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 .
Priority
Receipt is acknowledged that application is a National Stage application of PCT/EP2021/068445. Receipt is acknowledged that application claims priority to foreign application with application number RU2020123043 dated 07/10/2020. Copies of certified papers required by 37 CFR 1.55 have been received on 07/25/2025. Priority is acknowledged under 35 USC 119(e) and 37 CFR 1.78.
Response to Amendment
The amendment filed 03/17/2026 has been entered. Applicant’s amendments to the claims have overcome each and every objection previously set forth in the Non-Final Office Action mailed 12/23/2025. Claims 1, 3-5, 8-11, 13, 15-16, 18, and 20-23 remain pending in the application, with claims 21-23 being newly added.
Response to Arguments
Applicant's arguments filed 03/17/2026 have been fully considered but they are not persuasive. Applicant argues the following on pg. 9-10:
Examiner disagrees. The relied upon teachings from Estrada do not contrast with the claim limitation. Though Estrada does mask the pixels surrounding the synthetic feature, this method is still included within the broadest reasonable interpretation of the claim limitation “assign pixels of the synthetic feature a different pixel value than other pixels in the feature image”. In the masked image containing the synthetic feature, 1302 in FIG. 13 of Estrada, every pixel is assigned pixel data values, including those of the synthetic feature. In image 1302, the pixels of the synthetic feature are assigned a different pixel value than other pixels in the image (different than the masked transparent pixels). There is no requirement of order in assignment of pixel values expressed in the claim. Regardless, the data representation that results from the creation of image 1302 includes the assignment of pixel values corresponding to the synthetic feature. Furthermore, this narrow interpretation does not have written description support in the specification of the original disclosure, and thus would fail to comply with 35 U.S.C. 112(a). There is no definition provided by the applicant further describing the meaning and/or process of “assign pixels”. The specification solely describes a synthetic feature image where the synthetic feature has different pixel values than other pixels in the image (¶0031, ¶0034).
Applicant also argues the following on pg. 10:
PNG
media_image5.png
210
633
media_image5.png
Greyscale
Examiner disagrees. Estrada’s use of “overlaying” the demarcated footprint onto the real image is simply a visual description of the result of the image processing operations performed. Estrada clearly describes in ¶76 that the synthetic feature image and real image are merged (Estrada, para 76: “a synthetically manipulated geospatial image 1303 showing synthetic image 1302 scaled, aligned, masked and overlain onto geospatial image 1301 merged into synthetic image overlay area 1330”, emphasis added). It is clear in the difference between images 1301 and 1303 that the synthetic feature is added to a real image that did not include the synthetic feature beforehand. Estrada merges the pixel values of the synthetic image with pixel values in the real image corresponding to where the synthetic feature will be placed. Additionally, OpenCV also teaches this limitation. OpenCV merges two images of the same size, thus adding pixel values of one image to the corresponding pixel positions in the second image.
Lastly, Applicant argues the following on pg. 10:
PNG
media_image6.png
354
637
media_image6.png
Greyscale
Examiner disagrees. As described above, the solution of “overlaying” is a description of the visual result of the image processing operations performed by Estrada. There are no teachings in Comaniciu or Estrada that explicitly teach away from the relied upon combination, nor are any provided by Applicant in the Remarks. Comaniciu does not specifically describe how the pixels of the images are processed to create the synthetic training image. Therefore, it is unclear how Estrada’s solution, which is also consistent with a provided example given in ¶31 of the original disclosure (“…for the feature image, pixel values of pixels outside of pixels corresponding to a synthetic feature are given a value of zero so that they represent, e.g., air or empty space.”), teaches away from the recited claim limitation.
Merging transparent pixels with values of an existing image, 1301 of Estrada, leaves the pixel value as originally provided in the real image. It is clearly demonstrated by image 1303 that the pixel values where the synthetic feature is added are altered (compared to those in image 1301) and all other pixels (where transparent pixels are added) remain unchanged. “Overlaying” pixels in the method of Estrada results in a new image, demonstrated by the digital representation of new image 1303 in FIG. 13.
In view of the foregoing, the rejections are maintained in light of the most recent amendments.
