DETAILED ACTION
This final office action is responsive to the amendment filed on February 25, 2026. Claims 1-16 are pending. Claims 1, 8, and 15 are independent. Claim 16 is added.
The objection to claim 2 is withdrawn in light of applicant’s amendment.
Claim rejections under 35 USC §103 are withdrawn in light of applicant’s arguments. However, a new grounds of rejection is made in light of applicant’s IDS provided art.
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
All references in the information disclosure statement (IDS) filed on September 7, 2023 has been considered by the examiner except for CN 108830796 which is not submitted in a legible format. A copy of CN 108830796 is required to be resubmitted in a legible format that can be read prior to being considered by the examiner.
The information disclosure statement (IDS) submitted on February 23, 2026 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 § 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.
Claims 1, 2, 4-5, 8-9, 11-12, and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Schlemper et al. (US20200289019), hereinafter Schlemper, in view of Pasupuleti et al. (WO2021182798), hereinafter Pasupuleti.
Regarding claim 1, Schlemper teaches the method:
obtaining an input image signal, wherein the input image signal is a spatial domain signal; (Schlemper, paragraph 0083: “For example, one or more of the neural networks described herein may be configured to receive as input, data in the “sensor domain”, “spatial-frequency domain” (also known as k-space), and/or the image domain.”)
converting the received input image signal into a corresponding spectral domain signal; (Schlemper, paragraph 0145: “FIG. 3C shows an example implementation of data consistency block 320, in which the image domain input 322, is transformed to the spatial frequency domain through a series of transformations 324, 326, and 328, whose composition is used to implement a non-uniform fast Fourier transformation from the image domain to the spatial frequency domain.” And paragraph 0176: “The neural networks 220, 224, and 226 may be implemented in any suitable domain. For example, in some embodiments, each of one or more of these networks may be applied in the sensor domain, spectral domain, log spectral domain, time domain, spatial frequency domain, wavelet domain, and/or any other suitable domain, as aspects of the technology described herein are not limited in this respect.” – The image domain input is being transformed, which can be into any suitable domain including the spectral domain.)
calculating a final loss based on a first loss, a second loss, and a third loss; and (Schlemper, paragraph 0082: “In some embodiments, the common loss function is a weighted combination of a first loss function for the pre-reconstruction neural network, a second loss function for the reconstruction neural network, and a third loss function for the post-reconstruction neural network.” – The common loss function is analogous to the final loss.)
training the neural network using the calculated final loss. (Schlemper, paragraph 0082: “In some embodiments, the pre-reconstruction neural network, the reconstruction neural network, and the post-reconstruction neural network are jointly trained with respect to a common loss function.”)
Schlemper does not explicitly teach:
extracting a first set of predetermined learning features from the spectral domain signal and a second set of predetermined learning features from the spatial domain signal;
converting the first set of predetermined learning features extracted from the spectral domain signal into the spatial domain signal;
concatenating the extracted second set of predetermined learning features and the converted first set of predetermined learning features;
However, Pasupuleti teaches:
extracting a first set of predetermined learning features from the spectral domain signal and a second set of predetermined learning features from the spatial domain signal; (Pasupuleti, paragraph 5: “Another object of the embodiments herein is to analyze features of the image in multiple domains, i.e. pixel domain and frequency domain. The method allows the electronic device to extract the features of the image in the multiple domains and choose optimal features from the extracted features for generating the high resolution image.” And paragraph 60: “The edge guidance CNN (419) generates the first set of feature maps by filtering the third set of feature maps and the HR edge map using the second set of ESBs (404-406).” And paragraph 63: “The refinement CNN (425) generate the fifth set of feature maps by filtering the fourth set of feature maps using a second set of ESBs (410-412) and the frequency components of the first set of feature maps.” – Generating the feature maps is analogous to extracting the sets of predetermined learning features where the features being able to be extracted in multiple domains indicates that the method is capable of extracting a first set from one domain, i.e. the spectral domain, and a second set from another domain, i.e. the spatial domain.)
converting the first set of predetermined learning features extracted from the spectral domain signal into the spatial domain signal; (Pasupuleti, paragraph 63: “The frequency to pixel transformer (426) converts the frequency components of the fifth set of feature maps to the pixels to form the second set of feature maps.”)
concatenating the extracted second set of predetermined learning features and the converted first set of predetermined learning features; (Pasupuleti, paragraph 63: “The joint refinement NN block (113) generates the HR image (305) using the first set of feature maps and the second set of feature maps. The joint refinement NN block (113) processes the first set of feature maps and the second set of feature maps and updates the kernel weights based on important features of both feature maps to enhances output quality of the HR image (305).” – The joint refinement block processing the first set and second set is analogous to concatenating the extracted second set and converted first set.)
