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 .
Drawings
Figure 1 should be designated by a legend such as --Prior Art-- because only that which is old is illustrated. See MPEP § 608.02(g). See specification [0011] stating that this is an MRI machine and omitting discussion of the invention. Corrected drawings in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. The replacement sheet(s) should be labeled “Replacement Sheet” in the page header (as per 37 CFR 1.84(c)) so as not to obstruct any portion of the drawing figures. If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Specification
The title of the invention is not descriptive. A new title is required that is clearly indicative of the invention to which the claims are directed.
The following title is suggested: SYSTEM AND METHOD FOR ENHANCING PROPELLER IMAGE (MRI RECONSTRUCTION) QUALITY BY UTILIZING MULTI-LEVEL DENOISING.
The examiner’s goal is that title conveys that this is about denoising MRI images, not literal propellers.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 1-20 (all claims) are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Claims 1, 9, and 17 recite “to generate a … artifact free gridded image.” However, Fig. 8 shows artifacts (note the three black arrows added by the examiner):
PNG
media_image1.png
370
307
media_image1.png
Greyscale
MPEP 2163(II)(A)(3)(a) states “Estee Lauder Inc. v. L’Oreal, S.A., 129 F.3d 588, 593, 44 USPQ2d 1610, 1614 (Fed. Cir. 1997) (“[A] reduction to practice does not occur until the inventor has determined that the invention will work for its intended purpose.”).” The denoising at issue here (see, e.g., the title, claim 1 and the abstract) is not about the aesthetics of the image, but rather whether or not the image is accurate.
A review of the specification suggests that experimentation to determine if this technology works as expected was limited to the analysis shown in Figs. 13 and 14. Fig. 13 is measuring the structural similarity index (specification [0084]), but SSIM is only for perceived quality – not underlying accuracy.1 See, e.g., https://en.wikipedia.org/wiki/Structural_similarity_index_measure (attached). Similarly, Fig. 14 appears to be a visual inspection that the images are less noisy. However, neither Fig. 13 nor 14 discuss whether or not the resulting images introduce artifacts as shown in Fig. 8. In other words, there is not evidence that the inventors determined if the resulting images were actually more or less accurate.
This accuracy is particularly important for generative artificial intelligence. See, e.g.:
Cohen JP, Luck M, Honari S. Distribution matching losses can hallucinate features in medical image translation. In International conference on medical image computing and computer-assisted intervention 2018 Sep 16 (pp. 529-536). Cham: Springer International Publishing. (“Cohen,” attached)
As the above Cohen article explains in the abstract, “Therefore, we recommend that these translated images should not be used for direct interpretation (e.g. by doctors) because they may lead to misdiagnosis of patients based on hallucinated image features by an algorithm that matches a distribution. However there are many recent papers that seem as though this is the goal.”
The closing line of Cohen’s abstract, quoted above, fits here. In other words, the artifacts in Fig. 8 could lead to a misdiagnosis (e.g., if they are believed to be cancerous), but this is not addressed by the specification.
Claims 1, 9, and 17 are therefore rejected for lack of written description support for the related reasons that A) the drawings show that the technology at issue generates images that are not artifact free, and because there is not evidence that the inventors determine whether this technology denoises in a way that increases accuracy as opposed to just being perceived as more accurate.
Claims 1, 9, and 17 recite “to denoise each blade of the plurality of blades in an image domain to generate a plurality of denoised blades,” but this is unlimited functional claiming. MPEP 2173.05(g). The issue is that this is claiming a result, rather than the steps to achieve the result, and here there is a wide array of different approaches to achieving this result.
Claims 1, 9, and 17 also recite “to remove individual-based denoising-induced artifacts” and this is unlimited functional claiming. MPEP 2173.05(g).
Claims 1, 9, and 17 recite “to utilize a PROPELLER reconstruction algorithm to generate a denoised-gridded image from the plurality of denoised blades.” This sounds as though it is referring strictly to algorithms that are widely known in the art, and thus is understood as a claim step (rather than a result). However, dependent claims, such as claim 2, recite “wherein the PROPELLER reconstruction algorithm comprises adjoint non-uniform fast Fourier transform blocks configured to grid the plurality of denoised blades into a Cartesian grid.” These additional limitations are not understood to be prior art, but are within the scope of the parent claims. Thus, because the parent claims encompass techniques that are not widely known in the art, and the parent claims recite results rather than steps, the parent claims are unlimited functional claiming. MPEP 2173.05(g). Note that showing that the PROPELLER techniques from the dependent claims are widely known in the art is one way to overcome this rejection.
