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 .
Election/Restrictions
Applicant’s election without traverse of claims 11-18 in the reply filed on 05/15/2026 is acknowledged. Claims 21-29 are new. Claims 11-18 and 21-29 are pending.
Information Disclosure Statement
The information disclosure statement(s) filed 03/22/2024, 06/13/2024, and 01/22/2025 has/have been considered by the Examiner.
Claim Objections
Claim 12 is objected to because of the following informalities:
Claim 12 recites, “The method of claim 11, wherein the physiological parameter comprises an electrocardiogram (ECG), a plethysmograph, a capnograph, an electroencephalogram (EEG), or an electromyograph (EMG), or wherein the condition comprises an arrhythmia a seizure…”, where the phrase “an arrhythmia a seizure” is missing a comma. Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 11-18 and 21-29 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter (abstract ideas) without significantly more.
The framework for establishing a prima facie case of lack of subject matter eligibility requires that the Examiner determine: (1) Does the claim fall within the four categories of patent eligible subject matter; (2a) prong 1: Does the claim recite an abstract idea, law of nature, or natural phenomenon and (2a) prong 2: Does the claim recite additional elements that integrate the judicial exception into a practical application; and (2b) Does the claim recite additional elements that amount of significantly more than the judicial exception.
Step 1):
Claims 11-18, 21-26 recite a method, which satisfies the 4 statutory categories (process, machine, manufacture, or composition of matter) of patent-eligible subject matter.
Claims 27-29 recite a system, which satisfies the 4 statutory categories (process, machine, manufacture, or composition of matter) of patent-eligible subject matter.
Step 2a) Prong One:
Independent claim 11 recites:
A method, comprising:
identifying data indicative of measurements of a physiological parameter of a subject over a time period;
dividing the data into multiple segments comprising a first segment and a second segment;
determining, by detecting at least one first characteristic of the first segment, a first label indicating whether the first segment is indicative of a condition;
generating a first image that is indicative of the first segment;
determining, by inputting the first image into a trained machine learning (ML) model, that the first label is inaccurate;
determining, by detecting at least one second characteristic of the second segment, a second label indicating whether the second segment is indicative of the condition;
generating a second image that is indicative of the second segment;
determining, by inputting the second image into the trained ML model, that the second label is accurate; and
outputting an indication of whether the subject has the condition based on the second label.
Independent claim 11 is/are all directed to MENTAL PROCESSES (i.e. identifying, dividing, determining, outputting an indication), where nothing in the claimed steps precludes the steps from practically being performed in the human mind or by a human using pen and paper. In the instant case, a person could mentally identify data by simply observing and recognizing data. A person could mentally divide data by observing and mentally splicing data into segments, which could also be done by pen and paper. A person could mentally determine a data characteristic by making a mental decision or judgment based on observed data. The step of “outputting an indication” could also be interpreted as a mental process, in which a person could mentally output an indication by simply using a pen and paper to indicate or notify a condition based on observed data.
Dependent claims 12-18, 21-29 contain no additional elements that integrate the abstract ideas into practical application, or amount to significantly more than the abstract idea itself. Specifically, the dependent claims only further define the abstract ideas (mental processes), and do not amount to significantly more than the abstract idea itself. Accordingly, the dependent claims are also directed to non-statutory subject matter.
Step 2a) Prong Two:
This judicial exception is not integrated into a practical application because mere instruction to implement on a computer, or merely using a computer as a tool to perform the abstract idea, adding insignificant extra solution activity, and/or generally linking the use of the abstract idea to a technological environment or field of use is not considered integration into a practical application. The Court defines the phrase “integration into a practical application” to require an additional element or a combination of additional elements in the claim to apply, rely on, or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception, such that it is more than a drafting effort designed to monopolize the exception.
This judicial exception is not integrated into a practical application because claims 11-18 and 21-29 do not disclose using the result of the mental process steps for prophylactic treatment of a particular medical condition under MPEP 2106.05(e). In the instant case, there is no specific treatment in the form of stimulation/pacing pulses, drug therapy, radiation therapy, or other forms of treatment that is ultimately used to treat a particular condition as a result of the mental process steps as stated above. The step of “outputting an indication of whether the subject has the condition based on the second label” can be interpreted as a form of data transmission, and would therefore be considered an insignificant extra-solution activity by merely supplying a value or information to a system, and can be done without practical application of treating a particular condition as a result of the mental process steps. This insignificant extra-solution activity is considered an additional step that is merely performed after the primary mental process steps, and does not add any meaningful limitation to the claim. As such, these additional elements are merely nominal or tangential additions to the claims as they do not impose any meaningful limits on the claim, see MPEP 2106.05(g) Insignificant Extra-Solution Activity. In other words, there is no specific treatment delivered to treat a particular condition that is specified in the claims, but is only directed to a form of simple data transmission. Accordingly, claims 11-18 and 21-29 do not disclose using the result of the mental processes steps for prophylactic treatment of a particular medical condition under MPEP 2106.05(e).
