Notice of Pre-AIA or AIA Status
This action is made in response to the amendments/remarks filed on 10/28/2025. This action is made FINAL.
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 10/28/2025 has been entered. Claims 1-4, 6, and 10-14 remain pending in the application. Claims 5 and 7-9 are cancelled. Claims 15-16 are newly added. Applicant’s amendments to the claims have overcome the 112(b) rejection previously set forth in the Non-Final Office Action mailed 07/28/2025.
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 the first paragraph of pre-AIA 35 U.S.C. 112:
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.
Claims 1-4, 6, and 10-16 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), 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 applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Regarding Claim 1, the limitations “by optimizing weight parameters among heterogeneous preoperative variables and postoperative outcome data to minimize prediction loss caused by data imbalance or missing values” and “wherein the trained model predicts a quantitative risk score of postoperative acute kidney injury for an individual patient based on nonlinear correlations among the preoperative variables” recite elements without support in the original disclosure (i.e., introduces new matter). The specification lacks support for optimizing weight parameters among heterogeneous preoperative variables and postoperative outcome data to minimize prediction loss caused by data imbalance or missing values, and also lacks support for a quantitative risk score based on nonlinear correlations. Therefore, this limitation is new matter. See MPEP 608.04.
Regarding Claim 12, the limitation “to output a quantitative risk score through nonlinear weight correlations among heterogeneous preoperative variables and postoperative outcomes” recites elements without support in the original disclosure (i.e., introduces new matter). The disclosure lacks support for a quantitative risk score through nonlinear weight correlations among heterogeneous preoperative variables and postoperative outcomes. Therefore, this limitation is new matter.
Regarding Claim 14, the limitation “output a quantitative risk score through nonlinear correlations among preoperative variables” recites elements without support in the original disclosure (i.e., introduces new matter). The disclosure lacks supports for a quantitative risk score through nonlinear correlations among preoperative variables, therefore, this limitation is new matter.
Regarding Claims 15-16, the limitation “treating a subject based on the predicted risk of occurrence of acute kidney injury by administering a therapeutic regimen” recites elements without support in the original disclosure (i.e., introduces new matter). The disclosure lacks support for treating a subject in any manner based on the predicted risk by administering a therapeutic regimen, therefore, these limitations are new matter.
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 1-4, 6, and 10-16 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Independent Claims
Step 1 analysis:
Claims 1 and 12 are drawn to a method (i.e., process), and Claim 14 is drawn to a system, which are all within the four statutory categories. (Step 1 – Yes, the claims fall into one of the statutory categories).
Step 2A analysis – Prong One:
Claim 1 recites:
A method of predicting an occurrence of acute kidney injury, which is performed by at least one computing device, comprising:
preparing a dataset of a plurality of patients - wherein a dependent variable of the dataset relates to an occurrence of postoperative acute kidney injury, and independent variables of the dataset include variables relating to preoperative examination items of the patients - ; and
building a model configured to predict a risk of the occurrence of postoperative acute kidney injury using the prepared dataset,
wherein the preparing of the dataset comprises removing patient data satisfying predetermined kidney-related or surgery related conditions from an original patient dataset, correcting outliers, imputing missing values using multiple imputation by chained equations, and normalizing variables to generate a high-quality training dataset;
wherein the building of the model comprises training, by the computing device, a machine learning model including at least one of a neural network, logistic regression, and a light gradient boosting machine (LGBM), by optimizing weight parameters among heterogeneous preoperative variables and postoperative outcome data to minimize prediction loss caused by data imbalance or missing values; and
wherein the trained model predicts a quantitative risk score of postoperative acute kidney injury for an individual patient based on nonlinear correlations among the preoperative variables.
The series of steps as recited above describes managing personal behavior or relationships or interactions between people including following rules or instructions, and therefore fall within the scope of certain methods of organizing human activity. Fundamentally, the method is that of a person gathering preoperative and postoperative patient information and predicting the occurrence of acute kidney injury. Accordingly, the claim recites an abstract idea of managing interactions between people.
The series of steps as recited above also falls within the “mental processes” grouping of abstract ideas, and describes concepts that can be performed in the human mind through observation, evaluation, judgement, and opinion. Predicting a risk and preparing a dataset are both steps that can be performed in the human mind, with or without physical aid. Therefore, the claim recites an abstract idea of a mental process.
The series of steps as recited above also falls within the mathematical concepts grouping of abstract ideas. Imputing missing values using multiple imputation by chained equations, normalizing variables, and predicting a quantitative risk score recites mathematical calculations. Similarly, the step of training a neural network is fundamentally computing neural network parameters using a series of mathematical calculations. See MPEP 2106.04(a)(2)(I), “a step of "determining" a variable or number using mathematical methods or "performing" a mathematical operation may also be considered mathematical calculations when the broadest reasonable interpretation of the claim in light of the specification encompasses a mathematical calculation”.
