DETAILED ACTION
Notices to Applicant
This communication is a non-final rejection. Claims 1-10, as filed 09/18/2025, are currently pending and have been considered below.
Foreign priority is generally acknowledged to JP 2024-171955 which was filed 10/01/2024.
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . 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 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 Examiner identified a commonly-assigned application, US 19/350,429. No double patenting rejection is made because nothing in those claims recites clustering according to an age category, a loss operating on point-to-centroid and centroid-to-centroid distances, or movement of a projection point to a future age position.
Claim Objections
Claims 1, 5, 7 are objected to because of the following informalities. Claim 1 states “at least one first processor configured to execute the instructions to…optimizes the encoder”. The “s” on “optimizes” should be deleted for grammatical consistency. The same issue is present with “clusters obtained projection points” in claim 5. Claim 7 recites “the output second processor outputs” which the Examiner interprets to read “the second processor outputs”.
Appropriate correction is required.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-5 and 7-10 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. The claims do not fall within at least one of the four categories of patent eligible subject matter because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
Step 1
The claim(s) recite(s) subject matter within a statutory category as a process, machine, and/or article of manufacture which recite:
1. A training device comprising: at least one first memory configured to store instructions; and at least one first processor configured to execute the instructions to: (additional element – merely applying the abstract idea with a computer)
acquire an age and attribute data other than the age; (additional element; insignificant extra-solution activity; mere data-gathering)
project, by an encoder, the attribute data to a latent space according to a category of the age and cluster obtained projection points into a plurality of clusters; (abstract idea; mapping data values into a coordinate space and grouping points by distance relationships are mathematical concepts)
predict, by a predictor, disease risks based on positions of the projection points on the latent space; (abstract idea; deriving a value as a function of coordinate position is a mathematical concept; additionally, this prediction is a mental process that amounts to an evaluation)
optimizes the encoder and the predictor based on relationships between the projection points in the latent space and the plurality of clusters and a mutual relationship between the plurality of clusters. (abstract idea; adjusting parameters as a function of computed distances is a mathematical concept)
2. The training device according to claim 1, wherein the first processor optimizes the encoder and the predictor by using a loss function that decreases a loss as a distance between the projection point in the latent space and a center of gravity of the cluster to which the projection point belongs decreases and decreases the loss as a distance between the centers of gravity of the plurality of clusters increases. (further mathematical concepts)
5. A disease risk estimation device comprising: at least one second memory configured to store instructions; and at least one second processor configured to execute the instructions to: (additional element – merely applying the abstract idea with a computer)
acquire a current age, a future age, and current attribute data other than the age; (additional element; insignificant extra-solution activity; mere data-gathering)
project, by an encoder, the attribute data to a latent space according to a category of the age and clusters obtained projection points into a plurality of clusters; (abstract idea; mapping data values into a coordinate space and grouping points by distance relationships are mathematical concepts)
move, in the latent space, a projection point related to the current age to a position related to the future age; (mathematical concept of coordinate translation)
predict, by a predictor, a disease risk based on the position of the projection point on the latent space; and (abstract idea; deriving a value as a function of coordinate position is a mathematical concept; additionally, this prediction is a mental process that amounts to an evaluation)
output a prediction result of the disease risk. (insignificant extra-solution activity, mere data output)
7. The disease risk estimation device according to claim 5, wherein
the predictor predicts a current disease risk based on the projection point related to the current age, and predicts a future disease risk based on the projection point related to the future age, and
the output second processor outputs a comparison result between the current disease risk and the future disease risk. (further mathematical detail)
8. The disease risk estimation device according to claim 5, wherein
the predictor predicts a disease risk of a subject based on the projection point related to the current age, and predicts an average disease risk based on a projection point related to an average value of the attribute data, and
the second processor outputs a comparison result between the disease risk of the subject and the average disease risk. (further mathematical detail)
These claims are presented as exemplary but the same analysis applies to the other claims (except claim 6).
Step 2A Prong One
The broadest reasonable interpretation of these steps includes mathematical concepts as described above. The steps of mapping data values into a coordinate space and performing operations on them to predict a user’s health are mathematical concepts but for the mere recitation of generic computer devices.
Dependent claims recite additional mathematical concepts as described above.
