DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
I. Claim Rejections under 35 U.S.C. § 101
Applicant’s remarks concerning the previous § 101 rejections have been fully considered but are not persuasive. Applicant makes two arguments: (1) the obtaining and determining steps “cannot be performed by the human mind or with pen and paper,” and (2) the recitation of the automated identification of a change in the patient’s cardiac condition “provides a clear improvement in the functioning of a computer or other technology …” The Examiner respectfully disagrees.
Concerning argument (1), Applicant’s remarks fail to provide any explanation as to why these steps could not be carried out by the human mind or with pen and paper. The Examiner respectfully asserts that the human mind is fully capable of reviewing and understanding two types of physiological data and then performing calculations and determinations based on that data. For instance, a diagnostician could mentally review ECG and cardio-vibrational data of a patient and then manually (i.e., mentally with or without pen and paper) determine metrics (such as the claimed EMAT and/or LVST) and a cardiac condition measure without the aid of a computer. These are calculations and/or determinations/evaluations based on expertise and knowledge in this medical field, both of which are the types of data analyses capable of being carried out by the human mind. The claims here recite the type of diagnostic analysis that a human diagnostician could (and historically would) have performed mentally and/or with pen and paper, except that generic computing technology is being incorporated for its well known and understood benefits, such as speed, accuracy and convenience.
Concerning argument (2), Applicant fails to explain how any improvement to the actual underlying technology has taken place due to the claimed subject matter here. Applicant’s remarks simply conclude that because the claims recite generating an automated identification of a change in the patient’s cardiac condition via the device, that a “clear improvement” has taken place. However, this is not a technology improvement, or even a technological change. The claims here merely using generic processing technology to automated a diagnostic analysis that a human diagnostician could otherwise produce. Nothing in the claims amounts to an improvement to the technology itself. For instance, the claims do not set forth an improvement (or even a change) to the actual processor or machine learning technologies being claimed.
II. Claim Rejections under 35 U.S.C. § 102
Applicant’s remarks concerning the previous § 102 rejections have been fully considered and are persuasive to the extent that the independent claims have been amended to incorporate additional limitations not taught by the Liu reference. Accordingly, the § 102 rejections are withdrawn. However, the amendments have necessitated updated grounds of rejection under 35 U.S.C. § 103.
III. Claim Rejections under 35 U.S.C. § 103
No specific remarks are presented concerning the § 103 rejections. As discussed above, the § 103 rejections have been updated to reflect the claim amendments.
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 1-3, 5, 9, 11-14, 17-18, 23, 27, 34-35, 38, 40 and 48-50 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a mental process without significantly more.
Step 1: All claims are directed either to a method/process or to a system/machine.
Step 2A, Prong One: The claims recite a mental process including steps such as “determining … the cardiac condition measure” and “obtain a first result … obtain a second result …” which could be performed by the human mind and/or by a human with a physical aid such as pen and paper.
Step 2A, Prong Two: This judicial exception is not integrated into a practical application because the claims merely implement the mental process using generic processing technology and add insignificant extra-solution activity. Specifically: the steps of obtaining and collecting the physiological data is considered insignificant pre-solution activity of mere data gathering, since it merely collects the data necessary to carry out the mental process using conventional, generic sensors; the output steps of “to generate an automated identification …” in e.g. claims 1 and 35 and “providing … information” seen in e.g. claims 13-14 and 48-50 is considered insignificant post-solution activity since it merely outputs the result of the mental process using a generic output modality (such as a display). Furthermore, merely carrying out mental steps using generic computing technology such as “processing circuitry” and “a machine learning engine [having tiers and classifiers]” is well established to not amount to an integration into a practical application under the § 101 analysis. See, e.g., MPEP §§ 2106.04(a)(2)(III)(C) and 2106.04(d)(I) and 2106.05(f).
Step 2B: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the only additional elements recited in the claims are generic processing/computing components and generic data collection and output components. The Examiner previously took official notice that these are basic, generic components which are well-understood, routine and conventional in the medical diagnostic arts, and the claims here merely use them for their well-understood, routine and conventional functions. Applicant’s subsequent reply did not traverse the Examiner's assertion of official notice; therefore, the facts under official notice are now taken to be admitted prior art. See MPEP § 2144.03(C) (“If applicant does not traverse the examiner' s assertion of official notice or applicant' s traverse is not adequate, the examiner should clearly indicate in the next Office action that the common knowledge or well-known in the art statement is taken to be admitted prior art because applicant either failed to traverse the examiner' s assertion of official notice or that the traverse was inadequate.”). As such, those additional elements cannot be considered “significantly more” than the judicial exception in Step 2B of the § 101 analysis.
