of macro- and micro-recordings (Abstract). Thus, Fried is reasonably pertinent to the problem at hand.
Fried teaches an implantable system for recording and stimulating the brain which includes radiofrequency coils (Paragraph 0160). Fried teaches that the device may include a low-power wireless transceiver within the implantable device for near-field communication (Paragraph 0154 It is noted that near field communication is considered a proximity passive communication method).
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to combine the near field communication device of Fried with the device of Higgins in view of Osorio further in view of Echauz because Fried teaches that the near field communication module has low power requirements (Fried paragraph 0154: low-power radio blocks) and would thus be advantageous to implement into the implantable device of modified Higgins to preserve battery power during wireless communication for times where the shorter range and lower data transfer rate of near-field communication is acceptable.
Regarding claim 2 Higgins in view of Osorio in view of Echauz further in view of Fried teaches the epilepsy monitoring device of claim 1. Modified Higgins further discloses the device wherein the processor is configured to, generate an alarm signal when it is determined that the user has convulsions (Paragraphs 0152 and 0167: the system detects the user is at risk of a seizure or detects a seizure and issues seizure warning; Paragraph 0162; the implanted circuitry may generate a warning signal to the outside device).
Regarding claim 3 Higgins in view of Osorio in view of Echauz further in view of Fried teaches the epilepsy monitoring device of claim 2. Modified Higgins further discloses the device wherein the processor is configured to generate the brain stimulation signal to apply the cranial nerve stimulation treatment corresponding to the brain stimulation signal to the brain of the user, and provide the brain stimulation signal to the stimulation electrode (Paragraph 0170-0172: upon detection or prediction of a seizure event the therapy delivery assembly of the implantable assembly is configure to provide electrical stimulation. The stimulation may be to the cranial nerves through the electrode array implanted in the user’s head).
Regarding claim 4 Higgins in view of Osorio in view of Echauz further in view of Fried teaches the epilepsy monitoring device of claim 3. Modified Higgins further discloses the device wherein: the first transceiver is configured to communicate with the external device by a remote active communication method (Paragraph 0089: the telemetry transmitter). Higgins further discloses that the communication sub-assembly may include a magnetic reed switch (Paragraph 0085) which is a passive component operated by a proximity of a magnetic field, but such a switch does not reasonably “communicate with” the external device as it merely reacts to its presence and does not transmit or receive data itself.
Thus, modified Higgins fails to further disclose the device wherein the second transceiver is configured to communicate with the external device by a proximity passive communication method.
Fried teaches an implantable system for recording and stimulating the brain which includes radiofrequency coils (Paragraph 0160). Fried teaches that the device may include a low-power wireless transceiver within the implantable device for near-field communication (Paragraph 0154 It is noted that near field communication is considered a proximity passive communication method).
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to combine the near field communication device of Fried with the device of modified Higgins because Fried teaches that the near field communication module has low power requirements (Fried paragraph 0154: low-power radio blocks) and would thus be advantageous to implement into the implantable device of modified Higgins to preserve battery power during wireless communication for times where the shorter range and lower data transfer rate of near-field communication is acceptable.
Regarding claim 5 Higgins in view of Osorio in view of Echauz further in view of Fried teaches the epilepsy monitoring device of claim 4. Modified Higgins further discloses the device wherein: the first transceiver is configured to transmit and receive the alarm signal to the external device (Paragraphs 0152 and 0167: the system detects the user is at risk of a seizure or detects a seizure and issues seizure warning. Paragraph 0162: the implanted circuitry may send the alarm signal to the external device) by the remote active communication method (Paragraph 0089: the telemetry transmitter).
Modified Higgins fails to further disclose the device wherein the second transceiver is configured to transmit and receive the cranial nerve signals or the brain stimulation signal to the external device by the proximity passive communication method.
Fried teaches an implantable system for recording and stimulating the brain which includes radiofrequency coils (Paragraph 0160). Fried teaches that the device may include a low-power wireless transceiver within the implantable device for near-field communication (Paragraph 0154 It is noted that near field communication is considered a proximity passive communication method).
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to combine the near field communication device of Fried with the device of Higgins in view of Osorio in view of Echauz further in view of Fried because Fried teaches that the near field communication module has low power requirements (Fried paragraph 0154: low-power radio blocks) and would thus be advantageous to implement into the implantable device of modified Higgins to preserve battery power during wireless communication for times where the shorter range and lower data transfer rate of near-field communication is acceptable.