Step 2A Prong Two
This judicial exception is not integrated into a practical application. In particular, the additional elements do not integrate the abstract idea into a practical application, other than the abstract idea per se, because the additional elements:
amount to mere instructions to apply an exception. For example, the memory and processer language amounts to invoking computers as a tool to perform the abstract idea, see MPEP 2106.05(f)). Similarly, the encoder and predictor are recited functionally at a high level with no architecture or training mechanisms claimed. A generically recited trained model invoked as a tool does not remove the invention from being directed to a judicial exception.
add insignificant extra-solution activity to the abstract idea. For example, acquiring data and outputting a prediction result amount to mere data gathering and data output, see MPEP 2106.05(g))
Dependent claims recite additional subject matter which amount to limitations consistent with the additional elements in the independent claims. Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely provide conventional computer implementation and do not impose a meaningful limit to integrate the abstract idea into a practical application.
Step 2B
The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to discussion of integration of the abstract idea into a practical application, the additional elements amount to no more than mere instructions to apply an exception, add insignificant extra-solution activity to the abstract idea, and generally link the abstract idea to a particular technological environment or field of use. Additionally, the additional limitations, other than the abstract idea per se amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields such as receiving or transmitting data over a network, Symantec, MPEP 2106.05(d)(II)(i), performing repetitive calculations, Flook, MPEP 2106.05(d)(II)(ii), electronic recordkeeping, Alice Corp., MPEP 2106.05(d)(II)(iii), and/or storing and retrieving information in memory, Versata Dev. Group, MPEP 2106.05(d)(II)(iv).
Dependent claims recite additional subject matter which, as discussed above with respect to integration of the abstract idea into a practical application, amount to invoking computers as a tool to perform the abstract idea. Dependent claims recite additional subject matter which amount to limitations consistent with the additional elements in the independent claims. Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely provide conventional computer implementation.
Claim 6 is not rejected under § 101. Claim 6 recites the same exceptions as the other claims under Step 2A Prong One. The added limitation of moving a projected point such that its relationship with the center of gravity of the cluster is maintained between the current and future age clusters (i.e., offset-preservation) is also a mathematical concept. However, under Prong Two claim 6 recites a specific mechanism that improves the encoder and predictor and is not found in any other claims. Claim 5, for example, recites the result of moving the point from one position to another with no technical mechanism. Claim 6 provides this with the offset-preservation language which allows the individual’s geometric relationship to their own age cluster centroid to be carried over to the future age cluster centroid and produces individualized future age representations rather than a generic or population-average one. This fixes a technical failure mode of translations to future clusters in latent space. Because the improvement is a defined operation on the latent geometry that preserves discriminative information that the generic move discards, it improves the operation of the model itself rather than improving the risk number output in the end.
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 set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-4 are rejected under 35 U.S.C. 103 as being unpatentable over Yu (Zhihao Yu et al., "Predict and Interpret Health Risk using EHR through Typical Patients," arXiv:2312.10977 (Year: 2023)) in view of Wada (US20240266062A1).
Regarding claim 1, Yu discloses: A training device comprising: at least one first memory configured to store instructions; and at least one first processor configured to execute the instructions to (“our code is release at…” Abstract; to the extent that Yu lacks generic computer components, those teachings are provided by Wada as described below):
--acquire an age and attribute data other than the age (“Then we concatenate the embedded laboratory tests and demographic data to acquire overall the patient’s health status representation,” p. 2; age is part of demographic data in Table 4 on p. 4);
--project, by an encoder, the attribute data to a latent space (“where H denotes the dimension of latent space,” p. 2) and
--cluster obtained projection points into a plurality of clusters (“we cluster the latent representation h of all patients in the training set to obtain K clusters and take the patients closest to centroids as the prototypes…The clustering algorithm is applied to calculate centroids. The nearest patients to each centroid in the latent space will be new prototypes to replace old prototypes,” p. 2);
--predict, by a predictor, disease risks based on positions of the projection points on the latent space (“PPN first computes the similarity coefficient,” p. 2; “the prediction of health risks can be obtained,” p. 3); and
--optimizes the encoder and the predictor based on relationships between the projection points in the latent space and the plurality of clusters a mutual relationship between the plurality of clusters (“We use binary cross-entropy loss as the primary learning objective …two regularization terms to encourage a clustering structure in the latent space by minimizing the squared distance between an encoded sample and its closest prototype. But these terms may cause some samples closer to the cluster they do not belong to. Consequently, we introduce another separation term,” p. 3; “With loss terms formulated above, we can define the final objective of our model as [equation 6]… We use a gradient descent backpropagation algorithm to minimize the loss L and optimize the parameters,” p. 3).