Dependent claims are likewise rejected.
Claims 1-20 (all claims) are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, because the specification, while being enabling for trivial implementations, does not reasonably provide enablement for removing all artifacts. The specification does not enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the invention commensurate in scope with these claims.
MPEP 2164.06(a)(I) provides examples of enablement issues due to missing information for electrical and mechanical devices or processes. Here, the claimed inference is less predictable than other computer technologies, and thus the guidance from MPEP 2164.06(a)(I) applies here.
The claims are not limited to the level of performance that has been actually achieved. Specifically, as per above, while artificial intelligence may remove many artifacts, there is no disclosure of how to remove all artifacts. Rather, Fig. 8 shows artifacts after processing. See the first paragraph of MPEP 2164.06(a)(I) discussing MagSil Corp. v. Hitachi Global Storage Technologies, Inc., 687 F.3d 1377, 103 USPQ2d 1769 (Fed. Cir. 2012) and Auto. Techs. Int'l, Inc. v. BMW of N. Am., Inc., 501 F.3d 1274, 1283, 84 USPQ2d 1108, 1115 (Fed. Cir. 2007). Auto Techs. applies to the wide variety of models that could be used (e.g., if one were to use a vision transformer model instead of a convolutional neural network).
Summarizing the above in terms of the Wands factors:
(A) The breadth of the claims; - the claims cover all methods of artifact removal, which is very broad
(B) The nature of the invention; - not a significant factor
(C) The state of the prior art; - not a significant factor
(D) The level of one of ordinary skill; - not a significant factor
(E) The level of predictability in the art; - neural networks are less predictable than traditional software. For example, one needs to build and train a neural network to determine how well it will perform.
(F) The amount of direction provided by the inventor; - the direction provided does not address how to improve upon known shortcomings in artificial intelligence
(G) The existence of working examples; and – Fig. 8 still has artifacts
(H) The quantity of experimentation needed to make or use the invention based on the content of the disclosure. – as a result, an unreasonable amount of experimentation is required to reduce to practice artificial intelligence that does not hallucinate (i.e., there are not artifacts).
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-20 (all claims) are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claims 1, 9, and 17 recite utilizing a denoising network “to” perform various tasks, but it is unclear if this interpreted as an intended use or if these are required steps.
Claims 2, 4, 5-8, 10, 12-16, and 18-20 recite “to” various steps and each of these raise the same issue.
Claims 1, 9, and 17 recite a “deep learning-based multi-level denoising network,” but this is new terminology. MPEP 2173.05(a). One option to overcome this rejection is to use known terminology, such as that found in Fig. 2 (e.g., residual dense network).
Claims 2, 10, and 18 recite “deep learning-based artifact removing model,” and this raises the same issue.
Claims 1, 9, and 17 recite generation of an “artifact-free gridded image.” It is unclear if this is intended to have antecedent basis in the “denoising-induced artifacts,” or if it should be interpreted as removing all artifacts of any kind. MPEP 2173.05(e).
Claims 4, 12, and 19 recite “train … end to end.” However, these claims also recite that the training includes blade level loss, which is not an end point of the network. Additionally, this conflicts with what is shown in Fig. 2. Thus, it is unclear what is meant.
Claims 5, 13, and 20 recite training the artifact removing model using “both the blade level loss and the grid level loss,” however, this conflicts with what is shown in Fig. 2. Thus, it is unclear what is meant.
Dependent claims are likewise rejected.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-20 (all claims) are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea (mental process) without significantly more.
Step 1: Claim 1 (and its dependents) recite a method, and processes satisfy Step 1 of the eligibility test.
Claim 9 (and its dependents) recite a system, and machines satisfy Step 1 of the eligibility test.
Claim 17 (and its dependents) recite a non-transitory computer-readable medium, and manufactures satisfy Step 1 of the eligibility test.
