DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on May 16 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea (mental process and mathematical concept) without significantly more.
Claim 1:
Regarding claim 1, in step 1 of the 101-analysis set forth in MPEP 2106, the claim recites “A computer-implemented method for graph analysis”, and method or process is one of the four statutory categories of invention.
In step 2A prong 1 of the 101-analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
“identifying trainable control parameters of a graph refinement function;” (this is a mental process, a person could mentally evaluate identifying what parameters of a refinement function are trainable based on a mentally determined threshold, see MPEP § 2106.04(a)(2)(III)),
“generating a plurality of sample graph refinements of an input graph, using control parameters sampled from a variational distribution;” (this is a mental process, a person can also generate a plurality of sample graph refinements mentally using control parameters, see MPEP § 2106.04(a)(2)(III)),
“selecting graph refinement control parameters associated with a sample of the plurality of sample graph refinements that has a highest performance score when used…” (this is a mental process, a person can mentally evaluate graph refinement control parameters to select that are associated with a sample of the plurality of sample graph refinements that has a mentally calculated highest performance score, see MPEP § 2106.04(a)(2)(III)),
“performing graph analysis on the input graph using the selected graph refinement parameters to produce a refined graph on new test samples;” (this is a mental process, a person can mentally analyze a graph using the mentally selected refinement parameters to produce a new graph that can be done via pen and paper on new test samples, see MPEP § 2106.04(a)(2)(III)),
“and performing an action responsive to the graph analysis.” (this is a mental process, a person can do an action that’s responsive the graph analysis, see MPEP § 2106.04(a)(2)(III)),
If claim limitations, under the broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process and mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A prong 2 of the 101-analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
“A computer-implemented method for graph analysis, comprising:” (A computer being used to implement a method is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)),
“…to train a graph neural network (GNN);” (Training a GNN is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)),
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101-analysis set forth in the 2019 PEG, the examiner has determined that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above, additional element vi recites generic computer component being used as tool to perform functions of the judicial exception, and additional element vii recites mere instructions to apply an exception using generic computer, which is not indicative of significantly more.
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
Claim 2:
Regarding claim 2, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 2 recites the following additional elements:
“The method of claim 1, further comprising training the GNN and graph refinement parameters using the plurality of samples using smoothing-based optimization and local gradient descent.” (this is a mathematical concept, using smoothing-based optimization and local gradient descent to train a GNN and graph refinement parameters is a mathematical calculation and formula, see Specification paragraphs [0027-0033], see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept but for the recitation of generic computer components, then it falls within the mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 3:
Regarding claim 3, it is dependent upon claim 2, and thereby incorporates the limitations of, and corresponding analysis to claim 2. Further, claim 3 recites the following additional elements:
“The method of claim 2, wherein the performance score is a weighted average of a mean squared error and a Pearson correlation coefficient for a graph regression task.” (this is a mathematical concept, a performance score being a weighted average of a mean squared error and a Pearson correlation coefficient is a mathematical calculation, see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept but for the recitation of generic computer components, then it falls within the mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 4:
Regarding claim 4, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 4 recites the following additional elements:
“The method of claim 1, wherein the input graph is a graph indicating a spatial proximity of capturing spots, with nodes that represent capturing spots and edges that represent spatial adjacency.” (this is a mathematical concept, a graph that indicates spatial proximity is a mathematical relationship, see Specification paragraph [0020], see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept but for the recitation of generic computer components, then it falls within the mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 5:
Regarding claim 5, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 5 recites the following additional elements:
“The method of claim 1, wherein the graph analysis includes predicting spatially resolved gene expression via tissue morphology in hematoxylin and eosin (H&E) stained images.” (In step 2A, prong 2, this is considered as mere field of use or technological environment in which to apply a judicial exception, see MPEP § 2106.05(h)). (In step 2B, this is also considered mere field of use or technological environment in which to apply a judicial exception – see MPEP § 2106.05(h)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 6:
Regarding claim 6, it is dependent upon claim 5, and thereby incorporates the limitations of, and corresponding analysis to claim 5. Further, claim 6 recites the following additional elements:
“The method of claim 5, wherein the predicted spatially resolved gene expression relates to HER2 positivity and wherein the action includes a treatment for breast cancer.” (this is a mental process, a person can mentally evaluate a predicted spatially resolved gene expression being in relation to HER2 positivity and an action that includes a treatment for breast cancer, see MPEP § 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 7:
Regarding claim 7, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 7 recites the following additional elements:
“The method of claim 1, wherein the graph neural network is a machine learning model that generates a prediction to aid in medical decision making.” (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 8:
Regarding claim 8, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 8 recites the following additional elements:
“The method of claim 1, wherein the action includes automatically altering a patient’s treatment.” (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 9:
Regarding claim 9, it is dependent upon claim 1, and thereby incorporates the limitations of, and corresponding analysis to claim 1. Further, claim 9 recites the following additional elements:
“The method of claim 1, wherein the graph refinement function applies a threshold to linear projections of node features, using control parameters
PNG
media_image1.png
28
100
media_image1.png
Greyscale
:
PNG
media_image2.png
38
170
media_image2.png
Greyscale
, where
PNG
media_image3.png
30
404
media_image3.png
Greyscale
, and wherein message-passing on the GNN uses updates implemented as
PNG
media_image4.png
102
320
media_image4.png
Greyscale
for levels l < L, where l is a layer, L a number of layers in the GNN, m and n are nodes, E’ is a set of edges, Wl is weight matrix for the layer l, σ(x) = max(0,x) is a rectified linear unit (ReLU) function, and deg(•) is a degree of a node.” (this is a mathematical concept, the graph refinement function is a mathematical formula as shown, see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept but for the recitation of generic computer components, then it falls within the mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Claim 10:
Regarding claim 10, it is dependent upon claim 9, and thereby incorporates the limitations of, and corresponding analysis to claim 9. Further, claim 10 recites the following additional elements:
“The method of claim 9, wherein generating the plurality of sample graph refinements includes sampling a Gaussian distribution with mean and standard deviation parameters that are iteratively determined by smoothing-based optimization.” (this is a mathematical concept, the generating plurality of sample graph refinements that includes sampling a Gaussian distribution with mean and standard deviation parameters is a mathematical calculation, see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept but for the recitation of generic computer components, then it falls within the mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Claim 11:
Regarding claim 11, in step 1 of the 101-analysis set forth in MPEP 2106, the claim recites “A system for graph analysis, comprising:”, and system or machine is one of the four statutory categories of invention.
In step 2A prong 1 of the 101-analysis set forth in the MPEP 2106, the examiner has determined that the following limitations recite a process that, under the broadest reasonable interpretation, covers a mental process but for recitation of generic computer components:
“identify trainable control parameters of a graph refinement function;” (this is a mental process, a person could mentally evaluate identifying what parameters of a refinement function are trainable based on a mentally determined threshold, see MPEP § 2106.04(a)(2)(III)),
“generate a plurality of sample graph refinements of an input graph, using control parameters sampled from a variational distribution;” (this is a mental process, a person can also generate a plurality of sample graph refinements mentally using control parameters, see MPEP § 2106.04(a)(2)(III)),
“select graph refinement control parameters associated with a sample of the plurality of sample graph refinements that has a highest performance score when used…” (this is a mental process, a person can mentally evaluate graph refinement control parameters to select that are associated with a sample of the plurality of sample graph refinements that has a mentally calculated highest performance score, see MPEP § 2106.04(a)(2)(III)),
“perform graph analysis on the input graph using the selected graph refinement parameters to produce a refined graph on new test samples;” (this is a mental process, a person can mentally analyze a graph using the mentally selected refinement parameters to produce a new graph that can be done via pen and paper on new test samples, see MPEP § 2106.04(a)(2)(III)),
“and perform an action responsive to the graph analysis.” (this is a mental process, a person can do an action that’s responsive the graph analysis, see MPEP § 2106.04(a)(2)(III)),
If claim limitations, under the broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process and mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
In step 2A prong 2 of the 101-analysis set forth in MPEP 2106, the examiner has determined that the following additional elements do not integrate this judicial exception into a practical application:
“A system for graph analysis, comprising: a hardware processor;” (A system comprising a hardware processor is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)),
“and a memory that stores a computer program which, when executed by the hardware processor, causes the hardware processor to:” (Using a memory is considered generic computer component being used as tool to perform functions of the judicial exception – see MPEP § 2106.05(f)),
“…to train a graph neural network (GNN);” (Training a GNN is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)),
Since the claim as a whole, looking at the additional elements individually and in combination, does not contain any other additional elements that are indicative of integration into a practical application, the claim is “directed” to an abstract idea.
