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 .
Drawings
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(4) because reference character “D” is used to designate the depth of housing 201 [Applicant’s Specification ¶¶0077, 0079, Figs. 2B] and the depth of sensor device 310 [¶0087, Fig. 3]; “W” is used to designate the width of housing 201 [¶0079, Fig. 2A] and the width of sensor device 310 [¶0087, Fig. 3]; “L” is used to designate the length of housing 201 [¶0079, Fig. 2A] and the length of sensor device 310 [¶0087, Fig. 3], wherein the Examiner notes that each of “D”, “W”, and “L” are used to designate dimensions of what appears to be physically different embodiments.
Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
The drawings are objected to because: “O1” and “O2” as recited in the Applicant’s Specification ¶0075 appear to be depicted in Applicant’s Fig. 1E as “01” and “02” [zeros instead of O’s], respectively [see font of “O” in “NASION” and “INION” of Fig. 1E].
Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as “amended.” If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Interpretation
Examiner Notes: currently, NO limitation invokes interpretation under § 112(f).
Claim Rejections - 35 USC § 112
Examiner’s Note Regarding Machine Learning: the claimed “machine learning (ML) model” of claim(s) 1, 7-8, 11, 16-17, and 20 was considered under § 112(a), wherein the Examiner notes that the disclosure of an artificial intelligence model of the Applicant’s Specification [Processing circuitry 402 applies the EEG signal(s) to an artificial intelligence model (e.g., machine learning, neural networks, etc.) to determine one or more characteristic(s) of a brain event (706). While the following description will refer to the artificial intelligence model as a machine learning (ML) model, other types of artificial intelligence models may be used in place of the ML model (Applicant’s Specification ¶0131)] is considered to provide sufficient written description support for the ML model as presently claimed for one of ordinary skill in the art to understand that the Applicant possessed the instant invention at the time of filing.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim(s) 1-20 is/are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception without significantly more. Each claim has been analyzed to determine whether it is directed to any judicial exceptions.
Representative claim(s) 1 [representing all independent claims] recite(s):
A system comprising:
a memory;
a plurality of electrodes;
sensing circuitry configured to:
sense, via at least two electrodes of the plurality of electrodes, electrical signals from a patient; and
generate, based on the electrical signals, one or more electroencephalography (EEG) signals; and
processing circuitry configured to:
receive, from the sensing circuitry, one or more EEG signals; and
apply the one or more EEG signals to a machine learning (ML) model to determine one or more characteristics of a brain event, the ML model being trained on training EEG data and simulated EEG data.
(Emphasis added: abstract idea, additional element)
Step 2A Prong 1
Representative claim(s) 1 recites the following abstract ideas, which may be performed in the mind or by hand with the assistance of pen and paper:
“receive, from the sensing circuitry, one or more EEG signals” – may be performed by merely observing known or previously collected data, as the Examiner notes that the instant limitation is not a step of gathering data and the EEG signals being recited as being “from the sensing circuitry” merely limits the type of data
“apply the one or more EEG signals to a… model to determine one or more characteristics of a brain even, the… model being trained on training EEG data and simulated EEG data t” – may be performed by merely observing at least a limited amount of known or previously collected data and drawing mental conclusions therefrom based on prior knowledge or relationships between data [at least ¶0111 of the Applicant’s Specification defines the model as receiving an input and determining an output based on known or previously collected data or information]
If a claim, under BRI, covers performance of the limitations in the mind but for the mere recitation of extra-solutionary activity (and otherwise generic computer elements) then the claim falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea under Step 2A Prong 1 of the Mayo framework as set forth in the 2019 PEG.
No limitations are provided that would force the complexity of any of the identified evaluation steps to be non-performable by pen-and-paper practice.
Alternatively or additionally, these steps describe the concept of using implicit mathematical formula(s) [i.e., “apply the one or more EEG signals to a… model to determine one or more characteristics of a brain event”] to derive a conclusion based on input of data, which corresponds to concepts identified as abstract ideas by the courts [Diamond v. Diehr. 450 U.S. 175, 209 U.S.P.Q. 1 (1981), Parker v. Flook. 437 U.S. 584, 19 U.S.P.Q. 193 (1978), and In re Grams. 888 F.2d 835, 12 U.S.P.Q.2d 1824 (Fed. Cir. 1989)]. The concept of the recited limitations identified as mathematical concepts above is not meaningfully different than those mathematical concepts found by the courts to be abstract ideas.
The dependent claims merely include limitations that either further define the abstract idea [e.g. limitations relating the data gathered or particular steps which are entirely embodied in the mental process] and amount to no more than generally linking the use of the abstract idea to a particular technological environment or field of use because they are merely incidental or token additions to the claims that do not alter or affect how the process steps are performed.
Thus, these concepts are similar to court decisions of abstract ideas of itself: collecting, displaying, and manipulating data [Int. Ventures v. Cap One Financial], collecting information, analyzing it, and displaying certain results of the collection and analysis [Electric Power Group], collection, storage, and recognition of data [Smart Systems Innovations].
Step 2A Prong 2
The judicial exception is not integrated into a practical application.