Claims 12 and 14 recite/describe nearly identical steps as claim 1 (and therefore also recite limitations that fall within this subject matter grouping of abstract ideas), and these claims are therefore determined to recite an abstract idea under the same analysis.
Step 2A analysis – Prong 2:
This judicial exception is not integrated into a practical application. Specifically, independent claims 1, 12, and 14 recite the following additional elements beyond the abstract idea: at least one computing device, a machine learning model including at least one of a neural network, logistic regression, and alight gradient boosting machine (LGBM), one or more processors, and a memory. These limitations are recited at a high level of generality and amount to no more than mere instructions to apply the exception using generic computer components. The limitations do not impose any meaningful limits on practicing the abstract idea, and therefore do not integrate the abstract idea into a practical application (see MPEP 2106.05(f)).
Specifically, the processor may include a central processing unit (CPU), a micro processor unit (MPU), a micro controller unit (MCU), a graphic processing unit )GPU), or any other form of processor (see specification par. [0110]). The memory may be implemented as volatile memory, such as RAM, but is not limited thereto (see specification par. [0111]). The computing device may include any device having computing (processing) functions (see specification par. [0054]). The predictive model may be designed and implemented based on deep learning/machine learning models such as an artificial neural network (see 60 in FIG. 6), logistic regression, light gradient boosting machine (LGBM), naive bayes, support vector machine, decision tree, random forest, and the like. However, the scope of the present disclosure is not limited by these examples, and the predictive model may be implemented based on other types of models (e.g., deep learning models such as a convolutional neural network, recurrent neural network, transformer, etc.) (See specification par. [0078]).
The additional elements do not show an improvement to the functioning of a computer or to any other technology, rather the additional elements perform general computing functions and do not indicate how the particular combination improves any technology or provides a technical solution to a technical problem. Accordingly, these additional elements, when considered separately and as an ordered combination, do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea. Therefore, Claims 1, 12, and 14 are directed to an abstract idea without practical application. (Step 2A – Prong 2: No, the additional elements are not integrated into a practical application).
Step 2B analysis:
As discussed above in “Step 2A analysis – Prong 2”, the identified additional elements in Independent Claims 1, 12, and 14 are equivalent to adding the words “apply it” on a generic computer, and/or generally link the use of the judicial exception to a particular technological environment or field of use. Therefore, the claims as a whole do not amount to significantly more than the judicial exception itself.
For the role of a computer in a computer implemented invention to be deemed meaningful in the context of this analysis, it must involve more than performance of “well- understood, routine, [and] conventional activities previously known to the industry.” Further, “the mere recitation of a generic computer cannot transform a patent ineligible abstract idea into a patent-eligible invention.”
The applicant’s specification discloses: Specifically, the processor may include a central processing unit (CPU), a micro processor unit (MPU), a micro controller unit (MCU), a graphic processing unit )GPU), or any other form of processor (see specification par. [0110]). The memory may be implemented as volatile memory, such as RAM, but is not limited thereto (see specification par. [0111]). The computing device may include any device having computing (processing) functions (see specification par. [0054]). The predictive model may be designed and implemented based on deep learning/machine learning models such as an artificial neural network (see 60 in FIG. 6), logistic regression, light gradient boosting machine (LGBM), naive bayes, support vector machine, decision tree, random forest, and the like. However, the scope of the present disclosure is not limited by these examples, and the predictive model may be implemented based on other types of models (e.g., deep learning models such as a convolutional neural network, recurrent neural network, transformer, etc.) (See specification par. [0078]).
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because, as discussed above with respect to integration of the abstract idea into a practical application, using the additional elements to perform the steps for predicting acute kidney injury amounts to no more than using computer related devices to implement the abstract idea.
The use of a computer or processor to merely automate or implement the abstract idea cannot provide significantly more than the abstract idea itself. (See MPEP 2106.05(f) where mere instructions to apply an exception does not render an abstract idea patent eligible). There is no indication that the additional limitations alone or in combination improves the functioning of a computer, improves another technology or technical field, or effects a transformation or reduction of a particular article to a different state or thing. Therefore, the claims are not patent eligible. The Examiner has therefore determined that no additional element, or combination of additional claims elements is/are sufficient to ensure the claims amount to significantly more than the abstract idea identified above (Step 2B: Independent claims - NO).
Dependent Claims
Dependent Claims 2-4, 6, 10-11, 13 and 15-16 are directed towards elements used to describe the independent variables, preparing the dataset, and treating a patient. In particular, these elements include: the preoperative examination items; variables relating to disease history, medication history, and types and duration of surgeries undergone; removing patient data based on a predetermined kidney-related condition; augmenting and correcting the datasets, and treating a subject by administering a therapeutic regimen. These elements describe managing personal behavior or relationships or interactions between people including following rules or instructions, and therefore fall within the scope of certain methods of organizing human activity. Specifically, the dependent claims recite steps that involve gathering data from a patient before and after surgery.
