DETAILED ACTION
Notice of Pre-AIA or AIA Status
1. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . The Amendment filed 28 January, 2026 (hereinafter “the Amendment’) has been entered and considered. Claims 1, 8-9, 17, 23, and 29-30 have been amended. Claim 31 has been added. Claims 1-31 are rejected. All modifications to the grounds of rejection set forth in the present action were necessitated by Applicants’ claim amendments
REMARKS
Response to Amendment
2. On page 10-11 of the Amendment, the Applicant contends that the combination of Shkolnisky in view of Cucuringu does not explicitly disclose the newly added feature to the independent claim which recites in some form, “wherein at least one edge of the one or more edges comprises a weighted edge including a weight indicating the similarity metric between the plurality of images”, and the Applicant further contends that this feature is not explicitly disclosed by the combination of Shkolnisky in view of Cucuringu in further view of Fan.
The Examiner agrees that the applied prior art is deficient in disclosing “wherein at least one edge of the one or more edges comprises a weighted edge including a weight indicating the similarity metric between the plurality of images”.
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.
3. Claims 1-5, 8-11, 13-21, 23-25, and 27-31 are rejected under 35 U.S.C. 103 as being unpatentable over “Viewing Direction Estimation in Cryo-EM Using Synchronization” by Yoel Shkolnisky et al., (herein after “Shkolnisky”) in view of “Eigenvector Synchronization, Graph rigidity and the Molecule Problem” by Mihai Cucuringu et al., (herein after “Cucuringu”), and in further view of EP 3382644 A1: Yash Singh, (herein after “Singh”).
Regarding claim 1,
A computer-implemented method of pose estimation (Shkolnisky, §6.1, P[004]: “on a dual Intel Xeon X5560 CPU”), comprising: receiving image data, wherein the image data comprises a plurality of images taken of varying poses (Shkolnisky, §1, P[001]: “the structure is determined from images of randomly oriented and positioned identical copies of the investigated molecule.”, where the random orientations are the varying poses.);
identifying one or more pairs of spatially related images within the plurality of images (Shkolnisky, §Abstract: “Information about the spatial relation between pairs of images is extracted from common lines-”);
and estimating a pose of the object depicted in the plurality of images (Shkolnisky, Fig. 7 discloses reconstruction from estimated orientations, where the estimated orientation is the estimated pose of the object.) based on the synchronization graph. Shkolnisky does not explicitly disclose that the estimated orientations of the object were based on a “synchronization graph”.
However, Cucuringu discloses in the §Abstract: “for every node, a subgraph of its 1-hop neighborhood graph, which can be accurately embedded in its own coordinate system … robust to high levels of noise in the measured distances and to sparse connectivity in the measurement graph,”, and furthermore in Fig. 1.2 which discloses “Using the pairwise alignments, in Step 1 we estimate both the reflection and rotation matrix from an eigenvector synchronization computation over O(3), while in Step 2 we find the estimated coordinates by solving an overdetermined system of linear equations. If there is available information on the reflection or rotations of some patches, one may choose to further divide Step 1 into two consecutive steps. Step 1a is synchronization over Z2, while Step 1b is synchronization over SO(3), in which the missing reflections and rotations are estimated.”
generating a synchronization graph indicative of at least one similarity metric between the plurality of images, based at least in part on the identified one or more pairs of spatially related images is disclosed by Cucuringu in Fig. 1.2 where it discloses a synchronization graph where the lines represent non-negative distance measurements associated with each edge. , wherein the synchronization graph comprises one or more edges which correspond to estimated relative poses of the object in the identified one or more pairs of spatially related images (Cucuringu, §I, P[005]: “General properties of proteins such as bond lengths and angles can be translated into accurate distance constraints.”, where the bond lengths and angles represent the estimated pose for that node, and the edges that link the nodes together are determined based off of the distance constraints, which contain the estimated pose information, therefore, the edges correspond to estimated poses of the nodes.);
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Shkolnisky to utilize a synchronization graph to aid in estimating the orientation of a particle/molecule, as taught by Cucuringu, to arrive at the claimed invention discussed above. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. It is predictable that the proposed modification would have provided the benefit of increasing the orientation estimation algorithm’s robustness against noise in images (Cucuringu, Abstract, “Our extensive numerical simulations show that 3D-ASAP and 3D-SP-ASAP are very robust to high levels of noise in the measured distances and to sparse connectivity in the measurement graph, and compare favorably to similar state-of-the art localization algorithms.”).