This judicial exception is not integrated into a practical application because claims 11-18 and 21-29 do not provide improvements to the functioning of a computer or to any the technical field under MPEP 2106.05(a). Specifically, the claims recite additional elements of a wearable device comprising generic computer elements (i.e. processor), but these elements have not been described with sufficient detail to constitute an improvement in the tech field, as such these features merely define the field of use for the current invention by generally linking mental processes to generic computer elements as a tool to execute the abstract ideas (mental processes). By failing to explain how these elements are different from conventional computer elements, it is reasonable that the broadest reasonable interpretation of the additional elements is just a conventional computer performing generic functions (e.g., data analysis and data transfer). Conventional computer elements performing basic data analysis is directed to the components of a system amounting to merely field of use type limitations and/or extra solution activity to implement the abstract idea as identified above, and merely including instructions to implement abstract ideas on a computer does not integrate the judicial exception into practical application, see MPEP 2106.04(d) Integration of a Judicial Exception into a Practical Application.
Accordingly, dependent claims 12-18 and 21-29 do not recite additional elements which practically integrate the judicial exception(s) of the current invention.
Step 2b)
Step 2B in the analysis requires us to determine whether the claims do significantly more than
simply describe that abstract method. Mayo, 132 S. Ct. at 1297. We must examine the limitations of the
claims to determine whether the claims contain an "inventive concept" to "transform" the claimed
abstract idea into patent-eligible subject matter. Alice, 134 S. Ct. at 2357 (quoting Mayo, 132 S. Ct. at
1294, 1298). The transformation of an abstract idea into patent-eligible subject matter "requires 'more
than simply stat[ing] the [abstract idea] while adding the words 'apply it."' Id. (quoting Mayo, 132 S. Ct.
at 1294) (alterations in original). "A claim that recites an abstract idea must include 'additional features'
to ensure 'that the [claim] is more than a drafting effort designed to monopolize the [abstract idea].'" Id.
(quoting Mayo, 132 S. Ct. at 1297) (alterations in original). Those "additional features" must be more
than "well-understood, routine, conventional activity." Mayo, 132 S. Ct. at 1298.
The claims also do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the recited wearable device comprising the processor is/are recognized as generic computer interfaces and generic computers (or computer components), because the claims do not describe these features as having distinguishing element(s) over their generic counterparts, of which are well-understood, routine and conventional activities previously known in the industry as shown in the reference as taught by Masuda (US 20220028060 A1) used in the rejection below, which teaches a wearable ECG monitoring system comprising a processor (figure 11A) comprising a processor (paragraph 0078).
Additionally, Kim (US 20210052181 A1) similarly teaches a wearable cardioverter defibrillator (paragraph 0019; figure 1) comprising a processor (paragraphs 0030-0031).
Thus, the present claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. When looked at individually and as a whole, the claim limitations are determined to be an abstract idea without significantly more, and thus claims 11-18 and 21-29 are not patent eligible under 35 USC § 101.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claim(s) 11-13, 17-18, 21-22, and 27-29 is/are rejected under 35 U.S.C. 102 (a)(1)/(a)(2) as being anticipated by Masuda (US 20220028060 A1 – hereinafter Masuda).
Re. claim 11, Masuda teaches a method, comprising:
identifying data indicative of measurements of a physiological parameter of a subject over a time period (paragraph 0019 – “The method may include obtaining, in real-time by processing circuitry, a number of ECG signals of a patient from at least two ECG electrodes of a wearable medical device…”; paragraph 0082 – “The set of the cardio-vibrational measurements may represent a selected period of time (e.g., predetermined duration) such as, in some examples, 15 seconds, 30 seconds, 45 seconds, or 60 seconds to be real-real-time responsive to cardiac conditions occurring in the heart of the patient such as arrhythmia…”);
dividing the data into multiple segments comprising a first segment and a second segment (paragraph 0019 – “…transforming the ECG measurements of the predetermined duration into an ECG image matrix, where transforming includes segmenting the ECG measurements of the predetermined duration into a number of adjacent ECG portions each having a duration smaller than the predetermined duration…”; figure 1, step 120 allows additional segments to be made if necessary);
PNG
media_image1.png
462
494
media_image1.png
Greyscale
determining, by detecting at least one first characteristic of the first segment, a first label indicating whether the first segment is indicative of a condition (beat-to-beat label, or RR interval is used to determine and classify arrythmia, paragraph 0131 – “Patient-derived classifier(s) 804, for example, may be advantageous in training the arrhythmia machine learning engine(s) 802 to recognize arrhythmia states in comparison to the unique features of the patient's cardiac cycles”; paragraph 0132 – “Examples of ECG metrics that can be used to aid in the classification process include average heart rate, minimum heart rate, maximum heart rate, average RR interval in milliseconds or seconds, minimum RR interval in milliseconds or seconds, maximum RR interval in milliseconds or seconds, standard deviation of RR intervals in milliseconds or seconds, number of successive RR intervals greater than a predetermined duration (e.g., 45 ms) per minute…”);
generating a first image that is indicative of the first segment (paragraph 0019 - “…transforming the ECG measurements of the predetermined duration into an ECG image matrix, where transforming includes segmenting the ECG measurements of the predetermined duration into a number of adjacent ECG portions each having a duration smaller than the predetermined duration, and plotting the number of ECG portions using a number of pixel characteristic values mapped to parameter values of corresponding ECG measurements to produce the ECG image matrix”; paragraph 0007 – “The method may include outputting the cardio-vibrational image matrix as an image file for use in monitoring the progression of the cardiac disease in the patient”);