Representative claim 1 only recites additional elements of extra-solutionary activity – in particular, extra-solution activity [generic computer functions, data gathering] – without further sufficient detail that would tie the abstract portions of the claim into a specific practical application (2019 PEG p. 55 – the instant claim, for example does not tie into a particular machine, a sufficiently particular form of data or signal collection – via the claimed extra-solution activity identified above, or a sufficiently particular form of display or computing architecture/structure).
Dependent claim(s) 2-7, 9-10, 12-16, and 18-19 merely add detail to the abstract portions of the claim but do not otherwise encompass any additional elements which tie the claim(s) into a particular application/integration [the dependent claim(s) recite generic ‘units’ or ‘steps’ which encompass mere computer instructions to carry out an otherwise wholly abstract idea].
Dependent claim(s) 8, 17 encounter substantially the same issues as the independent claim(s) from which they depend in that they encompass further generic extra-solutionary activity [generic data gathering] and/or generic computer elements [storage, memory per se].
Accordingly, the claim(s) are not integrated into a practical application under Step 2A Prong 2.
Step 2B
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Independent claims 1, 11, and 20 as individual wholes fail to amount to significantly more than the judicial exception at Step 2B. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of extra-solutionary activity [i.e., generic computer function, data gathering] and generic computer elements cannot amount to significantly more than an abstract idea [MPEP § 2106.05(f)] and is further considered to merely implement an abstract idea on a generic computer [MPEP § 2106.05(d)(II) establishes computer-based elements which are considered to be well understood, routine, and conventional when recited at a high level of generality].
For the independent claim portions and dependent claims which provide additional elements of extra-solutionary data gathering, MPEP § 2106.05(g) establishes that mere data gathering for determining a result does not amount to significantly more. The extra-solutionary activity of processor steps [acquiring, storing signals, etc.] as presently recited, cannot provide an inventive concept which amounts to significantly more than the recited abstract idea.
For the independent claims as well as the dependent claims merely reciting generic computer elements and functions [memory and processing circuitry, each recited at a high level of generality, and corresponding generic functions therein], MPEP § 2106.05(d)(II) establishes computer-based elements which are considered to be well understood, routine, and conventional when recited at a high level of generality.
Accordingly, the generic computer elements and corresponding functions, as presently limited, cannot provide an inventive concept since they fall under a generic structure and/or function that does not add a meaningful additional feature to the judicial exception(s) of the claim(s).
Claims 1, 11, and 20 recite “a plurality of electrodes” and “sensing circuitry configured to: sense, via at least two electrodes of the plurality of electrodes, electrical signals from a patient; and generate, based on the electrical signals, one or more electroencephalography (EEG) signals”. Such a plurality of electrodes and sensing circuitry is/are considered well-understood, routine, and conventional, as known by at least:
Applicant’s disclosure is not particular regarding the particular structure of the generically claimed plurality of electrodes and sensing circuitry, and recites the plurality of electrodes and sensing circuitry at a high level of generality [Conventional EEG electrodes are typically positioned over a large portion of a user's scalp. While electrodes in this region are well positioned to detect electrical activity from the patient's brain (Applicant’s Specification ¶0045), wherein the Examiner notes that the Applicant’s acknowledgement of conventional EEG electrodes is considered to be applicable to the claimed broad recitation of “plurality of electrodes” and “sensing circuitry”; The example techniques may additionally, or alternatively, be used with a medical device not illustrated in FIG. 1A such as another type of IMD, a patch monitor device, a wearable device (e.g., smart watch), or another type of external medical device (Applicant’s Specification ¶0051)]. This lack of disclosure is acceptable under 35 U.S.C. 112(a) since this hardware performs non-specialized functions known by those of ordinary skill in the medical technology arts. Thus, Applicant's specification essentially admits that this hardware is conventional and performs well understood, routine and conventional activities in the field of electrical activity monitoring. In other words, Applicant’s specification demonstrates the well-understood, routine, conventional nature of the above-identified additional element because it describes such an additional element in a manner that indicates that the additional element is sufficiently well-known that the specification does not need to describe the particulars of such additional elements to satisfy 35 U.S.C. 112(a) [see Berkheimer memo from April 19, 2018, Page 3, (III)(A)(1), not attached]. Adding hardware that performs “well understood, routine, conventional activit[ies]’ previously known to the industry” will not make claims patent-eligible [TLI Communications].