The elements as recited above also falls within the “mental processes” grouping of abstract
ideas, and describes concepts that can be performed in the human mind through observation,
evaluation, judgement, and opinion. Removing patient data and augmenting the datasets are all tasks that can be performed in the human mind. Therefore, the dependent claims fall within the same abstract idea of a mental process as the independent claims.
This judicial exception is not integrated into a practical application. Specifically, the dependent
claims do not recite any additional elements beyond the abstract idea and the limitations do not impose any meaningful limits on practicing the abstract idea. Therefore, the dependent claims do not integrate the abstract idea into a practical application and do not provide significantly more than the abstract idea (see MPEP 2106.05(f)).
The Examiner has therefore determined that no additional element, or combination of
additional claims elements is/are sufficient to ensure the claims amount to significantly more than the
abstract idea identified above.
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.
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.
Claims 12-13 are rejected under 35 U.S.C. 103 as being unpatentable over Lee et al. (US 2022/0008018) (Hereinafter Lee) in view of Demirjian (US 2013/0330829).
Regarding Claim 12, Lee teaches the following:
A method of predicting an occurrence of acute kidney injury ([0068] a method for predicting a risk of acute kidney injury), comprising:
acquiring a model trained to predict a risk of an occurrence of postoperative acute kidney injury ([0089] a sequential order to construct a prediction model that includes three outcome classifications as follows: "No AKI", "Low-stage AKI" and "Critical AKI) - wherein the model is trained using a dataset of a plurality of patients, a dependent variable of the dataset is the occurrence of postoperative acute kidney injury ([0045] factors associated with an occurrence of acute kidney injury), and independent variables of the dataset include variables relating to preoperative examination items of the patients ([0069], [0086]: non-cardiac surgery preclinical data of patients subjected to non-cardiac surgery as variables; Information that could be collected or planned before surgery was included… Detailed information on the collected variables); and
predicting a risk of an occurrence of acute kidney injury to a specific patient after a target surgery using the trained model ([0006] predicting a risk of acute kidney injury after non-cardiac surgery),
wherein the model is configured to output a quantitative risk score (Lee Claims 5-7: the prediction of the risk of acute kidney injury is classified into total four (4) grades including A, B, C, and D, based on a sum of indexes determined according to the index set preset for each variable. Grade A is classified when the sum of the indexes is less than 20, the grade B is classified when the sum of the indexes is 20 or more and less than 40, the grade C is classified when the sum of the indexes is 40 or more and less than 60, and the grade D is classified when the sum of the indexes is 60 or more), and
outputting a risk score representing a probability of postoperative acute kidney injury occurrence ([0089] a sequential order to construct a prediction model that includes three outcome classifications as follows: "No AKI", "Low-stage AKI" and "Critical AKI).
However, Lee does not explicitly disclose the following which is met by Demirjian:
at least one computing device (See Demirjian [0011], [0016], [0024]: a computer system that can be employed to implement systems and methods described herein)
wherein the model is a machine learning model comprising at least one of a neural network or a light gradient boosting machine (LGBM) ([0015] the predictive model can be implemented as any appropriate classification or regression model, such as a polynomial model provided via least squares regression procedure, an artificial neural network, a statistical classifier, a support vector machine, or other, similar model) configured to output [a predicted occurrence of acute kidney injury] through nonlinear weight correlations among heterogeneous preoperative variables and postoperative outcomes ([0016] since the features themselves can be non-linear functions of the measured concentrations and rates of change, the results of the predictive model can be a non-linear function of the measured concentrations and rates of change), and
wherein the predicting comprises normalizing input data of the specific patient’s preoperative examination results and surgery-related parameters according to the trained model parameters (Demirjian Claim 1, [0019]: an input interface configured to receive a plurality of features derived from the results of a post-surgical metabolic blood panel and one of a pre-surgical metabolic blood panel and a presurgical metabolic blood panel; the system uses laboratory data from a first postoperative metabolic panel to calculate a change in serum creatinine (ΔCr) and blood urea nitrogen (ΔBUN) compared to values from a preoperative or perioperative metabolic blood panel. Each of ΔCr, and ΔBUN can be normalized.), and
It would have been obvious to one of ordinary skill in the art before the effective filing date to have combined the method as taught by Lee to include a computing device and a machine learning model that predicts a risk based on nonlinear correlations among the preoperative variables, as described in Demirjian, because the inclusion of different elements in training the machine learning model contributes to the predictive capacity of the model because they capture clinically relevant downstream complications of renal injury in a multidimensional approach (See Demirjian [0019]).