Step 2A, prong one: All of the elements of the claims are a mental process because a person can look at MRI images and imagine what they might look like with less noise (see, e.g., Figs. 7-11). Further, the various models are also mental processes, see example 47, claim 2, element (d) (from the July 2024 AI subject matter eligibility examples). MPEP 2106.04(a)(2)(III)(C) explains that use of a generic computer or in a computer environment is still a mental process. In particular, this section begins by citing Gottschalk v. Benson, 409 US 63 (1972). “The Supreme Court recognized this in Benson, determining that a mathematical algorithm for converting binary coded decimal to pure binary within a computer’s shift register was an abstract idea.” In Benson the Supreme Court did not separately analyze the computer hardware at issue; the specifics of what hardware was claimed is only included in an appendix to the decision.
Because there are no additional elements, no further analysis is required for Step 2A, prong two or Step 2B.
The examiner notes that Applicant’s “Background” section qualifies as evidence that the technologies described therein are well-understood, routine, conventional. Not only is this section labeled background, it also uses phrases such as “one of many well-known,” “are reported” and “typically.” Further, these technologies are understood as providing the technological environment of the invention. MPEP 2106.05(h). Recentive Analytics, Inc. v. Fox Corp., 134 F. 4th 1205,1213 (Fed. Cir. 2025) explains if the improvement is the use of machine learning in a new environment, rather than, say, an improvement in machine learning, the idea is abstract.
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 1-7, 9-15, and 17-20 (all claims except those rejected over Wikipedia, below) rejected under 35 U.S.C. 103 as being unpatentable over Chang Y, Pipe JG, Karis JP, Gibbs WN, Zwart NR, Schär M. The effects of SENSE on PROPELLER imaging. Magnetic resonance in medicine. 2015 Dec;74(6):1598-608 (“Chang”) in view of Wu D, Kim K, Fakhri GE, Li Q. A cascaded convolutional neural network for x-ray low-dose CT image denoising. arXiv preprint arXiv:1705.04267. 2017 May 11 (“Wu”).
1. A computer-implemented method for improving image quality of periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) imaging, comprising: (Chang, title, “Propeller Imaging”)
acquiring, via a processor, a plurality of blades of k-space data of a region of interest in a rotational manner around a center of k-space via a magnetic resonance imaging (MRI) scanner from a coil during a PROPELLER sequence, (Chang, Fig. 1a. The top box recites “blade”.)
wherein each blade of the plurality of blades of k-space data comprises a plurality of parallel phase encoding lines sampled in a phase encoding order; and (Chang, p. 1601, right side, “on each blade along the phase encoding direction.”)
to utilize a PROPELLER reconstruction algorithm to generate a denoised-gridded image from the plurality of denoised blades, and (Chang, p. 1600, left side, “To rotate the extrapolated map, it was inverse Fourier-transformed to k-space, rotated with a set of angles, regridded, and Fourier-transformed back to image domain.”)
to remove individual-based denoising-induced artifacts from the denoised-gridded image to generate a denoised, artifact-free gridded image. (Chang, p. 1605, right side, “PROPELLER SENSE with both 24 and eight blade reconstructions were resistant to motion artifacts”)
Chang is not relied on for the below claim language.
However, Wu teaches utilizing, via the processor, a deep learning-based multi-level denoising network to denoise each blade of the plurality of blades in an image domain to generate a plurality of denoised blades, (Wu, abstract, “Machine learning based denoising methods have shown great potential in removing the complex and spatial-variant noises in CT images.”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply the teachings of Wu to the teachings of Chang such that Wu’s denoising is used with Chang’s PROPELLER technology for the purpose of denoising PROPELLER images. As explained in Chang’s abstract, it is looking to reduce noise (e.g., improve the signal to noise ratio), when reconstructing blades. Wu, as per the title, denoises this type of image.
Based on the above, this is an example of “combining prior art elements according to known methods to yield predictable results.” MPEP 2143.
2. The computer-implemented method of claim 1, wherein the deep learning-based multi-level denoising network comprises a deep learning-based denoising model to denoise each blade of the plurality of blades in the image domain to generate the plurality of denoised blades and (Wu, abstract, “Machine learning based denoising methods have shown great potential in removing the complex and spatial-variant noises in CT images.”)
a deep learning-based artifact removing model to remove the individual-based denoising-induced artifacts from the denoised-gridded image to generate the denoised, artifact-free gridded image, and
(Wu, abstract, “A cascaded training network was proposed in this work, where the trained CNN was applied on the training dataset to initiate new trainings and remove artifacts induced by denoising.”)
wherein the PROPELLER reconstruction algorithm comprises adjoint non-uniform fast Fourier transform blocks configured to grid the plurality of denoised blades into a Cartesian grid. (Chang, p. 1600, left side, “To rotate the extrapolated map, it was inverse Fourier-transformed to k-space, rotated with a set of angles, regridded, and Fourier-transformed back to image domain.”)
