DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
The amendment filed July 6, 2026 has been entered. Claims 1 and 9-11 remain pending in the application. Applicant’s amendments to the claims have overcome most objections previously set forth in the Non-Final Office Action mailed April 6, 2026. More details provided below.
Response to Arguments
35 USC § 112
Applicant has cancelled claim 12, the previous 112(d) rejection has been withdrawn.
35 USC § 101
Applicant's arguments filed July 6, 2026 regarding 35 USC 101 have been fully considered but they are not persuasive.
On pg. 7 of Applicant’s response, applicant argues that even under the broadest reasonable interpretation the amended claims cannot be interpreted as mental processes performed in a human mind, with the additional limitations of "an electrocardiogram measuring device configured to record electrocardiogram data of a subject through a plurality of leads" and "display a waveform wherein a section of interest in the evaluation is indicated on the waveform." Examiner agrees that the additional limitations are not directed towards mental processes, but towards insignificant extra-solution activity. As recited the “electrocardiogram device” merely acquires data for use in the analysis which amounts to nothing more than insignificant extra solution activity. Likewise, the “display a waveform wherein a section of interest is indicated on the waveform” merely presents the results of an abstract analysis. Additionally, examiner respectfully disagrees that the recited limitation “wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value” are directed towards mental processes. A human, provided ECG data, pen and paper, could apply known mathematical techniques to ECG data for a target disease evaluation by the coefficient satisfying a predetermined value. Therefore, the additional limitations are directed towards mental processes, as well as mathematical calculations, which fall within the ‘mathematical concepts’ grouping of abstract ideas.
On pg. 8 of Applicant's response, applicant argues that the features of "an electrocardiogram measuring device configured to record electrocardiogram data of a subject through a plurality of leads" and "evaluate the electrocardiogram data regarding the target disease of examination based on the selected lead data using a model to which an attention mechanism is added such that output data prior to a full connection layer of a convolutional neural network is inputted to the attention mechanism, and obtain evaluation result" as well as "display a waveform based on the selected lead data on a display device, wherein a section of interest in the evaluation is indicated on the waveform, wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value” are integrated to an evaluation device not to merely perform an alleged abstract idea, but rather, to integrate the limitations into a practical application by providing a technical solution for improving the accuracy of electrocardiograms. Examiner respectfully disagree as there is nothing in the applicant’s specification that recites a structurally technologically improved computer system as the claims are directed to the use of a model having an attention mechanism to analyze ECG lead, and identifying a section of interest based on a coefficient being equal to greater than a predetermined model a model applying gathered data. These added limitations could be done by a human performing mathematical calculations from the selected lead data to identify a section of interest, by being equal to or greater than a predetermined value, and applying that knowledge to determine the presence or absence of a particular disease of interest.
35 USC § 102
Applicant’s arguments, see pg. 11, filed July 6, 2026, with respect to the rejection(s) of claims 9-10 under 102(a)(1) have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Yu et al. (US 20060173368 A1) and Charles et al. (US 20190374166 A1).
35 USC § 103
Applicant’s arguments, see pg. 9-11, filed July 6, 2026, with respect to the rejection(s) of claims 1 and 11 under 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Yu et al. (US 20060173368 A1) and Charles et al. (US 20190374166 A1).
Claim Rejections - 35 USC § 112
Claims 1 and 9-10 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claims 1 and 9-10 recite the limitation "output data" in line 10. There is insufficient antecedent basis for this limitation in the claim.
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 1 and 9-11 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Independent claims 1, 9, and 10 recite an electrocardiogram evaluation device, a method, and a computer storage medium. Thus, they are directed to statutory categories of invention.
Step 2A, prong 1:
Claims 1, 9, and 10 recite the following claim limitations:
acquire electrocardiogram data regarding an electrocardiogram of a subject;
select, from the electrocardiogram data, lead data of one or more leads corresponding to a target disease of examinations;
evaluate the electrocardiogram regarding the target disease of examination based on the selected lead data
section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value
These limitations under their broadest reasonable interpretation, cover concepts that can be practically preformed in the human mind. A human, provided ECG lead data, could observe and evaluate the ECG waveform to identify and detect a target disease. For example, these limitations are nothing more than a medical professional, selecting specific ECG leads known to correspond to specific disease conditions, analyzing the waveform data from these leads, and determining the presence or absence of a target disease. Applicant's specification further discloses that certain lead positions correspond to particular diseases (e.g. I lead, Il lead, and V1 lead correlating with atrial fibrillation (see paragraph [0023])). A medical professional could select and evaluate specific lead data to assess different target diseases. Thus, the claims recite limitations which fall within the 'mental processes' grouping of abstract ideas.