Claim Rejections - 35 USC § 103
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claims 1-3, 5, 9, 13-14, 17-18, 23, 27, 34-35, 38, 40 and 48-50 are rejected under 35 U.S.C. 103 as being unpatentable over Liu et al. entitled “Multiple-feature-branch convolutional neural network for myocardial infarction diagnosis using electrocardiogram” (NPL No. 10 cited by Applicant in the IDS dated 04/28/2023 and a copy of which exists in this application file wrapper with that same date) (hereinafter “Liu”) in view of US 2019/0282178 A1 to Volosin et al. (hereinafter “Volosin”).
Regarding Claims 1, 5 and 35, Liu teaches: A method for determining a cardiac condition measure of a patient wearing a wearable medical device based on machine learning analysis (see, e.g., the title and abstract and FIG. 2), the method comprising:
obtaining, by processing circuitry from the wearable medical device, physiological data representing a sample timeframe, the physiological data comprising ECG data representing a plurality of ECG signals of a patient, wherein the plurality of ECG signals was collected by at least two ECG electrodes of the wearable medical device monitoring a heart of the patient (see, e.g., “12-lead ECG” on page 24, left column, and in FIG. 2)
applying, by the processing circuitry, the physiological data to a machine learning engine (see, e.g., “Multiple-Feature-Branch Convolutional Neural Network (MFB-CNN)” in the abstract) to determine the cardiac condition measure of the patient, wherein applying comprises
applying the ECG data to a first tier of the machine learning engine comprising one or more first machine learning classifiers to obtain a first result (see generally Section 3.1 entitled “Independent feature branch”; also see the “feature branches” in FIG. 2), and
applying the first result along with i) clinical information regarding the patient and/or ii) physiological metrics determined from signals collected by the wearable medical device to a second tier of the machine learning engine comprising one or more second machine learning classifiers to obtain a second result (see generally Section 3.2 entitled “Global fully-connected layer”; also see the “fully-connected layers” in FIG. 2; the feature vectors themselves correspond to both the ”first result” as well as either “clinical information” and/or “physiological metrics”); and
determining, by the processing circuitry at least in part from the second result, the cardiac condition measure (see, e.g., the final output vector Y described in section 3.2; see table 1 on page 23 showing the various classes of cardiac conditions which are
encompassed by the final output).
Liu fails to explicitly teach also obtaining cardio-vibrational data by a vibrational sensor of the wearable device, and a cardio-vibrational trained machine learning classifier. Another reference, Volosin, teaches an analogous system in which both ECG and cardio-vibrational data (see “cardio-vibration” throughout Volosin; in e.g. Para. 93, it is seen that “bio-vibration” may include “cardio-vibrations”) may be used as training data for a machine learning model for cardiac condition measures (see, e.g., Para. 267). It would have been obvious to one of ordinary skill in the art as of Applicant's effective filing date to modify Liu to also incorporate cardio-vibrational data into the training of the machine learning engine as well as a subsequent input to the model because doing so would predictably and advantageously increase the accuracy of the overall determination.
Liu fails to explicitly teach that the physiological metrics comprise EMAT and/or LVST. Volosin teaches an analogous system in which these metrics be factored into the analysis for cardiac condition measures (see, e.g., Paras. 93 and 148 discussing EMAT and LVST and SDI). It would have been obvious to one of ordinary skill in the art as of Applicant's effective filing date to further modify Liu to input one or more of these types of data into the machine learning engine because doing so would predictably and advantageously increase the accuracy of the result.
Concerning the limitation “to generate an automated identification of a change in the patient’s cardiac condition via the wearable medical device,” Volosin teaches this in an analogous system (see, e.g., Paras. 74, 107 and 181). It would have been obvious to one of ordinary skill in the art as of Applicant's effective filing date to modify Liu to generate an automated identification of a change in the condition because this is extremely commonplace in the medical diagnostic arts and because doing so would allow the data to be properly utilized and put into practical use.
Regarding Claims 2-3, see, e.g., FIGS. 1-2, and page 24, last paragraph before section 3.1, and section 3.2.
Regarding Claim 9, Volosin teaches an analogous system in which the determined cardiac condition measure can include a NYHA I, II, III or IV heart failure classification (see, e.g., Paras. 176-180 and Table 9 of Volosin). It would have been obvious to one of ordinary skill in the art as of Applicant's effective filing date to modify Liu to determine a cardiac condition measure as one of the NYHA heart failure classifications, as taught by Volosin, because this would predictably and advantageously yield diagnostically useful information in a scoring/classification system that is already known and understood to medical professionals, thereby allowing for easier interpretation of the data.
Regarding Claims 13-14 and 48-50, it is extraordinarily well known to output the data and results of a diagnostic analysis to a medical professional’s device, such as on a display screen of a remote computing device. As one example, Volosin teaches this in an analogous system (see, e.g., Paras. 74, 107 and 181). It would have been obvious to one of ordinary skill in the art as of Applicant's effective filing date to modify Liu to provide the data and results to an output device, including a visual display, for review by a medical professional or caregiver etc. because this is extremely commonplace in the medical diagnostic arts and because doing so would allow the data to be properly utilized and put into practical use.
