DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after 16 March 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendment
The amendment to the claims filed 26 March 2026 has been entered. Claim(s) 1 and 4 is/are currently amended. Claim(s) 2-3 and 5-6 has/have been canceled. Claim(s) 1 and 4 is/are pending.
Objections Withdrawn
Objections to the claims not reproduced below have been withdrawn in view of Applicant's amendments to the claims and/or submitted remarks.
Claim Interpretation
As noted in the prior Office action (mailed 30 December 2025, "Non-Final Rejection"), the "analysis apparatus" limitation(s) of claim 1 and the "interface device" and "storage device" limitations of claim 4 have been interpreted to invoke 35 U.S.C. 112(f) (or pre-AIA 35 U.S.C. 112, sixth paragraph).
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of pre-AIA 35 U.S.C. 112, first paragraph:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claim(s) 1 and 4 is/are rejected under 35 U.S.C. 112(a) or pre-AIA 35 U.S.C. 112, first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for pre-AIA the inventor(s), at the time the application was filed, had possession of the claimed invention.
Regarding claim 1, claim 4 and claims dependent thereon, Applicant cites the originally-filed dependent claims and paragraphs [0034]-[0069] of the specification (presumably as filed) for support of the newly-presented limitation(s). The examiner respectfully disagrees either said cited portions, or the remainder of the application as filed, provides sufficient support for the newly-added limitations. Applicant discloses, like Bang, a plurality of deep learning models utilizing various time-bins lengths were trained and evaluated (¶¶ [0048]-[0051]). However, Applicant fails to disclose an analysis apparatus, or processor thereof, stores or otherwise has access to such a plurality of models, wherein the apparatus/processor selects a model from the plurality based on time length. The most relevant paragraphs of the specification state, "The computing device 300 can input (preprocessed) pain score data into a deep learning model trained in advance and can predict whether breakthrough pain of the corresponding subject will be generated on the basis of a probability value that is output from the deep learning model. In this case, as the deep learning model, a deep learning model suitable for input data can be used in accordance with the time lengths of past data collected from patients and/or the lengths of time section units dividing pain data (¶¶ [0068]-[0069]). Applicant provides no further clarification as to what "suitable for input data can be used in accordance with the time lengths of past data collected from patients and/or the lengths of time section units dividing pain data" means. Accordingly, to the best of the examiner's understanding, Applicant appears to disclose the deep learning model into which the preprocessed pain score data is input is a deep learning model suitable for the pre-processed input data in accordance with its time lengths and/or bin lengths (i.e., data used to train said model was preprocessed in the same manner as the preprocessed input data), a recognized, crucial design consideration when employing trained models (see, e.g., Bansal, cited herewith). Lacking in Applicant's disclosure, however, is any indication that the computer/processor selects the model to be used from a plurality of models based on the time length of the bin of preprocessed pain score data as required by the present claim. Therefore, the limitation "wherein the processor selects the deep learning model from a plurality of deep learning models based on the time length, wherein each of the plurality of deep learning models is trained using training data collected in a respective time unit, wherein the selected deep learning model is trained using training data collected in units corresponding to the time length of the bin" of claim 1 and the comparable limitations of claim 4 lack support in the application as filed.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim(s) 1 and 4 is/are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception(s) without significantly more.
Claims 1 and 4 recite the steps of preprocessing time-series pain score data collected from a subject by assigning the pain score data to each bin of a time length; selecting a model from a plurality of models based on the time length; and predicting whether breakthrough pain will be generated in the subject in a point in time in the future on the basis of the preprocessed pain score data and the selected model.