determining, by inputting the first image into a trained machine learning (ML) model, that the first label is inaccurate (paragraph 0018 – “The operations may be configured to apply the cardio-vibrational image matrix to a machine learning classifier to determine an arrhythmia condition in the patient, where the machine learning classifier is trained to identify at least an existence and a nonexistence of an arrhythmia condition in ECG image matrices”);
determining, by detecting at least one second characteristic of the second segment, a second label indicating whether the second segment is indicative of the condition (beat-to-beat label, or RR interval is used to determine and classify arrythmia, paragraph 0131 – “Patient-derived classifier(s) 804, for example, may be advantageous in training the arrhythmia machine learning engine(s) 802 to recognize arrhythmia states in comparison to the unique features of the patient's cardiac cycles”; paragraph 0132 – “Examples of ECG metrics that can be used to aid in the classification process include average heart rate, minimum heart rate, maximum heart rate, average RR interval in milliseconds or seconds, minimum RR interval in milliseconds or seconds, maximum RR interval in milliseconds or seconds, standard deviation of RR intervals in milliseconds or seconds, number of successive RR intervals greater than a predetermined duration (e.g., 45 ms) per minute…”);
generating a second image that is indicative of the second segment (paragraph 0019 - “…transforming the ECG measurements of the predetermined duration into an ECG image matrix, where transforming includes segmenting the ECG measurements of the predetermined duration into a number of adjacent ECG portions each having a duration smaller than the predetermined duration, and plotting the number of ECG portions using a number of pixel characteristic values mapped to parameter values of corresponding ECG measurements to produce the ECG image matrix”; paragraph 0007 – “The method may include outputting the cardio-vibrational image matrix as an image file for use in monitoring the progression of the cardiac disease in the patient”);
determining, by inputting the second image into the trained ML model, that the second label is accurate (paragraph 0018 – “The operations may be configured to apply the cardio-vibrational image matrix to a machine learning classifier to determine an arrhythmia condition in the patient, where the machine learning classifier is trained to identify at least an existence and a nonexistence of an arrhythmia condition in ECG image matrices”);
and outputting an indication of whether the subject has the condition based on the second label (paragraph 0089 – “The cardio-vibrational image matrix, for example, may be provided to a display of a computing device for review by a clinician or incorporated in a report for review by an end user. For example, the image may be displayed via a user interface mounted on an external housing of the wearable medical device. In implementations, the image may be transmitted via a network interface in the wearable medical device (described in further detail below) to a remote server, for display at a remote computer screen viewed by a technician, a caregiver (e.g., a nurse, a physician, a physician's assistant, or other authorized medical representative) or other authorized person, in communication with the remote server”).
Re. claim 12, Masuda further teaches wherein the physiological parameter comprises an electrocardiogram (ECG) (paragraph 0019 – “The method may include obtaining, in real-time by processing circuitry, a number of ECG signals of a patient from at least two ECG electrodes of a wearable medical device…”), a plethysmograph, a capnograph, an electroencephalogram (EEG), or an electromyograph (EMG),
or wherein the condition comprises an arrhythmia (paragraph 0012 – “The method may include applying, by the processing circuitry, the cardio-vibrational image matrix to a machine learning classifier to determine an arrhythmia condition in the patient…”), a seizure, atrial fibrillation (AF), atrial flutter, ventricular tachycardia, supraventricular tachycardia, or an atrioventricular block (paragraph 0022 – “In some embodiments, the type of arrhythmia condition includes at least one of a supraventricular tachycardia (SVT), a ventricular tachycardia, ventricular fibrillation, tachycardia, bradycardia, asystole, a heart pause condition, pulseless electrical activity, or atrial fibrillation”).
Re. claim 13, Masuda further teaches wherein the measurements of the physiological parameter of the subject are generated by a wearable device worn by the subject (paragraph 0197 – “For example, the wearable medical device can be configured to be worn by a patient for as many as twenty-four hours a day”; paragraph 0055 – “FIGS. 11A-11D illustrate example wearable medical devices for monitoring a cardiac condition of a patient”) or a hand-held device.
PNG
media_image2.png
420
410
media_image2.png
Greyscale
Re. claim 17, Masuda further teaches generating the trained ML model by optimizing parameters of an untrained ML model based on training data (paragraph 0027 – “In some embodiments, the machine learning classifier is trained to identify at least the existence and the nonexistence of the arrhythmia condition by identifying noise conditions in the ECG image matrices”), the training data comprising:
training images indicative of other measurements of the physiological parameter of other subjects (paragraph 0034 – “In some embodiments, the at least one machine learning classifier was trained at least in part using a number of ECG image matrices generated from a number of historic ECG readings obtained from the patient”);
and ground truth labels (arrythmia classifications, paragraph 0027) indicating whether the other measurements indicate the condition (paragraph 0027 – “Identifying the arrhythmia condition may include classifying the arrhythmia condition based on applying the ECG image matrix to the machine learning classifier, where the machine learning classifier is trained to identify the type of arrhythmia condition using a number of arrhythmia classifications”, further detailed in paragraph 0135).