An obvious variation of Higgins in view of Osorio in view of Echauz further in view of Fried as presented above would be to transmit and receive the cranial nerve signal or the brain stimulation signal to an external device by a proximity passive communication method. Higgins discloses that the implanted circuitry may communicate EEG data to an external assembly within a given range and at a variety of transfer speeds (Higgins: Paragraph 0086). Since the embodiment of Higgins utilized for the above rejections is an embodiment where the EEG data is processed within the implantable device (Higgins: Paragraph 0162), the transmission of EEG data may be considered low priority but still desirable to obtain for the patient medical records, or for review by a clinician. As such, the low power proximity passive communication method of Higgins in view of Echauz further in view of Fried would be an ideal method of communicating this low-priority data given its lower range and transmission speed for the benefit of reduced power consumption. Higher priority data such as extracted features or the alarm signals may be transmitted using the more power intensive system to ensure fast communication even at increased range.
Regarding claim 15 Higgins in view of Osorio in view of Echauz further in view of Fried teaches the epilepsy monitoring device of claim 1. Modified Higgins fails to further disclose the device, wherein the specific frequency band has a predetermined range that includes the frequency having the greatest separation width.
Osorio teaches the device, wherein the specific frequency band has a predetermined range that includes the frequency having the greatest separation width (Paragraph 0107: the frequencies with the greatest separation between ictal and interictal PSD are weighted more heavily; Paragraph 0144: the frequency band with the greatest difference in PSD between ictal and interictal is focused on; the device starts with a generic filter to capture a wide range of frequency bands then focuses on the frequency bands with the greatest difference; Paragraph 0209: the frequency bands are a range of frequencies).
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to configure the device of modified Higgins to have a predetermined frequency band that encompasses the frequency with the greatest difference and then selectively narrow the frequency bands based on the frequency with the greatest separation as taught by Osorio because Osorio teaches that the frequencies and frequency band with the greatest differentiation between ictal and interictal states should be focused on and weighted more heavily (Osorio: Paragraphs 0107 and 0144) and starting with a wide filter that encompasses these ranges allows the system to narrow down its focus for to the most relevant bands for a particular patient while ensuring that no critical frequencies are missed which would be a danger if the device started with a narrow frequency band for consideration. Focusing on these bands may help improve the accuracy and detection speed of the algorithm by focusing it on the most discriminative features.
Regarding claim 8, Higgins discloses an epilepsy monitoring system (Abstract) comprising:
a first unit (Paragraph 0152: the electrode array; Fig. 13 reference 12) comprising:
a first body portion configured to be placed on or implanted within a head of a user (Paragraph 0152; fig. 13 reference 12 depicts the electrode arrays including a surface and depth electrode array. These electrode arrays have individual connecting wires which join together into a single wire (reference 16) that leads back to the implantable assembly (reference 14). The point where the wires connect is considered to be “a first unit” and “a first body portion” because the connecting point of these two wires satisfies all requirements set for by the claim),
a sensor unit connected to the first body portion and including a first sensor configured to measure actual normal cranial nerve signals and actual convulsive cranial nerve signals (Paragraphs 0059-0060: the electrode arrays may include cranial nerve electrodes; Paragraphs 0053 and 0055: the systems monitors EEG signals during normal periods and seizure events), and
a stimulation electrode connected to the first body portion and configured to apply cranial nerve stimulation treatment to a brain of the user based on a provided brain stimulation signal (Paragraph 0063: the stimulation cuff may provide stimulation to cranial nerves; Paragraph 0066: the implanted assembly provides the electrical stimulation signal; Paragraph 0152: the electrodes implanted in the brain may be used for stimulation; Paragraphs 0161-0162, 0166, and 0170: the processing sub-assembly located within the implantable assembly may execute the detection algorithms and provide the stimulation signal);
a second unit electrically connected to the first unit (Paragraphs 0063-0064: the implanted assembly is in wired or wireless communication with the electrode arrays) and comprising:
a second body portion configured to be placed on or implanted within a body of the user other than the head of the user (Paragraph 0064: the implanted assembly may be implanted in a sub-clavicular pocket or abdomen),
a battery housed within the second body portion and configured to supply power to the first unit (Paragraph 0071: the implanted assembly includes a power source; Paragraphs 0066: the implanted assembly may provide the electrical stimulation and thus supplies power to the first unit), and
communication circuitry comprising a first transceiver, the communication circuitry configured for wireless communication with an external device (Paragraphs 0054 and 0071-0073: the implanted assembly wirelessly communicates with an external assembly; Paragraph 0089: the telemetry transmitter).