These limitations, as drafted, are a process that, under its broadest reasonable interpretation (BRI), cover performance of the limitations in the mind but for the recitation of generic computer components. That is, other than reciting the steps are performed "by means of an analysis apparatus," "deep learning model" and/or a "processor," nothing in the claim elements preclude the steps from practically being performed in the mind. For example, but for the generic computer component language, preprocessing the pain data encompasses a user mentally/manually binning collected pain score data according to the day (24-hour period) and the time (or time interval) of each day it was collected to determine or set a pain score for each of a plurality of time section units (i.e., bins of a time length). Predicting whether breakthrough pain will be generated in the subject in a point in time in the future on the basis of a prediction value, such as a probability value, encompasses a user considering the preprocessed/binned pain data, in a manner suitable for the manner in which the data was binned to form a judgement about when/what time of day a subject will next experience pain. If claim limitations, under a BRI thereof, cover performance of the limitations in the mind but for the recitation of generic computer components, then they fall within the mental processes grouping of abstract ideas. Accordingly, the claim recites an abstract idea.
This judicial exception is not integrated into a practical application. The claims recite the additional elements of using a generic computer (analysis apparatus, processor, etc.) for performing the abstract idea, a generic deep learning model for receiving input and performing the prediction step stored on a generic storage device, and a generic interface device or receiving step of receiving the pain data necessary to perform the abstract idea.
The analysis device, including the interface device, storage device, and computing device components, are recited so generically (i.e., no details whatsoever are provided other than that they are an interface, storage, and computing or analysis device) that they represent no more than mere instructions, or the generic components necessary, to apply the judicial exception on a computer. These limitations can also be viewed as nothing more than an attempt generally link the use of the judicial exception to the technological environment of a computer. Alternatively/Additionally, the step of receiving the collected pain data, and/or the generic interface device therefor, is necessary data gathering, is comparable to concepts identified by the courts as insignificant extra-solution activity (see MPEP 2106.05(g), "performing clinical tests on individuals to obtain input for an equation").
With respect to the deep learning model limitation(s), the trained deep learning model is used to receive preprocessed pain data and apply the abstract idea, or prediction step thereof, without placing any limits on how the trained model functions. Rather, this limitation(s) only recites the outcome of outputting a prediction value indicative of whether breakthrough pain will be generated in the subject in a point in time in the future based on input preprocessed pain data and does not include any details about how the prediction is accomplished. See MPEP 2106.05(f). The recitation of selecting and using the selected deep learning model (i.e., inputting data therein, and using the output thereof) alternatively/additionally merely indicates a field of use or technological environment in which the judicial exception is performed. Although the additional element using a deep learning model limits the identified mental process in the limitation "predicting whether breakthrough pain will be generated in the subject in a point in time in the future on the basis of a prediction value output from the deep learning model" this type of limitation merely confines the use of the abstract idea to a particular technological environment (deep learning model(s)) and thus fails to add an inventive concept to the claims. See MPEP 2106.05(h).
Even when viewed in combination, the additional elements do no more than merely invoke a computer as a tool to perform the above-noted abstract idea, and is comparable to concepts identified by the courts as such (e.g., MPEP 2106.05(f), "requiring the use of software to tailor information and provide it to the user on a generic computer"). Accordingly, claims 1-6 are directed to a judicial exception.
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application, the additional elements of using generic computer component and a generic deep learning model to perform the abstract idea (or particular steps thereof) at best mere instructions to "apply" the abstract idea, which cannot provide an inventive concept. See MPEP 2106.05(f). Further, the steps of/components for receiving the collected data are recited at a high level of generality encompassing well-understood, routine, conventional computer functions (see MPEP 2106.05(d), "receiving or transmitting data over a network," "storing and retrieving information in memory," etc.). Even when considered in combination, these additional elements represent mere instructions to implement an abstract idea on a computer and insignificant extra-solution activity, which do not provide an inventive concept. Accordingly, claims 1-6 are not patent eligible.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
Determining the scope and contents of the prior art.
Ascertaining the differences between the prior art and the claims at issue.
Resolving the level of ordinary skill in the pertinent art.
Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1 and 4 is/are rejected under 35 U.S.C. 102(a)(1) as anticipated by "Clinical relevance of deep learning models in predicting the onset timing of breakthrough cancer pain" (previously cited, "Bang") in view of US 20220108124 A1 ("Umezawa") and "Why your Machine Learning model may not work in production?" ("Bansal").