Re. claim 18, Masuda further teaches wherein identifying the data comprises:
receiving a signal indicating the data from an external device (paragraph 0089 – “The cardio-vibrational image matrix, for example, may be provided to a display of a computing device for review by a clinician or incorporated in a report for review by an end user. For example, the image may be displayed via a user interface mounted on an external housing of the wearable medical device. In implementations, the image may be transmitted via a network interface in the wearable medical device (described in further detail below) to a remote server, for display at a remote computer screen viewed by a technician, a caregiver (e.g., a nurse, a physician, a physician's assistant, or other authorized medical representative) or other authorized person, in communication with the remote server”);
OR
generating the data by detecting the physiological parameter (paragraph 0021 – “In some embodiments, generating the ECG measurements includes detecting, from the number of ECG signals, a number of ECG features including one or more of a set of R peaks, a set of P peaks, a set of T peaks, or a set of QRS complexes”).
Re. claim 21, Masuda further teaches wherein the first segment and the second segment each have a length that is greater than or equal to 5 seconds and less than or equal to 30 seconds (paragraph 0065 – “The ECG image matrix format is advantageous in providing a comparison mechanism that easily highlights to an end user substantial changes in timing and/or intensity of ECG fiducial points, such as P, Q, R, S, T, U, V values, and/or other ECG parameters and characteristics, such as QP, QR, ST, TU segment changes, among others. For example, such information contained in ECG signals of a predetermined duration can be depicted in the ECG image matrix. For instance, such predetermined duration can be of 30 seconds, 60 seconds, 90 seconds, 180 seconds, 300 seconds, or values therebetween, in some examples, to transform the ECG measurements into an ECG image matrix, ECG measurements of a predetermined duration are generated from the ECG signals. The ECG measurements may be segmented into adjacent cardiac portion”).
Re. claim 22, Masuda further teaches in response to determining that the second label is accurate, outputting the second segment without outputting the first segment (figures 3A-3B, ECG segments are made in figure 3A, steps 312-316 and are input in machine learning classifiers in steps 318 and 320, where figure 3B, step 324 outputs when an arrhythmia is determined, and goes back to step figure 3A, step 302 when arrhythmia is not determined).
PNG
media_image3.png
744
524
media_image3.png
Greyscale
PNG
media_image4.png
676
542
media_image4.png
Greyscale
Re. claim 27, Masuda teaches a system (abstract – “Systems and methods are provided for monitoring progression of a cardiac disease in a patient by providing cardio-vibrational image matrixes and/or ECG image matrices generated using sensor data supplied by a medical device”) comprising: at least one processor configured to perform the method of claim 11 (paragraph 0078 – “In some examples, the method 100 may be performed by processing circuitry of a medical device such as a wearable medical device, by one or more processors of a server or server system, or by one or more processors of a cloud computing platform”).
Re. claim 28, Masuda further teaches a wearable device (paragraph 0197 – “For example, the wearable medical device can be configured to be worn by a patient for as many as twenty-four hours a day”; paragraph 0055 – “FIGS. 11A-11D illustrate example wearable medical devices for monitoring a cardiac condition of a patient”),
PNG
media_image2.png
420
410
media_image2.png
Greyscale
configured to generate physiological parameter data by detecting the physiological parameter of the subject wearing the wearable device at a sampling rate over the time period (paragraph 0019 – “The method may include obtaining, in real-time by processing circuitry, a number of ECG signals of a patient from at least two ECG electrodes of a wearable medical device…”; paragraph 0082 – “The set of the cardio-vibrational measurements may represent a selected period of time (e.g., predetermined duration) such as, in some examples, 15 seconds, 30 seconds, 45 seconds, or 60 seconds to be real-real-time responsive to cardiac conditions occurring in the heart of the patient such as arrhythmia…”).
Re. claim 29, Masuda further teaches wherein the time period is greater than one minute and the first segment OR the second segment corresponds to a time interval of less than 30 seconds (paragraph 0019 – “The method may include obtaining, in real-time by processing circuitry, a number of ECG signals of a patient from at least two ECG electrodes of a wearable medical device…”; paragraph 0082 – “The set of the cardio-vibrational measurements may represent a selected period of time (e.g., predetermined duration) such as, in some examples, 15 seconds, 30 seconds, 45 seconds, or 60 seconds to be real-real-time responsive to cardiac conditions occurring in the heart of the patient such as arrhythmia…”);
wherein the physiological parameter comprises an electrocardiogram (ECG), a plethysmograph, a capnograph, an electroencephalogram (EEG), or an electromyograph (EMG) (paragraph 0019 – “The method may include obtaining, in real-time by processing circuitry, a number of ECG signals of a patient from at least two ECG electrodes of a wearable medical device…”);
wherein the trained ML model comprises a convolutional neural network (CNN) (paragraph 0133 – “In some embodiments, at least a portion of the arrhythmia machine learning engine(s) 802 include one or more convolution neural network (CCN) models configured to apply at least a portion of the classifiers 804”);
OR
wherein the at least one processor is further configured to output the one or more correctly classified segments (paragraph 0089 – “The cardio-vibrational image matrix, for example, may be provided to a display of a computing device for review by a clinician or incorporated in a report for review by an end user. For example, the image may be displayed via a user interface mounted on an external housing of the wearable medical device. In implementations, the image may be transmitted via a network interface in the wearable medical device (described in further detail below) to a remote server, for display at a remote computer screen viewed by a technician, a caregiver (e.g., a nurse, a physician, a physician's assistant, or other authorized medical representative) or other authorized person, in communication with the remote server”).