a processor and a memory storing a classification model (Paragraphs 0069 and 0072: the memory sub-assembly and processing sub-assembly; Paragraphs 0161-0162 and 0166: the processing sub-assembly located within the implantable assembly may store and execute the classifier)
wherein the processor is configured to:
receive the actual normal cranial nerve signals and the actual convulsive cranial nerve signals in time series from the first sensor of the first unit (Paragraphs 0059-0060: the electrode arrays may include cranial nerve electrodes; Paragraphs 0053 and 0055: the systems monitors EEG signals during normal periods and seizure events; Paragraph 0068-0069: the second unit receives signals from the electrodes of the first unit.)
wherein the classification model has been pre-trained using, as training data (1) the actual normal cranial nerve signals and the actual convulsive cranial nerve signals (Paragraphs 0157-0159: the feature extraction and classification models are customized to the user to characterize a subject’s condition; Paragraphs 0164-0165: the seizure detection algorithms may be trained on the annotated EEG data collected from the patient),
wherein the processor is further configured to execute the trained classification model on a cranial nerve signal received from the first sensor to determine whether the user has convulsions (Paragraphs 0161-0162 and 0166: the processing sub-assembly located within the implantable assembly may execute the detection algorithms; Paragraphs 0157-0159: the feature extraction and classification models are customized to the user to characterize a subject’s condition; Paragraphs 0153-0154 and 0166-0167: EEG data is received and processed in substantially real time to detect seizures)
Higgins fails to further disclose the system wherein the communication circuitry comprises a second transceiver, and the processor is configured to: analyze the power spectrum density to detect a specific frequency band that includes a frequency having the greatest separation width between the actual normal cranial nerve signals and the actual convulsive cranial nerve signals, and generate virtual normal cranial nerve signals and virtual convulsive cranial nerve signals having features similar to features extracted from the actual normal cranial nerve signals and the actual convulsive cranial nerve signals, the features including the specific frequency band detected from the actual normal cranial nerve signals and the actual convulsive cranial nerve signals and the classification model also being pre-trained using (2) the virtual normal cranial nerve signals and the virtual convulsive cranial nerve signals having the specific frequency band.
Osorio teaches the analysis of brainwave signals by performing power spectral density (PSD) analysis and matching the PSD signal to patterns representative of interictal, or normal, and ictal, or seizure, states. Osorio teaches that the system focuses on the PSD of frequency bands of past seizures that are maximally different from their respective normal segments. These bands of greatest difference are weighted more heavily in the determination of ictal states. (Paragraphs 0107 and 0144). Thus, Osorio teaches the calculation of PSD values for the normal and convulsive signal and the analysis of the PSD values to determine the frequency band with the greatest difference between normal and convulsive signals.
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to implement the PSD analysis and focus on the frequency bands with the greatest difference as taught by Osorio into the system of Higgins because weighting these frequency band more heavily than frequency bands with smaller changes as taught by Osorio (Paragraphs 0107 and 0144: the focus on maximally different frequency bands) may help improve seizure detection speed and accuracy since the most discriminative frequency bands are being more heavily considered than the less discriminative frequency bands.
Higgins in view of Osorio fails to further teach the wherein the communication circuitry comprises a second transceiver, and the processor is configured to: generate virtual normal cranial nerve signals and virtual convulsive cranial nerve signals having features similar to features extracted from the actual normal cranial nerve signals and the actual convulsive cranial nerve signals, the features including the specific frequency band detected from the actual normal cranial nerve signals and the actual convulsive cranial nerve signals and the classification model also being pre-trained using (2) the virtual normal cranial nerve signals and the virtual convulsive cranial nerve signals having the specific frequency band.