Regarding Claim 13, the combination of Lee and Demirjian teaches the method of claim 12, and Lee further teaches:
The method of claim 12, wherein the independent variables of the dataset further comprise variables regarding types and durations of surgeries undergone by the patients ([0087] data were collected in regard to the actual surgical duration (hours) and expected surgical duration (hours) entered by the physician who attended the collection prior to performing the surgery), and
wherein the predicting comprises:
constituting input data based on a type and duration of the target surgery ([0048], [0057]: The factors associated with the occurrence of acute kidney injury after non-cardiac surgery may be collected; the unit of the expected surgical duration among the variables may be time (hour), and the expected surgical duration may be defined to an index of 5 times the corresponding period.), and examination results of the specific patient for the preoperative examination items ([0051] The factors associated with the occurrence of acute kidney injury after non-cardiac surgery may include at least one selected from the group consisting of age, estimated glomerular filtration rate ( eGFR), dipstick albuminuria, sex, expected surgical duration, emergency operation, diabetes mellitus, use of renin-aldosterone-angiotensin-system blocker (use of RAAS blocker), hypoalbuminemia, anemia and hyponatremia); and
predicting the risk by inputting the input data into the trained model ([0050] the variable selection unit may construct a multivariable model in regard to ordinal variables composed of negative prognoses related to the acute kidney injury using the proportional odds regression technique with the selected variables and, at the same time, may preset an index set for each variable so that a sum of the indexes preset in each variable reflects the risk by converting model coefficients into an integer).
Claims 1, 4, 6, 10-11, and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Lee et al. (US 2022/0008018) (Hereinafter Lee) in view of Demirjian (US 2013/0330829), further in view of Park et al., Predicting acute kidney injury in cancer patients using heterogeneous and irregular data (Hereinafter Park).
Regarding Claim 1, Lee discloses the following:
A method of predicting an occurrence of acute kidney injury ([0068] a method for predicting a risk of acute kidney injury), comprising:
preparing a dataset of a plurality of patients - wherein a dependent variable of the dataset relates to an occurrence of postoperative acute kidney injury ([0045] factors associated with an occurrence of acute kidney injury), and independent variables of the dataset include variables relating to preoperative examination items of the patients ([0069], [0086]: non-cardiac surgery preclinical data of patients subjected to non-cardiac surgery as variables; Information that could be collected or planned before surgery was included… Detailed information on the collected variables) - ; and
building a model configured to predict a risk of the occurrence of postoperative acute kidney injury using the prepared dataset ([0089] a sequential order to construct a prediction model that includes three outcome classifications as follows: "No AKI", "Low-stage AKI" and "Critical AKI).
wherein the [trained] model predicts a quantitative risk score of the postoperative acute kidney injury for an individual patient (Lee Claims 5-7: the prediction of the risk of acute kidney injury is classified into total four (4) grades including A, B, C, and D, based on a sum of indexes determined according to the index set preset for each variable. Grade A is classified when the sum of the indexes is less than 20, the grade B is classified when the sum of the indexes is 20 or more and less than 40, the grade C is classified when the sum of the indexes is 40 or more and less than 60, and the grade D is classified when the sum of the indexes is 60 or more).
However, Lee does not explicitly disclose the following which is met by Demirjian:
at least one computing device (See Demirjian [0011], [0016], [0024]: a computer system that can be employed to implement systems and methods described herein)
wherein the building of the model comprises training ([0017] During a training process, records can be retrieved from an electronic health records (EHR) database via a database interface, and predictor variables and the outcomes can be extracted from these records at an associated feature extractor), by the computing device, a machine learning model including at least one of a neural network, logistic regression, and a light gradient boosting machine (LGBM) ([0015] the predictive model can be implemented as any appropriate classification or regression model, such as a polynomial model provided via least squares regression procedure, an artificial neural network, a statistical classifier, a support vector machine, or other, similar model) by optimizing weight parameters among heterogeneous preoperative variables and postoperative outcome data to minimize prediction loss caused by data imbalance or missing values ([0015], [0019]: each of the plurality of features has an associated weight, and the predictive model provides a linear combination of the plurality of features. The model can also be an extended model which includes preoperative values of serum creatinine, serum sodium, potassium, bicarbonate, and albumin which contributes to the predictive capacity of the model.); and
wherein the trained model predicts a quantitative risk score of the postoperative acute kidney injury for an individual patient based on nonlinear correlations among the preoperative variables ([0016] since the features themselves can be non-linear functions of the measured concentrations and rates of change, the results of the predictive model can be a non-linear function of the measured concentrations and rates of change).