Christensen (US-20220061743-A1, cited by Applicant) [Sensing circuitry 406 may monitor signals from electrodes 418A-418C in order to monitor electrical activity of the brain (e.g., to produce an EEG) (Christensen ¶0101, Fig. 4)]
Yamagata (US-20200297231-A1) [the IMD may include a housing that carries multiple electrodes directly on the housing. Using these housing electrodes, the IMD may sense electrical signals from one or more vectors and generate physiological information representative of patient condition. The physiological information may be indicative of brain activity (Yamagata ¶0038); Accordingly, in some embodiments, the sensor data may be filtered or otherwise manipulated to separate the brain activity data (e.g., EEG signals) (Yamagata ¶0041); For example, IMD 106 may extract features from EEG signals indicative of brain activity or cardiac activity. IMD 106 may then determine whether or not the patient has experienced a stroke or seizure based on these extracted feature (Yamagata ¶0047)]
Murphy (US-20210370064-A1) [The disclosed device comprising a wearable device housing or frame including one or more EEG electrodes, one of more ultrasound transducer arrays. The disclosed device can further comprise one or more EEG signal amplifiers and/or a digital analog converter (Murphy ¶0007)]
Claims 1, 7-8, 11, 16-17, and 20 each recite a “machine learning (ML) model”. Such a machine learning (ML) model is considered well-understood, routine, and conventional, as known by at least:
Hu (“Intelligent Sensor Networks”, NPL attached) [In supervised learning, the learner is provided with labeled input data. This data contains a sequence of input/output pairs of the form xi, yi, where xi is a possible input and yi is the correctly labeled output associated with it. The aim of the learner in supervised learning is to learn the mapping from inputs to outputs. The learning program is expected to learn a function f that accounts for the input/output pairs seen so far, f (xi) = yi, for all i. This function f is called a classifier if the output is discrete and a regression function if the output is continuous. The job of the classifier/regression function is to correctly predict the outputs of inputs it has not seen before (Hu, Page 5)]
Huang (“Kernel Based Algorithms for Mining Huge Data Sets”, NPL attached) [In supervised learning, the learner is provided with labeled input data. This data contains a sequence of input/output pairs of the form xi, yi, where xi is a possible input and yi is the correctly labeled output associated with it. The aim of the learner in supervised learning is to learn the mapping from inputs to outputs. The learning program is expected to learn a function f that accounts for the input/output pairs seen so far, f (xi) = yi, for all i. This function f is called a classifier if the output is discrete and a regression function if the output is continuous. The job of the classifier/regression function is to correctly predict the outputs of inputs it has not seen before (Huang, Page 1)]
Mitchell (“The Discipline of Machine Learning”, NPL attached) [For example, we now have a variety of algorithms for supervised learning of classification and regression functions; that is, for learning some initially unknown function f : X [Calibri font/0xE0] Y given a set of labeled training examples {xi; yi} of inputs xi and outputs yi = f(xi) (Mitchell, Pages 3-4)]
Examiner’s Note Regarding Particular Treatment or Prophylaxis: Claim(s) 1, 11, and 20 recite subject matter regarding “determine one or more characteristics of a brain event”, wherein claims 2-3, 10, 12-13“wherein the one or more characteristics of the brain event includes a type of stroke” [claims 2, 12], “wherein the type of stroke includes one or more of an ischemic stroke, hemorrhagic stroke, cryptogenic stroke, or stroke mimic” [claims 3, 13], and “wherein the brain event comprises a stroke, a brain ischemic event, a brain hypoxia event, or a seizure” [claim 10], which the Examiner notes is not considered to be a particular treatment or prophylaxis, as none of the identified claims positively recite or include language that is considered to be a particular treatment or prophylaxis as an additional element to integrate the judicial exception into a practical application or allow the identified claims to amount to significantly more than the judicial exception [MPEP § 2106.04(d)(2)].
Accordingly, the claim(s) as whole(s) fail amount to significantly more than the judicial exception under Step 2B.
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.
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.
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.
Claim(s) 1-6, 10-15, and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Christensen (US-20220061743-A1, cited by Applicant) in view of Soni (US-20200342362-A1).
Regarding claim 1, Christensen teaches
A system comprising:
a memory [storage device 410 may be referred to as a memory and include computer-readable instructions that, when executed by processing circuitry 402, cause IMD 400 and processing circuitry 402 to perform various functions attributed to IMD 400 and processing circuitry 402 herein (Christensen ¶0105)];
a plurality of electrodes [electrodes 418A-418C (Christensen Fig. 4)];
sensing circuitry configured to:
sense, via at least two electrodes of the plurality of electrodes, electrical signals from a patient [Sensing circuitry 406 may monitor signals from electrodes 418A-418C in order to monitor electrical activity of the brain (e.g., to produce an EEG) (Christensen ¶0101, Fig. 4)]; and
generate, based on the electrical signals, one or more electroencephalography (EEG) signals [Christensen ¶0101, Fig. 4]; and
processing circuitry configured to:
receive, from the sensing circuitry, one or more EEG signals [Sensing circuitry 406 may monitor signals from electrodes 418A-418C in order to monitor electrical activity of the brain (e.g., to produce an EEG) and/or hearth (e.g., to product an ECG) from which processing circuitry 402 may generate stroke metrics and seizure metrics (Christensen ¶0101, Fig. 4)]; and
apply the one or more EEG signals to a machine learning (ML) model to determine one or more characteristics of a brain event, the ML model being trained on training EEG data [the processing circuitry 402 is configured to analyze data from one or more electrode combinations using electrodes 418 to extract brain activity data (Christensen ¶0107); In some examples, sensing circuitry 406 senses a brain signal via electrodes 418. The brain signal may represent the electrical activity of the brain, and may be an EEG. Processing circuitry 402 may determine parameter values from the brain signal, such values determined based on magnitudes of the signal in one or more frequency bands. Sensing circuitry 406 may include filters and other circuitry to isolate the brain signal of interest (Christensen ¶0108); Processing circuitry 402 may be configured to calculate physiological characteristics relating to one or more electrical signals received from the electrodes 418, such as stroke metrics. For example, processing circuitry 402 may be configured to algorithmically determine the presence or absence of a stroke (via generation of a stroke metric) or other neurological condition from the electrical signal (Christensen ¶0115); Processing circuitry 402 may employ various techniques to determine the stroke metric and seizure metric. For example, processing circuitry 402 may generate the stroke metric using one or more different algorithms, such as using machine learning algorithms (Christensen ¶0134); The accuracy of any classifier can be improved by training the algorithm on larger sets of data corresponding to stroke and non-stroke EEG readings (Christensen ¶0136)].