Echauz teaches that additional data may be generated from the EEG data uploaded from an implanted neurostimulator by slightly modifying the collected data to create new datasets. The modifications may include adjustments to amplitude, noise, playback speed, and similar modifications. The modifications result in a much larger dataset which includes actual recorded data and the generated data surrogates which facilitates the generation of a set of detection parameters that maintains high sensitivity but is more specific than a set of detection parameters generated from only real recorded data. The generated data may be added to the data sets and used for generating a patient specific template (Page 11 lines 11-29; Page 31 line 32 – Page 33 line 6). Echauz further teaches that only signals that do not contain artifacts outliers or too much noise should be used in training (Page 53 lines 8-31: checking received signals for artifacts, noise, or post ictal activity; page 54 lines 26-31: detecting and discarding outlier signals; page 55 lines 20-25: marking signals as invalid for training if they contain substantial levels of artifacts)
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to implement the generation of virtual training data to aid in training a seizure detection algorithm as taught by Echauz into the system of Higgins in view of Osorio such that the generated data maintains the changes in PSD of the respective frequency bands because Echauz teaches that generating virtual data produces a larger dataset that allows the model to be trained to be more specific than a model trained only using real data while still maintaining high sensitivity (Echauz: Page 11 lines 11-29; Page 31 line 32 – Page 33 line 6) and ensuring that the generated data maintains the same PSD relationships in the most discriminative frequency bands would ensure that the virtual data maintains similar characteristics to true data and thus be more useful in training an accurate model by not introducing false relationships into the PSD analysis since Osorio teaches that the changes in the PSD are an important feature for seizure detection (Osorio: Paragraphs 0107 and 0144) , and Echauz teaches that signals with too much noise, artifacts, or outliers should not be used for training (Echauz: Page 53 lines 8-31: checking received signals for artifacts, noise, or post ictal activity; page 54 lines 26-31: detecting and discarding outlier signals; page 55 lines 20-25: marking signals as invalid for training if they contain substantial levels of artifacts).
Higgins in view of Osorio further in view of Echauz fails to further teach the system wherein the communication circuitry comprises a second transceiver
Fried teaches an implantable system for recording and stimulating the brain which includes radiofrequency coils (Paragraph 0160). Fried teaches that the device may include a low-power wireless transceiver within the implantable device for near-field communication (Paragraph 0154 It is noted that near field communication is considered a proximity passive communication method).
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to combine the near field communication device of Fried with the system of Higgins in view of Osorio further in view of Echauz because Fried teaches that the near field communication module has low power requirements (Fried paragraph 0154: low-power radio blocks) and would thus be advantageous to implement into the implantable device of modified Higgins to preserve battery power during wireless communication for times where the shorter range and lower data transfer rate of near-field communication is acceptable.
Regarding claim 9 Higgins in view of Osorio in view of Echauz further in view of Fried teaches the epilepsy monitoring system of claim 8. Modified Higgins further discloses the system wherein the processor is configured to generate an alarm signal when it is determined that the user has convulsions (Paragraph 0162; the implanted circuitry may generate a warning signal to the outside device).
Regarding claim 10 Higgins in view of Osorio in view of Echauz further in view of Fried teaches the epilepsy monitoring system of claim 9. Modified Higgins further discloses the system wherein the processor is configured to generate the brain stimulation signal to apply the cranial nerve stimulation treatment corresponding to the brain stimulation signal to the brain of the user, and provide the brain stimulation signal to the stimulation electrode (Paragraph 0170-0172: upon detection or prediction of a seizure event the therapy delivery assembly of the implantable assembly is configure to provide electrical stimulation. The stimulation may be to the cranial nerves through the electrode array implanted in the user’s head).
Regarding claim 11 Higgins in view of Osorio in view of Echauz further in view of Fried teaches the epilepsy monitoring system of claim 10. Modified Higgins further discloses the system wherein the first transceiver is configured to communicate with the external device (Paragraphs 0152 and 0167: the system detects the user is at risk of a seizure or detects a seizure and issues seizure warning. Paragraph 0162: the implanted circuitry may send the alarm signal to the external device) by a remote active communication method (Paragraph 0089: the telemetry transmitter). Higgins further discloses that the communication sub-assembly may include a magnetic reed switch (Paragraph 0085) which is a passive component operated by a proximity of a magnetic field, but such a switch does not reasonably “communicate with” the external device as it merely reacts to its presence and does not transmit or receive data itself.
Thus, modified Higgins fails to further disclose the system comprising the second transceiver is configured to communicate with the external device by a proximity passive communication method.
Fried teaches an implantable system for recording and stimulating the brain which includes radiofrequency coils (Paragraph 0160). Fried teaches that the device may include a low-power wireless transceiver within the implantable device for near-field communication (Paragraph 0154 It is noted that near field communication is considered a proximity passive communication method).