Regarding Claim 17, Liu teaches sample timeframes which include “at least 30 seconds” (see Section 2, first paragraph). The recitation of “at least 30 seconds” is interpreted to include at least some sample timeframes between 30 seconds and 60 seconds (i.e. overlapping with several of Applicant’s claimed ranges), since if all the timeframes were at least 60 seconds (for example), then Liu would simply have said “at least 60 seconds.” The reference to “30 seconds” in particular necessarily implies that at least some of the sample timeframes were down near that value.
Regarding Claim 18, Volosin further teaches one or more of the additional metrics listed in this claim (see, e.g., Paras. 93 and 148 discussing EMAT and LVST and SDI). It would have been obvious to one of ordinary skill in the art as of Applicant's effective filing date to further modify Liu to input one or more of these additional types of data into the machine learning engine because doing so would predictably and advantageously increase the accuracy of the result.
Regarding Claim 23, Volosin teaches an analogous system in which the analysis is carried out on data collected over one or more timeframes overlapping with Applicant’s ranges here (see, e.g., Paras. 7, 297 and claim 3). It would have been obvious to one of ordinary skill in the art as of Applicant's effective filing date to modify Liu to utilize data collected from any desired suitable prior timeframe, such as those taught in Volosin, since any of them would be predictably suitable for this analysis with predictable benefits and drawbacks to various lengths of time.
Regarding Claim 27, Volosin teaches an analogous system in which historic cardiac condition measures are compared to current ones to identify changes (see, e.g., Paras. 61, last two sentences, 64, 99, 283, 307). It would have been obvious to one of ordinary skill in the art as of Applicant's effective filing date to modify Liu to continually measure and identify changes in the patient’s cardiac condition because this would be clearly useful and valuable diagnostic information.
Regarding Claim 34, see, e.g., page 29, Section 4.4, and FIGS. 8-9 and Table 13.
Regarding Claim 38, see e.g. FIG. 2 (each tier has multiple stages of classifiers).
Regarding Claim 40, see e.g. Table 2, column entitled “Kernel size.” Furthermore, it would have been obvious to one of ordinary skill in the art as of the filing date of Applicant’s invention to engage in routine experimentation to discover the optimal size(s) of the convolutional filters. See MPEP § 2144.05(II)(A)( “[W]here the general conditions of a claim are disclosed in the prior art, it is not inventive to discover the optimum or workable ranges by routine experimentation”) (citing In re Aller, 220 F.2d 454, 456, 105 USPQ 233, 235 (CCPA 1955)). Liu also notes that this parameter can be experimented with and determined via trial and error (see Section 4.1: “there are also some parameters determined by trial and error, such as the kernel size …”).
Claims 11-12 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Liu in view of Volosin as applied to claim 1 above, and further in view of US 2020/0397313 A1 to Attia et al. (hereinafter “Attia”).
Regarding Claims 11-12, Liu as modified teaches the method of claim 1 as discussed above but fails to explicitly teach that the cardiac condition measure is an ejection fraction classification, and having multiple classifications as greater than and less than a certain percentage such as those listed in claim 12 here. Another reference, Attia, teaches an analogous machine learning method in which ECG data is input to obtain an ejection fraction classification (see, e.g., the title and abstract) including separate classifications for above and below a particular ejection fraction percentage, such as above and below 30%, 35%, 40% or 45% (see, e.g., Para. 36, 52 and 74). It would have been obvious to one of ordinary skill in the art as of Applicant's effective filing date to further modify Liu to determine an ejection fraction classification as above and below one or more threshold percentages as the cardiac condition measure, as taught in Attia, because this is a known measure for evaluating aspects of a patient’s cardiovascular health (see, e.g., Paras. 3-5 of Attia) and thus determining it would predictably and advantageously provide useful diagnostic information.
Regarding Claim 17, Liu teaches sample timeframes of “at least 30 seconds” and “typically of ~2 min” (see Section 2, first paragraph). The recitation of “at least 30 seconds” is interpreted to include at least some sample timeframes between 30 seconds and e.g. 60 seconds, since if all the timeframes were at least 60 seconds, then Liu would simply have said “at least 60 seconds.” The reference to “30 seconds” necessarily implies that at least some of the sample timeframes were down near that value. Nevertheless, in the interest of being thorough, Attia teaches an analogous system which uses a variety of possible ECG sample timeframes overlapping with Applicant’s claimed values here (see, e.g., Paras. 55, 74). It would have been obvious to one of ordinary skill in the art as of Applicant's effective filing date to further modify Liu to utilize a sample timeframe overlapping with Applicant’s claimed ranges here, as seen in Attia, because Attia demonstrates that these are known suitable timeframes for collecting ECG data to yield a diagnostically relevant result via a machine learning analysis.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOHN R DOWNEY whose telephone number is (571)270-7247. The examiner can normally be reached Monday-Friday 8:30am-5:00pm ET.
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.
/JOHN R DOWNEY/Primary Examiner, Art Unit 3792
Prosecution Insights