In step 2B of the 101-analysis set forth in the 2019 PEG, the examiner has determined that the claim does not include additional elements that are sufficient to amount to significantly more than the judicial exception.
As discussed above, additional elements vi and vii recites generic computer component being used as tool to perform functions of the judicial exception, and additional element viii recites mere instructions to apply an exception using generic computer, which is not indicative of significantly more.
Considering the additional elements individually and in combination, and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Therefore, the claim is not patent eligible.
Claim 12:
Regarding claim 12, it is dependent upon claim 11, and thereby incorporates the limitations of, and corresponding analysis to claim 11. Further, claim 12 recites the following additional elements:
“The system of claim 11, wherein the computer program further causes the hardware processor to train the GNN and graph refinement control parameters using the plurality of samples using smoothing-based optimization and local gradient descent.” (this is a mathematical concept, using smoothing-based optimization and local gradient descent to train a GNN and graph refinement parameters is a mathematical calculation and formula, see Specification paragraphs [0027-0033], see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept but for the recitation of generic computer components, then it falls within the mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 13:
Regarding claim 13, it is dependent upon claim 12, and thereby incorporates the limitations of, and corresponding analysis to claim 12. Further, claim 13 recites the following additional elements:
“The system of claim 12, wherein the performance score is a weighted average of a mean squared error and a Pearson correlation coefficient for a graph regression task.” (this is a mathematical concept, a performance score being a weighted average of a mean squared error and a Pearson correlation coefficient is a mathematical calculation, see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept but for the recitation of generic computer components, then it falls within the mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 14:
Regarding claim 14, it is dependent upon claim 11, and thereby incorporates the limitations of, and corresponding analysis to claim 11. Further, claim 14 recites the following additional elements:
“The system of claim 11, wherein the input graph is a graph indicating a spatial proximity of capturing spots, with nodes that represent capturing spots and edges that represent spatial adjacency.” (this is a mathematical concept, a graph that indicates spatial proximity is a mathematical relationship, see Specification paragraph [0020], see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept but for the recitation of generic computer components, then it falls within the mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 15:
Regarding claim 15, it is dependent upon claim 11, and thereby incorporates the limitations of, and corresponding analysis to claim 11. Further, claim 15 recites the following additional elements:
“The system of claim 11, wherein the graph analysis includes predicting spatially resolved gene expression via tissue morphology in hematoxylin and eosin (H&E) stained images.” (In step 2A, prong 2, this is considered as mere field of use or technological environment in which to apply a judicial exception, see MPEP § 2106.05(h)). (In step 2B, this is also considered mere field of use or technological environment in which to apply a judicial exception – see MPEP § 2106.05(h)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 16:
Regarding claim 16, it is dependent upon claim 15, and thereby incorporates the limitations of, and corresponding analysis to claim 15. Further, claim 16 recites the following additional elements:
“The system of claim 15, wherein the predicted spatially resolved gene expression relates to HER2 positivity and wherein the action includes a treatment for breast cancer.” (this is a mental process, a person can mentally evaluate a predicted spatially resolved gene expression being in relation to HER2 positivity and an action that includes a treatment for breast cancer, see MPEP § 2106.04(a)(2)(III)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mental process but for the recitation of generic computer components, then it falls within the mental process concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 17:
Regarding claim 17, it is dependent upon claim 11, and thereby incorporates the limitations of, and corresponding analysis to claim 11. Further, claim 17 recites the following additional elements:
“The system of claim 11, wherein the graph neural network is a machine learning model that generates a prediction to aid in medical decision making.” (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 18:
Regarding claim 18, it is dependent upon claim 11, and thereby incorporates the limitations of, and corresponding analysis to claim 11. Further, claim 18 recites the following additional elements:
“The system of claim 11, wherein the action includes automatically altering a patient’s treatment.” (In step 2A, prong 2, this is considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)). (In step 2B, this is also considered mere instructions to apply an exception using generic computer – see MPEP § 2106.05(f)).
Since the claim does not recite additional elements that either integrate the judicial exception into a practical application, nor provide significantly more than the judicial exception, the claim is not patent eligible.
Claim 19:
Regarding claim 19, it is dependent upon claim 11, and thereby incorporates the limitations of, and corresponding analysis to claim 11. Further, claim 19 recites the following additional elements:
“The system of claim 11, wherein the computer program further causes the hardware processor to apply a threshold to linear projections of node features, using control parameters
PNG
media_image1.png
28
100
media_image1.png
Greyscale
:
PNG
media_image2.png
38
170
media_image2.png
Greyscale
, where
PNG
media_image3.png
30
404
media_image3.png
Greyscale
, and wherein message-passing on the GNN uses updates implemented as
PNG
media_image4.png
102
320
media_image4.png
Greyscale
for levels l < L, where l is a layer, L a number of layers in the GNN, m and n are nodes, E’ is a set of edges, Wl is weight matrix for the layer l, σ(x) = max(0,x) is a rectified linear unit (ReLU) function, and deg(•) is a degree of a node.” (this is a mathematical concept, the graph refinement function is a mathematical formula as shown, see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept but for the recitation of generic computer components, then it falls within the mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Claim 20:
Regarding claim 20, it is dependent upon claim 19, and thereby incorporates the limitations of, and corresponding analysis to claim 19. Further, claim 20 recites the following additional elements:
“The system of claim 19, wherein the generation of the plurality of sample graph refinements includes sampling of a Gaussian distribution with mean and standard deviation parameters that are iteratively determined by smoothing-based optimization.” (this is a mathematical concept, the generating plurality of sample graph refinements that includes sampling a Gaussian distribution with mean and standard deviation parameters is a mathematical calculation, see MPEP § 2106.04(a)(2)(I)),
If claim limitations, under their broadest reasonable interpretation, covers performance of the limitations as a mathematical concept but for the recitation of generic computer components, then it falls within the mathematical concept grouping of abstract ideas. Accordingly, the claim “recites” an abstract idea.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries 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.
Claim(s) 1, 9, 11 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Zhang C et al., (US. Patent Application Publication 20240338559 A1) effectively filed on June 29, 2022, (hereafter Zhang), in view of Fu H. et al, (US. Patent Application Publication 20240220830 A1) effectively filed on December 29, 2022, (hereafter Fu).
