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 .
Claim Interpretation
Claim 11 and dependents require “processing circuitry”. Per MPEP 2181, certain terms typically denote structure. As identified in Mass. Inst. of Tech. v. Abacus Software, 462 F.3d 1344, and for the same reasons given in that decision, “the recitation of "aesthetic correction circuitry" is sufficient to avoid pre-AIA 35 U.S.C. 112, paragraph 6, treatment because the term circuit, combined with a description of the function of the circuit, connoted sufficient structure to one of ordinary skill in the art.” Circuitry elements in the present claims are construed as requiring hardware.
Claim Rejections - 35 USC § 112
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.
The following is a quotation of 35 U.S.C. 112(d):
(d) REFERENCE IN DEPENDENT FORMS.—Subject to subsection (e), a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
The following is a quotation of pre-AIA 35 U.S.C. 112, fourth paragraph:
Subject to the following paragraph [i.e., the fifth paragraph of pre-AIA 35 U.S.C. 112], a claim in dependent form shall contain a reference to a claim previously set forth and then specify a further limitation of the subject matter claimed. A claim in dependent form shall be construed to incorporate by reference all the limitations of the claim to which it refers.
Claims 1-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second and fourth paragraphs, 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 and/or failing to further limit the parent claim.
In claim 1, 11 and 20, it is unclear what “specific to a particular domain to which the medical imaging system corresponds” is unclear. Does “specific” mean more than that there is a dedicated second neural network?
In Claim 1, 11, 20, the second neural network is pretrained; in Claim 2, 12 the second neural network is being trained. This appears to be either removing the “pretrained” from claim 1, 11, 20 or duplicative of that training. What significance does “pretrained” have in Claim 1 where the training happens later in Claim 2, 12?
Claim 7 and 17 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. Claim 7 includes “the obtained second image dataset”, but a second image dataset appears only in claim 2 which is not a parent claim. To the extent claim 7 were to depend on Claim 2 it would be removing the second image dataset as a distinct element (i.e. there would no longer be a second dataset where the first dataset was “used” as the second dataset).
The remaining claims are rejected for their dependency on a rejected parent claim.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1-3, 5-7. 9-13, 15-17, and 19-20 as best understood is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Varshitha et al., “Enhancing Medical Imaging Resolution: Exploring SRGAN for High-Quality Medical Image Reconstruction”, added to IEEE explore 23 April 2024
With respect to Claims 1, 11 and 20, Varshitha et al. shows collecting a first image dataset (either of Retinal OCT images, Ultrasound Fetus Images at III A 1-2); the dataset is the LR images at III B (b));
using the collected first image dataset, training a first neural network, based on a loss function having a perceptual component (see figure at the bottom of p. 3, “perceptual loss computation” and “perceptual loss”);
and using the trained first neural network, inferring output image data from input image data obtained by the medical imaging system, such that the inferred output image data has an image quality better than an image quality of the obtained input image data, (abstract, “This study examines the use
of Super-Resolution Generative Adversarial Networks (SRGANs) to improve the resolution of medical images. SRGANs, known for enhancing conventional images, are adapted for medical imaging using a deep convolutional neural network (CNN) generator and discriminator, with a perceptual loss function guiding the reconstruction process.”)
wherein the training of the first neural network uses a pretrained second neural network (III C 1, “Before feeding the images to the SRGAN, initial feature extraction is performed using various Convolutional Neural Networks (CNNs). These features provide essential details that assist the Generator in creating super-resolved images.”), and the pretrained second neural network is specific to a particular domain to which the medical imaging system corresponds. (“Different architectures resulted in distinct outcomes, painting a clear picture of the optimal feature extractor for the specific medical imaging datasets at hand.” As understood, Varshitha et al. tested the two datasets against each of the four models in Table IV, and concluded that the combination of VGG19/Ultrasound was better and ResNet50V2/OCT was better, from IV. Conclusion. )
Alternatively, the generator/discriminator of the figure on P. 3 constitute the two neural networks; the low resolution images are fed to a generator to generate high resolution imaged, where the generator was trained by the discriminator on a second set of images.
With respect to claims 2-3 and 12-13, the nature of training a GAN is that the generator and discriminator are iteratively trained until a satisfactory level of authenticity is reached. This is shown in the figure at page 7, which shows training through 1000 epochs.
With respect to Claim 5-7 and 15-17, Varshitha et al. is understood to collect high-resolution images of Retinal OCT and Ultrasound fetus images from real-world conditions and also downsample those images to produce two datasets. See page 3 at “B”.
With respect Claims 9 and 19, Varshitha et al. shows the perceptual loss function (“perceptual loss computation”) in the figure on page 3.
With respect to Claim 10 Varshitha et al. shows datasets in Computed Tomography and Ultrasound (III. Methodology, A. Dataset Description)
Allowable Subject Matter
Claims 4 and 14 would be allowable if rewritten to overcome the rejection(s) under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), 2nd paragraph, set forth in this Office action and to include all of the limitations of the base claim and any intervening claims. While Varshitha et al. gives examples of four CNNs suitable for discriminators (VGG19, ResNet50, ResNet50Vs and MobileNetV2), and at least two of them appear to be suitable for use as encoders in a U-net setup, (ResNet50 and MobileNetV2), they would not be part of a complete U-net in the context of the system of Varshitha et al.
Claims 8 and 18 would be allowable if rewritten to overcome the rejection(s) under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), 2nd paragraph, set forth in this Office action and to include all of the limitations of the base claim and any intervening claims. While contrastive loss is known and has a role in medical image analysis (see Mahbod et al., “Evaluating Pre-trained Convolutional Neural Networks and Foundation Models as Feature Extractors…”, page 3, 7, referring to the Foundation models.) it would not be an obvious addition to the GAN of Varshitha et al. on this record. Varshitha et al. already has perceptual and adversarial losses in the figure of page 3.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Wu et al. “Modal adaptive super-resolution for medical images via continual learning” describes a system suitable for multiple modalities.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GREGORY A MORSE whose telephone number is (571)272-3838. The examiner can normally be reached M-F 7:30-4.
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.
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.
/GREGORY A MORSE/ Supervisory Patent Examiner, Art Unit 2698
Prosecution Insights