The combination of Shkolnisky and Cucuringu does not explicitly disclose “and wherein at least one edge of the one or more edges comprises a weighted edge including a weight indicating the similarity metric between the plurality of images;” However, Singh discloses and wherein at least one edge of the one or more edges comprises a weighted edge including a weight indicating the similarity metric between the plurality of images in P[0013]: “
a) collecting said images;
b) extracting keypoints from each image and generating a descriptor for each keypoint;
c) organizing the images in a proximity graph;
d) pairwise image matching and generating keypoints connecting tracks according to the maximum proximity between the keypoints themselves;
e) performing an autocalibration between image clusters to extract the internal and external parameters of the acquisition devices,
f) performing a Euclidean reconstruction of the object in form of said sparse 3D point cloud based on the parameters extracted at the preceding step.”
, and further in P[0046]: “Therefore, the graph may be defined by the specification of set of vertices and a set of weighted edges set as follows: the vertices are the images and the weighted edge between two images is defined as the number of matches with larger scale between the images. Then couples of neighbor images may be extracted from this fore said graph by iterating a maximum spanning tree algorithm several times in order to guarantee a k-edge connection on the resulting sub-graph.”, where Singh discloses that they utilize a weighted edge which includes a weight that indicates the levels of similarity between images that represent objects.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Shkolnisky and Cucuringu to utilize a weighted edge indicative of a similarity metric, as taught by Singh, to arrive at the claimed invention discussed above. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. It is predictable that the proposed modification would have provided the benefit of reducing computational complexity, and to more accurately represent similarity.
Regarding claim 2, wherein at least one pair of the one or more pairs of spatially related images comprises two mirrored images is disclosed by Cucuringu in §1, P[003]: “Applying a rigid transformation (composition of rotation, translation and possibly reflection)”, therefore Cucuringu discloses mirrored images through reflection, which is further disclosed in Fig. 1.2 and the description of the Figure.
Claim 18 recites features nearly identical to those recited in claim 2. Claim 18 is rejected for reasons analogous to those discussed above in conjunction with claim 2.
Regarding claim 3, wherein at least one pair of the one or more pairs of spatially related images comprises planar rotated images is disclosed by Cucuringu in §1, P[007]: “an eigenvector method for solving the synchronization problem over the group SO(2) of planar rotations. This algorithm serves as the basic building block for our 2DASAP and 3D-ASAP algorithms.”
Claim 19 recites features nearly identical to those recited in claim 3. Claim 19 is rejected for reasons analogous to those discussed above in conjunction with claim 3.
Regarding claim 4, wherein the pose is an SO(3) pose is disclosed by Shkolnisky in §3, P[001]: “RN are unknown rotations corresponding to the orientation of the molecule at the moment of freezing (elements of SO(3), the group of rotations of the three dimensional space). Each image is thus formed by projecting φ in some random direction and is therefore known as a “projection image.””
Claim 20 recites features nearly identical to those recited in claim 4. Claim 20 is rejected for reasons analogous to those discussed above in conjunction with claim 4.
Regarding claim 5, further comprising providing the estimated pose of the object to a 3D reconstruction algorithm is disclosed by Shkolnisky in Fig. 7 where it discloses: “NE, reconstruction from noisy projections and estimated orientations; CE, reconstruction from clean projections and estimated orientations; NT, reconstruction from noisy projections and true orientations.” Therefore, the estimated poses are provided to the 3D reconstruction algorithm.
Claim 21 recites features nearly identical to those recited in claim 5. Claim 21 is rejected for reasons analogous to those discussed above in conjunction with claim 5.