However, while Christensen discloses that the accuracy of any classifier can be improved by training an algorithm of the classifier on larger sets of data [Christensen ¶0136], Christensen fails to explicitly disclose wherein the ML model is also trained on simulated EEG data.
Soni discloses systems and methods for using machine learning to detect events, including stroke events, wherein Soni discloses training a machine learning model using simulated data [An example framework includes a computer and/or other processor executing one or more deep generative models such as a Generative Adversarial Network, etc., trained on aggregated medical machine time series data converted into a single standardized data structure format. The data can be organized in an ordered arrangement per patient to generate synthetic data samples and corresponding synthetic events and/or to generate missing data for time-series real data imputation, for example. Thus, additional, synthetic data/events can be generated to provide more data for training, testing, etc., of artificial intelligence network models, and/or data missing from a time series can be imputed and/or otherwise interpolated to provide a time series of data for modeling, analysis, etc. (Soni ¶0029); Certain examples provide systems and methods for missing data imputation of machine and/or physiological vitals data using AI model(s). For example, as shown in an example system 200 of FIG. 2A, machine data 210 and physiological (e.g., vitals, etc.) data 212, 214 can be captured from one or more medical devices 220, mobile digital health monitors 222, one or more diagnostic cardiology (DCAR) devices 224, etc., is provided in a data stream 230, 235 (e.g., continuous streaming, live streaming, periodic streaming, etc.) to a preprocessor 240, 245 to pre-process the data and apply one or more machine learning models 250, 255 (e.g., AI models, such as a DL model, a hybrid RL model, a DL+hybrid RL model, etc.) to detect events (e.g., heart attack, stroke, high blood pressure, accelerated heart rate, etc.) in a set of real data 260, 265 formed from the data stream 230, 235, etc., for example (Soni ¶0057)].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the system of Christensen to employ wherein the ML model is also trained on simulated EEG data, so as to allow for additional training and testing of the ML model.
Regarding claim 2, Christensen in view of Soni teaches
The system of claim 1, wherein the one or more characteristics of the brain event includes a type of stroke [Additionally, in some examples, a classifier can be used to discriminate between ischemic and hemorrhagic strokes (Christensen ¶0136)].
Regarding claim 3, Christensen in view of Soni teaches
The system of claim 2, wherein the type of stroke includes one or more of an ischemic stroke, hemorrhagic stroke [Christensen ¶0136], cryptogenic stroke, or stroke mimic.
Regarding claim 4, Christensen in view of Soni teaches
The system of claim 1, wherein the one or more characteristics of the brain event includes a location of the brain event in a brain of the patient [In some examples, such an etiology classifier can make a determination (probabilistic or definitive) of the origin of the stroke (e.g., ischemic or hemorrhagic). Such determinations can be made based on collected EEG sensor data alone or in conjunction with additional physiological parameters or patient data. For example, the etiology classifier may determine a location of the stroke. For example, the location determination can include a left-versus-right hemisphere determination (e.g., a binary output or probabilistic result)… the system may utilize information regarding which hemisphere of the brain the seizure or stroke originates from as at least part of the location of the stroke, as one example. This hemisphere specific information may be obtained from locations other than T3 and T4 in other examples. In some examples, the location determination can include a more precise mapping of brain regions with particular probabilities assigned, for example a 70% probability of the stroke location being at a particular point on the patient's brain. The stroke location may be output along a spherical surface map or other suitable coordinate system for identifying the location in the patient's brain (Christensen ¶0138)].
Regarding claim 5, Christensen in view of Soni teaches
The system of a claim 1, wherein the one or more characteristics of the brain event include a magnitude of the brain event [Processing circuitry 402 may determine parameter values from the brain signal, such values determined based on magnitudes of the signal in one or more frequency bands (Christensen ¶0108)].
Regarding claim 6, Christensen in view of Soni teaches
The system of claim 1, wherein the simulated EEG data fills in coverage gaps of the training EEG data [See § 103 modification of claim 1 above; Soni ¶0029].
Regarding claim 10, Christensen in view of Soni teaches
The system of claim 1, wherein the brain event comprises a stroke, a brain ischemic event [Christensen ¶0136], a brain hypoxia event, or a seizure [Processing circuitry 402 may search for one or more features in the physiological information that are indicative of one or more types of seizures (Christensen ¶0116)].
