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 .
Response to Arguments
The reply filed on 2/10/2026 has been entered. Applicant’s arguments with respect to claims 1-20 have been considered but are moot in view of new ground(s) of rejection caused by the amendments.
Claims 1-20 are pending in this application and have been considered below.
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Information Disclosure Statement
The IDS dated 11/15/2023 that have been previously considered remain placed in the application file.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 2, 3, 4, 5, 11, 13, 14, 15, 16, and 17 are rejected under 35 U.S.C. 103 as being unpatentable by (Pack JD, Manohar A, Ramani S, et al. Four-dimensional computed tomography of the left ventricle, Part I: Motion artifact reduction. Med Phys. 2022; 49: 4404-4418.) (Pack et al.) in view of U.S. Patent Publication US-20160256127, (Lee et al.) and U.S. Patent Publication US-20190328341, (Katsevich et al.)
Claim 1
Regarding claim 1, Pack et al. teach a medical image processing method (pg.
4408, sec. 3, "ResyncCT Method"), comprising: obtaining a set of projection data (pg. 4410, fig. 5, "x-ray source position"; pg. 4415, sec. 5, "projection views") acquired in a computed tomography (CT) scan of a three-dimensional region of an object to be examined (pg.4406, sec. 2.1, "The data coming from a CT scanner actually has excellent temporal resolution as each sinogram..."); generating for each time point of a plurality of time points (pg. 4409, sec. 3.2, "temporal series") of the CT scan based on a part of the obtained set of projection data corresponding to the time point, a pair of feature maps (pg. 4410, sec. 3.3, "conjugate-PAR image pairs"; pg.4410, fig. 5) for estimating motion at the time point so as to generate a plurality of pairs of feature maps (pg.4410, fig. 5, "A1 B1", "A2 B2", "A3 B3"), each feature map representing a feature of an image (pg.4410, sec. 3.2, "moving features") reconstructed from the part of the obtained set of projection data; estimating, based on the generated plurality of pairs of feature maps (pg. 4411, sec. 3.3, "The result is a correlation map: a quantitative indicator showing how well the conjugate-PAR images match at the subvolume A..."), a four-dimensional motion field, wherein the four-dimensional motion field indicates change of a motion of the object in the three-dimensional region over time (pg. 4409, sec. 3, "4D motion field").
Pack et al. does not disclose wherein the generating step comprises utilizing a nonlinear transform or utilizing a feature-extraction machine-learning model and reconstructing, based on the estimated four-dimensional motion field and the obtained set of projection data, a CT image of the object.
However, Katsevich et al. teach wherein the generating step comprises utilizing a nonlinear transform or utilizing a feature-extraction machine-learning model ("A nonlinear activation function is then applied to the intermediate images, which are then convolved using a second set of filters 720 and the procedure is repeated several times, depending upon the exact architecture of the CNN," par. 67).
Additionally, Lee et al. teach reconstructing, based on the estimated four-
dimensional motion field and the obtained set of projection data, a CT image of the
object (par. 351, "the first information may be acquired using the motion vector field",
par. 354, "reconstructing a target image, by using the first information, the image
processing unit may perform motion correction on a surface or region using pieces of
projection data").
It would have been obvious to a person having ordinary skill in the art before the
time of the effective filing date of the claimed invention of the instant application to
modify the medical image processing method as taught by Pack et al. to reconstruct a
CT image based on the estimated four-dimensional motion field and the projection data
as taught by Lee et al.