Regarding claim 8, wherein estimating the pose of the object depicted in the plurality of images further comprises: generating one or more matrices indicative of the synchronization graph (Cucuringu, Table 3.1 discloses generating matrices indicative of the synchronization graph); denoising the one or more matrices (Cucuringu, Table 3.1 discloses median-based denoising.); and estimating the pose of the object based on the one or more matrices (Cucuringu, Table 3.1 discloses estimating pose and coordinates based on the matrices.)
Claim 23 recites features nearly identical to those recited in claim 8. Claim 23 is rejected for reasons analogous to those discussed above in conjunction with claim 8.
Regarding claim 9, the one or more matrices comprise Graph-Connection Laplacians (GCLs); and at least one of the GCLs associated with the one or more matrices is indicative of a frequency of the images is disclosed by Cucuringu in §3.1, P[005]: “the normalized discrete graph Laplacian L is defined as L = I − Δ−1*AP”, where the graph Laplacian is the matrix representation that captures the structural properties and connectivity patterns of the network where the eigenvalues characterize the networks frequency spectrum, where the nodes represent images and the lines are distance measurements that connect the nodes.
Regarding claim 10, wherein estimating the pose of the object based on the plurality of images comprises performing an eigenvalue decomposition is disclosed by Cucuringu in §4.6, P[001]: “the best rigid transformation between two sets of points is obtained by various matrix manipulations and eigenvalue/eigenvector decomposition.”, where the transformations are used to estimate the pose of the object.
Claim 24 recites features nearly identical to those recited in claim 10. Claim 24 is rejected for reasons analogous to those discussed above in conjunction with claim 10.
Regarding claim 11, wherein estimating the pose of the object further comprises computing top three eigenvectors of a tangent vector bundle (Cucuringu, Table 3.1 discloses computing the top three eigenvectors of the vertical bundle, where the vertical bundles are specific sub-bundles within the tangent bundle.), computing two bases based on the eigenvectors and computing a third basis based on the two bases, where the bases/basis are found by verifying that the eigenvectors are linearly independent, which is disclosed by Cucuringu in Table 3.1 where they apply a least squares solution for each coordinate to confirm linear independence. Furthermore, when the matrix is symmetric there will always be a basis with eigenvectors, and if the eigenvectors are linearly independent, the eigenvalues are distinct. Even further, computing a third basis from the first two bases is interpreted as finding a third base that is linearly independent from the first two, and Cucuringu disclosed that they apply the least squares solution for each coordinate (out of three) to verify linear independence.
Claim 25 recites features nearly identical to those recited in claim 11. Claim 25 is rejected for reasons analogous to those discussed above in conjunction with claim 11.
Regarding claim 13, wherein the synchronization graph further comprises a plurality of vertices and wherein a vertex of the plurality of vertices indicates an image is disclosed by Cucuringu in §1, P[001]: “graph G = (V,E) consisting of a set of |V | = n nodes and |E| = m edges, together with a non-negative distance measurement dij associated with each edge,”, and furthermore in §3.1, P[002]: “We denote by GP = (V P ,EP ) the patch graph whose vertices V P are the patches P1, . . . , PN, and two patches Pi and Pj are adjacent, (Pi, Pj) ∈ EP , iff they have enough vertices in common to be aligned such that the ratio hih−1 j can be estimated. We let AP denote the adjacency matrix of the patch graph, i.e., AP ij = 1 if (Pi, Pj) ∈ EP, and AP ij = 0 otherwise. Obviously, two patches that are far apart and have no common nodes cannot be aligned-”, where common nodes indicate a similarity metric between the nodes which is defined by the edges/distance between the nodes.
Claim 27 recites features nearly identical to those recited in claim 13. Claim 27 is rejected for reasons analogous to those discussed above in conjunction with claim 13.
Regarding claim 14, wherein the similarity metric indicates a maximum similarity between two images of the plurality of images is disclosed by Cucuringu in §1, P[001]: “graph G = (V,E) consisting of a set of |V | = n nodes and |E| = m edges, together with a non-negative distance measurement dij associated with each edge,”, where the similarity metric is the distance between the nodes, where the maximum similarity is when two nodes overlap entirely or when the distance between the nodes is zero.