Claim 1:
Regarding claim 1, Zhang teaches “A computer-implemented method for graph analysis, comprising: identifying trainable control parameters of a graph refinement function;”
See Zhang in paragraph [0048] describing, “Each node 104 represents a function of the input value(s) received on its input edges(s) 106i, the outputs of the function being output on the output edge(s) 106o of the respective node 104, such that the value(s) output on the output edge(s) 106o of the node 104 depend on the respective input value(s) according to the respective function. The function of each node 104 is also parametrized by one or more respective parameters w, sometimes also referred to as weights (not necessarily weights in the sense of multiplicative weights, though that is certainly one possibility). Thus, the relation between the values of the input(s) 106i and the output(s) 106o of each node 104 depends on the respective function of the node and its respective weight(s).” Further, see Zhang in paragraph [0051] describing, “The different weights of the various nodes 104 in the neural network 100 can be gradually tuned based on a set of experience data (e.g. training data), so as to tend towards a state where the output 108o of the network will produce a desired value for a given input 108i. For instance, before being used in an actual application, the neural network 100 may first be trained for that application. Training comprises inputting experience data in the form of training data to the inputs 108i of the graph and then tuning the weights w of the nodes 104 based on feedback from the output(s) 108o of the graph. The training data comprises multiple different input data points, each comprising a value or vector of values corresponding to the input edge or edges 108i of the graph 100.”Here, Zhang establishes parameters as weights and these weights being tuned or trained from various nodes. These nodes represent functions which can be seen as graph refinement function.
Further, Zhang teaches “generating a plurality of sample graph refinements of an input graph, using control parameters sampled from a variational distribution;”
See Zhang in paragraph [0047] describing, “A neural network 100 comprises a graph of interconnected nodes 104 and edges 106 connecting between nodes, all implemented in software. Each node 104 has one or more input edges and one or more output edges. The input edges of one or more of the nodes 104 form the overall input 108i to the graph (typically an input vector, i.e. there are multiple input edges).” Here, Zhang establishes an input graph with the overall input to a graph. Further, see Zhang in paragraph [0051] describing, “The different weights of the various nodes 104 in the neural network 100 can be gradually tuned based on a set of experience data (e.g. training data), so as to tend towards a state where the output 108o of the network will produce a desired value for a given input 108i.”Here, Zhang establishes a plurality of graph refinements of the input graph here with the tuning of the weights of the nodes which are established to be parameters. Further, see Zhang in paragraph [0051] describing, “Each weight could simply be a scalar value. Alternatively, as shown in FIG. 1(b), at some or all of the nodes 104 in the network 100, the respective weight may be modelled as a probabilistic distribution such as a Gaussian. In such cases the neural network 100 is sometimes referred to as a Bayesian neural network. Optionally, the value input/output on each of some or all of the edges 106 may each also be modelled as a respective probabilistic distribution. For any given weight or edge, the distribution may be modelled in terms of a set of samples of the distribution, or a set of parameters parameterizing the respective distribution, e.g. a pair of parameters specifying its centre point and width (e.g. in terms of its mean and standard deviation σ or variance σ2).”Here, Zhang establishes the parameters being sampled from a distribution which can be seen as variational distribution.
Further, Zhang teaches “and performing an action responsive to the graph analysis.”
See Zhang in paragraph [0164] describing, “FIG. 13(b) shows a Graph predicted by VICAUSE (when the one on FIG. 13(a) is used as the true one). VICAUSE predicts all the true relationships plus some additional ones (dashed edges).” Here, Zhang establishes predicting, which is the action here, true relationships in relation to a graph predicted by VICAUSE known in the art to perform graph analysis with relationships.
Further, Zhang teaches “…to produce a refined graph on new test samples;”
See Zhang in paragraph [0051] describing, “Training comprises inputting experience data in the form of training data to the inputs 108i of the graph and then tuning the weights w of the nodes 104 based on feedback from the output(s) 108o of the graph. The training data comprises multiple different input data points, each comprising a value or vector of values corresponding to the input edge or edges 108i of the graph 100.” Here, Zhang establishes an output graph produced using experience data of inputs where the inputs of experience data is interpreted as the test samples and the output of the graph is the refined graph.
Zhang does not appear to explicitly teach “selecting graph refinement control parameters associated with a sample of the plurality of sample graph refinements that has a highest performance score when used to train a graph neural network (GNN); performing graph analysis on the input graph using the selected graph refinement parameters…;”,
However in the same field of art, Fu teaches “selecting graph refinement control parameters associated with a sample of the plurality of sample graph refinements that has a highest performance score when used to train a graph neural network (GNN);”
See Fu in paragraph [0016] describing, “In a sample configuration, the linkage score learning algorithm conducts r cycles of calculation, where a cycle includes 3 steps: 1) calculate linkage scores, 2) add top t proportion of new links with highest linkage scores to the bipartite graph, and 3) update degrees of each user and item. t and r are learnable parameters as are α, β, γ, and ?, which means that the optimal values for these parameters is dependent on the training data being considered. The optimal set of values for α, β, γ, ?, t, and r is the one that achieves the best performance on validating data on some metrics. The linkage score learning algorithm pre-defines candidate values for each parameter and selects the best combination of them.”
Further, Fu teaches “performing graph analysis on the input graph using the selected graph refinement parameters…”
See Fu in paragraph [0016] describing, “In a sample configuration, the linkage score learning algorithm conducts r cycles of calculation, where a cycle includes 3 steps: 1) calculate linkage scores, 2) add top t proportion of new links with highest linkage scores to the bipartite graph, and 3) update degrees of each user and item. t and r are learnable parameters as are α, β, γ, and ?, which means that the optimal values for these parameters is dependent on the training data being considered. The optimal set of values for α, β, γ, ?, t, and r is the one that achieves the best performance on validating data on some metrics. The linkage score learning algorithm pre-defines candidate values for each parameter and selects the best combination of them.” Here, Fu establishes performing graph analysis with the linkage score learning algorithm finding linkage scores to the bipartite graph using selected parameters combinations.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of Zhang with the teachings of Fu by using Zhang’s teachings of a computer-implemented method for graphs and incorporate with Fu’s teachings of a linkage score learning algorithm with graphs.
One of ordinary skill in the art would be motivated to do so because by integrating Fu’s frameworks into the methods of Zhang, which are both in the same field of art of graph analysis, one of ordinary skill in the art would bring “The linkage score learning algorithm [that] addresses the problem of link prediction by predicting new links (e.g., missing addresses) in a graph that do not already exist in training data. The link prediction aspect of collaborative filtering as described herein provides significant performance gains by aligning link prediction with item recommendations” (Fu, paragraph [0014]).
Claim 9:
Regarding claim 9, Zhang in view of Fu teaches the limitations of claim 1.
Further, Zhang teaches “The method of claim 1, wherein the graph refinement function applies a threshold to linear projections of node features, using control parameters
PNG
media_image1.png
28
100
media_image1.png
Greyscale
:
PNG
media_image2.png
38
170
media_image2.png
Greyscale
, where
PNG
media_image3.png
30
404
media_image3.png
Greyscale
, and wherein message-passing on the GNN uses updates implemented as
PNG
media_image4.png
102
320
media_image4.png
Greyscale
for levels l < L, where l is a layer, L a number of layers in the GNN, m and n are nodes, E’ is a set of edges, Wl is weight matrix for the layer l, σ(x) = max(0,x) is a rectified linear unit (ReLU) function, and deg(•) is a degree of a node.”