At the time the invention was made, it would have been obvious to one of
ordinary skill in the art to modify the medical image processing method to include
reconstructing a CT image based on the estimated four-dimensional motion field and
the projection data because such a modification is the result of combining prior art
elements according to known methods to yield predictable results. More specifically, the medical image processing method as modified with the step of reconstructing a CT image based on the estimated four-dimensional motion field and the projection data can yield a predictable result of improved image quality, reduced motion artifacts, and more accurate representation of dynamic anatomy compared to conventional CT reconstruction since incorporating the 4D motion field, generated by the medical image processing method of Pack et al., the reconstruction algorithm can compensate for the motion and yield a more accurate image. Thus, a person of ordinary skill would have appreciated including in the medical image processing method the ability to reconstruct a CT image based on the estimated four-dimensional motion field and the projection data since the claimed invention is merely a combination of old elements, and in the combination each element merely would have performed the same function as it did separately, and one of ordinary skill in the art would have recognized that the results of the combination were predictable.
Therefore, taking the teachings of Pack et al., Lee et al., and Katsevich et al. as a whole, it would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify the medical image processing method as taught by Pack et al. and the estimated four-dimensional motion field as taught by Lee et al. to use a nonlinear transform as taught by Katsevich et al. The suggestion/motivation for doing so would have been that, “Activation functions are also incorporated into various layers to introduce nonlinearity and enable the network to learn complex predictive relationships.” as noted by the Katsevich et al. disclosure in paragraph [0072], which also motivates combination because the combination would predictably have a higher efficiency as there is a reasonable expectation that incorporating such a nonlinear transform enables enhanced non-rigid motion modeling and improved accuracy in image reconstruction by modeling the complex relationships between motion fields and detector signals; and/or because doing so merely combines prior art elements according to known methods to yield predictable results.
The rejection of system of claim 1 above applies mutatis mutandis to the
corresponding limitations of claim 13. Claim 13 is mapped below for clarity of the record
and to specify any new limitations not included in claim 1.
Claim 2
Regarding claim 2, Pack et al., Lee et al., and Katsevich et al. teach the medical image processing method of claim 1.
Pack et al. teach wherein the generating step further comprises: applying feature extraction processing (pg. 4412, fig. 8, extraction step) to at least the part of the obtained set of projection data to obtain feature data (pg. 4410, sec. 3.2, "difference map"); and reconstructing the plurality of pairs of feature maps based on the obtained feature data (pg. 4412, fig. 8, masking/modified cross correlation step).
Pack et al., Lee et al., and Katsevich et al. are combined as per claim 1.
Claim 3
Regarding claim 3, Pack et al., Lee et al., and Katsevich et al. teach the medical image processing method of claim 2.
Pack et al. do not explicitly teach wherein the feature extraction processing comprises utilizing the nonlinear transform or utilizing a feature-extraction machine-learning model.
However, Katsevich et al. wherein the feature extraction processing comprises utilizing the nonlinear transform or utilizing a feature-extraction machine-learning model ("A nonlinear activation function is then applied to the intermediate images, which are then convolved using a second set of filters 720 and the procedure is repeated several times, depending upon the exact architecture of the CNN," par. 67).
Pack et al., Lee et al., and Katsevich et al. are combined as per claim 1.
Claim 4
Regarding claim 4, Pack et al., Lee et al., and Katsevich et al. teach the medical image processing method of claim 1.
Pack et al. teach wherein the generating step further comprises: reconstructing a plurality of partial angle reconstruction (PAR) images based on the part of the obtained set of projection data (pg. 4409, sec. 3.2, "PAR images that are reconstructed from parallel rebinned data"); and applying feature extraction processing on the plurality of PAR images to obtain the plurality of pairs of feature maps (pg. 4410, sec. 3.2, "For each pair, a high pass filter is applied filtration Gaussian kernel").
Pack et al., Lee et al., and Katsevich et al. are combined as per claim 1.
Claim 5
Regarding claim 5, Pack et al., Lee et al., and Katsevich et al. teach the medical image processing method of claim 1.
Pack et al. teach wherein the estimating step further comprises: performing, for each pair of the plurality of pairs of feature maps, registration processing between the feature maps of the pair to generate a plurality of three-dimensional motion fields, one at each time point of the plurality of time points (pg. 4411, sec. 3.4, "produce a dense, voxelized estimate of this motion for each time frame of the PAR-based motion compensation". "This is accomplished by interpolating [3D velocity and acceleration] that describe the motion onto a dense grid"); and fitting the generated plurality of the three-dimensional motion fields to obtain the four-dimensional motion field (pg.4408-9, sec. 3, "interpolates these motions to build a dense 4D motion field").
