Prosecution Insights
Last updated: April 19, 2026
Application No. 18/553,698

Systems and Methods for Multi-Kernel Synthesis and Kernel Conversion in Medical Imaging

Non-Final OA §103
Filed
Oct 02, 2023
Examiner
MEMON, OWAIS IQBAL
Art Unit
2663
Tech Center
2600 — Communications
Assignee
Mayo Foundation for Medical Education and Research
OA Round
2 (Non-Final)
74%
Grant Probability
Favorable
2-3
OA Rounds
3y 2m
To Grant
97%
With Interview

Examiner Intelligence

Grants 74% — above average
74%
Career Allow Rate
75 granted / 101 resolved
+12.3% vs TC avg
Strong +22% interview lift
Without
With
+22.4%
Interview Lift
resolved cases with interview
Typical timeline
3y 2m
Avg Prosecution
27 currently pending
Career history
128
Total Applications
across all art units

Statute-Specific Performance

§101
4.4%
-35.6% vs TC avg
§103
51.8%
+11.8% vs TC avg
§102
30.6%
-9.4% vs TC avg
§112
12.6%
-27.4% vs TC avg
Black line = Tech Center average estimate • Based on career data from 101 resolved cases

Office Action

§103
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 Applicants Remarks/Arguments In regards to the applicants remarks that Huber was authored by the same inventors and that the applicant has filed a 1.130 Declaration to affirm that the other authors (Nathan Huber, Hao Gong, and Scott Hsieh) are not the co-inventors of the subject matter that is claimed in the patent application, the office received the 1.130 declaration on 3/4/2026 which disqualifies Huber et al as prior art. The examiner conducted further search and found pertinent art Grbic teaching and a loss term that trains the neural network ([0114] "The processor may utilize the machine/deep learning network to select a kernel") for similarity ([0114] "Gaussian function" is understood to be the same as similarity in light of instant specifications [0084]) with a smooth kernel target. ([0114] "the machine/deep learning network is trained to emphasize resolution for lesions with relatively smaller features and emphasize a kernel with better noise properties for lesions with a relatively weak contrast." Kernel for weak contrast is understood to be the same as the claimed smooth kernel target in light of instant specifications [0035]) 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 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. Claims 1-2, 4, 7-11, 13 and 16-18 are rejected under 35 U.S.C. 103 as being unpatentable over Missert et al. (NPL “Synthesizing images from multiple kernels using a deep convolutional neural network”, hereinafter “Missert”) and in view of Grbic et al (US20190021677, hereinafter "Gbric") Claim 10. (Original) Missert teaches A system for synthesizing computed tomography (CT) image series (pg1PDF “combine features from CT images reconstructed with different kernels to produce a single synthesized image series”) comprising: a computer system (pg8pdf “The Titan Xp used for this research was donated by the NVIDIA Corporation.” Is a computer processor https://www.nvidia.com/content/geforce-gtx/NVIDIA_TITAN_Xp_User_Guide.pdf)configured to: i) reconstruct at least two CT image series, wherein the at least two image series are reconstructed with different reconstruction kernels; (pg1PDF “The CNN inputs consisted of two images produced with different reconstruction kernels,”) ii) synthesize at least one new CT image series by applying the at least two CT image series to an artificial neural network (pg2PDF “using a deep CNN to synthesize the clinically useful features from multiple images, each reconstructed with different kernels, into a single image representation.”) that has been trained on training data (pg3PDF “the CNN was trained”) using a task-based loss function comprising a sharp loss term (pg3PDF “CNN model was performed using modified stochastic gradient descent with a scalar loss function. We defined our loss function as simply the pixelwise mean squared error between the output image and the reference image (the routine-dose sharp kernel image) for each training example.”) that trains the neural network for similarity with a sharp kernel target ( pg2PDF “sharp (SiemensD50” and pg3PDF “single channel for sharp kernel inputs only.”) and trains the neural network for similarity with a smooth kernel target. (pg2PDF “Multiple reconstruction kernels were used to produce images for this study…smooth (Siemens D10)”) Missert does not explicitly teach and a loss term that trains the neural network for similarity with a smooth kernel target. Grbic teaches and a loss term that trains the neural network ([0114] "The processor may utilize the machine/deep learning network to select a kernel") for similarity ([0114] "Gaussian function" is understood to be the same as similarity in light of instant specifications [0084]) with a smooth kernel target. ([0114] "the machine/deep learning network is trained to emphasize resolution for lesions with relatively smaller features and emphasize a kernel with better noise properties for lesions with a relatively weak contrast." Kernel for weak contrast is understood to be the same as the claimed smooth kernel target in light of instant specifications [0035]) It would have been obvious to persons of ordinary skill in the art before the effective filing date of the claimed invention to modify Missert to have a task based loss function which has a sharp loss term with a smooth loss term to train the neural network as taught by Grbic to arrive at the claimed invention discussed above. The motivation for the proposed modification would have been so that (Grbic [0012]" Thus, a time to read and diagnose images of trauma patients should be reduced. Reducing the overall time for diagnosis would help to increase the probability of patient survival.") Claim 11. (Original) Missert and Grbic teach The system of claim 10, Missert teaches wherein the different reconstruction kernels include a smooth kernel (pg2PDF “Multiple reconstruction kernels were used to produce images for this study…smooth (Siemens D10)”) and a sharp kernel, ( pg2PDF “sharp (SiemensD50” and pg3PDF “single channel for sharp kernel inputs only.”) such that the at least two CT image series (pg2PDF “using a deep CNN to synthesize the clinically useful features from multiple images, each reconstructed with different kernels, into a single image representation.”) comprise at least a smooth kernel image series (pg6 “smooth-kernel images” and pg1PDF “smooth input images”) and a sharp kernel image series. (pg2PDF “sharp-kernel images” and pg1PDF “sharp input images”) PNG media_image1.png 756 1010 media_image1.png Greyscale Claim 13. (Original) Missert and Grbic teach The system of claim 10, Missert teaches wherein the computer system is further configured to train the artificial neural network on a training dataset, (“training dataset was constructed”) wherein the training dataset is created by inserting noise (pg2PDF “generated for each case using a validated projection domain noiseinsertion algorithm”) into full-dose CT data (pg2PDF “routine-dose CT (RDCT) images”) to create low- dose CT data (pg2PDF “we generated simulated low-dose CT (LDCT) images”) Claim 16. (Original) Missert and Grbic teach The system of claim 10, Missert teaches wherein the computer system is further configured to synthesize the at least one CT image series by combining image qualities of the at least two CT image series (pg2PDF “using a deep CNN to synthesize the clinically useful features from multiple images, each reconstructed with different kernels, into a single image representation.”) and wherein image qualities include at least one of spatial resolution, (pg2PDF “reconstructed with the sharpest kernel, and higher spatial resolution”) noise, (pg2PDF “lower nois”) contrast to noise ratio, or signal to noise ratio. Claim 17. (Original) Missert and Grbic teach The system of claim 10, Missert teaches wherein the computer system is further configured to determine domains in which the task-based loss function operates by determining a range of Hounsfield units (HU) between anatomic regions of interest in the at least one new CT image series using intensity thresholding. (pg4PDF “The standard deviation of the pixel HU values in uniform ROIs was used to compare the noise levels.”) Claim 18. (Original) Missert and Grbic teach The system of claim 10, Missert teaches wherein the computer system is further configured to display the synthesized at least one new CT image series(pg1PDF “If a clinical task involves interpreting multiple anatomic regions in the same scan, then multiple image series, each reconstructed with a different kernel, must be produced for the CT exam.”) for a user. (pg8PDF “from the technologists acquiring the images to the radiologists (and computerized algorithms) interpreting them.”) Claim 1. (Original) The method herein has been executed and performed by the system of claim 10 and is likewise rejected. Claim 2. (Original) The method herein has been executed and performed by the system of claim 11 and is likewise rejected. Claim 4. (Original) The method herein has been executed and performed by the system of claim 13 and is likewise rejected. Claim 7. (Original) The method herein has been executed and performed by the system of claim 16 and is likewise rejected. Claim 8. (Original) The method herein has been executed and performed by the system of claim 17 and is likewise rejected. Claim 9. (Original) The method herein has been executed and performed by the system of claim 18 and is likewise rejected. Allowable Subject Matter Claims 3, 5-6, 12, 14-15 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. Conclusion The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure: Yi et al US20220172353 teaches a sharp kernel and a smooth kernel for CT scans Any inquiry concerning this communication or earlier communications from the examiner should be directed to OWAIS MEMON whose telephone number is (571)272-2168. The examiner can normally be reached M-F (7:00am - 4:00pm) CST. 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. /OWAIS I MEMON/Examiner, Art Unit 2663
Read full office action