See Zhang in paragraph [0168] describing, “In this small synthetic experiment, it is possible to visually inspect the predicted graph. FIG. 14 shows the posterior probability of each edge (i.e. the estimated matrix G) for the simulated dataset that uses the true graph in FIG. 13(a). By using a threshold of 0.5, we obtain the predicted graph in FIG. 13(b). We observe that all the true edges are captured by VICAUSE, which also predicts some additional edges. Some of them can be explained by looking at the relationships between the corresponding variables, recall FIG. 18 in the appendix. For instance, the one connecting the third and fifth variables can be explained through the linear relationship that exist between both variables.” Here, Zhang establishes applying a 0.5 threshold for linear prediction or projections, as established in previous limitations edges are a part of nodes and therefore can be seen as node features. Further see Zhang in paragraph [0048] describing, “Each node 104 represents a function of the input value(s) received on its input edges(s) 106i, the outputs of the function being output on the output edge(s) 106o of the respective node 104, such that the value(s) output on the output edge(s) 106o of the node 104 depend on the respective input value(s) according to the respective function. The function of each node 104 is also parametrized by one or more respective parameters w, sometimes also referred to as weights (not necessarily weights in the sense of multiplicative weights, though that is certainly one possibility). Thus, the relation between the values of the input(s) 106i and the output(s) 106o of each node 104 depends on the respective function of the node and its respective weight(s).” Here, Zhang establishes nodes being controlled by parameters or weights and further establishes the connection with edges and nodes. Further see Zhang in paragraph [0139] describing, “
PNG
media_image5.png
378
414
media_image5.png
Greyscale
” Here, Zhang establishes message passing on the GNN with VICAUSE. Further see Zhang in paragraph [0187] describing, “As for the decoder, we perform T=3 iterations of GNN message passing. All the MHLPs in the decoder (i.e. MLPf, MLPb, MLPe2n and g) have two linear layers with ReLU non-linearity. The dimensionality of the hidden layer, which is the dimensionality of each latent subspace, is 256. Regarding the encoder, it is given by a multi-head neural network that defines the mean and standard deviation of the latent representation. The neural network is a MLP with two standard linear layers with ReLu non-linearity.” Here, Zhang establishes ReLU in the message passing of GNN and establishes the layers further. With all of these in combination what is claimed here can be achieved.
Claim 11:
Regarding claim 11, Zhang teaches “A system for graph analysis, comprising: a hardware processor; and a memory that stores a computer program which, when executed by the hardware processor, causes the hardware processor to: identify trainable control parameters of a graph refinement function;”
See Zhang in paragraph [0048] describing, “Each node 104 represents a function of the input value(s) received on its input edges(s) 106i, the outputs of the function being output on the output edge(s) 106o of the respective node 104, such that the value(s) output on the output edge(s) 106o of the node 104 depend on the respective input value(s) according to the respective function. The function of each node 104 is also parametrized by one or more respective parameters w, sometimes also referred to as weights (not necessarily weights in the sense of multiplicative weights, though that is certainly one possibility). Thus, the relation between the values of the input(s) 106i and the output(s) 106o of each node 104 depends on the respective function of the node and its respective weight(s).” Further, see Zhang in paragraph [0051] describing, “The different weights of the various nodes 104 in the neural network 100 can be gradually tuned based on a set of experience data (e.g. training data), so as to tend towards a state where the output 108o of the network will produce a desired value for a given input 108i. For instance, before being used in an actual application, the neural network 100 may first be trained for that application. Training comprises inputting experience data in the form of training data to the inputs 108i of the graph and then tuning the weights w of the nodes 104 based on feedback from the output(s) 108o of the graph. The training data comprises multiple different input data points, each comprising a value or vector of values corresponding to the input edge or edges 108i of the graph 100.”Here, Zhang establishes parameters as weights and these weights being tuned or trained from various nodes. These nodes represent functions which can be seen as graph refinement function. Further, see Zhang in paragraph [0043] describing, “Each of the controller 202, interface 204 and AI algorithm 206 may be implemented in the form of software code embodied on computer readable storage and run on processing apparatus comprising one or more processors such as CPUs, work accelerator co-processors such as GPUs, and/or other application specific processors, implemented on one or more computer terminals or units at one or more geographic sites. The storage on which the code is stored may comprise one or more memory devices employing one or more memory media (e.g. electronic or magnetic media), again implemented on one or more computer terminals or units at one or more geographic sites.”
Further, Zhang teaches “generate a plurality of sample graph refinements of an input graph, using control parameters sampled from a variational distribution;”
See Zhang in paragraph [0047] describing, “A neural network 100 comprises a graph of interconnected nodes 104 and edges 106 connecting between nodes, all implemented in software. Each node 104 has one or more input edges and one or more output edges. The input edges of one or more of the nodes 104 form the overall input 108i to the graph (typically an input vector, i.e. there are multiple input edges).” Here, Zhang establishes an input graph with the overall input to a graph. Further, see Zhang in paragraph [0051] describing, “The different weights of the various nodes 104 in the neural network 100 can be gradually tuned based on a set of experience data (e.g. training data), so as to tend towards a state where the output 108o of the network will produce a desired value for a given input 108i.”Here, Zhang establishes a plurality of graph refinements of the input graph here with the tuning of the weights of the nodes which are established to be parameters. Further, see Zhang in paragraph [0051] describing, “Each weight could simply be a scalar value. Alternatively, as shown in FIG. 1(b), at some or all of the nodes 104 in the network 100, the respective weight may be modelled as a probabilistic distribution such as a Gaussian. In such cases the neural network 100 is sometimes referred to as a Bayesian neural network. Optionally, the value input/output on each of some or all of the edges 106 may each also be modelled as a respective probabilistic distribution. For any given weight or edge, the distribution may be modelled in terms of a set of samples of the distribution, or a set of parameters parameterizing the respective distribution, e.g. a pair of parameters specifying its centre point and width (e.g. in terms of its mean and standard deviation σ or variance σ2).”Here, Zhang establishes the parameters being sampled from a distribution which can be seen as variational distribution.
Further, Zhang teaches “and perform an action responsive to the graph analysis.”
See Zhang in paragraph [0164] describing, “FIG. 13(b) shows a Graph predicted by VICAUSE (when the one on FIG. 13(a) is used as the true one). VICAUSE predicts all the true relationships plus some additional ones (dashed edges).” Here, Zhang establishes predicting, which is the action here, true relationships in relation to a graph predicted by VICAUSE known in the art to perform graph analysis with relationships.
Further, Zhang teaches “…to produce a refined graph on new test samples;”
See Zhang in paragraph [0051] describing, “Training comprises inputting experience data in the form of training data to the inputs 108i of the graph and then tuning the weights w of the nodes 104 based on feedback from the output(s) 108o of the graph. The training data comprises multiple different input data points, each comprising a value or vector of values corresponding to the input edge or edges 108i of the graph 100.” Here, Zhang establishes an output graph produced using experience data of inputs where the inputs of experience data is interpreted as the test samples and the output of the graph is the refined graph.
Zhang does not appear to explicitly teach “selecting graph refinement control parameters associated with a sample of the plurality of sample graph refinements that has a highest performance score when used to train a graph neural network (GNN); performing graph analysis on the input graph using the selected graph refinement parameters…;”,
However in the same field of art, Fu teaches “selecting graph refinement control parameters associated with a sample of the plurality of sample graph refinements that has a highest performance score when used to train a graph neural network (GNN);”
See Fu in paragraph [0016] describing, “In a sample configuration, the linkage score learning algorithm conducts r cycles of calculation, where a cycle includes 3 steps: 1) calculate linkage scores, 2) add top t proportion of new links with highest linkage scores to the bipartite graph, and 3) update degrees of each user and item. t and r are learnable parameters as are α, β, γ, and ?, which means that the optimal values for these parameters is dependent on the training data being considered. The optimal set of values for α, β, γ, ?, t, and r is the one that achieves the best performance on validating data on some metrics. The linkage score learning algorithm pre-defines candidate values for each parameter and selects the best combination of them.”
Further, Fu teaches “performing graph analysis on the input graph using the selected graph refinement parameters…”
See Fu in paragraph [0016] describing, “In a sample configuration, the linkage score learning algorithm conducts r cycles of calculation, where a cycle includes 3 steps: 1) calculate linkage scores, 2) add top t proportion of new links with highest linkage scores to the bipartite graph, and 3) update degrees of each user and item. t and r are learnable parameters as are α, β, γ, and ?, which means that the optimal values for these parameters is dependent on the training data being considered. The optimal set of values for α, β, γ, ?, t, and r is the one that achieves the best performance on validating data on some metrics. The linkage score learning algorithm pre-defines candidate values for each parameter and selects the best combination of them.” Here, Fu establishes performing graph analysis with the linkage score learning algorithm finding linkage scores to the bipartite graph using selected parameters combinations.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base reference of Zhang with the teachings of Fu by using Zhang’s teachings of a computer-implemented method for graphs and incorporate with Fu’s teachings of a linkage score learning algorithm with graphs.