Additionally, the limitation of “section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value” recite performing mathematical calculations (comparing data to a threshold), and as such, the claims recite limitations which also fall within the ‘mathematical concepts’ grouping of abstract ideas.
Step 2A, prong 2:
Claims 1, 9, and 10 recite the following additional elements:
an electrocardiogram measuring device configured to record electrocardiogram data of a subject through a plurality of leads;
at least one memory configured to store instructions; (claim 1)
at least one processor configured to execute the instructions; (claim 1)
a model to which an attention mechanism is added such that output data prior to a full connection layer of a convolutional neural network is inputted to the attention mechanism, and obtain an evaluation result
display a waveform based on the selected lead data on a display device wherein a section of interest in the evaluation is indicated on the waveform
storage medium storing a program that, when executed by a computer, implement a method (claim 10)
The recited “electrocardiogram measuring device” amounts to nothing more than insignificant extra solution activity for data gathering.
Claim 1's recitation of "at least one processor configured to execute the instructions" fails to recite any additional element or combination of additional elements that apply, rely on, or use the judicial exception in a manner that imposes a meaningful limitation on the judicial exception. As recited the processor, which upon review of the applicant's specification is an example of a computer (see paragraph [18]) is a conventional component that does not impose any meaningful structural limitations on the apparatus used to implement the judicial exception. The recitation of a processor in the claim does not integrate the judicial exception into a practical application because the claim merely uses the processor as a tool to perform the abstract idea.
The additional elements of storing the instructions in memory and/or displaying the section of a interest of the selected lead data in waveform on a display device, amount to nothing more than functions performed by a generic computer. Merely storing instructions/data in a memory and displaying results on a display device does not integrate a judicial exception into practical application.
The additional element of “a model to which an attention mechanism is added such that output data prior to a full connection layer of a convolutional neural network is inputted to the attention mechanism, and obtain an evaluation result” which as above, is a processor in conjunction with an algorithm. The limitations are merely used as tools to perform the abstract idea.
Claim 10's recitation of a storage medium with computer-usable instructions that, when executed by a computer, implement a method are merely reciting the storage medium at a high-level of generality, and the storage medium merely instructs the computer to carry out the steps of the method. In other words, the computer components are being used as a tool to carry out the method (See MPEP 2106.05(f)).
Thus, the abstract idea is not integrated into a practical application. The combination of these additional elements is no more than insignificant extra solution activity, and mere instructions to apply the exception using generic computer components (the processors and computer readable storage media). Accordingly, even in combination, these additional elements do not integrate the abstract idea into a practical application. The claim is directed to an abstract idea.
Step 2B:
As discussed above with respect to Step 2A Prong 2, the additional elements of storing the instructions in memory and/or displaying the section of interest of the selected lead data in waveform on a display device and a storage medium in the claim, amount to no more than insignificant extra solution activity and mere instructions to apply the exception using a generic computer component. The same analysis applies here in 2B and does not provide an inventive concept.
The recitation of an electrocardiogram measuring device is not sufficient to amount to significantly more than the judicial exception because they are recited at a high level of generality, there is no meaningful limitation, such as a particular or unconventional structure that distinguishes the elements from well-known, routine, and/or conventional elements. Recitation of an electrocardiogram measuring device as a tool to perform the abstract idea does not add significantly more than what is well-known, routine, and/or conventional in view of Krusor et al. (US 20210096712 A1), see para. 0002, (An electrocardiogram (ECG) is a plot of the electrical activity of a heart over time. The electrical activity of an individual is identified, for instance, by measuring relative voltages between various electrodes placed on the body of the individual), and Xia et al. (US 20210338135 A1) see para. 0046 (An ECG can be obtained from multiple lead electrodes, which are attached to the surface of a patient's body to record, and provide spatial information about, the heart's electrical activity).