Pack et al., Lee et al., and Katsevich et al. are combined as per claim 1.
Claim 11
Regarding claim 11, Pack et al., Lee et al., and Katsevich et al. teach the medical image processing method of claim 1.
Pack et al. teach wherein the estimating step further comprises obtaining the set of projection data using a computed tomography (CT) scanner apparatus (pg. 4406, sec. 2.1, "The data coming from a CT scanner as each sinogram view is acquired").
Pack et al., Lee et al., and Katsevich et al. are combined as per claim 1.
Claim 13
Regarding claim 13, Pack et al. teach a medical image processing apparatus, comprising: obtain a set of projection data (pg. 4410, fig. 5, "x-ray source position"; pg. 4415, sec. 5, "projection views") acquired in a computed tomography (CT) scan of a three-dimensional region of an object to be examined (pg.4406, sec. 2.1, "The data coming from a CT scanner actually has excellent temporal resolution as each sinogram..."); generate for each time point of a plurality of time points (pg. 4409, sec. 3.2, "temporal series") of the CT scan based on a part of the obtained set of projection data corresponding to the time point, a pair of feature maps (pg. 4410, sec. 3.3, "conjugate-PAR image pairs"; pg.4410, fig. 5) for estimating motion at the time point so as to generate a plurality of pairs of feature maps (pg.4410, fig. 5, "A1 B1", "A2 B2", "A3 B3"), each feature map representing a feature of an image (pg.4410, sec. 3.2, "moving features") reconstructed from the part of the obtained set of projection data; estimate, based on the generated plurality of pairs of feature maps (pg. 4411, sec. 3.3, "The result is a correlation map: a quantitative indicator showing how well the conjugate-PAR images match at the subvolume A..."), a four-dimensional motion field, wherein the four-dimensional motion field indicates change of a motion of the object in the three-dimensional region over time (pg. 4409, sec. 3, "4D motion field").
Pack et al. does not disclose processing circuitry; wherein the processing circuitry is configured to generate the feature maps utilizing a nonlinear transform or utilizing a feature-extraction machine-learning model and reconstruct, based on the estimated four-dimensional motion field and the obtained set of projection data, a CT image of the object.
However, Katsevich et al. teach processing circuitry configured to; ("These computer program instructions may be provided to a processor of a general-purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified," par. 84) wherein the processing circuitry is configured to generate the feature maps utilizing a nonlinear transform or utilizing a feature-extraction machine-learning model ("A nonlinear activation function is then applied to the intermediate images, which are then convolved using a second set of filters 720 and the procedure is repeated several times, depending upon the exact architecture of the CNN," par. 67).
Additionally, Lee et al. teach reconstruct, based on the estimated four-
dimensional motion field and the obtained set of projection data, a CT image of the
object (par. 351, "the first information may be acquired using the motion vector field",
par. 354, "reconstructing a target image, by using the first information, the image
processing unit may perform motion correction on a surface or region using pieces of
projection data").
Pack et al., Lee et al., and Katsevich et al. are combined as per claim 1.
Claim 14
Regarding claim 14, as previously mentioned, Pack et al., Lee et al., and Katsevich et al. teach the medical image processing apparatus of claim 13.
Pack et al. teach wherein in generating the pair of feature maps: apply feature extraction processing (pg. 4412, fig. 8, extraction step) to at least the part of the obtained set of projection data to obtain feature data (pg. 4410, sec. 3.2, "difference map"); and reconstruct the plurality of pairs of feature maps based on the obtained feature data (pg. 4412, fig. 8, masking/modified cross correlation step).
Pack et al. do not explicitly teach the processing circuitry is further configured.