Pasupuleti is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Schlemper, which already teaches obtaining an input image, converting the image into a different domain and calculating a final loss to train the neural network but does not explicitly teach extracting features from different domains and converting one set of features to the other domain before concatenating the feature sets, to include the teachings of Pasupuleti which does teach extracting features from different domains and converting one set of features to the other domain before concatenating the feature sets in order to allow computing systems with limited resources to "easily perform operations required for generating the high resolution image by using the ESBs.” (Pasupuleti, paragraph 4)
Regarding claim 2, Schlemper and Pasupuleti teach the method of claim 1, as cited above.
Schlemper further teaches:
calculating the first loss based on the first set of predetermined learning features and a predefined spectral domain ground truth associated with the input image signal. (Schlemper, paragraph 0082: “…first loss function for the pre-reconstruction neural network…” and paragraph 0205: “
L
1
=
f
y
h
-
x
h
” – Where the pre-reconstruction neural network includes the first set of predetermined learning features as those are the learning features from the spectral domain, e.g., the pre-reconstruction features, and
x
h
represents the spectral ground truth while
f
y
h
represents the learning features.)
Regarding claim 4, Schlemper and Pasupuleti teach the method of claim 1, as cited above.
Schlemper does not explicitly teach:
blending the set of concatenated learning features; and
calculating the third loss based on the set of blended learning features and a predefined ground truth associated with the received input image signal.
However, Pasupuleti further teaches:
blending the set of concatenated learning features; and (Pasupuleti, paragraph 46: “In an embodiment, the joint refinement NN block (113) learns kernel weights during a training phase, and updates the kernel weight based on a best feature in the first set of feature maps and the second set of feature maps, wherein the joint refinement NN block (113) comprises a set of ESBs (413-415).” – Using the best features in the first and second feature sets indicates that the learning features are being blended.)
calculating the third loss based on the set of blended learning features and a predefined ground truth associated with the received input image signal. (Pasupuleti, paragraph 46: “An objective loss function measures the error between CNN output and ground truth, wherein, the objective of supervised training is to minimize the loss function and thereby the error.” – where the objective loss function is analogous to the third loss function taught by Schlemper above.)
Regarding claim 5, Schlemper and Pasupuleti teach the method of claim 1, as cited above.
Schlemper further teaches:
wherein the final loss is determined based on weighted sum of a first weight, a second weight, and a third weight. (Schlemper, paragraph 0082: “In some embodiments, the common loss function is a weighted combination of a first loss function for the pre-reconstruction neural network, a second loss function for the reconstruction neural network, and a third loss function for the post-reconstruction neural network.” – The weighted combination of first loss function, second loss function, and third loss function is analogous to a final loss based on a weighted sum of a first weight, second weight, and third weight.)
Regarding claim 8, Claim 8 has all the same limitations of claim 1 which are taught by Schlemper and Pasupuleti – see claim 1 above.
Schlemper further teaches:
An electronic device for training a neural network using feature augmentation, the electronic device comprising: (Schlemper, paragraph 0416: “In some embodiments, computing device 2204 may be any electronic device(s) configured to process acquired MR data and generate image(s) of the subject being imaged.”)
memory, comprising one or more storage media, storing instructions; and at least one processor communicatively coupled to the memory, wherein the instructions, when executed by the at least one processor individually or collectively, cause the electronic device to: (Schlemper, paragraph 0427: “The computer system 2600 may include one or more processors 2610 and one or more articles of manufacture that comprise non-transitory computer-readable storage media (e.g., memory 2620 and one or more non-volatile storage media 2630).”)
Regarding claim 9, Schlemper and Pasupuleti teach the electronic device of claim 8, as cited above.
Claim 9 additionally has the same limitations of claim 2 which are taught by Schlemper and Pasupuleti – see claim 2 above.
Regarding claim 11, Schlemper and Pasupuleti teach the electronic device of claim 8, as cited above.
Claim 11 additionally has the same limitations of claim 4 which are taught by Schlemper and Pasupuleti – see claim 4 above.
Regarding claim 12, Schlemper and Pasupuleti teach the electronic device of claim 8, as cited above.
Claim 12 additionally has the same limitations of claim 5 which are taught by Schlemper and Pasupuleti – see claim 5 above.
Regarding claim 15, claim 15 has all the same limitations of claim 1 which are taught by Schlemper and Pasupuleti – see claim 1 above.
Schlemper further teaches:
One or more non-transitory computer-readable storage media storing one or more computer programs including computer-executable instructions that, when executed by one or more processors of an electronic device individually or collectively, cause the electronic device to perform operations, the operations comprising: (Schlemper, paragraph 0429: “One or more aspects and embodiments of the present disclosure involving the performance of processes or methods may utilize program instructions executable by a device (e.g., a computer, a processor, or other device) to perform, or control performance of, the processes or methods. In this respect, various inventive concepts may be embodied as a computer readable storage medium (or multiple computer readable storage media) (e.g., a computer memory, one or more floppy discs, compact discs, optical discs, magnetic tapes, flash memories, circuit configurations in Field Programmable Gate Arrays or other semiconductor devices, or other tangible computer storage medium) encoded with one or more programs that, when executed on one or more computers or other processors, perform methods that implement one or more of the various embodiments described above.”)