Regarding claim 15, wherein the image data comprises electron microscopy image data is disclosed by Shkolnisky in the §Abstract: “A central task in recovering the structure of a macromolecule from cryo-electron microscopy (cryo- EM) images is to determine a three-dimensional model of the macromolecule given many of its two-dimensional projection images.”
Regarding claim 16, wherein the object is a molecule is disclosed by Shkolnisky in the §Abstract: “A central task in recovering the structure of a macromolecule from cryo-electron microscopy (cryo- EM) images is to determine a three-dimensional model of the macromolecule given many of its two-dimensional projection images.”, and furthermore in §1, P[001]: “One of the fundamental tasks in structural biology is to recover the three-dimensional structure of molecules. Electron microscopy is one of the popular methods for that task, and in particular single particle reconstruction (SPR), where the structure is determined from images of randomly oriented and positioned identical copies of the investigated molecule.”
Regarding claim 17, An apparatus comprising: memory configured to store computer-executable instructions; and at least one processor configured to execute the computer-executable instructions is disclosed by Shkolnisky in §6.1, P[004]: “All tests were executed on a dual Intel Xeon X5560 CPU (8 cores in total), with 48GB of RAM-” The rest of the features of claim 17 are recited nearly identically to those recited in claim 1. Claim 17 is rejected for reasons analogous to those discussed above in conjunction with claim 1.
Regarding claim 28, wherein: the image data comprises electron microscopy image data; and the object is a molecule is disclosed by Shkolnisky in the §Abstract: “A central task in recovering the structure of a macromolecule from cryo-electron microscopy (cryo- EM) images is to determine a three-dimensional model of the macromolecule given many of its two-dimensional projection images.”, and furthermore in §1, P[001]: “One of the fundamental tasks in structural biology is to recover the three-dimensional structure of molecules. Electron microscopy is one of the popular methods for that task, and in particular single particle reconstruction (SPR), where the structure is determined from images of randomly oriented and positioned identical copies of the investigated molecule.”
Regarding claim 29, An apparatus comprising: is disclosed by Cucuringu in §8, P[009]: “PC machine equipped with an Intel(R) Core(TM)2 Duo CPU E8500 @ 3.16GHz 4 GB RAM.”, and furthermore in the §Abstract: “importance in applications such as wireless sensor networks”, where the data from the sensor networks are sent wirelessly to a computer which is the user equipment. The rest of the features of claim 29 are recited nearly identically to those recited in claim 1. Claim 29 is rejected for reasons analogous to those discussed above in conjunction with claim 1.
Regarding claim 30, A non-transitory computer-readable medium having instructions stored thereon for is disclosed by Cucuringu in §8, P[009]: “PC machine equipped with an Intel(R) Core(TM)2 Duo CPU E8500 @ 3.16GHz 4 GB RAM.” The rest of the features of claim 30 are recited nearly identically to those recited in claim 1. Claim 30 is rejected for reasons analogous to those discussed above in conjunction with claim 1.
Regarding claim 31, wherein the similarity metric indicates a rotational similarity of the object between the plurality of images is disclosed by Cucuringu in the §Abstract: “for every node, a subgraph of its 1-hop neighborhood graph, which can be accurately embedded in its own coordinate system … robust to high levels of noise in the measured distances and to sparse connectivity in the measurement graph,”, and furthermore in Fig. 1.2 which discloses “Using the pairwise alignments, in Step 1 we estimate both the reflection and rotation matrix from an eigenvector synchronization computation over O(3), while in Step 2 we find the estimated coordinates by solving an overdetermined system of linear equations. If there is available information on the reflection or rotations of some patches, one may choose to further divide Step 1 into two consecutive steps. Step 1a is synchronization over Z2, while Step 1b is synchronization over SO(3), in which the missing reflections and rotations are estimated.”, where Fig. 1.2 discloses a synchronization graph where the lines represent non-negative distance measurements associated with each edge, and Cucuringu, §I, P[005]: “General properties of proteins such as bond lengths and angles can be translated into accurate distance constraints.” Therefore, the similarity metric is guided by bond lengths and angles which are indicative of (not based on), rotational similarity.