However, Katsevich et al. teach the processing circuitry is further configured ("These computer program instructions may be provided to a processor of a general-purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified," par. 84).
Pack et al., Lee et al., and Katsevich et al. are combined as per claim 1.
Claim 15
Regarding claim 15, as previously mentioned, Pack et al., Lee et al., and Katsevich et al. teach the medical image processing apparatus of claim 14.
Pack et al. do not explicitly teach wherein the feature extraction processing performed by the processing circuitry comprises utilizing the nonlinear transform or utilizing the feature-extraction machine-learning model.
However, Katsevich et al. teach wherein the feature extraction processing performed by the processing circuitry ("These computer program instructions may be provided to a processor of a general-purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified," par. 84) comprises utilizing the nonlinear transform or utilizing the feature-extraction machine-learning model ("A nonlinear activation function is then applied to the intermediate images, which are then convolved using a second set of filters 720 and the procedure is repeated several times, depending upon the exact architecture of the CNN," par. 67).
Pack et al., Lee et al., and Katsevich et al. are combined as per claim 1.
Claim 16
Regarding claim 16, as previously mentioned, Pack et al., Lee et al., and Katsevich et al. teach the medical image processing apparatus of claim 13.
Pack et al. teach in generating the pair of feature maps: reconstruct a plurality of partial angle reconstruction (PAR) images based on the part of the obtained set of projection data (pg. 4409, sec. 3.2, "PAR images that are reconstructed from parallel rebinned data"); and apply feature extraction processing on the plurality of PAR images to obtain the plurality of pairs of feature maps (pg. 4410, sec. 3.2, "For each pair, a high pass filter is applied filtration Gaussian kernel").
Pack et al. do not explicitly teach the processing circuitry is further configured.
However, Katsevich et al. teach the processing circuitry is further configured ("These computer program instructions may be provided to a processor of a general-purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified," par. 84).
Pack et al., Lee et al., and Katsevich et al. are combined as per claim 1.
Claim 17
Regarding claim 17, as previously mentioned, Pack et al., Lee et al., and Katsevich et al. teach the medical image processing apparatus of claim 13.
Pack et al. teach in estimating the four-dimensional motion field: performing, for each pair of the plurality of pairs of feature maps, registration processing between the feature maps of the pair to generate a plurality of three-dimensional motion fields, one at each time point of the plurality of time points (pg. 4411, sec. 3.4, "produce a dense, voxelized estimate of this motion for each time frame of the PAR-based motion compensation". "This is accomplished by interpolating [3D velocity and acceleration] that describe the motion onto a dense grid"); and fitting the generated plurality of the three-dimensional motion fields to obtain the four-dimensional motion field (pg.4408-9, sec. 3, "interpolates these motions to build a dense 4D motion field").
Pack et al. do not explicitly teach the processing circuitry is further configured.
However, Katsevich et al. teach the processing circuitry is further configured ("These computer program instructions may be provided to a processor of a general-purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified," par. 84).
Pack et al., Lee et al., and Katsevich et al. are combined as per claim 1.
2nd Claim Rejections - 35 USC § 103
Claims 6, 7, 8, 9, 18, 19, and 20 are rejected under 35 U.S.C. 103 as being unpatentable by (Pack JD, Manohar A, Ramani S, et al. Four-dimensional computed tomography of the left ventricle, Part I: Motion artifact reduction. Med Phys. 2022; 49: 4404-4418.) (Pack et al.), U.S. Patent Publication US-20160256127, (Lee et al.) and U.S. Patent Publication US-20190328341, (Katsevich et al.) in further view of US Patent Publication 20230206477 A1, (Akahori).
Claim 6
Regarding claim 6, Pack et al. teach the medical image processing method of claim 1 and fitting the generated plurality of the three-dimensional motion fields to obtain the four-dimensional motion field (pg.4408-9, sec. 3, "interpolates these motions to build
a dense 4D motion field").