The recitation of a processor is not sufficient to amount to significantly more than the judicial exception because they are recited at a high level of generality, there is no meaningful limitation, such as a particular or unconventional structure that distinguishes the elements from well-known, routine, and/or conventional elements. Recitation of a processor as a tool to perform the abstract idea does not add significantly more than what is well-known, routine, and/or conventional in view of Alice Corp. Pty. Ltd. V. CLS Bank Int'l, 573 U.S. 208, 223, 110 USPQ2d 1976, 1983 (2014).
For these reasons, there is no inventive concept. The claim is not patent eligible. Even when viewed as a whole, nothing in the claim adds significantly more to the abstract idea.
Dependent claims
Claim 11 adds the additional element of "machine learning" which as above, a processor in conjunction with an algorithm. Recitation of a generic processor as a tool to perform the abstract idea does not add significantly more than what is well-known, routine, and/or conventional in view of Alice Corp. Pty. Ltd. V. CLS Bank Int'l, 573 U.S. 208, 223, 110 USPQ2d 1976, 1983 (2014). Even when viewed as whole in combination with independent claim 1, the machine learning in claim 11 fails to add significantly more to the abstract idea.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1 and 9-11 are rejected under 35 U.S.C. 103 as being unpatentable over Sasano et al. (JP 2021109049 A1), and in view of Yu et al. (US 20060173368 A1), hereinafter Yu, and in further view of Krusor et al. (US 20210096712A1), hereinafter Krusor, and Charles et al. (US 20190374166 A1), hereinafter Charles.
Regarding Claim 1, Sasano teaches an electrocardiogram evaluation device comprising (Fig. 1, element 100):
an electrocardiogram measuring device configured to record electrocardiogram data of a subject through a plurality of leads (Fig. 2, element S201, para. 0014 (device acquires electrocardiograms), 0015 (plurality of leads));
at least one memory configured to store instructions (Fig. 1, element 102, para. 0010 (stores programs and data)); and
at least one processor configured to execute the instructions (Fig. 1, element 101, para. 0010 (controls the device)) to:
acquire electrocardiogram data regarding an electrocardiogram of the subject (Fig. 2, element S201, para. 0014 (acquires electrocardiograms));
evaluate the electrocardiogram data regarding the target disease of examination based on the selected lead data (Fig. 2, element S203, para. 0016 (presence or absence of the target disease), 0018 (determination result by the model based on the respective leads), 0025 (each of the small models 601a to 601l determines whether or not the subject has the target disease based on the waveform of the individual lead included in the ECG of the subject)) using a model (Fig. 6, element 600) to which an attention mechanism is added such that output data prior to a full connection layer of a convolutional neural network is inputted to the attention mechanism, and obtain an evaluation result (para. 0016 (device 100 determines a model for determining the presence or absence of the target disease for each lead by performing the mechanical learning by individually using the waveform of each of the plurality of leads included in the ECG of the learning data. This machine learning may be performed, for example, using a convolutional neural network, CNN), 0026 (synthesizer outputs a determination result obtained by synthesizing the determination results by the plurality of small models 601a to 601l… may combine a plurality of determination results by majority decision), see annotated elements in Fig. 1 below)).
Sasano does not teach selecting from the electrocardiogram data, lead data of a subset of the plurality of leads corresponding to a target disease of examination, displaying a waveform based on the selected lead data on a display device wherein a section of interest in the evaluation is indicated on the waveform, wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value.
Yu teaches selecting from the electrocardiogram data (abstract (A system for detecting heart diseases, especially the coronary artery diseases, comprises the steps of obtaining twelve (12) lead cardiac electrical signals from a patient), Fig. 2, element 20), lead data of a subset of the plurality of leads corresponding (Fig. 2, element 214, para. 0022 (analyze the frequency components of the cardiac electrical signals from lead II and lead V5)) to a target disease of examination (para. 0022 (The reason of selecting these two leads is because the cardiac electrical signals detected by these two lead travel through the frontal left ventricular area of a heart where coronary diseases would cause more serious consequences than that of other part of the heart)).
Yu does not teach displaying a waveform based on the selected lead data on a display device wherein a section of interest in the evaluation is indicated on the waveform, wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value.