4. Claims 6-7 and 22 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Shkolnisky and Cucuringu in view of Singh, and in further view of “CryoPoseNet: End-to-End Simultaneous Learning of Single-particle Orientation and 3D Map Reconstruction from Cryo-electron Microscopy Data” by Youssef Nashed, (herein after “Nashed”).
Regarding claim 6, the combination of Shkolnisky and Cucuringu does not explicitly disclose wherein the 3D reconstruction algorithm is based on Expectation Maximization (EM).
However, Nashed discloses wherein the 3D reconstruction algorithm is based on Expectation Maximization (EM) in §2, P[004]: “The expectation step of the k-th iteration estimates a conditional distribution on the space of poses for each projection given the current estimate of the structure xk … The maximization step then uses these poses to update the structures-”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Shkolnisky and Cucuringu to base their reconstruction algorithm on expectation maximization, as taught by Nashed, to arrive at the claimed invention discussed above. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. It is predictable that the proposed modification would have provided the benefit of maximizing the likelihood of the acquired data while still meeting a map condition for map reconstruction.
Claim 22 recites features nearly identical to those recited in claim 6. Claim 22 is rejected for reasons analogous to those discussed above in conjunction with claim 6.
Regarding claim 7, wherein the 3D reconstruction algorithm is implemented by a neural network is disclosed by Nashed in §1: “we discuss a new approach to solving the single-particle orientation problem using a neural-network”
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Shkolnisky and Cucuringu to utilize a neural network, as taught by Nashed, to arrive at the claimed invention discussed above. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. It is predictable that the proposed modification would have provided the benefit of limiting the number of variables that need to be estimated and prevents the number of variables from growing as the number of measurements grow.
5. Claims 12 and 26 are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Shkolnisky, Cucuringu, and Singh in view of “Vector Diffusion Maps and the Connection Laplacian” by A. Singer et al., (herein after “Singer”).
Regarding claim 12, wherein estimating the pose of the object based on the one or more matrices comprises: combining the denoised one or more matrices to estimate a denoised relative pose (Cucuringu, Table 3.1 discloses denoising before outputting the coordinates for reconstruction.) and a denoise weight for at least one edge in the synchronization graph is disclosed by Cucuringu in §9, P[007]: “Second, in the step that synchronizes the translations using least squares, one may choose to give more weight to equations involving the “good” edges, keeping in mind however that such equations are not noise free since the direction of an edge may be noisy as a result of steps 1 and 2, even if the distances are accurate. However, note that 3D-ASAP does use the “good” edges as hard constraints in the gradient descent refinement at the end of step 3.”;
constructing a tangent bundle for at least one image in the plurality of images, wherein the tangent bundle is a fiber bundle is disclosed by Cucuringu in Table 3.1 where it discloses computing the top three eigenvectors of the vertical bundle, where the vertical bundles are specific sub-bundles within the calculated tangent bundle.;
constructing one or more vector bundles associated with the tangent bundle is disclosed by Cucuringu in Table 3.1 where it discloses computing the top three eigenvectors of the vertical bundle, where the vertical bundles are specific sub-bundles within the calculated tangent bundle.;
for at least one of the one or more vector bundles, constructing a discretized Vector-Diffusion Laplacian operator is described by Cucuringu in §3.1, P[005]: “thus the eigenvalues of D−1AP are the same as the eigenvalues of Δ−1AP , each with multiplicity 3. In addition, if _ denotes the 3N × 3N matrix with diagonal blocks hi, i = 1, . . . ,N, then the normalized alignment matrix H can be written as H = ΔD−1APΔ−1, (3.6) and thus H and D−1AP have the same eigenvalues, which are also the eigenvalues of Δ−1AP , each with multiplicity 3 … the normalized discrete graph Laplacian L is defined as L = I − Δ−1*AP”,;
The combination of Shkolnisky and Cucuringu does not explicitly disclose that the discretized vector-diffusion Laplacian operators are denoised.