Pack et al. do not explicitly teach applying, to a trained machine-learning model for motion estimation, each pair of the plurality of pairs of the feature maps to generate a plurality of three-dimensional motion fields, one at each time point of the plurality of time points.
However, Akahori teaches applying, to a trained machine-learning model for
motion estimation, each pair of the plurality of pairs of the feature maps to generate a
plurality of three-dimensional motion fields, one at each time point of the plurality of time
points (par. 21, "the first neural network and the second neural network may be trained
models and a process of machine learning may be performed such that a combination of feature maps of two images obtained by inputting each of the two images outputs the deformation vector field").
Therefore, taking the teachings of Pack et al., Lee et al., Katsevich et al., and Akahori as a whole, it would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify the medical image processing method as taught by Pack et al., the estimated four-dimensional motion field as taught by Lee et al., and a nonlinear transform as taught by Katsevich et al., to use trained machined learning models to process pairs of feature maps and output three-dimensional motion fields as taught by Akahori. The suggestion/motivation for doing so would have been that, “a trained model can suppress calculation resources required for registration between a plurality of images” as noted by the Akahori disclosure in paragraph [0015], which also motivates combination because the combination would predictably have a higher efficiency as there is a reasonable expectation that the resulting trained model would efficiently produce accurate deformation fields for registration, thereby reducing the computational time and resources needed for iterative optimization techniques; and/or because doing so merely combines prior art elements according to known methods to yield predictable results.
The rejection of system of claim 6 above applies mutatis mutandis to the corresponding limitations of claim 18. Claim 18 is mapped below for clarity of the record
and to specify any new limitations not included in claim 6.
Claim 7
Regarding claim 7, Pack et al., Lee et al., Katsevich et al., and Akahori teach the medical image processing method of claim 6 as shown above.
Park et al. does not teach the trained machine learning model for motion estimation is a 3D deep convolutional neural network.
However, Akahori teaches the trained machine-learning model for motion
estimation is a 3D deep convolutional neural network (par. 59, "a trained model that is
configured using a convolutional neural network and is trained to output the deformation
vector field").
Pack et al., Lee et al., Katsevich et al., and Akahori are combined as per claim 6.
Claim 8
Regarding claim 8, Pack et al., Lee et al., Katsevich et al., and Akahori teach the medical image processing method of claim 6 as shown above.
Park et al. does not teach training a neural network using training data and a function that represents a disagreement between pairs of data as an error value, the
training data including pairs in which a pair includes defect-exhibiting data paired with
corresponding defect-minimized data, and the neural network including is trained by
performing, for each of the pairs, the steps of applying the neural network to defect-
exhibiting data of a pair to generate network processed data; calculating, using the
function, the error value between the network processed data and the defect-minimized
data of the pair; updating, based on the calculated error value, the weighting coefficients
of the neural network; and repeating the steps of applying, calculating, and updating
using respective pairs of the training data until one or more stopping criteria are satisfied.
However, Akahori teaches training a neural network using training data (par. 48, fig. 12, "that trains a learning model using training data") and a function that represents
a disagreement between pairs of data as an error value (par. 150, "the calculation of the result of loss indicating an error between the output of the learning model and a teacher signal"), the training data including pairs in which a pair includes defect-exhibiting data paired with corresponding defect-minimized data (par. 145, "training data including a pair of the augmentation training image and the augmentation deformation training image"), and the neural network including is trained by performing, for each of the pairs, the steps of applying the neural network to defect-exhibiting data of a pair (par. 149 "Each of the augmentation training image and the augmentation deformation training image is input to the first neural network") to generate network processed data (par. 155 , "amount of deformation output from the learning model"); calculating, using the function, the error value between the network processed data and the defect-minimized data of the pair (par. 150, "the calculation of the result of loss indicating an error between the output of the learning model and a teacher signal"); updating, based on the calculated error value, the weighting coefficients of the neural network (par. 155, "The parameters of the learning model are updated", par. 150, "the parameters of the learning model include a weight for the coupling between nodes"); and repeating the steps of applying, calculating, and updating using respective pairs of the training data until one or more stopping criteria are satisfied (fig. 12, par. 151, a learning process is performed to optimize the parameters of the learning model and to generate the registration model having the desired performance"). One could reasonably infer that reaching the desired performance would constitute stopping the learning model.