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to combine the near field communication device of Fried with the system of modified Higgins because Fried teaches that the near field communication module has low power requirements (Fried paragraph 0154: low-power radio blocks) and would thus be advantageous to implement into the implantable device of modified Higgins to preserve battery power during wireless communication for times where the shorter range and lower data transfer rate of near-field communication is acceptable.
Regarding claim 12 Higgins in view of Osorio in view of Echauz further in view of Fried teaches the epilepsy monitoring system of claim 11. Modified Higgins further discloses the system wherein: the first transceiver is configured to transmit and receive the alarm signal to the external device (Paragraphs 0152 and 0167: the system detects the user is at risk of a seizure or detects a seizure and issues seizure warning. Paragraph 0162: the implanted circuitry may send the alarm signal to the external device) by the remote active communication method (Paragraph 0089: the telemetry transmitter), and
Modified Higgins fails to further disclose the system wherein the second transceiver is configured to transmit and receive the cranial nerve signals or the brain stimulation signal to the external device by the proximity passive communication method.
Fried teaches an implantable system for recording and stimulating the brain which includes radiofrequency coils (Paragraph 0160). Fried teaches that the device may include a low-power wireless transceiver within the implantable device for near-field communication (Paragraph 0154 It is noted that near field communication is considered a proximity passive communication method).
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to combine the near field communication device of Fried with the system of Higgins in view of Osorio in view of Echauz further in view of Fried because Fried teaches that the near field communication module has low power requirements (Fried paragraph 0154: low-power radio blocks) and would thus be advantageous to implement into the implantable system of modified Higgins to preserve battery power during wireless communication for times where the shorter range and lower data transfer rate of near-field communication is acceptable.
An obvious variation of Higgins in view of Osorio in view of Echauz in view of Fried as presented above would be to transmit and receive the cranial nerve signal or the brain stimulation signal to an external device by a proximity passive communication method. Higgins discloses that the implanted circuitry may communicate EEG data to an external assembly within a given range and at a variety of transfer speeds (Paragraph 0086). Since the embodiment of Higgins utilized for the above rejections is an embodiment where the EEG data is processed within the implantable device (Paragraph 0162), the transmission of EEG data may be considered low priority but still desirable to obtain for the patient medical records, or for review by a clinician. As such, the low power proximity passive communication method of Higgins in view of Fried would be an ideal method of communicating this low-priority data given its lower range and transmission speed for the benefit of reduced power consumption. Higher priority data such as extracted features or the alarm signals may be transmitted using the more power intensive system to ensure fast communication even at increased range.
Claims 6-7 and 13-14 are rejected under 35 U.S.C. 103 as being unpatentable over Higgins US Patent Application Publication Number US 2011/0201944 A1 hereinafter Higgins in view of Osorio US Patent Application Publication Number US 2005/0197590 A1 hereinafter Osorio and in view of Echauz international Patent Application Publication Number WO 03/030734 A2 hereinafter Echauz, further in view of Fried US Patent Application Publication Number US 2017/0113046 A1 hereinafter Fried as applied to claims 1 and 8 above and further in view of Giftakis US Patent Application Publication Number US 2010/0121213 A1 hereinafter Giftakis.
Regarding claim 6 Higgins in view of Osorio in view of Echauz further in view of Fried teaches the epilepsy monitoring device of claim 1. Modified Higgins further discloses the device configured to detect a biological signal different from the cranial nerve signals (Paragraph 0181)
Modified Higgins fails to further disclose the location of such a sensor and thus fails to disclose the sensor unit further comprises a second sensor.
Giftakis teaches systems and methods for monitoring trends in the intracranial pressure over time, e.g., to detect changes to the patient's condition. In addition, in some examples, a seizure metric may be generated for a detected seizure based on sensed intracranial pressures. The seizure metric may indicate, for example, an average, median, or highest relative intracranial pressure value observed during a seizure, a percent change from a baseline value during the seizure, or the time for the intracranial pressure to return to a baseline state after the occurrence of a seizure. In addition to or instead of intracranial pressure, patient motion or posture may be monitored in order to assess the patient's seizure disorder. For example, a seizure type or severity may be determined based on patient motion sensed during a seizure (Abstract). Thus, Giftakis falls within the same field of endeavor as Applicant’s invention.