One of ordinary skill in the art would be motivated to do so because by integrating Fu’s frameworks into the methods of Zhang, which are both in the same field of art of graph analysis, one of ordinary skill in the art would bring “The linkage score learning algorithm [that] addresses the problem of link prediction by predicting new links (e.g., missing addresses) in a graph that do not already exist in training data. The link prediction aspect of collaborative filtering as described herein provides significant performance gains by aligning link prediction with item recommendations” (Fu, paragraph [0014]).
Claim 19:
Regarding claim 19, Zhang in view of Fu teaches the limitations of claim 11.
Further, Zhang teaches “The system of claim 11, wherein the computer program further causes the hardware processor to apply a threshold to linear projections of node features, using control parameters
PNG
media_image1.png
28
100
media_image1.png
Greyscale
:
PNG
media_image2.png
38
170
media_image2.png
Greyscale
, where
PNG
media_image3.png
30
404
media_image3.png
Greyscale
, and wherein message-passing on the GNN uses updates implemented as
PNG
media_image4.png
102
320
media_image4.png
Greyscale
for levels l < L, where l is a layer, L a number of layers in the GNN, m and n are nodes, E’ is a set of edges, Wl is weight matrix for the layer l, σ(x) = max(0,x) is a rectified linear unit (ReLU) function, and deg(•) is a degree of a node.”
See Zhang in paragraph [0168] describing, “In this small synthetic experiment, it is possible to visually inspect the predicted graph. FIG. 14 shows the posterior probability of each edge (i.e. the estimated matrix G) for the simulated dataset that uses the true graph in FIG. 13(a). By using a threshold of 0.5, we obtain the predicted graph in FIG. 13(b). We observe that all the true edges are captured by VICAUSE, which also predicts some additional edges. Some of them can be explained by looking at the relationships between the corresponding variables, recall FIG. 18 in the appendix. For instance, the one connecting the third and fifth variables can be explained through the linear relationship that exist between both variables.” Here, Zhang establishes applying a 0.5 threshold for linear prediction or projections, as established in previous limitations edges are a part of nodes and therefore can be seen as node features. Further see Zhang in paragraph [0048] describing, “Each node 104 represents a function of the input value(s) received on its input edges(s) 106i, the outputs of the function being output on the output edge(s) 106o of the respective node 104, such that the value(s) output on the output edge(s) 106o of the node 104 depend on the respective input value(s) according to the respective function. The function of each node 104 is also parametrized by one or more respective parameters w, sometimes also referred to as weights (not necessarily weights in the sense of multiplicative weights, though that is certainly one possibility). Thus, the relation between the values of the input(s) 106i and the output(s) 106o of each node 104 depends on the respective function of the node and its respective weight(s).” Here, Zhang establishes nodes being controlled by parameters or weights and further establishes the connection with edges and nodes. Further see Zhang in paragraph [0139] describing, “
PNG
media_image5.png
378
414
media_image5.png
Greyscale
” Here, Zhang establishes message passing on the GNN with VICAUSE. Further see Zhang in paragraph [0187] describing, “As for the decoder, we perform T=3 iterations of GNN message passing. All the MHLPs in the decoder (i.e. MLPf, MLPb, MLPe2n and g) have two linear layers with ReLU non-linearity. The dimensionality of the hidden layer, which is the dimensionality of each latent subspace, is 256. Regarding the encoder, it is given by a multi-head neural network that defines the mean and standard deviation of the latent representation. The neural network is a MLP with two standard linear layers with ReLu non-linearity.” Here, Zhang establishes ReLU in the message passing of GNN and establishes the layers further. With all of these in combination what is claimed here can be achieved.
Claim(s) 2, 10, 12 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Zhang C et al., in view of Fu H. et al, further in view of Selvan R. et al, "Graph Refinement based Airway Extraction using Mean-Field Networks and Graph Neural Networks" available at https://arxiv.org/pdf/1811.08674, effectively published on June 2, 2020, (hereafter Selvan).
Claim 2:
Regarding claim 2, Zhang in view of Fu teaches the limitations of claim 1.
Further, Zhang teaches ” The method of claim 1, further comprising training the GNN and graph refinement parameters using the plurality of samples using…and local gradient descent.”
See Zhang in paragraphs [0100-0103] describing, “In some examples, tuning the values of one or more of G (e.g. edge probabilities of graph G 921), θGNN (parameters of decoder GNN 921), θ (parameters of decoder 938) and φ (parameters of encoder 934) in order to minimise the loss function L can comprise performing a gradient step for G, θGNN, θ and φ to minimise the loss function. The gradient step may involve the following steps: Computing the loss L; Differentiating L and computing the gradient of each parameter with respect to L (using backpropagation); Updating the values of each parameter by subtracting a small value proportional to the gradient.” Here, Zhang establishes training of the parameters of the GNN computing gradient for each parameter using backpropagation which as known in the art can be seen as using local gradient descent.
Neither Zhang or Fu appear to explicitly teach “…smoothing-based optimization…”,
However in the same field of art, Selvan teaches “…smoothing-based optimization…”
See Selvan on page 10 section 4 Discussion and Conclusions describing, “The input to the two graph refinement models was based on a preprocessing step that used the Bayesian smoothing method in (Selvan et al., 2017) as described in Section 3.1.1.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Selvan by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Selvan’s teachings of graph refinement using Graph Neural Networks.
One of ordinary skill in the art would be motivated to do so because by integrating Selvan’s frameworks into the methods of Zhang and Fu, which are both in the same field of art of graph analysis, one of ordinary skill in the art would bring “two approaches to performing graph refinement using MFNs and GNNs.” (Selvan, section 2 Methods on page 3).
Claim 10:
Regarding claim 10, Zhang in view of Fu teaches the limitations of claim 9.
Further, Zhang teaches “The method of claim 9, wherein generating the plurality of sample graph refinements includes sampling a Gaussian distribution with mean and standard deviation parameters that are iteratively determined by…”
See Zhang in paragraph [0049] describing, “Each weight could simply be a scalar value. Alternatively, as shown in FIG. 1(b), at some or all of the nodes 104 in the network 100, the respective weight may be modelled as a probabilistic distribution such as a Gaussian. In such cases the neural network 100 is sometimes referred to as a Bayesian neural network. Optionally, the value input/output on each of some or all of the edges 106 may each also be modelled as a respective probabilistic distribution. For any given weight or edge, the distribution may be modelled in terms of a set of samples of the distribution, or a set of parameters parameterizing the respective distribution, e.g. a pair of parameters specifying its centre point and width (e.g. in terms of its mean and standard deviation σ or variance σ2).”
Neither Zhang or Fu appear to explicitly teach “…smoothing-based optimization.”,
However in the same field of art, Selvan teaches “…smoothing-based optimization.”
See Selvan on page 10 section 4 Discussion and Conclusions describing, “The input to the two graph refinement models was based on a preprocessing step that used the Bayesian smoothing method in (Selvan et al., 2017) as described in Section 3.1.1.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Selvan by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Selvan’s teachings of graph refinement using Graph Neural Networks.
One of ordinary skill in the art would be motivated to do so because by integrating Selvan’s frameworks into the methods of Zhang and Fu, which are both in the same field of art of graph analysis, one of ordinary skill in the art would bring “two approaches to performing graph refinement using MFNs and GNNs.” (Selvan, section 2 Methods on page 3).