Pack et al., Lee et al., Katsevich et al., and Akahori are combined as per claim 6.
Claim 9
Regarding claim 9, Pack et al., Lee et al., Katsevich et al., and Akahori teach the medical image processing method of claim 1.
Pack et al. teach each one of the plurality of pairs of the feature maps to generate the four-dimensional motion field (pg.4408-9, sec. 3, "interpolates these motions to build a dense 4D motion field").
Pack et al. do not explicitly teach a trained machine-learning model for motion
estimation.
However, Akahori teaches a trained machine learning model for motion
estimation (par. 21, "the first neural network and the second neural network may be
trained models and a process of machine learning may be performed such that a
combination of feature maps").
Pack et al., Lee et al., Katsevich et al., and Akahori are combined as per claim 6.
Claim 18
Regarding claim 18, as previously mentioned, Pack et al., Lee et al., and Katsevich et al. teach the medical image processing apparatus of claim 13.
Pack et al. teach to fit the generated plurality of the three-dimensional motion fields to obtain the four-dimensional motion field (pg.4408-9, sec. 3, "interpolates these motions to build a dense 4D motion field").
Pack et al. do not explicitly teach processing circuitry or applying, to a trained machine-learning model for motion estimation, each pair of the plurality of pairs of the feature maps to generate a plurality of three-dimensional motion fields, one at each time point of the plurality of time points.
However, Katsevich et al. teach processing circuitry ("These computer program instructions may be provided to a processor of a general-purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified," par. 84).
Additionally, Akahori teaches applying, to a trained machine-learning model for motion estimation, each pair of the plurality of pairs of the feature maps to generate a plurality of three-dimensional motion fields, one at each time point of the plurality of time points (par. 21, "the first neural network and the second neural network may be trained
models and a process of machine learning may be performed such that a combination of feature maps of two images obtained by inputting each of the two images outputs the deformation vector field").
Pack et al., Lee et al., Katsevich et al., and Akahori are combined as per claim 6.
Claim 19
Regarding claim 19, Pack et al., Lee et al., Katsevich et al., and Akahori teach the medical image processing apparatus of claim 18.
Park et al. do not explicitly teach processing circuitry or the trained machine
learning model for motion estimation is a 3D deep convolutional neural network.
However, Katsevich et al. teach processing circuitry ("These computer program instructions may be provided to a processor of a general-purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified," par. 84).
Additionally, Akahori teaches the trained machine-learning model for motion estimation is a 3D deep convolutional neural network (par. 59, "a trained model that is
configured using a convolutional neural network and is trained to output the deformation
vector field").
Pack et al., Lee et al., Katsevich et al., and Akahori are combined as per claim 6.
Claim 20
Regarding claim 20, as previously mentioned, Pack et al., Lee et al., and Katsevich et al. teach the medical image processing apparatus of claim 13.
Park et al. teach using the plurality of pairs of the feature maps to generate the four-dimensional motion field (pg.4408-9, sec. 3, "interpolates these motions to build a dense 4D motion field").
Park et al. do not explicitly teach processing circuitry or a trained machine- learning model for motion estimation.
However, Katsevich et al. teach processing circuitry ("These computer program instructions may be provided to a processor of a general-purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified," par. 84).
Additionally, Akahori teaches trained machine learning model for motion estimation (par. 21, "the first neural network and the second neural network may be trained models and a process of machine learning may be performed such that a combination of feature maps of two images obtained by inputting each of the two images outputs the deformation vector field").
Pack et al., Lee et al., Katsevich et al., and Akahori are combined as per claim 6.