Regarding claim 11, Christensen teaches
A method comprising:
sensing, by sensing circuitry and via at least two electrodes of a plurality of electrodes, electrical signals from a patient [Sensing circuitry 406 may monitor signals from electrodes 418A-418C in order to monitor electrical activity of the brain (e.g., to produce an EEG) (Christensen ¶0101, Fig. 4)];
generating, by the sensing circuitry and based on the electrical signals, one or more electroencephalography (EEG) signals [Christensen ¶0101, Fig. 4];
receiving, by processing circuitry and from the sensing circuitry, the one or more EEG signals [Sensing circuitry 406 may monitor signals from electrodes 418A-418C in order to monitor electrical activity of the brain (e.g., to produce an EEG) and/or hearth (e.g., to product an ECG) from which processing circuitry 402 may generate stroke metrics and seizure metrics (Christensen ¶0101, Fig. 4)]; and
applying the one or more EEG signals to a machine learning (ML) model to determine one or more characteristics of a brain event, the ML model being trained on training EEG data [the processing circuitry 402 is configured to analyze data from one or more electrode combinations using electrodes 418 to extract brain activity data (Christensen ¶0107); In some examples, sensing circuitry 406 senses a brain signal via electrodes 418. The brain signal may represent the electrical activity of the brain, and may be an EEG. Processing circuitry 402 may determine parameter values from the brain signal, such values determined based on magnitudes of the signal in one or more frequency bands. Sensing circuitry 406 may include filters and other circuitry to isolate the brain signal of interest (Christensen ¶0108); Processing circuitry 402 may be configured to calculate physiological characteristics relating to one or more electrical signals received from the electrodes 418, such as stroke metrics. For example, processing circuitry 402 may be configured to algorithmically determine the presence or absence of a stroke (via generation of a stroke metric) or other neurological condition from the electrical signal (Christensen ¶0115); Processing circuitry 402 may employ various techniques to determine the stroke metric and seizure metric. For example, processing circuitry 402 may generate the stroke metric using one or more different algorithms, such as using machine learning algorithms (Christensen ¶0134); The accuracy of any classifier can be improved by training the algorithm on larger sets of data corresponding to stroke and non-stroke EEG readings (Christensen ¶0136)].
However, while Christensen discloses that the accuracy of any classifier can be improved by training an algorithm of the classifier on larger sets of data [Christensen ¶0136], Christensen fails to explicitly disclose wherein the ML model is also trained on simulated EEG data.
Soni discloses systems and methods for using machine learning to detect events, including stroke events, wherein Soni discloses training a machine learning model using simulated data [An example framework includes a computer and/or other processor executing one or more deep generative models such as a Generative Adversarial Network, etc., trained on aggregated medical machine time series data converted into a single standardized data structure format. The data can be organized in an ordered arrangement per patient to generate synthetic data samples and corresponding synthetic events and/or to generate missing data for time-series real data imputation, for example. Thus, additional, synthetic data/events can be generated to provide more data for training, testing, etc., of artificial intelligence network models, and/or data missing from a time series can be imputed and/or otherwise interpolated to provide a time series of data for modeling, analysis, etc. (Soni ¶0029); Certain examples provide systems and methods for missing data imputation of machine and/or physiological vitals data using AI model(s). For example, as shown in an example system 200 of FIG. 2A, machine data 210 and physiological (e.g., vitals, etc.) data 212, 214 can be captured from one or more medical devices 220, mobile digital health monitors 222, one or more diagnostic cardiology (DCAR) devices 224, etc., is provided in a data stream 230, 235 (e.g., continuous streaming, live streaming, periodic streaming, etc.) to a preprocessor 240, 245 to pre-process the data and apply one or more machine learning models 250, 255 (e.g., AI models, such as a DL model, a hybrid RL model, a DL+hybrid RL model, etc.) to detect events (e.g., heart attack, stroke, high blood pressure, accelerated heart rate, etc.) in a set of real data 260, 265 formed from the data stream 230, 235, etc., for example (Soni ¶0057)].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Christensen to employ wherein the ML model is also trained on simulated EEG data, so as to allow for additional training and testing of the ML model.
Regarding claim 12, Christensen in view of Soni teaches
The method of claim 11, wherein the one or more characteristics of the brain event includes a type of stroke [Christensen ¶0136].
Regarding claim 13, Christensen in view of Soni teaches
The method of claim 12, wherein the type of stroke includes one or more of an ischemic stroke, hemorrhagic stroke [Christensen ¶0136], cryptogenic stroke, or stroke mimic.
Regarding claim 14, Christensen in view of Soni teaches
The method of claim 11, wherein the one or more characteristics of the brain event includes one or more of a location of the brain event in a brain of the patient [Christensen ¶0138] or a magnitude of the brain event [Christensen ¶0108].
Regarding claim 15, Christensen in view of Soni teaches
The method of claim 11, wherein the simulated EEG data fills in coverage gaps of the training EEG data [See § 103 modification of claim 11 above; Soni ¶0029].