Prosecution Timeline

Oct 02, 2023
Application Filed
Oct 24, 2025
Non-Final Rejection — §103
Feb 19, 2026
Response Filed
Mar 04, 2026
Response after Non-Final Action
Mar 11, 2026
Non-Final Rejection — §103 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12597224
SYSTEM AND METHOD FOR FEATURE SUB-IMAGE DETECTION AND IDENTIFICATION IN A GIVEN IMAGE
2y 5m to grant Granted Apr 07, 2026
Patent 12591989
METHOD FOR DEPTH ESTIMATION AND HEAD-MOUNTED DISPLAY
2y 5m to grant Granted Mar 31, 2026
Patent 12592013
REAL SCENE IMAGE EDITING METHOD BASED ON HIERARCHICALLY CLASSIFIED TEXT GUIDANCE
2y 5m to grant Granted Mar 31, 2026
Patent 12586338
SYSTEM FOR UPDATING NEURAL NETWORK PARAMETERS BASED ON OBJECT DETECTION AREA OVERLAP SCORE
2y 5m to grant Granted Mar 24, 2026
Patent 12573069
SYSTEMS AND METHODS FOR GENERATING AND CODING MULTIPLE FOCAL PLANES FROM TEXTURE AND DEPTH
2y 5m to grant Granted Mar 10, 2026
Study what changed to get past this examiner. Based on 5 most recent grants.

AI Strategy Recommendation

Get an AI-powered prosecution strategy using examiner precedents, rejection analysis, and claim mapping.
Powered by AI — typically takes 5-10 seconds

Prosecution Projections

2-3
Expected OA Rounds
74%
Grant Probability
97%
With Interview (+22.4%)
3y 2m
Median Time to Grant
Moderate
PTA Risk
Based on 101 resolved cases by this examiner. Grant probability derived from career allow rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month