Claim 12:
Regarding claim 12, Zhang in view of Fu teaches the limitations of claim 11.
Further, Zhang teaches ” The system of claim 11, further comprising training the GNN and graph refinement parameters using the plurality of samples using…and local gradient descent.”
See Zhang in paragraphs [0100-0103] describing, “In some examples, tuning the values of one or more of G (e.g. edge probabilities of graph G 921), θGNN (parameters of decoder GNN 921), θ (parameters of decoder 938) and φ (parameters of encoder 934) in order to minimise the loss function L can comprise performing a gradient step for G, θGNN, θ and φ to minimise the loss function. The gradient step may involve the following steps: Computing the loss L; Differentiating L and computing the gradient of each parameter with respect to L (using backpropagation); Updating the values of each parameter by subtracting a small value proportional to the gradient.” Here, Zhang establishes training of the parameters of the GNN computing gradient for each parameter using backpropagation which as known in the art can be seen as using local gradient descent.
Neither Zhang or Fu appear to explicitly teach “…smoothing-based optimization…”,
However in the same field of art, Selvan teaches “…smoothing-based optimization…”
See Selvan on page 10 section 4 Discussion and Conclusions describing, “The input to the two graph refinement models was based on a preprocessing step that used the Bayesian smoothing method in (Selvan et al., 2017) as described in Section 3.1.1.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Selvan by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Selvan’s teachings of graph refinement using Graph Neural Networks.
One of ordinary skill in the art would be motivated to do so because by integrating Selvan’s frameworks into the methods of Zhang and Fu, which are both in the same field of art of graph analysis, one of ordinary skill in the art would bring “two approaches to performing graph refinement using MFNs and GNNs.” (Selvan, section 2 Methods on page 3).
Claim 20:
Regarding claim 20, Zhang in view of Fu teaches the limitations of claim 19.
Further, Zhang teaches “The system of claim 19, wherein generating the plurality of sample graph refinements includes sampling a Gaussian distribution with mean and standard deviation parameters that are iteratively determined by…”
See Zhang in paragraph [0049] describing, “Each weight could simply be a scalar value. Alternatively, as shown in FIG. 1(b), at some or all of the nodes 104 in the network 100, the respective weight may be modelled as a probabilistic distribution such as a Gaussian. In such cases the neural network 100 is sometimes referred to as a Bayesian neural network. Optionally, the value input/output on each of some or all of the edges 106 may each also be modelled as a respective probabilistic distribution. For any given weight or edge, the distribution may be modelled in terms of a set of samples of the distribution, or a set of parameters parameterizing the respective distribution, e.g. a pair of parameters specifying its centre point and width (e.g. in terms of its mean and standard deviation σ or variance σ2).”
Neither Zhang or Fu appear to explicitly teach “…smoothing-based optimization.”,
However in the same field of art, Selvan teaches “…smoothing-based optimization.”
See Selvan on page 10 section 4 Discussion and Conclusions describing, “The input to the two graph refinement models was based on a preprocessing step that used the Bayesian smoothing method in (Selvan et al., 2017) as described in Section 3.1.1.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Selvan by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Selvan’s teachings of graph refinement using Graph Neural Networks.
One of ordinary skill in the art would be motivated to do so because by integrating Selvan’s frameworks into the methods of Zhang and Fu, which are both in the same field of art of graph analysis, one of ordinary skill in the art would bring “two approaches to performing graph refinement using MFNs and GNNs.” (Selvan, section 2 Methods on page 3).
Claim(s) 3-7, and 13-17 are rejected under 35 U.S.C. 103 as being unpatentable over Zhang C. et al., in view of Fu H. et al, and further in view of Zeng Y. et al, "Graph Refinement based Airway Extraction using Mean-Field Networks and Graph Neural Networks" available at https://academic.oup.com/bib/article/23/5/bbac297/6645485, effectively published on July 17, 2022, (hereafter Zeng).
Claim 3:
Regarding claim 3, Zhang in view of Fu, and further in view of Selvan teaches the limitations of claim 2.
None of Zhang, Fu, or Selvan appear to explicitly teach “The method of claim 2, wherein the performance score is a weighted average of a mean squared error and a Pearson correlation coefficient for a graph regression task.”,
However in the same field of art, Zeng teaches “The method of claim 2, wherein the performance score is a weighted average of a mean squared error and a Pearson correlation coefficient for a graph regression task.”
See Zeng on page 5 Self-distillation mechanism section describing, “To fuse the knowledge information between the anchor image patch and the augmented image patches, we calculate the MSE loss as follows:
PNG
media_image6.png
64
282
media_image6.png
Greyscale
Finally, the total loss functions of our model are as follows:
PNG
media_image7.png
52
246
media_image7.png
Greyscale
where
PNG
media_image8.png
24
52
media_image8.png
Greyscale
are the hyperparameters to balance contribution from the self-distillation and ZINB. In this setting, we set
PNG
media_image9.png
18
156
media_image9.png
Greyscale
for all datasets.” Further, see Zeng on page 5 Pearson correlation coefficients section describing, “We measure the accuracy of predicted gene expression by Pearson correlation coefficient (PCC) [53].
PNG
media_image10.png
108
350
media_image10.png
Greyscale
” Here, establishes calculating MSE loss and gives a function, this function is being interpreted as the performance score and explicitly does calculate MSE which is mean squared error known in the art to calculate a weighted average, Pearson correlation coefficient is also explicitly stated in this same embodiment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang, Fu, and Selvan with the teachings of Zeng by using Zhang’s teachings of a computer-implemented method for graphs, Selvan’s teachings of graph refinement using Graph Neural Networks, and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Zeng’s teachings of spatial gene prediction using graph neural networks.
One of ordinary skill in the art would be motivated to do so because by integrating Zeng’s frameworks into the methods of Zhang, Selvan, and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring “Hist2ST [that]was tested on several ST datasets and shown to outperform existing methods in terms of both gene expression prediction and spatial region identification using the predicted expression. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.” (Zeng, Introduction section on page 2).
Claim 4:
Regarding claim 4, Zhang in view of Fu teaches the limitations of claim 1.
Neither Zhang or Fu appear to explicitly teach “The method of claim 1, wherein the input graph is a graph indicating a spatial proximity of capturing spots, with nodes that represent capturing spots and edges that represent spatial adjacency.”,
However in the same field of art, Zeng teaches “The method of claim 1, wherein the input graph is a graph indicating a spatial proximity of capturing spots, with nodes that represent capturing spots and edges that represent spatial adjacency.”
See Zeng on page 4 section The GNN module describing, “The GNN module is applied to explicitly learn the local spatial dependencies. In this study, the backbone of GNN employs the GraphSAGE [48], which needs the nodes with corresponding features and the nearest neighboring graph
PNG
media_image11.png
26
86
media_image11.png
Greyscale
as the input. In our setting, the node is the spot, and the corresponding feature is from the Transformer module. Meanwhile, we construct the nearest neighboring graph by selecting four nearest neighbors for each spot, where
PNG
media_image12.png
28
80
media_image12.png
Greyscale
means the number of spots, and indicates the edges connecting with the nearest neighbors. We use spatial locations in Euclidean distance as the intrinsic edges [49].”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Zeng by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Zeng’s teachings of spatial gene prediction using graph neural networks.
One of ordinary skill in the art would be motivated to do so because by integrating Zeng’s frameworks into the methods of Zhang and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring “Hist2ST [that]was tested on several ST datasets and shown to outperform existing methods in terms of both gene expression prediction and spatial region identification using the predicted expression. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.” (Zeng, Introduction section on page 2).
Claim 5:
Regarding claim 5, Zhang in view of Fu teaches the limitations of claim 1.