Yu does not expressly disclose projecting and clustering the attribute data according to a category of the age. Wada teaches this. Wada excludes age itself from the clustered parameters while clustering separately within each age bracket (“second filtering processing step…age and gender parameters, which are demographic data, are excluded,” [0100]; Age is excluded in FIGS. 6, 8, and 10; “initialization (center point setting) is performed for 40% of learning data with disease labels, and clustering about whether a high risk and a low risk for each age group at the center point (each non-disease category) is performed,” [0086]; computer in [0070]; processor in claim 18; computer storage in [0154]).
One of ordinary skill in the art would have been motivated before the effective filing date to expand Yu’s prototype-based representational learning to include Wada’s computerized age-conditioned clustering because this improves patient health predictions by making “it possible to determine the degree of the risk of developing a specific disease from medical checkup data in a healthy stage by a method in which two stages of a first stage of classifying people…” (Wada [0165]) and by improving predictions for young, healthy people (Wada [0065]).
Additionally, it can be seen that each element is taught by either Yu or Wada. Wada’s age clustering does not affect the normal functioning of the elements of the claim which are taught by Yu. Because the elements do not affect the normal functioning of each other, the results of their combination would have been predictable. Therefore, before the effective filing date of the claimed invention, it would have been obvious to combine the teachings of Wada with the teachings of Yu since the result is merely a combination of old elements, and, since the elements do not affect the normal functioning of each other, the results of the combination would have been predictable.
Claims 3 and 4 are substantially similar to claim 1 and are rejected with the same reasoning.
Regarding claim 2, Yu discloses: wherein the first processor optimizes the encoder and the predictor by using a loss function that decreases a loss as a distance between the projection point in the latent space and a center of gravity of the cluster to which the projection point belongs decreases and decreases the loss as a distance between the centers of gravity of the plurality of clusters increases (“We use binary cross-entropy loss as the primary learning objective …two regularization terms to encourage a clustering structure in the latent space by minimizing the squared distance between an encoded sample and its closest prototype. But these terms may cause some samples closer to the cluster they do not belong to. Consequently, we introduce another separation term,” p. 3; “With loss terms formulated above, we can define the final objective of our model as [equation 6]… We use a gradient descent backpropagation algorithm to minimize the loss L and optimize the parameters,” p. 3; “we cluster the latent representation h of all patients in the training set to obtain K clusters and take the patients closest to centroids as the prototypes…The clustering algorithm is applied to calculate centroids. The nearest patients to each centroid in the latent space will be new prototypes to replace old prototypes,” p. 2).
Claims 5 and 7-10 are rejected under 35 U.S.C. 103 as being unpatentable over Wada in view of Yu and Alaluf (Yuval Alaluf et al., "Only a Matter of Style: Age Transformation Using a Style-Based Regression Model," arXiv:2102.02754 (Year: 2021)) .
Regarding claim 5, Wada discloses: A disease risk estimation device comprising: at least one second memory configured to store instructions; and at least one second processor configured to execute the instructions (computer in [0070]; processor in claim 18; computer storage in [0154]) to:
--acquire a current age, a future age (“it is possible to evaluate a risk after many years from the distribution for each age group,” [0083]), and current attribute data other than the age (“, a computer has a learning data acquisition step S10 of acquiring at least two kinds of category data,” [0070]; “For target person data of the estimation target person used at the determination step S70,” [0082]);
--project… according to a category of the age and clusters obtained projection points into a plurality of clusters (“clustering about whether a high risk and a low risk for each age group at the center point (each non-disease category) is performed,” [0086]; age and gender exclusions in FIGS. 6, 8, and 10);
--predict, by a predictor, a disease risk based on the position of the projection point on the latent space (“perform clustering into at least two groups, and a disease risk is estimated for an estimation target person who is in a healthy stage by determining which group the estimation target person belongs to or is close to,” [0069]); and
--output a prediction result of the disease risk (“When the process ends for all the targeted diseases, scores of individual subjects are created at S6. Scores are displayed for the high incidence risk group, and quantified incidence levels showing diseases the incidence risk of which is high are displayed. The scores are numerically displayed, for example, with values from 0 to 100 inclusive,” [0058]).
Wada does not expressly disclose, but Yu teaches: projection by an encoder to a latent space (pages 2-3 described above in the rejection of claim 1).
One of ordinary skill in the art would have been motivated before the effective filing date to expand Wada’s cluster-based patient predictions to include Yu’s latent space encoding because encoding patient attribute data into a learned latent representation captures information that operating traditional approaches “learn sufficient representations and lead to poor performance when it comes to patients with few visits or sparse records” and thus this technique “brings improvement on all metrics” (Yu Abstract).