It would have been obvious to one of ordinary skill in the art before the effective filing date to have combined the method as taught by Lee to include a computing device and training of a machine learning model that predicts a risk based on nonlinear correlations among the preoperative variables, as described in Demirjian, because the inclusion of different elements in training the machine learning model contributes to the predictive capacity of the model because they capture clinically relevant downstream complications of renal injury in a multidimensional approach (See Demirjian [0019]).
However, the combination of Lee and Demirjian does not disclose the following that is met by Park:
wherein the preparing of the dataset comprises removing patient data satisfying predetermined kidney-related or surgery-related conditions from an original patient dataset (Park Pg. 6, par. 3-9: to maximize the predictive performance, they exclude some variables or some user’s data based on the following criteria: an estimated glomerular filtration rate (eGFR) value of less than a certain threshold, whether a patient had hemodialysis, peritoneal dialysis, continuous renal replacement therapy (CRRT), kidney transplantation (KTPL), etc.), correcting outliers (Park Pg. 8, par. 5: apply squeezing and normalization to the feature vectors for postprocessing. First, we compute the lower and upper bounds for each feature, and set the value of those that lie outside of the boundary to the boundary value. This is to reduce the impact of outlier values), imputing missing values using multiple imputation by chained equations (Park Pg. 6, par. 9: perform multiple imputation for the missing values of the remaining variables. Since the decision to measure biomarkers is based on existing patient records, and data are not missing completely at random in general, we assume that missing values in our dataset are missing at random; under this assumption, we apply multivariate imputation by chained equations (MICE) [24] to the non-temporal data of entire patients), and normalizing variables to generate a high-quality training dataset (Park Pg. 8, par. 5: apply squeezing and normalization to the feature vectors for postprocessing);
It would have been obvious to one of ordinary skill in the art before the effective filing date to have combined the prediction system as taught by Lee and Demirjian with the correction techniques outlined in Park because preprocessing the dataset makes it available as an input to the regression algorithms, as well as maximized their predictive performance (See Park Pg. 6, par. 3).
Regarding Claim 4, the combination of Lee, Demirjian, and Park teaches the method of claim 1, and Lee further discloses:
The method of claim 1, wherein the independent variables of the dataset further comprise variables regarding types and duration of surgeries undergone by the patients ([0087] data were collected in regard to the actual surgical duration (hours) and expected surgical duration (hours) entered by the physician who attended the collection prior to performing the surgery).
Regarding Claim 6, the combination of Lee, Demirjian, and park teaches the method of claim 1, and Lee further discloses:
The method of claim 1, wherein the predetermined kidney-related condition of surgery-related conditions comprise at least one of:
A history of renal replacement therapy ([0084] patients with pre-operative renal dysfunction which is defined by: history of kidney replacement therapy;
A preoperative eGFR value below a predetermined threshold ([0084] patients with pre-operative renal dysfunction which is defined by: estimated glomerular filtration rate (eGFR) of 15 mL/min/1.73 m2);
A preoperative creatinine (Cr) level above a predetermined threshold or a degree of elevation of the creatinine (Cr) level within a predetermined period of time prior to surgery ([0084] patients with pre-operative renal dysfunction which is defined by: preoperative serum creatinine (sCr) level of 4 mg/dL or higher or an increase in baseline of sCr by 0.3 mg/dL or more or 1.5 times or more from the minimum value 2 weeks prior to surgery); or
A surgery type or surgery duration satisfying a predefined exclusion criterion ([0084] Exclusion criteria includes cardiac surgery, surgery of a deceased patient (e.g., transplantation of a deceased donor), patients with nephrectomy or kidney transplantation, and small surgical procedures defined with the surgical duration of less than 1 hour).
Regarding Claim 10, the combination of Lee, Demirjian, and Park teaches the method of claim 1, and Park further discloses:
The method of claim 1, wherein the preparing of the dataset comprises:
correcting for outliers in the original patient dataset (Park Pg. 8, par. 5: apply squeezing and normalization to the feature vectors for postprocessing. First, we compute the lower and upper bounds for each feature, and set the value of those that lie outside of the boundary to the boundary value. This is to reduce the impact of outlier values);
correcting for missing values in the original patient dataset using multiple imputation by chained equations (Park Pg. 6, par. 9: perform multiple imputation for the missing values of the
remaining variables. Since the decision to measure biomarkers is based on existing patient records, and data are not missing completely at random in general, we assume that missing values in our dataset are missing at random; under this assumption, we apply multivariate imputation by chained equations (MICE) [24] to the non-temporal data of entire patients.); and
normalizing the original patient dataset corrected for the outliers and the missing values (Park Pg. 8, par. 5: apply squeezing and normalization to the feature vectors for postprocessing).