Neither Zhang or Fu appear to explicitly teach “The method of claim 1, wherein the graph analysis includes predicting spatially resolved gene expression via tissue morphology in hematoxylin and eosin (H&E) stained images.”,
However in the same field of art, Zeng teaches “The method of claim 1, wherein the graph analysis includes predicting spatially resolved gene expression via tissue morphology in hematoxylin and eosin (H&E) stained images.”
See Zeng on page 2 Introduction section describing, “In this study, we have developed Hist2ST, a novel spatial information-guided deep learning method, to predict spatial transcriptomic from WSIs. As shown in Figure 1, around each sequenced spot, the corresponding histology image is cropped into an image patch. The image patches are then sequentially fed into three modules (the Convmixer, Transformer and GNN). The learned features are then used to predict the gene expression by following the zero-inflated negative binomial (ZINB) distribution [41].” Further, see Zeng on page 6 Performance on the spatial transcriptomics datasets section describing, “In addition, we further tested our model trained from ST-seq data on four randomly selected breast cancer H&E images from The Cancer Genome Atlas (TCGA) that did not contain spatial expression information.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Zeng by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Zeng’s teachings of spatial gene prediction using graph neural networks.
One of ordinary skill in the art would be motivated to do so because by integrating Zeng’s frameworks into the methods of Zhang and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring “Hist2ST [that]was tested on several ST datasets and shown to outperform existing methods in terms of both gene expression prediction and spatial region identification using the predicted expression. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.” (Zeng, Introduction section on page 2).
Claim 6:
Regarding claim 6, Zhang in view of Fu, and further in view of Zeng teaches the limitations of claim 5.
Neither Zhang or Fu appear to explicitly teach “The method of claim 5, wherein the predicted spatially resolved gene expression relates to HER2 positivity and wherein the action includes a treatment for breast cancer.”,
Further, Zeng teaches “The method of claim 5, wherein the predicted spatially resolved gene expression relates to HER2 positivity and wherein the action includes a treatment for breast cancer.”
See Zeng on page 7 Visualization of the predicted gene expression section describing, “To further evaluate the predicted gene expression, we ranked genes based on the highest values of mean −log10 (P-values) across all tissue sections in each dataset, where the P-value for each tissue section was obtained according to the correlation between the predicted and observed gene expression. For the HER2+ dataset, we visualized the top four genes FN1, GANS, SCD and MYL12B (Supplementary Table S2 available online at http://bib.oxfordjournals.org/). The visualized results showed that our model achieved the highest PCC on each gene (Figure 4). Additionally, all these four top predicted genes were found to be related with breast cancer [29, 60–62].”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Zeng by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Zeng’s teachings of spatial gene prediction using graph neural networks.
One of ordinary skill in the art would be motivated to do so because by integrating Zeng’s frameworks into the methods of Zhang and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring “Hist2ST [that]was tested on several ST datasets and shown to outperform existing methods in terms of both gene expression prediction and spatial region identification using the predicted expression. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.” (Zeng, Introduction section on page 2).
Claim 7:
Regarding claim 7, Zhang in view of Fu, teaches the limitations of claim 1.
Neither Zhang or Fu appear to explicitly teach “The method of claim 1, wherein the graph neural network is a machine learning model that generates a prediction to aid in medical decision making.”,
However in the same field of art, Zeng teaches “The method of claim 1, wherein the graph neural network is a machine learning model that generates a prediction to aid in medical decision making.”
See Zeng on page 3 The architecture of Hist2ST section describing, “Hist2ST is a deep learning-based model to predict the spatial gene expression from histology images. As shown in Figure 1, the input is the histology image that is divided into multiple patches around the sequenced spots. To process the input, Hist2ST consists of three modules: the Convmixer, Transformer and GNN. The Convmixer aims to capture internal relations of the 2D vision features within the image patch through convolution operations. The Transformer module aims to capture the global spatial dependencies through the self-attention mechanism. Meanwhile, the GNN module explicitly captures the neighborhood relationships between spots. Finally, the learned features are used to predict the gene expression by following the ZINB distribution.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Zeng by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Zeng’s teachings of spatial gene prediction using graph neural networks.
One of ordinary skill in the art would be motivated to do so because by integrating Zeng’s frameworks into the methods of Zhang and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring “Hist2ST [that]was tested on several ST datasets and shown to outperform existing methods in terms of both gene expression prediction and spatial region identification using the predicted expression. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.” (Zeng, Introduction section on page 2).
Claim 13:
Regarding claim 13, Zhang in view of Fu, and further in view of Selvan teaches the limitations of claim 12.
None of Zhang, Fu, or Selvan appear to explicitly teach “The system of claim 12, wherein the performance score is a weighted average of a mean squared error and a Pearson correlation coefficient for a graph regression task.”,
However in the same field of art, Zeng teaches “The system of claim 12, wherein the performance score is a weighted average of a mean squared error and a Pearson correlation coefficient for a graph regression task.”
See Zeng on page 5 Self-distillation mechanism section describing, “To fuse the knowledge information between the anchor image patch and the augmented image patches, we calculate the MSE loss as follows:
PNG
media_image6.png
64
282
media_image6.png
Greyscale
Finally, the total loss functions of our model are as follows:
PNG
media_image7.png
52
246
media_image7.png
Greyscale
where
PNG
media_image8.png
24
52
media_image8.png
Greyscale
are the hyperparameters to balance contribution from the self-distillation and ZINB. In this setting, we set
PNG
media_image9.png
18
156
media_image9.png
Greyscale
for all datasets.” Further, see Zeng on page 5 Pearson correlation coefficients section describing, “We measure the accuracy of predicted gene expression by Pearson correlation coefficient (PCC) [53].
PNG
media_image10.png
108
350
media_image10.png
Greyscale
” Here, establishes calculating MSE loss and gives a function, this function is being interpreted as the performance score and explicitly does calculate MSE which is mean squared error known in the art to calculate a weighted average, Pearson correlation coefficient is also explicitly stated in this same embodiment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang, Fu, and Selvan with the teachings of Zeng by using Zhang’s teachings of a computer-implemented method for graphs, Selvan’s teachings of graph refinement using Graph Neural Networks, and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Zeng’s teachings of spatial gene prediction using graph neural networks.
One of ordinary skill in the art would be motivated to do so because by integrating Zeng’s frameworks into the methods of Zhang, Selvan, and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring “Hist2ST [that]was tested on several ST datasets and shown to outperform existing methods in terms of both gene expression prediction and spatial region identification using the predicted expression. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.” (Zeng, Introduction section on page 2).
Claim 14:
Regarding claim 14, Zhang in view of Fu teaches the limitations of claim 11.
Neither Zhang or Fu appear to explicitly teach “The system of claim 11, wherein the input graph is a graph indicating a spatial proximity of capturing spots, with nodes that represent capturing spots and edges that represent spatial adjacency.”,
However in the same field of art, Zeng teaches “The system of claim 11, wherein the input graph is a graph indicating a spatial proximity of capturing spots, with nodes that represent capturing spots and edges that represent spatial adjacency.”
See Zeng on page 4 section The GNN module describing, “The GNN module is applied to explicitly learn the local spatial dependencies. In this study, the backbone of GNN employs the GraphSAGE [48], which needs the nodes with corresponding features and the nearest neighboring graph
PNG
media_image11.png
26
86
media_image11.png
Greyscale
as the input. In our setting, the node is the spot, and the corresponding feature is from the Transformer module. Meanwhile, we construct the nearest neighboring graph by selecting four nearest neighbors for each spot, where
PNG
media_image12.png
28
80
media_image12.png
Greyscale
means the number of spots, and indicates the edges connecting with the nearest neighbors. We use spatial locations in Euclidean distance as the intrinsic edges [49].”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Zeng by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Zeng’s teachings of spatial gene prediction using graph neural networks.