Krusor teaches an electrocardiogram evaluation device comprising (Fig. 2, element 200):
an electrocardiogram measuring device configured to record electrocardiogram data of a subject through a plurality of leads (Fig. 1, element 114, para. 0038 (Sensor electrodes 110 are coupled to the patient 102… receive an electrical signal output by the heart of the patient 102. Sensor leads 112 electrically couple the sensor electrodes));
at least one memory configured to store instructions (Fig. 2, element 252, para. 0069 (memory stores instructions of the device)); and
at least one processor configured to execute the instructions (Fig. 2, element 251, para. 0069 (processor executes the instructions)) to:
acquire electrocardiogram data regarding an electrocardiogram of the subject (Fig. 2, element 232, para. 0057 (ECG signal));
display a waveform based on the selected lead data on a display device (Fig. 3, element 302) wherein a section of interest in the evaluation is indicated on the waveform (Fig. 3, element 320 (para. 0071 (individual lead display), Fig. 3, element 330 (para. 0071 (selected lead display)).
Krusor does not teach wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value.
Charles teaches using a model to which an attention mechanism is added such that output data prior to a full connection layer of a convolutional neural network (para. 0093 (neural network may be a convolutional neural network), 0099 (neural network comprises an input layer 201 followed by n superlayers… each superlayer 210 comprises a convolution layer 211 and activation layer 212)) is inputted to the attention mechanism, and obtain an evaluation result (para. 0094 (location of each of the points identified by the neural network may be used to classify each beat of the ECG as “AF” or “non-AF”), 0095 (the beat classification may provide an estimate of the probability for each beat of the ECG data being “AF” and “not-AF”. Since these example classifications are mutually exclusive, the sum of their probabilities may be equal to 1)); and
display a waveform wherein a section of interest in the evaluation is indicated on the waveform (Figs. 7-8),
wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value (para. 0098 (beats or regions classified as “AF” with a probability of greater than a predetermined threshold (e.g. 0.8) may be flagged or highlighted for further review by a cardiologist. A duration based threshold may also be used in determining regions of interest (e.g. AF with a probability of >0.8, over a time period of at least 2 ECG beats)).
Sasano, Yu, Krusor, and Charles are all considered to be analogous to the claimed invention because they are in the same field of analyzing ECG signals. Sasano teaches selecting and evaluating all 12 ECG leads to determine the absence of presence of a target disease (para. 0016). Yu teaches selecting only a subset of leads from the 12 ECG signals, that have the most detection on a coronary disease (para. 0022). Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Sasano to incorporate the teachings of Yu listed, to select a subset of the plurality of leads, to correspond to a target disease of examination. Doing so would optimize the device by solely focusing on a subset of leads related to a target disease, to improve the accuracy of the evaluation result, as opposed to including the weight of all the leads which may not have an influence on the target disease.
Sasano teaches a CNN that inputs the waveforms of the individual leads, to detect a target disease and outputs the judgement result. Based on majority vote, a final judgement result is reached (para. 0025). Charles teaches a CNN model that detects a section of interest, and defines the section of interest as a coefficient output by the attention mechanism being equal to or larger than a predetermined value. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Sasano’s CNN model to incorporate the teachings of Charles and provide a section of interest identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value. Doing so improves the model output by reducing the likelihood of possible artifacts/noise, that could give false weight on whether a lead has the presence or absence of a target disease. Additionally, having a section of interest highlight’s relevant signal features, which improves the accuracy of the overall model.
Sasano discloses an output device (Fig. 1, 104, para. 0011 (display)), but displays only the probability that the subject has the target disease (para. 0017). Krusor discloses a display device that includes the waveforms of the lead data. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Sasano to incorporate the teachings of Krusor and provide a display device showing the lead data waveforms. Doing so would allow a health professional to properly look at the waveforms from the corresponding lead data, to increase the accuracy of a target disease diagnosis.
Furthermore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the modified Sasano’s CNN model in view of Charles, and combine the teachings taught by Krusor, to display the identified section of interest on the lead waveform display taught by Krusor. Doing so allows a clinician to review the section of interest of the individual leads to evaluate the model’s output to determine the presence or absence of a target disease.