Regarding claim 1, Bang discloses/suggests a prediction method for breakthrough pain of a subject, the method comprising:
receiving, by an analysis apparatus (pg. 10, where scripts for pre-processing, training, and evaluation were written in Python, such that the "analysis apparatus" is the hardware component(s) by which said scripts were executed, indicating data was first necessarily received thereby; pgs. 4-5, GPU executing programming; etc.) pain score data including time-series data of pain scores collected from a subject for a pre-determined time (pg. 3, collecting pain records or logs for patients between July 2016 and February 2020; pain records/logs for representative cases illustrated in Fig. 5; etc.);
preprocessing, by the analysis apparatus (pg. 10; pgs. 4-5; etc.), the pain score data (pgs. 3-4, Fig. 2, etc., preprocessing) by assigning a pain score from the pain score data to each bin of a time length (pg. 4, binning the entire NRS record in arbitrary
τ
-length time-bins; pg. 4, pain records for
n
days with
τ
-length time-bins were transformed from a shape of
1
×
(
(
24
/
τ
)
×
n
vector form to a
(
24
/
τ
)
×
n
matrix form); and
inputting, by the analysis apparatus (pg. 10; pgs. 4-5; etc.), the preprocessed pain score data into a deep learning model trained in advance (Fig. 2, modeling) and predicting, by the analysis apparatus (pg. 10; pgs. 4-5; etc.), whether breakthrough pain will occur in the subject in a point at time in the future on the basis of a prediction value output from the deep learning model (Fig. 2, evaluation; Fig. 5, yellow dots and lines in the black-colored box showed the probabilities of BTcP on the forecast day).
Bang discloses a plurality of deep learning models may be generated, wherein each of the plurality of deep learning models is based on a predetermined time length of the bin and each of the plurality of deep learning models is trained using training data collected in a respective time unit (pgs. 6-7). Bang does not disclose the analysis apparatus selects the deep learning model from a plurality of deep learning models based on the time length of the bin, wherein the selected deep learning model is trained using training data collected in units corresponding to the time length of the bin.
Umezawa discloses/suggests a method comprising selecting a model from a plurality of models based on a degree of similarity with training data for a trained model and using the selected model for inference (e.g., ¶ [0055]; ¶ [0067]; etc.).
Bansal discloses preprocessing should be the same for training and testing data to prevent training-serving skew (pg. 6).
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 Bang with selecting the deep learning model from a plurality of deep learning models based on the time length of the bin, wherein each of the plurality of deep learning models is trained using training data collected in a respective time unit, wherein the selected deep learning model is trained using training data collected in units corresponding to the time length of the bin (i.e., training data for the selected model was pre-processed in the same manner as the recited preprocessing step) as taught/suggested by Umezawa and Bansal in order to facilitate selecting an appropriate trained model for given preprocessed data, e.g., that prevents training-serving skew (Umezawa, ¶ [0050]; Bansal, pg. 6; etc.).