However, Singer discloses denoising the at least one discretized Vector-Diffusion Laplacian operators in §10, P[003]: “the matrix that lies at the heart of the vector diffusion map framework approximates the connection- Laplacian operator.”, where the process of approximating denoises the operator, where approximating involves utilizing a matrix with less parameters which discards the less important noise.
constructing a denoised Vector-Diffusion Laplacian for the tangent bundle based on denoising the at least one discretized Vector-Diffusion Laplacian operators is disclosed by Singer in §10, P[003]: “the matrix that lies at the heart of the vector diffusion map framework approximates the connection- Laplacian operator.”, where the process of approximating denoises the operator, where approximating involves utilizing a matrix with less parameters which discards the less important noise, where the output is the denoised Laplacian.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Shkolnisky and Cucuringu to denoise the discretized vector-diffusion Laplacian operators, as taught by Singer, to arrive at the claimed invention discussed above. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. It is predictable that the proposed modification would have provided the benefit of increasing the efficiency of calculations through limiting the amount of data needed to be processed via denoising by approximation.
determining top eigenvector-fields of the denoised Vector-Diffusion Laplacian for the tangent bundle, where the top eigenvector-fields are represented by the eigenvectors with the greatest corresponding eigenvalues, which is disclosed by Singer in §5, P[004]: “the l-th eigenvector vl D1-1S1-I is a discrete approximation of the l-th eigen-vector field Xl of the connection-Laplacian ∇ 2 over M, which satisfies ∇ 2Xl = -λl Xl for some λl (greater than or equal to zero).” Furthermore, Cucuringu also discloses determining the top eigenvectors in Table 3.1 which correspond to the top eigenvalues.
and generating the pose of at least one of the plurality of images based on the top eigenvector-fields is disclosed by Cucuringu in Table 3.1 where it is disclosed that the top eigenvectors, which are representative of the top eigenvector fields, are utilized to generate pose/orientation information.
Claim 26 recites features nearly identical to those recited in claim 12. Claim 26 is rejected for reasons analogous to those discussed above in conjunction with claim 12.
6. Independent claims 1, 17, and 29-30 are rejected under 35 U.S.C. 103 as being unpatentable over Shkolnisky in view of Cucuringu in view of Singh and in further view of “Cryo-Electron Microscopy Image Analysis Using Multi-Frequency Vector Diffusion Maps” by Yifeng Fan et al., (herein after “Fan”), to make explicit to what may seem to be implicitly taught in Cucuringu.
Regarding independent claims 1, 17, and 29-30, each recite in some form, “wherein the synchronization graph comprises one or more edges which correspond to estimated relative poses of the identified one or more pairs of spatially related images”. For what the combination of Shkolnisky and Cucuringu implicitly disclose, Fan explicitly discloses wherein the synchronization graph comprises one or more edges which correspond to estimated relative poses of the identified one or more pairs of spatially related images in §III,§B, P[001]: “a local neighborhood graph (V;E) over the sphere with the nodes V representing the viewing directions and the edges E connecting images of similar views, i.e., (i; j) 2 E indicates hvi; vji _ 1. All the views of similar images are in a small spherical cap (see the upper middel panel of Fig. 3). In addition, the rotational alignment has cycle consistency among the neighbors”, and see Fig. 3.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combination of Shkolnisky, Cucuringu, and Singh to explicitly incorporate pose information related to the edges between the nodes, as taught by Fan, to arrive at the claimed invention discussed above. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. It is predictable that the proposed modification would have provided the benefit of improving the efficiency and accuracy of cryo-EM image classification.
Conclusion
7. Any inquiry concerning this communication or earlier communications from the examiner should
be directed to TY M BEATTY whose telephone number is (703)756-5370. The examiner can normally be
reached Mon-Fri: 8AM-4PM EST..
Examiner interviews are available via telephone, in-person, and video conferencing using a
USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use
the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor,
Gregory Morse can be reached at (571) 272 - 3838. 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.
/TY MITCHELL BEATTY/Examiner, Art Unit 2663
/GREGORY A MORSE/Supervisory Patent Examiner, Art Unit 2698