Regarding claim 20, Christensen teaches
A non-transitory computer-readable storage medium comprising instructions [Aspects of the technology described herein can be embodied in a special purpose computer or data processor that is specifically programmed, configured, or constructed to perform one or more of the computer-executable instructions explained in detail herein (Christensen ¶0043); Computer-implemented instructions, data structures, screen displays, and other data under aspects of the technology may be stored or distributed on computer-readable storage media, including magnetically or optically readable computer disks, as microcode on semiconductor memory, nanotechnology memory, organic or optical memory, or other portable and/or non-transitory data storage media (Christensen ¶0044)] that, when executed, cause processing circuitry to execute method comprising:
sensing, using sensing circuitry and via at least two electrodes of a plurality of electrodes, electrical signals from a patient [Sensing circuitry 406 may monitor signals from electrodes 418A-418C in order to monitor electrical activity of the brain (e.g., to produce an EEG) (Christensen ¶0101, Fig. 4)];
generating, using the sensing circuitry and based on the electrical signals, one or more electroencephalography (EEG) signals [Christensen ¶0101, Fig. 4];
receiving, from the sensing circuitry, the one or more EEG signals [Sensing circuitry 406 may monitor signals from electrodes 418A-418C in order to monitor electrical activity of the brain (e.g., to produce an EEG) and/or hearth (e.g., to product an ECG) from which processing circuitry 402 may generate stroke metrics and seizure metrics (Christensen ¶0101, Fig. 4)]; and
applying the one or more EEG signals to a machine learning (ML) model to determine one or more characteristics of a brain event, the ML model being trained on training EEG data [the processing circuitry 402 is configured to analyze data from one or more electrode combinations using electrodes 418 to extract brain activity data (Christensen ¶0107); In some examples, sensing circuitry 406 senses a brain signal via electrodes 418. The brain signal may represent the electrical activity of the brain, and may be an EEG. Processing circuitry 402 may determine parameter values from the brain signal, such values determined based on magnitudes of the signal in one or more frequency bands. Sensing circuitry 406 may include filters and other circuitry to isolate the brain signal of interest (Christensen ¶0108); Processing circuitry 402 may be configured to calculate physiological characteristics relating to one or more electrical signals received from the electrodes 418, such as stroke metrics. For example, processing circuitry 402 may be configured to algorithmically determine the presence or absence of a stroke (via generation of a stroke metric) or other neurological condition from the electrical signal (Christensen ¶0115); Processing circuitry 402 may employ various techniques to determine the stroke metric and seizure metric. For example, processing circuitry 402 may generate the stroke metric using one or more different algorithms, such as using machine learning algorithms (Christensen ¶0134); The accuracy of any classifier can be improved by training the algorithm on larger sets of data corresponding to stroke and non-stroke EEG readings (Christensen ¶0136)].
However, while Christensen discloses that the accuracy of any classifier can be improved by training an algorithm of the classifier on larger sets of data [Christensen ¶0136], Christensen fails to explicitly disclose wherein the ML model is also trained on simulated EEG data.
Soni discloses systems and methods for using machine learning to detect events, including stroke events, wherein Soni discloses training a machine learning model using simulated data [An example framework includes a computer and/or other processor executing one or more deep generative models such as a Generative Adversarial Network, etc., trained on aggregated medical machine time series data converted into a single standardized data structure format. The data can be organized in an ordered arrangement per patient to generate synthetic data samples and corresponding synthetic events and/or to generate missing data for time-series real data imputation, for example. Thus, additional, synthetic data/events can be generated to provide more data for training, testing, etc., of artificial intelligence network models, and/or data missing from a time series can be imputed and/or otherwise interpolated to provide a time series of data for modeling, analysis, etc. (Soni ¶0029); Certain examples provide systems and methods for missing data imputation of machine and/or physiological vitals data using AI model(s). For example, as shown in an example system 200 of FIG. 2A, machine data 210 and physiological (e.g., vitals, etc.) data 212, 214 can be captured from one or more medical devices 220, mobile digital health monitors 222, one or more diagnostic cardiology (DCAR) devices 224, etc., is provided in a data stream 230, 235 (e.g., continuous streaming, live streaming, periodic streaming, etc.) to a preprocessor 240, 245 to pre-process the data and apply one or more machine learning models 250, 255 (e.g., AI models, such as a DL model, a hybrid RL model, a DL+hybrid RL model, etc.) to detect events (e.g., heart attack, stroke, high blood pressure, accelerated heart rate, etc.) in a set of real data 260, 265 formed from the data stream 230, 235, etc., for example (Soni ¶0057)].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the non-transitory computer-readable storage medium comprising instructions of Christensen to employ wherein the ML model is also trained on simulated EEG data, so as to allow for additional training and testing of the ML model.
Claim(s) 7, 9, 16, and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Christensen in view of Soni, as applied to claims 1 and 11 above, in further view of Yamagata (US-20200297231-A1).
Regarding claim 7, Christensen in view of Soni teaches
The system of claim 1.
However, while Christensen discloses time-correlating EEG signals [The sensed electrical signals may be time-coded or otherwise correlated with time data, and stored in this form, so that the recency, frequency, time of day, time span, or date(s) of a particular signal data point or data series (or computed measures or statistics based thereon) may be determined and/or reported (Christensen ¶0077)] and wherein the processing circuitry is configured to apply the one or more EEG signals to the ML model to generate a three-dimensional map of the brain of the patient [In some examples, the location determination can include a more precise mapping of brain regions with particular probabilities assigned, for example a 70% probability of the stroke location being at a particular point on the patient's brain. The stroke location may be output along a spherical surface map or other suitable coordinate system for identifying the location in the patient's brain. Processing circuitry 402 may output the result of these classifiers in the form of a value of the patient metric or other value indicative of the detected type of stroke (Christensen ¶0138)], Christensen in view of Soni fails to explicitly disclose wherein the processing circuitry is configured to: apply the one or more EEG signals to the ML model to generate a four-dimensional map of the brain of the patient.