Claims 3 and 10 are rejected under 35 U.S.C. 103 as being unpatentable over Schlemper in view of Pasupuleti in view of Brehmer et al. (US20230074979), hereinafter Brehmer.
Regarding claim 3, Schlemper and Pasupuleti teach the method of claim 1, as cited above.
Schlemper and Pasupuleti do not explicitly teach:
extracting a set of reconstructed features from the extracted second set of predetermined learning features; and
calculating the second loss based on the extracted set of reconstructed features and a predefined ground truth associated with the received input image signal.
However, Brehmer teaches:
extracting a set of reconstructed features from the extracted second set of predetermined learning features; and (Brehmer, paragraph 0109: “The decoder 423 decompresses the code z and outputs an approximation
x
^
(which can be referred to as a reconstructed or decoded image) of the image content
x
.” – The approximation
x
^
is analogous to the reconstructed features.)
calculating the second loss based on the extracted set of reconstructed features and a predefined ground truth associated with the received input image signal. (Brehmer, paragraph 0112: “For example, the autoencoder 401 can compare
n
and
n
^
to determine a loss (e.g., represented by a distance vector or other difference value) between the first training image
n
and the reconstructed first training image
n
^
.” – Here
n
and
n
^
are analogous to
x
and
x
^
as recited above. Therefore, comparing them to determine a loss is analogous to the second loss function, which is already taught by Schlemper above, based on the extracted set of reconstructed features and a predefined ground truth.)
Brehmer is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Schlemper and Pasupuleti, which already teaches training a neural network using a total loss function but does not explicitly teach the second loss function is based on the extracted set of reconstructed features and a ground truth, to include the teachings of Brehmer which does teach the second loss function is based on the extracted set of reconstructed features and a ground truth in order to "analyze error in the output." (Brehmer, paragraph 0112)
Regarding claim 7, Schlemper and Pasupuleti teach the method of claim 1, as cited above.
Schlemper and Pasupuleti do not explicitly teach:
wherein the trained neural network is configured to perform video data compression.
However, Brehmer teaches:
wherein the trained neural network is configured to perform video data compression. (Brehmer, paragraph 0043: “In some examples, the machine learning techniques can provide image and/or video compression that produces high quality visual outputs.”)
Brehmer is considered analogous to the claimed invention as it is in the same field of endeavor, machine learning. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have modified Schlemper and Pasupuleti, which already teaches the trained neural network to compress MRI images but does not explicitly teach the trained neural network is configured to perform video data compression, to include the teachings of Brehmer which does teach the neural network can be used to compress both images and video in order to "reduce the size of image content—and thus the amount of storage involved to store image content and the amount of bandwidth involved in delivering video content—various compression algorithms (also referred to as coding algorithms or tools) may be applied to image and video content." (Brehmer, paragraph 0036)
Regarding claim 10, Schlemper and Pasupuleti teach the electronic device of claim 8, as cited above.
Claim 10 additionally has the same limitations of claim 3 which are taught by Schlemper, Pasupuleti, and Brehmer – see claim 3 above.
Regarding claim 13, Schlemper and Pasupuleti teach the electronic device of claim 8, as cited above.
Claim 13 additionally has the same limitations of claim 7 which are taught by Schlemper, Pasupuleti, and Brehmer – see claim 7 above.
Allowable Subject Matter
Claims 6 and 14 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Response to Arguments
Regarding applicant’s comment on CN 108830796 from the Information Disclosure Statement filed on September 7, 2023. 37 CFR 1.98 requires a legible copy of the foreign patent as well as a concise explanation of the relevance about the content of the information listed that is not in the English language. Examiner notes that the first page of the document, which contains the English abstract, is illegible and therefore cannot be considered by the examiner until a legible copy is submitted.
Applicant’s arguments with respect to claim(s) 1, 8, and 15 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Conclusion
Applicant's submission of an information disclosure statement under 37 CFR 1.97(c) with the timing fee set forth in 37 CFR 1.17(p) on February 23, 2026 prompted the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 609.04(b). 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 JACQUELINE MEYER whose telephone number is (703)756-5676. The examiner can normally be reached M-F 8:00 am - 4:30 pm 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, Tamara Kyle can be reached at 571-272-4241. 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.
/J.C.M./Examiner, Art Unit 2144
/TAMARA T KYLE/Supervisory Patent Examiner, Art Unit 2144
Prosecution Insights