It would have been obvious to one of ordinary skill in the art before the effective filing date to have combined the prediction system and taught by Lee/Demirjian with the correction techniques outlined in Park because preprocessing the dataset makes it available as an input to the regression algorithms, as well as maximized their predictive performance (See Park Pg. 6, par. 3)
Regarding Claim 11, the combination of Lee, Demirjian, and Park teaches the method of claim 1, and Lee further discloses:
The method of claim 1, wherein the preparing of the dataset comprises:
acquiring an original patient dataset - wherein the original patient dataset includes a first dataset for a patient group that has an occurrence of postoperative acute kidney injury and a second dataset for a patient group that does not have an occurrence of postoperative acute kidney injury – ([0089] The term "PO-AKI" as used herein includes all AKIs regardless of AKI severity. In order to address the severity and patient oriented outcomes of PO-AKI, the inventors defined results representing a sequential order to construct a prediction model that includes three outcome classifications as follows: "No AKI", "Low-stage AKI" and "Critical AKI".) ; and
augmenting the first dataset (Lee discloses in [0101] and [0102] several exclusion criteria that is used to augment the data for the model. This results in A total of 49,803 and 29,715 cases, respectively, in the discovery cohort and the validation cohort with complete information of the finally selected variables were used for further analysis in order to construct and verify simplified models (FIG. 2).
Regarding Claim 14, Lee discloses the following:
A system for predicting an occurrence of acute kidney injury ([0044] The present invention provides a system for predicting a risk of acute kidney injury) comprising:
acquiring a model trained to predict a risk of an occurrence of postoperative acute kidney injury ([0089] a sequential order to construct a prediction model that includes three outcome classifications as follows: "No AKI", "Low-stage AKI" and "Critical AKI) - wherein the model is trained using a dataset of a plurality of patients, a dependent variable of the dataset relates to the occurrence of postoperative acute kidney injury ([0045] factors associated with an occurrence of acute kidney injury), and independent variables of the dataset include variables relating to preoperative examination items of the patients ([0069], [0086]: non-cardiac surgery preclinical data of patients subjected to non-cardiac surgery as variables; Information that could be collected or planned before surgery was included… Detailed information on the collected variables) - ; and
predicting a risk of an occurrence of acute kidney injury to a patient after a target surgery using the trained model ([0050] the variable selection unit may construct a multivariable model in regard to ordinal variables composed of negative prognoses related to the acute kidney injury using the proportional odds regression technique with the selected variables and, at the same time, may preset an index set for each variable so that a sum of the indexes preset in each variable reflects the risk by converting model coefficients into an integer),
to execute a trained machine learning model configured to output a quantitative risk score through [nonlinear correlations] among preoperative variables representing the likelihood of postoperative acute kidney injury (Lee Claims 5-7: the prediction of the risk of acute kidney injury is classified into total four (4) grades including A, B, C, and D, based on a sum of indexes determined according to the index set preset for each variable. Grade A is classified when the sum of the indexes is less than 20, the grade B is classified when the sum of the indexes is 20 or more and less than 40, the grade C is classified when the sum of the indexes is 40 or more and less than 60, and the grade D is classified when the sum of the indexes is 60 or more).
However, Lee does not disclose the following that is met by Demirjian:
one or more processors ([0025] The computer system 200 includes a processor); and
a memory ([0025] a system memory) configured to store one or more instructions ([0026] The
long-term data storage 210 components provide nonvolatile storage of data, data structures, and computer-executable instructions for the computer system),
wherein the one or more processors ([0025] The computer system 200 includes a processor) perform:
executing the stored one or more instructions (computer-executable instructions for the computer system),
to execute a trained machine learning model configured to output a quantitative risk score through nonlinear correlations among preoperative variables representing the likelihood of postoperative acute kidney injury ([0016] since the features themselves can be non-linear functions of the measured concentrations and rates of change, the results of the predictive model can be a non-linear function of the measured concentrations and rates of change).
It would have been obvious to one of ordinary skill in the art before the effective filing date to have combined the method as taught by Lee to include a computing device including a processor and a memory as described in Demirjian, because the claimed invention is only a combination of these well-known elements which would have performed the same function in combination as each did separately. Lee already discloses various units that perform the functions of predicting acute kidney disease, and combining the units of Lee with the computing system of Demirjian would perform the same function of predicting the disease. Therefore, the results would have been predictable to one of ordinary skill in the art (MPEP 2143). Additionally, including training of a machine learning model that predicts a risk based on nonlinear correlations among the preoperative variables, as described in Demirjian, contributes to the predictive capacity of the model because they capture clinically relevant downstream complications of renal injury in a multidimensional approach (See Demirjian [0019]).