Yamagata discloses systems and methods for measuring and displaying biomedical signals, including at least EEG signals [The biomedical-signal measuring system 1 (an example of an information processing system) measures various kinds of biomedical signals of a test subject such as magneto-encephalography (MEG) signals and electro-encephalography (EEG) signals, and displays the results of measurement (Yamagata ¶0084)], wherein Yamagata discloses generating a four-dimensional map of a brain of a patient [The three-view head image 913 has functions similar to those of the three-view head image 613 of the time-frequency analysis screen 601, and includes sectional views 941 to 943 (an example of a sectional image) and a three-dimensional image 944. The dipole that is selected from the dipole list 916 and the result of time-frequency analysis that is selected from the time-frequency analysis result list 918 (i.e., a heat map that indicates the distribution of the signal strength of the biomedical signal at the specified time and frequency that corresponds to the activity of the brain selected from the time-frequency analysis result list 918) are superimposed on the three-view head image 913. A plurality of dipoles are selectable from the dipole list 916 in a similar manner to the dipole list 616 on the time-frequency analysis screen 601, and the dipole display control unit 231 controls the display to display a plurality of dipoles that are selected from the dipole list 916 on the three-view head image 913. In order to secure the viewability of the dipoles, for example, the dipole display control unit 231 may add a border to each of the dipoles, or may control the display to display the dipoles with the color selected from the color options displayed when dipoles are selected from the dipole list 916. Such measures to secure the viewability of the dipoles may also be performed on the above three-view head image 613 of the time-frequency analysis screen 601 in a similar manner to the above (Yamagata ¶0308)].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the system of Christensen in view of Soni to employ wherein the processing circuitry is configured to: apply the one or more EEG signals to the ML model to generate a four-dimensional map of the brain of the patient, so as to facilitate visualization of brain activity over time.
Regarding claim 9, Christensen in view of Soni and Yamagata teaches
The system of claim 7, wherein the processing circuitry is configured to generate the four-dimensional map of the brain of the patient using a dipole model [See § 103 modification of claim 7 above; Yamagata ¶0308].
Regarding claim 16, Christensen in view of Soni teaches
The method of claim 11.
However, while Christensen discloses time-correlating EEG signals [Christensen ¶0077] and wherein the processing circuitry is configured to apply the one or more EEG signals to the ML model to generate a three-dimensional map of the brain of the patient [Christensen ¶0138], Christensen in view of Soni fails to explicitly disclose further comprising: applying the one or more EEG signals to the ML model to generate a four-dimensional map of the brain of the patient.
Yamagata discloses systems and methods for measuring and displaying biomedical signals, including at least EEG signals [Yamagata ¶0084], wherein Yamagata discloses generating a four-dimensional map of a brain of a patient [Yamagata ¶0308].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Christensen in view of Soni to employ wherein the processing circuitry is configured to: apply the one or more EEG signals to the ML model to generate a four-dimensional map of the brain of the patient, so as to facilitate visualization of brain activity over time.
Regarding claim 18, Christensen in view of Soni and Yamagata teaches
The method of claim 16 further comprising: generating the four-dimensional map of the brain of the patient using a dipole model [See § 103 modification of claim 16 above; Yamagata ¶0308].
Claim(s) 8 and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Christensen in view of Soni and Yamagata, as applied to claims 7 and 16 above, in further view of Lee (US-20230386227-A1, EFD of 24 May 2023).
Regarding claim 8, Christensen in view of Soni and Yamagata teaches
The system of claim 7.
However, while Christensen in view of Soni and Yamagata is considered to disclose sending simulated data to train the ML model in the event of missing data [Soni ¶0029] to provide the four-dimensional map of the brain of the patient [Yamagata ¶0308], Christensen in view of Soni and Yamagata fails to explicitly disclose wherein the processing circuitry is configured to: determine, using the four-dimensional map, whether a coverage gap of the training EEG data and the simulated EEG data satisfy a gap threshold; in response to determining the coverage gap satisfies the gap threshold, generate additional simulated EEG data to fill in the coverage gap; send the additional simulated EEG data to further train and update the ML model; and apply the one or more EEG signals to the updated ML model to determine the one or more characteristics of the brain event.