PNG
media_image1.png
872
554
media_image1.png
Greyscale
Fig. 1 (from Sasano)
Regarding claim 9, Sasano teaches an electrocardiogram evaluation method (Figures 2 and 3) executed by a computer (Fig. 1, element 105, para. 0011 (device communicates with a computer), the electrocardiogram evaluation method comprising (para. 0001 (method of controlling the device)):
acquiring electrocardiogram data regarding an electrocardiogram of a subject from an electrocardiogram measuring device configured to record electrocardiogram data through a plurality of leads (Fig. 2, element S201, para. 0014 (device acquires electrocardiograms), 0015 (plurality of leads));
evaluate the electrocardiogram data regarding the target disease of examination based on the selected lead data (Fig. 2, element S203, para. 0016 (presence or absence of the target disease for each lead), 0018 (determination result by the model based on the respective leads), 0025 (each of the small models 601a to 601l determines whether or not the subject has the target disease based on the waveform of the individual lead included in the ECG of the subject)) using a model (Fig. 6, element 600) to which an attention mechanism is added such that output data prior to a full connection layer of a convolutional neural network is inputted to the attention mechanism, and obtain an evaluation result (para. 0016 (device 100 determines a model for determining the presence or absence of the target disease for each lead by performing the mechanical learning by individually using the waveform of each of the plurality of leads included in the ECG of the learning data. This machine learning may be performed, for example, using a convolutional neural network), 0026 (synthesizer outputs a determination result obtained by synthesizing the determination results by the plurality of small models 601a to 601l… may combine a plurality of determination results by majority decision), see annotated elements in Fig. 1 above)).
Sasano does not teach selecting, from the electrocardiogram data, lead data of a subset of the plurality of leads, displaying a waveform based on the selected lead data on a display device wherein a section of interest in the evaluation is indicated on the waveform, wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value.
Yu teaches selecting from the electrocardiogram data (abstract (A method for detecting heart diseases, especially the coronary artery diseases, comprises the steps of obtaining twelve (12) lead cardiac electrical signals from a patient), Fig. 2, element 20), lead data of a subset of the plurality of leads corresponding (Fig. 2, element 214, para. 0022 (analyze the frequency components of the cardiac electrical signals from lead II and lead V5)) to a target disease of examination (para. 0022 (The reason of selecting these two leads is because the cardiac electrical signals detected by these two lead travel through the frontal left ventricular area of a heart where coronary diseases would cause more serious consequences than that of other part of the heart)).
Yu does not teach displaying a waveform based on the selected lead data on a display device wherein a section of interest in the evaluation is indicated on the waveform, wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value.
Krusor teaches displaying a waveform based on the selected lead data on a display device (Fig. 3, element 302) wherein a section of interest in the evaluation is indicated on the waveform (Fig. 3, element 320 (para. 0071 (individual lead display), Fig. 3, element 330 (para. 0071 (selected lead display)).
Krusor does not teach wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value.
Charles teaches using a model to which an attention mechanism is added such that output data prior to a full connection layer of a convolutional neural network (para. 0093 (neural network may be a convolutional neural network), 0099 (neural network comprises an input layer 201 followed by n superlayers… each superlayer 210 comprises a convolution layer 211 and activation layer 212)) is inputted to the attention mechanism, and obtain an evaluation result (para. 0094 (location of each of the points identified by the neural network may be used to classify each beat of the ECG as “AF” or “non-AF”), 0095 (the beat classification may provide an estimate of the probability for each beat of the ECG data being “AF” and “not-AF”. Since these example classifications are mutually exclusive, the sum of their probabilities may be equal to 1)); and
display a waveform wherein a section of interest in the evaluation is indicated on the waveform (Fig.s 7-8),
wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value (para. 0098 (beats or regions classified as “AF” with a probability of greater than a predetermined threshold (e.g. 0.8) may be flagged or highlighted for further review by a cardiologist. A duration based threshold may also be used in determining regions of interest (e.g. AF with a probability of >0.8, over a time period of at least 2 ECG beats)).
Sasano teaches selecting and evaluating all 12 ECG leads to determine the absence of presence of a target disease (para. 0016). Yu teaches selecting only a subset of leads from the 12 ECG signals, that have the most detection on a coronary disease (para. 0022). Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Sasano to incorporate the teachings of Yu listed, to select a subset of the plurality of leads, to correspond to a target disease of examination. Doing so would optimize the device by solely focusing on a subset of leads related to a target disease, to improve the accuracy of the evaluation result, as opposed to including the weight of all the leads which may not have an influence on the target disease.