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.
Claim(s) 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Masuda (US 20220028060 A1 – hereinafter Masuda) in view of Kalidas (US 20220015711 A1 – hereinafter Kalidas).
Re. claim 14, Masuda teaches the claimed method of claim 11, including the first ECG segment as stated above, but does not expressly teach wherein the first segment comprises at least one artifact associated with motion of the subject when the measurements of the physiological parameter were obtained.
Kalidas teaches a similar system for sensing and treating atrial fibrillation (paragraph 0041 – “This step is followed by atrial fibrillation detection (210) …”),
and further teaches dividing an ECG signal into segments which includes at least one artifact associated with motion of the subject (paragraph 0041 – “An exemplary method (200) initiates with acquiring (201) an ECG signal and prior to performing any arrhythmia analysis, the incoming ECG signal is preprocessed (202) to remove low frequency and high frequency artifacts using Stationary Wavelet Transforms and Denoising Convolutional Autoencoders. This initial step is complemented by a signal quality assessment (203) using Convolutional Neural Networks where ECG segments corrupted by high grade motion artifacts are identified and suppressed from further arrhythmia analysis”; paragraph 0053 – “First, after acquiring or obtaining the ECG data, the data is processed to remove noise and low-quality segments, as indicated by denoising or noise removal step/stage 202 and signal quality analysis step/stage 203 of FIG. 2”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the ECG sensing as taught by Masuda, to incorporate the ECG motion artifact sensing as taught by Kalidas, since such modification would predictably result in, for example, better quality of arrythmia classification and reduce arrhythmia misclassifications (Kalidas’s paragraph 0065).
Claim(s) 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over Masuda (US 20220028060 A1 – hereinafter Masuda) in view of Clifton (US 20220249031 A1 – hereinafter Clifton).
Re. claim 15, Masuda teaches the method of claim 11 as stated above, but does not expressly teach wherein generating the first image that is indicative of the first segment comprises generating a first spectrogram of the first segment, and wherein generating the second image that is indicative of the second segment comprises generating a second spectrogram of the second segment.
Clifton teaches a system for measuring ECG signals using multiple ECG leads (paragraph 0007 – “Applying the machine-learning algorithm to the combined data from multiple ECG leads means that correlations between the data from different leads can be taken advantage of to improve classification of the ECG data of the patient”; paragraph 0117 – “The ECG signals for each patient contain the standard 12 leads, which are I, II, III, V1, V2, V3, V4, V5, V6, aVF, aVL, and aVR”),
and further teaches wherein generating the first image that is indicative of the first segment comprises generating a first spectrogram of the first segment, and wherein generating the second image that is indicative of the second segment comprises generating a second spectrogram of the second segment (paragraph 0118 – “For each ECG lead of a patient, a spectrogram was computed for a segment of 10 second window without overlap between successive windows, using the short time Fourier transform, with a Hamming window of 1 second and 95% overlaps”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the segment images of Masuda, to incorporate the spectrogram segment images as taught by Clifton, since such modification would predictably result in more robust ECG data analysis for arrhythmia identification (Clifton paragraph 0020 – “Analysis using spectrogram data has been shown to be more robust to variations in how the data is collected from a patient, such as changes in electrode position or sampling rate”).
Claim(s) 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Masuda (US 20220028060 A1 – hereinafter Masuda) in view of Mwikirize (US 20210330289 A1 – hereinafter Mwikirize).
Re. claim 16, Masuda further teaches wherein the trained ML model comprises a convolutional neural network (CNN) (paragraph 0133 – “In some embodiments, at least a portion of the arrhythmia machine learning engine(s) 802 include one or more convolution neural network (CCN) models configured to apply at least a portion of the classifiers 804”).
Masuda does not expressly teach the CNN comprises comprises:
a first block comprising a first convolutional layer, a first batch normalization layer, a first rectified linear unit (ReLu) activation layer, and a first max pooling layer, and a softmax layer connected in series to the third block.
Mwikirize teaches a similar image analysis system (paragraph 0002 – “The present disclosure relates generally to the field of computer vision technology. More specifically, the present disclosure relates to computer vision systems and methods for real-time localization of needles in ultrasound images”) which similarly uses machine-learning, specifically convolutional neural network (CNN) (paragraph 0017 – “FIG. 8 is a block diagram illustrating another embodiment of the system and method of the present disclosure, wherein a convolutional neural network localizes a needle tip in ultrasound images”).