However, the combination of Lee and Demirjian does not disclose the following that is met by Park:
wherein the one or more processors are configured to preprocess patient data by removing outliers (Park Pg. 8, par. 5: apply squeezing and normalization to the feature vectors for postprocessing. First, we compute the lower and upper bounds for each feature, and set the value of those that lie outside of the boundary to the boundary value. This is to reduce the impact of outlier values), imputing missing values by multiple imputation by chained equations (Park Pg. 6, par. 9: perform multiple imputation for the missing values of the remaining variables. Since the decision to measure biomarkers is based on existing patient records, and data are not missing completely at random in general, we assume that missing values in our dataset are missing at random; under this assumption, we apply multivariate imputation by chained equations (MICE) [24] to the non-temporal data of entire patients), and normalizing variables to generate input data (Park Pg. 8, par. 5: apply squeezing and normalization to the feature vectors for postprocessing); and
It would have been obvious to one of ordinary skill in the art before the effective filing date to have combined the prediction system as taught by Lee and Demirjian with the correction techniques outlined in Park because preprocessing the dataset makes it available as an input to the regression algorithms, as well as maximized their predictive performance (See Park Pg. 6, par. 3).
Claims 2-3 are rejected under 35 U.S.C. 103 as being unpatentable over Lee et al. (US 2022/0008018) (Hereinafter Lee) in view of Demirjian (US 2013/0330829), in further view of Park et al., Predicting acute kidney injury in cancer patients using heterogeneous and irregular data (Hereinafter Park), in further view of Sato et al. (WO 2019/026918) (Hereinafter Sato).
Regarding Claim 2, the combination of Lee and Demirjian teaches the method of claim 1, and Lee further discloses:
The method of claim 1, wherein the preoperative examination items comprise albumin ([0087] A serum albumin level), protein ([0087] The presence of baseline proteinuria as another kidney function variable was confirmed by a simple dipstick test.).
However, Lee does not disclose the following that is met by Demirjian:
wherein the preoperative examination items comprise creatinine (Cr) ([0022] a preoperative or perioperative level of serum creatinine is determined from the first blood serum sample), potassium (Fig 3., [0005], [0014]: Determine at least a preoperative creatinine level from the isolated first serum sample; the substances of interest can include two or more of serum creatinine, blood urea nitrogen, potassium)
It would have been obvious to one of ordinary skill in the art before the effective filing date to have combined the preoperative examination items described in Lee with the preoperative creatinine and potassium levels, as mentioned by Demirjian, because The inclusion of the additional elements in the panel such as potassium, sodium, and bicarbonate also independently contribute to the predictive capacity of the model because they capture clinically relevant downstream complications of renal injury in a multidimensional approach (See Demirjian [0019]).
However, the combination of Lee and Demirjian does not disclose the following that is met by Sato:
wherein the preoperative examination items comprise urinary specific gravity (Sato Pg. 4, par. 5: chronic kidney disease can be predicted from the average daily water intake or specific gravity of urine).
It would have been obvious to one of ordinary skill in the art before the effective filing date to have modified the combination of Lee and Demirjian to include the urinary specific gravity mentioned by Sato because the claimed invention is only a combination of these well-known elements which would have performed the same function in combination as each did separately. Since Lee/Demirjian already discloses the prediction of a disease using certain biomarkers, simply including urinary specific gravity into the preoperative data pool would perform the same function of predicting the disease. Therefore, the results would have been predictable to one of ordinary skill in the art (MPEP 2143).
Regarding Claim 3, the combination of Lee and Demirjian discloses the method of claim 1, and Lee further teaches:
The method of claim 1, wherein the independent variables of the dataset further comprise variables relating to disease history ([0087] Co-morbidities of heart disease were collected, which include a history of heart failure, coronary artery disease (e.g., angina or myocardial infarction), hypertension and diabetes) and medication history of the patients,
wherein the disease history comprises history of hypertension (HTN), cardiovascular disease (CVD),
wherein the medication history relates to antihypertensive drugs ([0087] pre-operative use of renin-aldosterone-angiotensin-system blockers was included in the variables of the present invention).
However, Lee/Demirjian does not disclose the following that is met by Sato:
wherein the disease history comprises history of chronic kidney disease (CKD), chronic obstructive pulmonary disease (COPD), and liver cirrhosis (LC) (Sato Pg. 4, par. 9; Pg. 16, par. 12: The subject may or may not be an individual having a history of impaired renal function or other renal disease).
It would have been obvious to one of ordinary skill in the art before the effective filing date to have modified the combination of Lee and Demirjian to include the various disease histories because the claimed invention is only a combination of these well-known elements which would have performed the same function in combination as each did separately. Since Lee/Demirjian already discloses the prediction of a disease including a disease history of HTN and CVD, simply including various other diseases into the data pool would perform the same function of predicting the disease, and would produce the same result. Therefore, the results would have been predictable to one of ordinary skill in the art (MPEP 2143).