Sasano teaches a CNN that inputs the waveforms of the individual leads, to detect a target disease and outputs the judgement result. Based on majority vote, a final judgement result is reached (para. 0025). Charles teaches a CNN model that detects a section of interest, and defines the section of interest as a coefficient output by the attention mechanism being equal to or larger than a predetermined value. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Sasano’s CNN model to incorporate the teachings of Charles and provide a section of interest identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value. Doing so improves the model output by reducing the likelihood of possible artifacts/noise, that could give false weight on whether a lead has the presence or absence of a target disease. Additionally, having a section of interest highlight’s relevant signal features, which improves the accuracy of the overall model.
Sasano discloses an output device (Fig. 1, 104, para. 0011 (display)), but displays only the probability that the subject has the target disease (para. 0017). Krusor discloses a display device that includes the waveforms of the lead data. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Sasano to incorporate the teachings of Krusor and provide a display device showing the lead data waveforms. Doing so would allow a health professional to properly look at the waveforms from the corresponding lead data, to increase the accuracy of a target disease diagnosis.
Furthermore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the modified Sasano’s CNN model in view of Charles, and combine the teachings taught by Krusor, to display the identified section of interest on the lead waveform display taught by Krusor. Doing so allows a clinician to review the section of interest of the individual leads to evaluate the model’s output to determine the presence or absence of a target disease.
Regarding Claim 10, Sasano teaches a non-transitory storage medium (Fig. 1, element 106) storing a program executed by a computer (Fig. 1, element 105, para. 0011 (device communicates with a computer), the program causing the computer to (para. 0012 (storage medium, stores data used for the operation of the device)):
acquire electrocardiogram data regarding an electrocardiogram of a subject from an electrocardiogram measuring device configured to record electrocardiogram data through a plurality of leads (Fig. 2, element S201, para. 0014 (device acquires electrocardiograms), 0015 (plurality of leads));
evaluate the electrocardiogram data regarding the target disease of examination based on the selected lead data (Fig. 2, element S203, para. 0016 (presence or absence of the target disease for each lead), 0018 (determination result by the model based on the respective leads), 0025 (each of the small models 601a to 601l determines whether or not the subject has the target disease based on the waveform of the individual lead included in the ECG of the subject)) using a model (Fig. 6, element 600) to which an attention mechanism is added such that output data prior to a full connection layer of a convolutional neural network is inputted to the attention mechanism, and obtain an evaluation result (para. 0016 (device 100 determines a model for determining the presence or absence of the target disease for each lead by performing the mechanical learning by individually using the waveform of each of the plurality of leads included in the ECG of the learning data. This machine learning may be performed, for example, using a convolutional neural network), 0026 (synthesizer outputs a determination result obtained by synthesizing the determination results by the plurality of small models 601a to 601l… may combine a plurality of determination results by majority decision), see annotated elements in Fig. 1 above)).
Sasano does not teach select, from the electrocardiogram data, lead data of a subset of the plurality of leads corresponding to a target disease of examination, displaying a waveform based on the selected lead data on a display device wherein a section of interest in the evaluation is indicated on the waveform, wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value.
Yu teaches selecting from the electrocardiogram data (abstract (A system for detecting heart diseases, especially the coronary artery diseases, comprises the steps of obtaining twelve (12) lead cardiac electrical signals from a patient), Fig. 2, element 20), lead data of a subset of the plurality of leads corresponding (Fig. 2, element 214, para. 0022 (analyze the frequency components of the cardiac electrical signals from lead II and lead V5)) to a target disease of examination (para. 0022 (The reason of selecting these two leads is because the cardiac electrical signals detected by these two lead travel through the frontal left ventricular area of a heart where coronary diseases would cause more serious consequences than that of other part of the heart)).
Yu does not teach displaying a waveform based on the selected lead data on a display device wherein a section of interest in the evaluation is indicated on the waveform, wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value.
Krusor teaches a non-transitory storage medium (para. 0137 (elements stored in the memory may be non-transitory); and
display a waveform based on the selected lead data on a display device (Fig. 3, element 302) wherein a section of interest in the evaluation is indicated on the waveform (Fig. 3, element 320 (para. 0071 (individual lead display), Fig. 3, element 330 (para. 0071 (selected lead display)).
Krusor does not teach wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value.