Mwikirize further teaches the CNN (figure 13, CNN 240; paragraph 0073 – “FIG. 13 is a block diagram illustrating a convolutional neural network 240 capable of performing processing steps for tip classification process 226 of FIG. 11”) which includes a:
convolutional layer (paragraph 0073 – “The network consists of a needle tip enhanced image 242, 6 blocks of convolution…”),
a batch normalization layer (paragraph 0073 – “…6 blocks of convolution, Rectified Linear Unit (ReLU) activation layers, batch normalization and max pooling layers 244, 2 blocks of convolution, ReLU and batch normalization layers 246…”),
a rectified linear unit (ReLu) activation layer (paragraph 0073 – “…6 blocks of convolution, Rectified Linear Unit (ReLU) activation layers, batch normalization and max pooling layers 244, 2 blocks of convolution, ReLU and batch normalization layers 246…”),
and a max pooling layer (paragraph 0073 – “…6 blocks of convolution, Rectified Linear Unit (ReLU) activation layers, batch normalization and max pooling layers 244…The max pooling layers utilize a 2×2 kernel”).
PNG
media_image5.png
470
697
media_image5.png
Greyscale
Mwikirize further teaches a softmax layer (paragraph 0074 – “A Softmax activation is applied to this feature map and the Log Loss is utilized to calculate deviations between the network output and the ground-truth”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the CNN of Masuda, to incorporate the CNN layers as taught by Mwikirize as stated above, since such modification would predictably result in greater efficacy in identifying arrhythmia through ECG image segment processing.
The combined invention of Masuda and Mwikirize (hereinafter the combined invention) teaches the first block CNN comprising the layers as stated above, but does not expressly teach the second and third blocks comprising the layers as stated above, specifically a second block connected in series to the first block, the second block comprising a second convolutional layer, a second batch normalization layer, a second ReLu activation layer, and a second max pooling layer; a third block connected in series to the second block, the third block comprising a third convolutional layer, a third batch normalization layer, and a third ReLu activation layer.
However, it has held by the court that “…mere duplication of parts has no patentable significance unless a new and unexpected result is produced”, see MPEP 2114.04, Duplication of Parts. In the instant case, duplicating the CNN as taught by Mwikirize comprising the convolutional layer, batch normalization layer, rectified linear unit (ReLu) activation layer, and max pooling layer as stated above, into second and third blocks would not change operation of the system/method of processing and analyzing segmented ECG images for identifying conditions. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the CNN of the combined invention, specifically the CNN as taught by Mwikirize, to try duplicating the CNN block into second and third blocks, since such modification would predictably result in greater efficacy in identifying arrhythmia through ECG image segment processing.
The combined invention teaches the obvious first, second and third CNN blocks as stated above, AS WELL AS the softmax layer as stated above, but does not expressly teach the second block connected in series to the first block, the third block connected in series to the second block the softmax layer connected in series to the third block.
Upon analysis of the present invention, the claims recite previously known elements. The combined invention, specifically Mwikirize, is considered to recite each and every claimed element of the present invention, including the obvious first-third CNN blocks comprising the convolutional layer, batch normalization layer, ReLu activation layer, and max pooling layer, as well as the softmax layer. However, despite the recitation of the essential working elements as claimed, a difference is found with respect to the layout and arrangement of three CNN blocks and the softmax layer within the CNN. The differences in the layout and arrangement of these essential working elements are not considered to comprise a critical advantage to the layout of elements as disclosed by Mwikirize of the combined invention and thus is not considered to constitute a patentable difference. Since Mwikirize of the combined invention demonstrates the capability of housing each of the working elements to fit within the CNN, then the architecture in which the elements are arranged is not deemed critical. This difference, wherein the only difference being the arrangement and location of essential working elements, has been held by the reviewing courts as being obvious to one of ordinary skill in the art since it is motivated by choice in design and routine skill, in re Japikse, 181 F.2d 1019, 86 USPQ 70 (CCPA 1950), see MPEP 2144.04. A skilled designer is considered to recognize the claimed arrangement of elements as one of many possibilities for the layout of elements within the limited space created by the CNN. The location of each element is further motivated by choice in design dictated by the limited available space.
Claim(s) 23 is/are rejected under 35 U.S.C. 103 as being unpatentable over Masuda (US 20220028060 A1 – hereinafter Masuda) in view of Chapman (US 20160331987 A1 – hereinafter Chapman).
Re. claim 23, Masuda further teaches wherein identifying the data indicative of the measurements of the physiological parameter of the subject comprises detecting the ECG by a Holter monitor worn by the subject (paragraph 0011 – “In some embodiments, the medical device includes a wearable cardiac monitoring device. The wearable cardiac medical device may include a cardiac holter monitor and associated number of ECG electrodes”).
Masuda does not expressly teach wherein the ECG comprises twelve leads or fewer than twelve leads.