Additionally, it can be seen that each element is taught by either Yu or Wada. Yu’s latent space encoding does not affect the normal functioning of the elements of the claim which are taught by Wada. Because the elements do not affect the normal functioning of each other, the results of their combination would have been predictable. Therefore, before the effective filing date of the claimed invention, it would have been obvious to combine the teachings of Wada with the teachings of Yu since the result is merely a combination of old elements, and, since the elements do not affect the normal functioning of each other, the results of the combination would have been predictable.
Yu and Wada do not disclose but Alaluf teaches:
--acquire a future age (“given a single input image and desired target age, we show that
our approach can successfully generate the corresponding image,” p. 2; “representing source identity at age a_t,” p. 3).
--move, in the latent space, a projection point related to the current age to a position related to the future age (Alaluf encodes the input to a latent code and learns a shift of that code within the latent space to the position corresponding to the target age: “the resulting latent codes are then edited in a semantically meaningful manner by traversing the latent space to obtain a new latent code,” p. 3; “In a sense, SAM is tasked with learning a shift in the latent space with respect to w∗,” p. 14).
One of ordinary skill in the art would have been motivated before the effective filing date to expand the latent-space encoding-based health predictions of Wada and Yu to include the target age analysis of Alaluf because this “allows for more fine-grained control over the desired transformation,” p. 2 and “one can directly specify the desired target age, which cannot be done when operating solely in latent space,” p. 9.
Regarding claim 7, Wada further discloses:
--the predictor predicts a current disease risk based on the projection point related to the current age, and predicts a future disease risk based on the projection point related to the future age (“. This kind of graph shows a rate of people having a disease or the risk of the disease in various risk groups for various ages,” [0141]; “it is possible to evaluate a risk after many years from the distribution for each age group,” [0083]), and
--the output second processor outputs a comparison result between the current disease risk and the future disease risk (“In analysis of a risk associated with aging, the x axis represents age, and the y axis represents prevalence rate,” [0141]; “by displaying a determination result about the estimation target person with a plot, the determination result can be compared with the line graphs of the low-risk and high-risk groups, and it is possible to determine which group the estimation target person is close to, that is, a risk position,” [0083]).
Regarding claim 8, Wada further discloses:
--the predictor predicts a disease risk of a subject based on the projection point related to the current age (“a disease risk is estimated for an estimation target person who is in a healthy stage by determining which group the estimation target person belongs to or is close to,” [0069]), and
--the second processor outputs a comparison result between the disease risk of the subject and the average disease risk (“A red solid line in FIG. 3A indicates a mean value of incidence rates of people classified as having a high risk, as a result of performing clustering with data that has been trained for cardiovascular disease, by age,” [0064]; “by displaying a determination result about the estimation target person with a plot, the determination result can be compared with the line graphs of the low-risk and high-risk groups, and it is possible to determine which group the estimation target person is close to, that is, a risk position,” [0083]).
Wada does not expressly disclose but Yu teaches:
--predicts an average disease risk based on a projection point related to an average value of the attribute data (“the clustering algorithm is applied to calculate centroids,” p. 2; “Since the prototypes can be regarded as particular patients, we predict the health risk y_j of prototypes j by p_j as health status and adopt a separation loss that encourages the prototypes to represent different subtypes of patients,” p. 3).
The motivation to combine is the same as in claim 5.
Claims 9 and 10 are substantially similar to claim 5 and are rejected with the same reasoning.
Allowable Subject Matter
Claim 6 is objected to as dependent on a rejected claim but would be allowable if rewritten in independent form.
The closest prior art is Wada, Yu, and Alaluf as described above. Wada evaluates risk at a later age by reference to the per-age-group distribution but is silent as to any constraint preserving an individual’s geometric relationship to a group centroid across age groups. Yu forms and separates prototypes but does not translate an individual’s representation from one prototype’s neighborhood to another’s. Alaluf moves a latent representation to a specified target-age position, and its regularization encourages latent codes “closer to the average latent vector” (p. 3), that is, toward a single global average, not a preserved per-cluster centroid-relative offset across two age-specific centroids as claim 6 requires.
The Examiner identifies no art that teaches or reasonably suggests moving the projection point related to the current age “in such a way that a positional relationship between the projection point related to the current age in the latent space and a center of gravity of a cluster related to the current age matches a positional relationship between a projection point related to the future age in the latent space and a center of gravity of a cluster related to the future age” in combination with the limitations of claim 5.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JOSHUA BLANCHETTE whose telephone number is (571)272-2299. The examiner can normally be reached on Monday - Thursday 7:30AM - 6:00PM, EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Shahid Merchant, can be reached on (571) 270-1360. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/JOSHUA B BLANCHETTE/Primary Examiner, Art Unit 3624
Prosecution Insights