Giftakis teaches a therapy system which includes an implantable motion sensor which may be located on one of the electrode lead implanted within the patient’s brain. The motion signals are analyzed to determine current patient activity levels and patient posture which may be used to evaluate detected seizures and determine typical patient movement activity during seizure events (Paragraphs 0058-0059; fig. 1 references 20A and 20B).
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to incorporate the motion sensor located on an implanted electrode lead as described by Giftakis into the device of modified Higgins because Higgins already considers the utilization of motion signals to help detect seizure events (Higgins: Paragraph 0181) but fails to disclose where such a sensor would be implemented. The device of Giftakis teaches that the electrodes in the user’s brain are acceptable locations for such a sensor and that the resultant data can be used to help characterize what movements are typical for a particular patient during a seizure event (Giftakis: paragraph 0059) which may help the user better plan and react for predicted seizure events
Regarding claim 7 Higgins in view of Osorio in view of Echauz in view of Fried further in view of Giftakis teaches the epilepsy monitoring device of claim 6. Modified Higgins further discloses the device wherein the second sensor is configured to detect a motion signal of the user (Paragraph 0181: accelerometer or movement recordings)
Regarding claim 13 Higgins in view of Osorio in view of Echauz further in view of Fried teaches the epilepsy monitoring system of claim 8. Modified Higgins further discloses the system comprising a second sensor configured to detect a biological signal different from the cranial nerve signals (Paragraph 0181).
Modified Higgins fails to further disclose the location of such a sensor and thus fails to disclose the sensor unit further comprises a second sensor.
Giftakis teaches a therapy system which includes an implantable motion sensor which may be located on one of the electrode lead implanted within the patient’s brain. The motion signals are analyzed to determine current patient activity levels and patient posture which may be used to evaluate detected seizures and determine typical patient movement activity during seizure events (Paragraphs 0058-0059; fig. 1 references 20A and 20B).
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to incorporate the motion sensor located on an implanted electrode lead as described by Giftakis into the system of modified Higgins because Higgins already considers the utilization of motion signals to help detect seizure events (Higgins: Paragraph 0181) but fails to disclose where such a sensor would be implemented. The device of Giftakis teaches that the electrodes in the user’s brain are acceptable locations for such a sensor and that the resultant data can be used to help characterize what movements are typical for a particular patient during a seizure event (Giftakis: paragraph 0059) which may help the user better plan and react for predicted seizure events
Regarding claim 14 Higgins in view of Osorio in view of Echauz in view of Fried further in view of Giftakis teaches the epilepsy monitoring system of claim 13. Modified Higgins further discloses the system wherein the second sensor is configured to detect a motion signal of the user (Paragraph 0181: accelerometer or movement recordings).
Claim 16 is rejected under 35 U.S.C. 103 as being unpatentable over Higgins US Patent Application Publication Number US 2011/0201944 A1 hereinafter Higgins in view of Osorio US Patent Application Publication Number US 2005/0197590 A1 hereinafter Osorio and in view of Echauz international Patent Application Publication Number WO 03/030734 A2 hereinafter Echauz further in view of Fried US Patent Application Publication Number US 2017/0113046 A1 hereinafter Fried as applied to claim 1 above and further in view of Kidmose US Patent Application Publication Number US 2013/0296731 A1 hereinafter Kidmose.
Regarding claim 16 Higgins in view of Osorio in view of Echauz in view of Fried teaches the epilepsy monitoring device of claim 1. Modified Higgins fails to further disclose the device, wherein the processor is configured to detect the specific frequency band using a Welch method.
Kidmose teaches a personal wearable EEG monitor is adapted to be carried at the head of a person. The EEG monitor comprises an EEG sensor part having skin surface electrodes for measuring EEG signals from said person. The EEG monitor comprises an EEG signal analyzer adapted for monitoring and analyzing the EEG signal. The EEG monitor performs at least one of the following: providing a stimulus to the person, requesting the person to perform a stimuli creating act, or identifying a stimuli creating ambient sound. The EEG monitor comprises means for identifying an induced response from the EEG signal caused by the stimuli, and a classifier for deciding whether the skin surface electrodes receive EEG signals. The invention further provides a method of monitoring EEG signals of a person (Abstract). Thus, Kidmose falls within the same field of endeavor as Applicant’s invention.
Kidmose teaches that power spectral density may be determined using a Welch method (Paragraph 0069).