Clifton teaches a similar wearable stimulation system for measuring ECG signals (abstract – “A wearable cardioverter defibrillator (“WCD”) system includes a support structure that can be worn by a patient, and a defibrillator coupled to the support structure. An ECG input, rendered from an ECG of the patient, may meet a primary shock criterion”),
PNG
media_image6.png
564
414
media_image6.png
Greyscale
And further teaches wherein the ECG comprises twelve leads OR fewer than twelve leads (paragraph 0052 – “Defibrillator 300 may optionally also have an ECG port 319 in housing 301, for plugging in ECG electrodes 309, which are also known as ECG leads…ECG electrodes 309 can help sense an ECG signal, e.g. a 12-lead signal, or a signal from a different number of leads, as long as they make good electrical contact with the body of the patient”).
PNG
media_image7.png
164
172
media_image7.png
Greyscale
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the segment images of Masuda, to incorporate the twelve ECG leads as taught by Chapman, since such modification would predictably result in, for example, receiving ECG signals and locating target cardiac arrhythmia sites via the ECG signals.
Claim(s) 24 is/are rejected under 35 U.S.C. 103 as being unpatentable over Masuda (US 20220028060 A1 – hereinafter Masuda) in view of Bashour (US 20080167567 A1 – hereinafter Bashour).
Re. claim 24, Masuda further teaches wherein determining the first label indicating whether the first segment is indicative of AF comprises:
detecting first beats in the first segment (paragraph 0132 – “Additionally, or alternatively, in implementations, the machine learning engines(s) 802 can receive inputs apart from the image matrices for use in the classification (illustrated in FIG. 8 using dotted arrows). Examples of ECG metrics that can be used to aid in the classification process include average heart rate, minimum heart rate, maximum heart rate…”);
identifying first beat-to-beat intervals defined between the first beats (paragraph 0132 – “Examples of ECG metrics that can be used to aid in the classification process include average heart rate, minimum heart rate, maximum heart rate, average RR interval in milliseconds or seconds, minimum RR interval in milliseconds or seconds, maximum RR interval in milliseconds or seconds, standard deviation of RR intervals in milliseconds or seconds, number of successive RR intervals greater than a predetermined duration (e.g., 45 ms) per minute…”);
and determining that the first segment is indicative of AF based on the first beat-to-beat intervals (RR intervals are used to help classify heart conditions as shown in figure 8, arrhythmia classifiers 804, further detailed in paragraph 0131 – “In some embodiments, the classifiers 804 include a normal heart condition classifier. The normal heart condition classifier, for example, may be trained using historic ECG image matrices and/or historic cardio-vibrational image matrices of the patient, thereby including the unique “fingerprint” of the cardiac cycles of the patient”, and paragraph 0132 – “Additionally, or alternatively, in implementations, the machine learning engines(s) 802 can receive inputs apart from the image matrices for use in the classification (illustrated in FIG. 8 using dotted arrows). Examples of ECG metrics that can be used to aid in the classification process include average heart rate, minimum heart rate, maximum heart rate, average RR interval in milliseconds or seconds, minimum RR interval in milliseconds or seconds, maximum RR interval in milliseconds or seconds, standard deviation of RR intervals in milliseconds or seconds, number of successive RR intervals greater than a predetermined duration (e.g., 45 ms) per minute…”),
and wherein determining the second label indicating whether the second segment is indicative of AF comprises:
detecting second beats in the second segment (paragraph 0132 – “Additionally, or alternatively, in implementations, the machine learning engines(s) 802 can receive inputs apart from the image matrices for use in the classification (illustrated in FIG. 8 using dotted arrows). Examples of ECG metrics that can be used to aid in the classification process include average heart rate, minimum heart rate, maximum heart rate…”; repeat ECG segments can be made as per figures 3A-3B);
identifying second beat-to-beat intervals defined between the second beats (paragraph 0132 – “Examples of ECG metrics that can be used to aid in the classification process include average heart rate, minimum heart rate, maximum heart rate, average RR interval in milliseconds or seconds, minimum RR interval in milliseconds or seconds, maximum RR interval in milliseconds or seconds, standard deviation of RR intervals in milliseconds or seconds, number of successive RR intervals greater than a predetermined duration (e.g., 45 ms) per minute…”; repeat ECG segments can be made as per figures 3A-3B);
and determining that the second segment is indicative of AF based on the second beat-to-beat intervals (RR intervals are used to help classify heart conditions as shown in figure 8, arrhythmia classifiers 804, further detailed in paragraph 0131 – “In some embodiments, the classifiers 804 include a normal heart condition classifier. The normal heart condition classifier, for example, may be trained using historic ECG image matrices and/or historic cardio-vibrational image matrices of the patient, thereby including the unique “fingerprint” of the cardiac cycles of the patient”, and paragraph 0132 – “Additionally, or alternatively, in implementations, the machine learning engines(s) 802 can receive inputs apart from the image matrices for use in the classification (illustrated in FIG. 8 using dotted arrows). Examples of ECG metrics that can be used to aid in the classification process include average heart rate, minimum heart rate, maximum heart rate, average RR interval in milliseconds or seconds, minimum RR interval in milliseconds or seconds, maximum RR interval in milliseconds or seconds, standard deviation of RR intervals in milliseconds or seconds, number of successive RR intervals greater than a predetermined duration (e.g., 45 ms) per minute…”; repeat ECG segments can be made as per figures 3A-3B).