Regarding claim 4, Bang discloses/suggests an analysis apparatus comprising:
an interface device configured to receive pain score data comprising time-series data of pain scores collected from a subject for a predetermined time (pg. 3, collecting pain records or logs for patients between July 2016 and February 2020; pain records/logs for representative cases illustrated in Fig. 5; etc., wherein the "interface device" is the hardware component(s) by which said pain records/logs are received by a computing device);
a storage device configured to storing a deep learning model configured to predict whether breakthrough pain will be generated by receiving pain information of a patient (pg. 10, where scripts for pre-processing, training, and evaluation were written in Python; pgs. 4-5, GPU executing programming; etc., wherein the "storage device" is the hardware component(s) by which said scripts and/or the model are stored for use by the computing device); and
a processor (pg. 10, where scripts for pre-processing, training, and evaluation were written in Python, such that processor is the hardware component(s) by which said scripts were executed; pgs. 4-5, GPU executing programming; etc., where the "processor" is the hardware component(s), e.g., GPU, etc., by which said scripts and/or the model are executed) configured to:
preprocess the pain score data (pgs. 3-4, Fig. 2, etc., preprocessing) by assigning a pain score from the pain score data to each bin of a time length (pg. 4, binning the entire NRS record in arbitrary
τ
-length time-bins; pg. 4, pain records for
n
days with
τ
-length time-bins were transformed from a shape of
1
×
(
(
24
/
τ
)
×
n
vector form to a
(
24
/
τ
)
×
n
matrix form); and
predict whether breakthrough pain will occur in the subject in a point at time in the future on the basis of a prediction value output by inputting the preprocessed pain score data into the deep learning model trained in advance (Fig. 2, evaluation; Fig. 5, yellow dots and lines in the black-colored box showed the probabilities of BTcP on the forecast day).
Bang discloses a plurality of deep learning models may be generated, wherein each of the plurality of deep learning models is based on a predetermined time length of the bin and each of the plurality of deep learning models is trained using training data collected in a respective time unit (pgs. 6-7). Bang does not disclose the processor selects the deep learning model from a plurality of deep learning models based on the time length of the bin, wherein the selected deep learning model is trained using training data collected in units corresponding to the time length of the bin.
Umezawa discloses/suggests a method comprising selecting a model from a plurality of models based on a degree of similarity with training data for a trained model and using the selected model for inference (e.g., ¶ [0055]; ¶ [0067]; etc.).
Bansal discloses preprocessing should be the same for training and testing data to prevent training-serving skew (pg. 6).
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 apparatus of Bang with selecting the deep learning model from a plurality of deep learning models based on the time length of the bin, wherein each of the plurality of deep learning models is trained using training data collected in a respective time unit, wherein the selected deep learning model is trained using training data collected in units corresponding to the time length of the bin (i.e., training data for the selected model was pre-processed in the same manner as the recited preprocessing step) as taught/suggested by Umezawa and Bansal in order to facilitate selecting an appropriate trained model for given preprocessed data, e.g., that prevents training-serving skew (Umezawa, ¶ [0050]; Bansal, pg. 6; etc.).
Regarding claims 1 and 4, though Bang does not expressly describe the analysis apparatus, or interface device, storage device and/or processor thereof, Bang discloses the use of a deep learning model(s), which, as a subset of machine learning, one of ordinary skill in the art would at once envision requires an analysis apparatus (i.e., machine) having the necessary hardware for receiving the data used by said model(s), means for storing the disclosed programming/model, and means for executing the model(s) (e.g., processor), as noted above.
Alternatively/Additionally, Umezawa discloses a similar system comprising an analysis apparatus (e.g., Fig. 2) including an interface device configured to receive data collected from a subject (acquisition unit 208), a storage device configured to store a model(s) (storage unit 204), and a processor configured to input the received data into a the model(s) trained in advance (inference unit 206; ¶ [0028]). 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/system of Bang to comprise an analysis apparatus for performing the prediction method, said apparatus comprising an interface device for receiving the pain score data collected from a subject for a predetermined time, a storage device for storing a deep learning model configured to predict whether breakthrough pain will be generated by receiving pain information of a patient; and a processor for preprocessing the received pain score data and predicting whether breakthrough pain of the patient will be generated at a point in time in the future on the basis of a prediction value that is output by inputting the preprocessed pain score data into the stored deep learning model trained in advance as taught and/or suggested by Umezawa in order to facilitate automating prediction of pain onset, enabling preemptive pain management (Bang, pg. 2).
Response to Arguments
Applicant's arguments have been fully considered but they are not persuasive.
With respect to claim interpretation under 35 U.S.C. 112(f), Applicant contends claims 1 and 4 recite sufficient structures to perform the recited functions and therefore do not invoke 35 U.S.C. 112(f) (Remarks, pg. 5).