3. The computer-implemented method of claim 2, wherein the deep learning-based artifact removing model has fewer parameters than the deep learning-based denoising model. (Wu, section 1, “Rather than initiating completely new training with deeper CNNs, we proposed a cascaded framework to boost performance of simple CNNs.” Wu’s “simple” versus “deeper” teaches the claimed fewer parameters. Alternatively, a network with more/fewer parameters is a known substitute.)
4. The computer-implemented method of claim 2, further comprising backpropagating, via the processor, a combination of both blade level loss and grid level loss to train the deep learning-based multi-level denoising network from end to end. (Wu, Fig. 2 and Fig. 2 caption, “cascaded CNN training scheme”)
5. The computer-implemented method of claim 4, wherein the combination of both the blade level loss and the grid level loss are backpropagated to train both the deep learning-based denoising model and the deep learning-based artifact removing model. (Wu, section 3.2, “Images from 7 patients were used as the training datasets and the CNN denoisers were trained on 2D slices.” Wu’s images teach the claimed grid level and Wu’s 2d slices teach the claimed blade level.)
6. The computer-implemented method of claim 2, further comprising utilizing, via the processor, the PROPELLER reconstruction algorithm to generate a noisy gridded image from the plurality of blades that have not been denoised. (Chang, Fig. 1a. Note that Chang shows “grid and combine all blades,” and lacks denoising.)
7. The computer-implemented method of claim 6, further comprising inputting, via the processor, both the denoised-gridded image and the noisy gridded image into the deep learning-based artifact removing model, wherein the deep learning-based artifact removing model utilizes both the denoised-gridded image and the noisy gridded image to generate the denoised, artifact-free gridded image. (Wu, Fig. 3. The lower set of inputs are the original, teaching the claimed noisy image. The upper set of inputs teach the claimed denoised image.)
Claims 9-15 and 17-20 are rejected as per their counterparts. Wu and Chang teach the claimed hardware because both disclose having actually implemented their technologies.
Claims 8 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Chang and Wu as applied to their parent claims above, and further in view of the Wikipedia article “Super resolution imaging” as of July 27, 2019, retrieved from https://en.wikipedia.org/w/index.php?title=Super-resolution_imaging&oldid=908130228 (“Wikipedia”).
8. The computer-implemented method of claim 1, (See the mapping of claim 1)
The combination of Chang and Wu are not relied on for the below claim language.
However, Wikipedia teaches further comprising applying, via the processor, super-resolution to the denoised, artifact-free gridded image to generate a higher resolution denoised, artifact-free gridded image that is further denoised. (Wikipedia, section “Research,” “There is promising research on using deep convolutional networks to perform super-resolution.” See also, first section, “In some radar and sonar imaging applications (e.g., magnetic resonance imaging (MRI) …” and “Super-resolution imaging techniques are used in general image processing.”)
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to apply the teachings of Wikipedia to the teachings of the combination of Chang and Wu such that Wikipedia’s super resolution is used on the images of the combination of Chang and Wu for the purpose of enhancing resolution. Wikipedia, first sentence.
Based on the above, this is an example of “combining prior art elements according to known methods to yield predictable results.” MPEP 2143.
Claim 16 is rejected as per claim 8. Wu and Chang teach the claimed hardware because both disclose having actually implemented their technologies.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 20240095889 B1 – “SYSTEMS AND METHODS FOR MAGNETIC RESONANCE IMAGE RECONSTRUCTION WITH DENOISING” – this is a very relevant reference
US 20210319539 A1 – “SYSTEMS AND METHODS FOR BACKGROUND AWARE RECONSTRUCTION USING DEEP LEARNING” – indicative of the wide variety of GE patents in this field
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DAVID ORANGE whose telephone number is (571)270-1799. The examiner can normally be reached Mon-Fri, 9-5.
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.
/DAVID ORANGE/Primary Examiner, Art Unit 2663
1 The Wikipedia article states “the measurement or prediction of image quality is based on an initial uncompressed or distortion-free image as reference.” In contrast, the initial image used here is noisy.
Prosecution Insights