Masuda teaches the first and second ECG signal segments as stated above in claims 11-12 as stated above, but does not expressly teach determining whether a first p wave is present in the first segment; determining that the first segment is indicative of AF based on whether the first p wave is present in the first segment; determining whether a second p wave is present in the second segment; and determining that the first segment is indicative of AF based on determining whether the second p wave is present in the second segment.
Bashour teaches a similar system for classifying atrial fibrillation (abstract – “Systems and methods are provided for predicting the onset of postoperative atrial fibrillation (AF) from electrocardiogram (ECG) data representing a patient”), and further teaches the known technique of collecting segmented ECG signals and determining p-waves in the segment (paragraph 0014 – “In accordance with an aspect of the present invention, computer based algorithms have been developed to collect and segment electrocardiogram (ECG) data and to identify and characterize premature atrial contraction (PAC) activity, heart rate variability (HRV), and P-wave morphology characteristics…”) and determining that the segment is indicative of AF based on determining whether the p wave is present in the segment (paragraph 0022 – “Once the signals have been analyzed, a feature extractor 20 computes features from the detected R-waves, premature atrial contractions (PACs), and P-waves that are useful in predicting postoperative atrial fibrillation”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the first and second ECG signal segments as taught by Masuda, to incorporate the P-wave and atrial fibrillation detection based on the P-wave as taught by Bashour, since such modification would predictably result in identifying atrial fibrillation, and in turn taking mitigating action.
Claim(s) 25 is/are rejected under 35 U.S.C. 103 as being unpatentable over Masuda (US 20220028060 A1 – hereinafter Masuda) in view of Berenfeld (US 20060122526 A1 – hereinafter Berenfeld).
Re. claim 25, Masuda teaches the step of determining the second label indicating whether the second segment is indicative of AF as stated above in claims 11-12, but does not expressly teach detecting a brief episode of AF in the second segment, the brief episode of AF occurring for greater than or equal to 5 seconds and less than or equal to 30 seconds.
Berenfeld teaches a similar atrial fibrillation detection system (paragraph 0003 – “The present invention generally relates to a method and algorithm for spatially identifying sources generative of cardiac fibrillation, and in particular, atrial fibrillation”), and further teaches the known technique of detecting a brief episode of AF in the second segment, the brief episode of AF occurring for greater than or equal to 5 seconds and less than or equal to 30 seconds (paragraph 0059 – “In a patient with fibrillation, the roving cardiac signal acquisition device may, for example, be used to record real-time episodes of atrial fibrillation over an acquisition time T of, for example of 5 seconds”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the ECG label determination of Masuda, to incorporate the 5 second AF episode detection as taught by Berenfeld, since such modification would predictably result in identifying atrial fibrillation, and in turn taking mitigating action.
Claim(s) 26 is/are rejected under 35 U.S.C. 103 as being unpatentable over Masuda (US 20220028060 A1 – hereinafter Masuda) in view of Verzal (US 20190175026 A1 – hereinafter Verzal).
Re. claim 26, Masuda teaches the method of claim 12 as stated above, but does not expressly teach determining an AF burden based on the second label, wherein the indication of whether the subject has the condition comprises the AF burden.
Verzal teaches a similar system for detecting atrial fibrillation (paragraph 0069 – “In some examples, the monitoring comprises determining further information or drawing a conclusion, such as whether a particular parameter may be associated with or at least partially define a condition. For instance, upon monitoring a particular cardiac parameter, the monitoring may determine that a cardiac condition (e.g. atrial fibrillation) is exhibited”),
And further teaches the known technique of determining an AF burden based on the second label (RR intervals are used to determine AF burden, paragraph 0325 – “In some examples, the atrial fibrillation burden can be quantified in at least two ways. For instance, the atrial fibrillation burden can be quantified via RR interval variability (where R refers to the R in a QRS complex of a cardiac waveform) or via atrial-atrial (AA) timing vs ventricle-ventricle (VV) timing”),
wherein the indication of whether the subject has the condition comprises the AF burden (paragraph 0324 – “For instance, if the atrial fibrillation burden persists (e.g. persistence parameter 784 in FIG. 22) even with a high value of SDB efficacy and a high value of SDB therapy patient compliance, this correlation may be an indication of a structural cardiac issue and that the patient may benefit from an interventional cardiac procedure such as, but not limited to, ablation to treat the atrial fibrillation behavior”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Masuda, to incorporate the atrial fibrillation burden determination as taught by Verzal, since such modification would predictably result in tailoring mitigating therapy more precisely to the patient.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Anh-Khoa N. Dinh whose telephone number is (571)272-7041. The examiner can normally be reached Mon-Fri 7:00am-4:00pm 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, CARL LAYNO can be reached at 571-272-4949. 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.
/ANH-KHOA N DINH/Examiner, Art Unit 3796
Prosecution Insights