Lee discloses systems and methods for assessing a physiological state using physiological sensors and machine learning, wherein Lee discloses the classification of the user undergoing a stroke and use of an EEG sensor [The data processing system can classify a status of an occupant of the vehicle based on the sensor data. For example, the data processing system may include or otherwise employ a classification model or ensemble of classification models. The classification models can be trained to classify a status of an occupant according to one or more predefined statuses, such as statuses relating to a magnitude of an anomaly, an ability to control a vehicle, or a particular medical status… The action can include conveying an indication of the condition to a provider for the occupant (e.g., convey an indication that the user has undergone a stroke, seizure (Lee ¶0030); The physiological sensor 106B can include an electroencephalogram (Lee ¶0047)], and wherein Lee discloses imputing simulated sensor data when an amount or quality of real sensor data satisfies a threshold [a data imputation engine 202 can densify received sensor data 122. For example, the data imputation engine 202 can define data according to linear, polynomial (e.g., spline), or other interpolation. For example, a pattern such as breathing, heartrate, or the like can be interpolated according to a portion thereof (e.g., missing data in a QRS complex may be substituted for an average of other QRS complexes). The data imputation engine 202 can determine a quantity of deficient sensor data 122, compare the quantity of deficient sensor data 122 to a threshold, and impute sensor data 122 responsive to the quantity of deficient sensor data 122 being less than a threshold (Lee ¶0061); the trained models 112A of the anomaly detection or classification circuit 112 can receive video derived features 302, physiological parameters 304, user parameters 208, vehicle parameters 210, identified conditions 212, or other sensor data 122 or derivatives thereof (Lee ¶0069)].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the system of Christensen in view of Soni and Yamagata to employ wherein the processing circuitry is configured to: determine, using the four-dimensional map, whether a coverage gap of the training EEG data and the simulated EEG data satisfy a gap threshold; in response to determining the coverage gap satisfies the gap threshold, generate additional simulated EEG data to fill in the coverage gap; send the additional simulated EEG data to further train and update the ML model; and apply the one or more EEG signals to the updated ML model to determine the one or more characteristics of the brain event, so as to remedy deficient sensor data and ensure sufficient data for further analysis and classification.
Regarding claim 17, Christensen in view of Soni and Yamagata teaches
The method of claim 16.
However, while Christensen in view of Soni and Yamagata is considered to disclose sending simulated data to train the ML model in the event of missing data [Soni ¶0029] to provide the four-dimensional map of the brain of the patient [Yamagata ¶0308], Christensen in view of Soni and Yamagata fails to explicitly disclose further comprising: determining, using the four-dimensional map, whether a coverage gap of the training EEG data and the simulated EEG data satisfy a gap threshold; in response to determining the coverage gap satisfies the gap threshold, generating additional simulated EEG data to fill in the coverage gap; sending the additional simulated EEG data to further train and update the ML model; and applying the one or more EEG signals to the updated ML model to determine the one or more characteristics of the brain event.
Lee discloses systems and methods for assessing a physiological state using physiological sensors and machine learning, wherein Lee discloses the classification of the user undergoing a stroke and use of an EEG sensor [Lee ¶¶0030, 0047], and wherein Lee discloses imputing simulated sensor data when an amount or quality of real sensor data satisfies a threshold [Lee ¶¶0061, 0069)].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Christensen in view of Soni and Yamagata to employ further comprising: determining, using the four-dimensional map, whether a coverage gap of the training EEG data and the simulated EEG data satisfy a gap threshold; in response to determining the coverage gap satisfies the gap threshold, generating additional simulated EEG data to fill in the coverage gap; sending the additional simulated EEG data to further train and update the ML model; and applying the one or more EEG signals to the updated ML model to determine the one or more characteristics of the brain event, so as to remedy deficient sensor data and ensure sufficient data for further analysis and classification.
Claim(s) 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Christensen in view of Soni, Yamagata, and Lee, as applied to claim 17 above, in further view of Mersmann (US-20190223779-A1).
Regarding claim 19, Christensen in view of Soni, Yamagata, and Lee teaches
The method of claim 17.
However, Christensen in view of Soni, Yamagata, and Lee fails to explicitly disclose further comprising: determining optimized electrode separation and implant location using the four- dimensional map of the brain of the patient.
Mersmann discloses systems for mapping electrical activity occurring within a patient’s brain using EEG electrodes, wherein Mersmann discloses determining optimized electrode separation and implant location using a four dimensional map of the brain of the patient [Forward models are used to map, for example, electrical activity or tissue oxygenation occurring within the brain volume at different times in different places to surface potentials, which may be measured by invasive or non-invasive surface electrodes during the recording of a SEEG, EEG or MEG. Since the analysis of SEEG and EEG data, in particular the comparison of different EEG and SEEG recordings, is easily accomplished by a plurality of commercial and non-commercial tools, simulated seizures can easily be compared to recorded seizure patterns. In particular, the implantation of electrodes can be postponed until a number of potential epileptogenic zones have been ruled out by the method as suggested in this application. Furthermore, the implantation scheme can be optimized with regard to best coverage and minimal invasiveness. This significantly improves the chances of performing a successful surgical resection (Mersmann ¶0023); To generate a brain network model for the patient, each voxel of the structural skeleton was replaced by a node. Each node of the resulting network was set with an epileptor as described above. The nodes were connected via permittivity coupling, which acting on a slow time scale and allowing the spread of the seizure though the network by recruiting regions not in the epileptogenic zone… To characterize the four-dimensional parameter space, we define quantities relevant for seizure description such as (i) regions involved in the seizure, (ii) seizure length, (iii) length of time delays before recruitment of other regions, (iv) seizure frequency in each region (Mersmann ¶0065)].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method of Christensen in view of Soni, Yamagata, and Lee to employ further comprising: determining optimized electrode separation and implant location using the four- dimensional map of the brain of the patient, so as to rule out potential epileptogenic zones to optimize coverage and minimize invasiveness.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to SEVERO ANTONIO P LOPEZ whose telephone number is (571)272-7378. The examiner can normally be reached M-F 9-6 EST.
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, Charles Marmor II can be reached at (571) 272-4730. 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.
/SEVERO ANTONIO P LOPEZ/Examiner, Art Unit 3791
Prosecution Insights