Charles teaches using a model to which an attention mechanism is added such that output data prior to a full connection layer of a convolutional neural network (para. 0093 (neural network may be a convolutional neural network), 0099 (neural network comprises an input layer 201 followed by n superlayers… each superlayer 210 comprises a convolution layer 211 and activation layer 212)) is inputted to the attention mechanism, and obtain an evaluation result (para. 0094 (location of each of the points identified by the neural network may be used to classify each beat of the ECG as “AF” or “non-AF”), 0095 (the beat classification may provide an estimate of the probability for each beat of the ECG data being “AF” and “not-AF”. Since these example classifications are mutually exclusive, the sum of their probabilities may be equal to 1)); and
display a waveform wherein a section of interest in the evaluation is indicated on the waveform (Fig.s 7-8),
wherein the section of interest is identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value (para. 0098 (beats or regions classified as “AF” with a probability of greater than a predetermined threshold (e.g. 0.8) may be flagged or highlighted for further review by a cardiologist. A duration based threshold may also be used in determining regions of interest (e.g. AF with a probability of >0.8, over a time period of at least 2 ECG beats)).
Sasano teaches selecting and evaluating all 12 ECG leads to determine the absence of presence of a target disease (para. 0016). Yu teaches selecting only a subset of leads from the 12 ECG signals, that have the most detection on a coronary disease (para. 0022). Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Sasano to incorporate the teachings of Yu listed, to select a subset of the plurality of leads, to correspond to a target disease of examination. Doing so would optimize the device by solely focusing on a subset of leads related to a target disease, to improve the accuracy of the evaluation result, as opposed to including the weight of all the leads which may not have an influence on the target disease.
Sasano teaches a CNN that inputs the waveforms of the individual leads, to detect a target disease and outputs the judgement result. Based on majority vote, a final judgement result is reached (para. 0025). Charles teaches a CNN model that detects a section of interest, and defines the section of interest as a coefficient output by the attention mechanism being equal to or larger than a predetermined value. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Sasano’s CNN model to incorporate the teachings of Charles and provide a section of interest identified as a section in which a coefficient output by the attention mechanism is equal to or larger than a predetermined value. Doing so improves the model output by reducing the likelihood of possible artifacts/noise, that could give false weight on whether a lead has the presence or absence of a target disease. Additionally, having a section of interest highlight’s relevant signal features, which improves the accuracy of the overall model.
Sasano discloses an output device (Fig. 1, 104, para. 0011 (display)), but displays only the probability that the subject has the target disease (para. 0017). Krusor discloses a display device that includes the waveforms of the lead data. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified Sasano to incorporate the teachings of Krusor and provide a display device showing the lead data waveforms. Doing so would allow a health professional to properly look at the waveforms from the corresponding lead data, to increase the accuracy of a target disease diagnosis.
Furthermore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the modified Sasano’s CNN model in view of Charles, and combine the teachings taught by Krusor, to display the identified section of interest on the lead waveform display taught by Krusor. Doing so allows a clinician to review the section of interest of the individual leads to evaluate the model’s output to determine the presence or absence of a target disease.
Regarding Claim 11, Sasano (in view of Yu, Krusor, and Charles) teaches an electrocardiogram evaluation device comprising the features of claim 1 as discussed above.
Sasano further discloses wherein the at least one processor (Fig. 1, element 101) is configured to execute the instructions to evaluate the electrocardiogram data using the model (Fig. 5), wherein the model is configured to use machine learning (Fig. 5, element S502, para. 0024 (machine-learning using the waveform datasets of the leads)) to output an evaluation result regarding a presence or absence of the target disease of examination when the lead data is inputted to the model (Fig. 6, element 601a 601I, para. 0025 (model determines the target disease based on the waveform of the individual lead included in the ECG of the subject), see annotated elements in Fig. 1 above)).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant' s disclosure. Chen et al. (US 20230293079 A1) and Xia et al. (US 20210338135 A1) are additional examples of a CNN applying weight to abnormal waveforms.
Applicant’s amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to EILEEN ROBLES whose telephone number is (571)429-9383. The examiner can normally be reached Monday-Friday: 8:00 - 5:00 PM.
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, Niketa Patel can be reached at (571) 272-4156. 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.
/EILEEN ROBLES/Examiner, Art Unit 3792
/William J Levicky/Primary Examiner, Art Unit 3796
Prosecution Insights