One of ordinary skill in the art would be motivated to do so because by integrating Zeng’s frameworks into the methods of Zhang and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring “Hist2ST [that]was tested on several ST datasets and shown to outperform existing methods in terms of both gene expression prediction and spatial region identification using the predicted expression. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.” (Zeng, Introduction section on page 2).
Claim 15:
Regarding claim 15, Zhang in view of Fu teaches the limitations of claim 11.
Neither Zhang or Fu appear to explicitly teach “The system of claim 11, wherein the graph analysis includes predicting spatially resolved gene expression via tissue morphology in hematoxylin and eosin (H&E) stained images.”,
However in the same field of art, Zeng teaches “The system of claim 11, wherein the graph analysis includes predicting spatially resolved gene expression via tissue morphology in hematoxylin and eosin (H&E) stained images.”
See Zeng on page 2 Introduction section describing, “In this study, we have developed Hist2ST, a novel spatial information-guided deep learning method, to predict spatial transcriptomic from WSIs. As shown in Figure 1, around each sequenced spot, the corresponding histology image is cropped into an image patch. The image patches are then sequentially fed into three modules (the Convmixer, Transformer and GNN). The learned features are then used to predict the gene expression by following the zero-inflated negative binomial (ZINB) distribution [41].” Further, see Zeng on page 6 Performance on the spatial transcriptomics datasets section describing, “In addition, we further tested our model trained from ST-seq data on four randomly selected breast cancer H&E images from The Cancer Genome Atlas (TCGA) that did not contain spatial expression information.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Zeng by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Zeng’s teachings of spatial gene prediction using graph neural networks.
One of ordinary skill in the art would be motivated to do so because by integrating Zeng’s frameworks into the methods of Zhang and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring “Hist2ST [that]was tested on several ST datasets and shown to outperform existing methods in terms of both gene expression prediction and spatial region identification using the predicted expression. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.” (Zeng, Introduction section on page 2).
Claim 16:
Regarding claim 16, Zhang in view of Fu, and further in view of Zeng teaches the limitations of claim 15.
Neither Zhang or Fu appear to explicitly teach “The system of claim 15, wherein the predicted spatially resolved gene expression relates to HER2 positivity and wherein the action includes a treatment for breast cancer.”,
Further, Zeng teaches “The system of claim 15, wherein the predicted spatially resolved gene expression relates to HER2 positivity and wherein the action includes a treatment for breast cancer.”
See Zeng on page 7 Visualization of the predicted gene expression section describing, “To further evaluate the predicted gene expression, we ranked genes based on the highest values of mean −log10 (P-values) across all tissue sections in each dataset, where the P-value for each tissue section was obtained according to the correlation between the predicted and observed gene expression. For the HER2+ dataset, we visualized the top four genes FN1, GANS, SCD and MYL12B (Supplementary Table S2 available online at http://bib.oxfordjournals.org/). The visualized results showed that our model achieved the highest PCC on each gene (Figure 4). Additionally, all these four top predicted genes were found to be related with breast cancer [29, 60–62].”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Zeng by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Zeng’s teachings of spatial gene prediction using graph neural networks.
One of ordinary skill in the art would be motivated to do so because by integrating Zeng’s frameworks into the methods of Zhang and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring “Hist2ST [that]was tested on several ST datasets and shown to outperform existing methods in terms of both gene expression prediction and spatial region identification using the predicted expression. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.” (Zeng, Introduction section on page 2).
Claim 17:
Regarding claim 17, Zhang in view of Fu, teaches the limitations of claim 11.
Neither Zhang or Fu appear to explicitly teach “The system of claim 11, wherein the graph neural network is a machine learning model that generates a prediction to aid in medical decision making.”,
However in the same field of art, Zeng teaches “The system of claim 11, wherein the graph neural network is a machine learning model that generates a prediction to aid in medical decision making.”
See Zeng on page 3 The architecture of Hist2ST section describing, “Hist2ST is a deep learning-based model to predict the spatial gene expression from histology images. As shown in Figure 1, the input is the histology image that is divided into multiple patches around the sequenced spots. To process the input, Hist2ST consists of three modules: the Convmixer, Transformer and GNN. The Convmixer aims to capture internal relations of the 2D vision features within the image patch through convolution operations. The Transformer module aims to capture the global spatial dependencies through the self-attention mechanism. Meanwhile, the GNN module explicitly captures the neighborhood relationships between spots. Finally, the learned features are used to predict the gene expression by following the ZINB distribution.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Zeng by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Zeng’s teachings of spatial gene prediction using graph neural networks.
One of ordinary skill in the art would be motivated to do so because by integrating Zeng’s frameworks into the methods of Zhang and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring “Hist2ST [that]was tested on several ST datasets and shown to outperform existing methods in terms of both gene expression prediction and spatial region identification using the predicted expression. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.” (Zeng, Introduction section on page 2).
Claim(s) 8, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Zhang C. et al., in view of Fu H. et al, and further in view of Giovannini A. et al, (US. Patent Application Publication 20230252268 A1) effectively filed on February 7, 2022, (hereafter Giovannini).
Claim 8:
Regarding claim 8, Zhang in view of Fu teaches the limitations of claim 1.
Neither Zhang or Fu appear to explicitly teach “The method of claim 1, wherein the action includes automatically altering a patient’s treatment.”,
However in the same field of art, Giovannini teaches “The method of claim 1, wherein the action includes automatically altering a patient’s treatment.”
See Giovannini in paragraph [0054] describing, “Embodiments may leverage graphs to predict a future patient status. Any treatment, intervention, or update, of the patient status would automatically translate into a change in the connectivity and nodes of a patient's graph, which can be exploited to model the temporal evolution of the patient (including disease prognosis) as a dynamically changing graph.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Giovannini by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Giovannini’s teachings of altering patient status with graphs.
One of ordinary skill in the art would be motivated to do so because by integrating Giovannini’s frameworks into the methods of Zhang and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring a “proposed method [that] can advantageously be applied to medical patients, by exploiting data in electronic health records (EHRs), with a view to predicting potential future states of such entities.” (Giovannini, paragraph [0025]).
Claim 18:
Regarding claim 18, Zhang in view of Fu teaches the limitations of claim 11.
Neither Zhang or Fu appear to explicitly teach “The system of claim 11, wherein the action includes automatically altering a patient’s treatment.”,
However in the same field of art, Giovannini teaches “The system of claim 11, wherein the action includes automatically altering a patient’s treatment.”
See Giovannini in paragraph [0054] describing, “Embodiments may leverage graphs to predict a future patient status. Any treatment, intervention, or update, of the patient status would automatically translate into a change in the connectivity and nodes of a patient's graph, which can be exploited to model the temporal evolution of the patient (including disease prognosis) as a dynamically changing graph.”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the base references of Zhang and Fu with the teachings of Giovannini by using Zhang’s teachings of a computer-implemented method for graphs and Fu’s teachings of a linkage score learning algorithm with graphs, and incorporate with Giovannini’s teachings of altering patient status with graphs.
One of ordinary skill in the art would be motivated to do so because by integrating Giovannini’s frameworks into the methods of Zhang and Fu, which are all in the same field of art of graph analysis, one of ordinary skill in the art would bring a “proposed method [that] can advantageously be applied to medical patients, by exploiting data in electronic health records (EHRs), with a view to predicting potential future states of such entities.” (Giovannini, paragraph [0025]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HASSAN R SESAY whose telephone number is (571)272-8493. The examiner can normally be reached Monday-Friday 8am-5pm.
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, Usmaan Saeed can be reached at (571) 272-4046. 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.
/HASSAN RAMADAN SESAY/Examiner, Art Unit 2146
/USMAAN SAEED/Supervisory Patent Examiner, Art Unit 2146
Prosecution Insights