It would have been obvious to one of ordinary skill in the art prior to the effective filling date of the invention to configure the device of modified Higgins to determine the PSD and thus the frequency and frequency band having the greatest difference between the ictal and interictal states by using a Welch method because Kidmose teaches that a Welch method may be used to determine the PSD of EEG signal (Paragraph 0069) and the use of such a method is a simple substitution of one known method (the method used by Osorio) for another known method (a Welch method as taught by Kidmose) with no surprising technical effect (the PSD is calculated).
Response to Arguments
Applicant’s arguments with respect to claims 1 and 8 have been fully considered but are not found to be persuasive.
In particular, Applicant’s amendments have overcome the previously presented 35 USC 112 rejections and the language no longer necessitates 112(f) interpretation.
In regards to the 35 USC 103 rejections:
Applicant argues that none of Higgins, Osorio, and/or Echauz either alone or in combination teach or reasonably suggest the generation of the virtual signals having the specific frequency band and the use of said virtual signals for training the classification model. Applicant asserts that: Higgins is directed towards a training process that only uses real data collected over a longer period of time and does not contemplate the generation of virtual signals. Osorio teaches the detection of frequency bands with greatest separation purely for filter design and not for data generation, and that Echauz generates data using generic time domain perturbations and not frequency band targeted signal synthesis. The signals of Echauz are then used to test threshold robustness rather than training a classifier.
In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986). In particular, Applicant does not argue describe how the teachings of various references, when considered in combination, do not result in the claimed invention. In particular the combination of references are considered as follows: Higgins sets forth a seizure detection system using a trained machine learning classifier which lacks the PSD analysis and generation of virtual signals to use in training the classifier. Osorio then teaches the PSD analysis and teaches that the frequencies of maximum differences are important for the detection of seizures and are given greater weight than other frequency bands (Osorio: Paragraphs 0107 and 0144). Echauz then teaches the generation of virtual data and the use of said data to generate patient-specific templates (Page 11 lines 11-29; Page 31 line 32 – Page 33 line 6). The generation of a patient specific template is considered analogous to training a classifier. Thus a combination of Higgins, Osorio, and Echauz teaches a device/system which performed seizure detection using a trained machine learning classifier which may be trained using virtual data as taught by Echauz; One of ordinary skill in the art would recognize that said virtual data should still maintain the important signal features/relationships used in classifying the seizures because feeding random signals into a machine learning model would result in poor performance, this is further indicated by Echauz which includes a signal selection process where signals with too many artifacts or too much noise are discarded as they are not useful in determining the patient template (Page 53 lines 8-31: checking received signals for artifacts, noise, or post ictal activity; page 54 lines 26-31: detecting and discarding outlier signals; page 55 lines 20-25: marking signals as invalid for training if they contain substantial levels of artifacts); and Osorio teaches that one such important feature to maintain is the PSD ratios.
Additionally, it is noted that the data manipulations taught by Echauz inherently maintain the PSD ratios, and thus inherently include the specific frequency bad, because, as Applicant acknowledges, the manipulations of Echauz used to generate additional data such as amplitude adjustment are in the time domain and thus do not alter the features of the frequency domain. Therefore the features of the frequency domain are preserved including the specific frequency band.
Applicant further asserts that the rationale of maintaining the PSD relationships of respective frequency bands is not taught or reasonably suggested by any of the prior art of record and is impermissible hindsight reasoning.
This argument is not fou8nd to be persuasive because one of ordinary skill in the art would recognize that feeding random signals into a machine learning model would degrade its performance. It is obvious to maintain the important relationships/features being used for identifying conditions when generating virtual data. Echauz further teaches this concept in page 53 lines 8-31, page 54 lines 26-31, and page 55 lines 20-25 where Echauz teaches that removing invalid signals containing artifacts and excessive noise is important to generate an accurate template for a patient. Applicant’s arguments are not found to be persuasive in light of the skill of one of ordinary skill in the art and the teachings of Higgins, Osorio, and Echauz regarding the importance of certain features in the training data.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MATTHEW ERIC OGLES whose telephone number is (571)272-7313. The examiner can normally be reached M-F 8:00AM - 5:30PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jason Sims can be reached on Monday-Friday from 9:00AM – 4:00PM at (571) 272 – 7540. 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.
/MATTHEW ERIC OGLES/Examiner, Art Unit 3791
/RENE T TOWA/Primary Examiner, Art Unit 3791
Prosecution Insights