The examiner respectfully disagrees. There is no structure recited in claim 1 beyond the "analysis apparatus," a nonce term modified by function. Similarly, no structure is recited for the "interface device" and "storage device" limitations of claim 4. Accordingly, these limitations meet the three-prong analysis and therefore invoke 35 U.S.C. 112(f) (see MPEP 2181(I)).
With respect to eligibility under 35 U.S.C. 101, Applicant contends the amended claims "recite limitations that foreclose mental performance," which the examiner understands to amount to an assertion that limitations cannot be practically performed in the mind. Specifically, Applicant contends "no human can form and populate such a data structure across days of NRS records with the precision and consistency required for deep learning input," "maintaining multiple trained deep learning models in storage and selecting the architecturally matched model based on input data characteristics is a specific computational operation," and the "architectural constraint on both the model ensemble and the selection logic…is far beyond anything a human mind could perform or even conceptualize as a mental process" (Remarks, pg. 6).
The examiner respectfully disagrees. Assigning the pain score encompasses a user looking at a time-series of pain score data and assigning time intervals a pain score based on said data, e.g., selecting the highest, average, etc. value within a time interval in the time-series. There is no and/or insufficient evidence of record to indicate a person is not capable of practically performing this function mentally and/or using pen and paper. There is no limit on the number of days of data collected, the frequency of the data collected, etc. This function could be as simple as assigning a highest pain score before noon and a highest pain score after noon for a given patient over the course of, e.g., 5 days, which a human could readily mentally/manually perform. Further, the manner in which the models are trained is not a function required by the claimed method or apparatus. All that is required is that a model is selected that was trained/learned using the same time bin lengths as the preprocessed data, of which a human is capable.
Applicant further submits that the claims integrate the judicial exception into a practical application, contending the model selection is a specific technical architecture in which the input data structure (bin length) controls the choice of inference model, the preprocessing step transforms input data into a form with concrete technical utility, and outputting a prediction is a practically applicable result with direct clinical utility in preemptive pain management (Remarks, pgs. 6-7).
The examiner respectfully disagrees. The relevant considerations for integration into a practical application is 1) identifying whether there are any additional elements recited in the claim beyond the judicial exception(s); and (2) evaluating those additional elements individually and in combination to determine whether they integrate the exception into a practical application. The steps of selecting a model and preprocessing the data are part of the judicial exception the claim recites and/or to which the claim is directed, not an additional element of the claim. Additionally, providing a "practically applicable result" does not require that said result is practically applied. For example, while Applicant appears to contend that the output of the model could be used in preemptive pain management, no such application is recited in the claim. Further, in order to practically apply a judicial exception, an additional element reciting a treatment based on said exception must be "particular" (see MPEP 2106.04(d)(2)). Accordingly, even if providing pre-emptive pain management treatment based on the judicial exception were required by the claim, said treatment is not particular, and would not practically apply the judicial exception.
Applicant further contends the limitations of the pending claims amount to significantly more than the judicial exception, contending "a system that is novel and non-obvious over the cited references necessarily comprises elements that are not well-understood or routine" (Remarks, pgs. 7-8).
The examiner respectfully disagrees that the claimed method/system is non-obvious over the cited references, or is even adequately disclosed by Applicant, as discussed in the rejection(s) of record above. However, as noted MPEP 2106.05(d), "The question of whether a particular claimed invention is novel or obvious is "fully apart" from the question of whether it is eligible."
Applicant's arguments with respect to the prior art rejections have been considered but are moot because the new ground(s) of rejection does/do not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
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 extension fee 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 date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Meredith Weare whose telephone number is 571-270-3957. The examiner can normally be reached Monday - Friday, 9 AM - 5 PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. Applicant is encouraged to use the USPTO Automated Interview Request at http://www.uspto.gov/interviewpractice to schedule an interview.
If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, Tse Chen, can be reached on 571-272-3672. 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.
/Meredith Weare/Primary Examiner, Art Unit 3791
Prosecution Insights