3rd Claim Rejections - 35 USC § 103
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Park et al.,
Lee et al., Katsevich et al., and Akahori in further view of Myronenko et al. (2020). 4D CNN for semantic segmentation of cardiac volumetric sequences. Lecture Notes in Computer Science, 72-80. (Myronenko et al.).
Regarding claim 10, Pack et al., Lee et al., Katsevich et al., and Akahori teach the medical image processing method of claim 9.
Akahori teaches a trained machine learning model for motion estimation (par. 21, "the first neural network and the second neural network may be trained models and a process of machine learning may be performed such that a combination of feature maps").
Park et al. and Akahori do not teach a 4D deep convolutional neural network.
However, Myronenko et al. teach a 4D deep convolutional neural network (pg. 7,
"4D CNN").
Therefore, taking the teachings of Pack et al., Lee et al., Katsevich et al., Akahori, and Myronenko et al. as a whole, it would have been obvious to a person having ordinary skill in the art before the time of the effective filing date of the claimed invention of the instant application to modify the medical image processing method as taught by Pack et al., the estimated four-dimensional motion field as taught by Lee et al., a nonlinear transform as taught by Katsevich et al., and trained machine learning models to process pairs of feature maps and output three-dimensional motion fields as taught by Akahori, to use a $D deep convolutional neural network as taught by Myronenko et al. The suggestion/motivation for doing so would have been that, “the temporal dimension [of 4D CNNs] offers valuable information that is otherwise lost when treating each volume independently” as noted by the Myronenko et al. disclosure in page [2], which also motivates combination because the combination would predictably have additional utility as there is a reasonable expectation that leveraging temporal information across the 4D (3D+time) sequence would improve segmentation accuracy, enhance motion estimation, and reduce noise compared to independent 3D frame analysis and/or because doing so merely combines prior art elements according to known methods to yield predictable results.
4th Claim Rejections - 35 USC § 103
Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over Park et al., Lee et al., and Katsevich et al. in further view of US Patent Publication 20210049734, (Wahrenberg et al.).
Claim 12
Regarding claim 12, Park et al., Lee et al., and Katsevich et al. teach the medical image processing method of claim 11.
Park et al. do not teach using a helical scan or a volume scan.
However, Wahrenberg et al. teach a helical scan or a volume scan (par. 44,
"volumetric CT imaging data").
At the time the invention was made, it would have been obvious to one of
ordinary skill in the art to modify the computed tomography (CT) scanner apparatus to
include a helical scan or a volume scan because such a modification is the result of
applying a known technique to a known device ready for improvement to yield
predictable results. More specifically, a helical scan or a volume scan permits a faster
and larger scan while also producing continuous data for 3D reconstruction. This known
benefit is applicable to the computed tomography (CT) scanner apparatus as they both
share characteristics and capabilities, namely, they are directed to diagnostic imaging.
Therefore, it would have been recognized that modifying the computed tomography
(CT) scanner apparatus to include a helical scan or a volume scan would have yielded
predictable results because (i) the level of ordinary skill in the art demonstrated by the
references applied shows the ability to incorporate a helical scan or a volume scan in
diagnostic imaging and (ii) the benefits of such a combination would have been
recognized by those of ordinary skill in the art.
Reference Cited
The prior art made of record and not relied upon is considered pertinent to
applicant's disclosure.
US Patent Publication 2020 US2020/0305806 A1 to Tang et al. discloses neural
network training methods configured to assist in the reconstruction of CT images.
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Karsten F. Lantz whose telephone number is (571)272-4564. The examiner can normally be reached Monday-Friday 8:00-4:00.
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, Ms. Jennifer Mehmood can be reached on 571-272-2976. 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.
/Karsten F. Lantz/Examiner, Art Unit 2664
Date: 4/20/2026
/JENNIFER MEHMOOD/Supervisory Patent Examiner, Art Unit 2664
Prosecution Insights