Claims 15-16 are rejected under 35 U.S.C. 103 as being unpatentable over Lee et al. (US 2022/0008018) (Hereinafter Lee) in view of Demirjian (US 2013/0330829), in further view of Sato et al. (WO 2019/026918) (Hereinafter Sato), in further view of Park et al., Predicting acute kidney injury in cancer patients using heterogeneous and irregular data (Hereinafter Park), further in view of Munoz et al. (WO 2018/223006) (Hereinafter Munoz).
Regarding Claim 15, the combination of Lee, Demirjian, and Park teaches the method according to claim 1, however, they each fail to teach the following which is met by Munoz:
The method according to claim 1, further comprising treating a subject based on the predicted risk of occurrence of acute kidney injury by administering a therapeutic regimen ([0018] The present invention relates to methods of inhibiting the development of acute kidney injury comprising: (a) determining the risk profile value of claim 1, and (b) administering medical treatment to the human subject prior to the onset of acute kidney injury symptoms. In some embodiments, the medical treatment is an administration of IV fluids or an administration of supplemental oxygen).
It would have been obvious to one of ordinary skill in the art before the effective filing date to have combined the teachings of Lee, Demirjian, and Park with the treatment of a subject as taught by Munoz because the administration of a treatment regimen may be used to further monitor a subject’s risk profile over time, and can be used to assess the effectiveness of treatments for acute kidney injury. The subject’s risk profile would be able to be assessed before, during, and after treatment (See Munoz [0068]).
Regarding Claim 16, the combination of Lee, Demirjian, and Park teaches the method according to claim 11, however, they each fail to teach the following which is met by Munoz:
The method according to claim 11, further comprising treating a subject based on the predicted risk of occurrence of acute kidney injury by administering a therapeutic regimen ([0018] The present invention relates to methods of inhibiting the development of acute kidney injury comprising: (a) determining the risk profile value of claim 1, and (b) administering medical treatment to the human subject prior to the onset of acute kidney injury symptoms. In some embodiments, the medical treatment is an administration of IV fluids or an administration of supplemental oxygen).
It would have been obvious to one of ordinary skill in the art before the effective filing date to have combined the teachings of Lee, Demirjian, and Park with the treatment of a subject as taught by Munoz because the administration of a treatment regimen may be used to further monitor a subject’s risk profile over time, and can be used to assess the effectiveness of treatments for acute kidney injury. The subject’s risk profile would be able to be assessed before, during, and after treatment (See Munoz [0068]).
Relevant Prior Art of Record Not Currently Being Applied
The relevant art made of record and not relied upon is considered pertinent to applicant’s disclosure.
Horsch et al. (WO 2017060525) discloses a method for predicting acute kidney injury by measuring the amount of IGFBP7 and Cystatin C in a sample collected before surgery. They use several other biomarkers for the prediction, including creatinine.
Yuan et al. (CN 110827992) discloses a preoperative method for predicting acute renal injury using a logistic regression model.
Response to Arguments
Applicant's arguments filed 10/28/2025 with regards to 35 U.S.C. 101 have been fully considered but they are not persuasive. Applicant argues the claims are directed to a patent-eligible application of computer technology, not an abstract idea. Examiner respectfully disagrees. Applicant argues that the operations of removing patient data satisfying predetermined conditions, correcting outliers, imputing missing values using multiple imputation by chained equations, and normalizing variables cannot be performed in the human mind, however, the examiner respectfully disagrees. Each of these functions recite concepts that can be performed in the human mind through observation, evaluation, judgement, and opinion with or without the use of a physical aid. Further, the step of training a machine learning model would be part of the abstract idea of mathematical concepts since it uses “chained equations” from the dataset and “normalizing variables” (See Example 47, claim 2 for reference). Additionally, the applicant argues the claims integrate the abstract idea into a practical application by providing a technological improvement, however, examiner respectfully disagrees. The data cleaning steps are part of the abstract idea, and therefore cannot integrate the abstract idea into a practical application or provide significantly more.
Applicant’s arguments, see Pg. 11-14 of applicant’s remarks, filed 10/28/2025, with respect to the rejections of claims 1, 4, 6, and 10-14 under 35 U.S.C. 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new grounds of rejection is made in view of Park et al., further in view of Munoz et al. Applicant argues Lee does not disclose removing data based on predefined conditions, correcting outliers or imputing missing values using MICE, normalizing datasets, or training a non-linear machine-learning model. By combining Lee with Park, the combination of these elements would have been obvious to one of ordinary skill in the art (See Park Pg. 6 and 8).
Conclusion
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 ALEXIS K VAN DUZER whose telephone number is (571)270-5832. The examiner can normally be reached Monday thru Thursday 8-5 CT.
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, Marc Jimenez can be reached at (571) 272-4530. 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.
/A.K.V./Examiner, Art Unit 3681
/MARC Q JIMENEZ/Supervisory Patent Examiner, Art Unit 3681