Claim Rejections - 35 USC § 103
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, 3-5, 8-11, 13, 15-16, 18, and 20-23 are rejected under 35 U.S.C. 103 as being unpatentable over Comaniciu et al. (U.S. Patent No. 2017/0357844 A1), hereinafter Comaniciu, in view of Estrada et al. (U.S. Patent No. 2017/0061625 A1), hereinafter Estrada, in further view of OpenCV. Arithmetic Operations on Images. August 7, 2017 [online], [retrieved on 2025-09-18]. Retrieved from the Internet <URL: https://web.archive.org/web/20170807005038/https://docs.opencv.org/3.0.0/d0/d86/tutorial_py_image_arithmetics.html>, hereinafter OpenCV.
Regarding claim 1, Comaniciu teaches a system for detecting a visual feature (Comaniciu, para 17: "The feature detection and learning paradigm is based on synthetic samples created using multi-scale models of tumor phenotypes”; the visual feature is a tumor) manifested by a rare disease in an image of a subject (Comaniciu, para 51: "The approach of using synthetic images and datasets has the advantage of being able to span pathological conditions that are relatively rare and hard to sample from the patient population in sufficient numbers"; para 49: “rare cancers”; the subject is the patient whom the image is taken of, see personalized treatment in para 16 and patient data in para 18), the system comprising:
at least one data repository configured to store images generated by an imaging modality of a healthcare entity (Comaniciu, para 103-104: “The memory 15 is configured to store medical scan data, other data, extracted features, examples (e.g., training data or data from other patients), and/or other information”; para 23: “The medical system of FIG. 4 or other medical system implements the acts. The system may be a medical imaging system, a hospital workstation, a patient medical records computer, a medical server, a cloud-based system or other secure medical data processing system”), wherein the images include at least one image that does not include the visual feature (Comaniciu, para 71: “scan data (e.g., 3D mesh or voxel representation) of healthy tissue”); and
a processor (Comaniciu, para 100: "processor 13" of Figure 4; para 105: “The processor 13 operates pursuant to stored instructions to perform various acts described herein”) configured to:
create training data based on the at least one image that does not include the visual feature (Comaniciu, library training data based on healthy scan data, para 71: "In one embodiment, the library is created based on a template image or scan data (e.g., 3D mesh or voxel representation) of healthy tissue or tissue with a tumor. Given the template scan data, automatic image parsing is performed to extract the main organ and/or tissues of interest. The user or processor picks a seed location within an organ or tissue for the tumor"; library contains training data, para 82: “The library, created at least in part or completely from simulated images based on synthetic tumor samples, is used for training”), wherein the processor is configured to:
generate a feature image with a synthetic feature that visually mimics the visual feature based on a model (Comaniciu, simulator image with a tumor based on the tumor model, para 50: “A tumor model is provided with different values of one or more parameters, resulting in different synthetic tumors. An image simulator then simulates generation of one or more sets of scan data from each of the tumor models”), and
repurpose the at least one image that does not include the visual feature and create a synthetic training image for the training data (Comaniciu, scan data is repurposed to include a synthetic tumor, para 71: "In one embodiment, the library is created based on a template image or scan data (e.g., 3D mesh or voxel representation) of healthy tissue…The user or processor picks a seed location within an organ or tissue for the tumor. The system would then recognize which organ has been selected, and the most likely tumor (e.g., based on population analysis) is automatically selected and virtually simulated at that location with various perturbations”; see next citation from para 82 regarding training); and
train an artificial intelligence module based on the training data (Comaniciu, para 82: “In act 44, a machine trains the machine-learnt classifier using, at least in part, the simulated medical images”) to detect the visual feature in the image of the subject based on the training data (Comaniciu, para 87: “The machine-learnt classifier is trained to output phenotype information, such as diagnosis, prognosis, and/or treatment outcome for the tumor associated with input data”).
Comaniciu discloses simulating the synthetic tumor, for example, in paragraph 71 (Comaniciu, para 71: “the most likely tumor (e.g., based on population analysis) is automatically selected and virtually simulated at that location with various perturbations”), but does not specifically describe how the pixels of the images are processed to create the synthetic training image. Thus, Comaniciu fails to explicitly teach a) assign pixels of the synthetic feature a different pixel value than other pixels in the feature image and b) adding the assigned different pixel values to corresponding pixel values in the at least one image that does not include the visual feature to add the synthetic feature in the at least one image that does not include the visual feature.
Estrada teaches a system for generating synthetic images (Estrada, para 76: “a synthetically manipulated geospatial image 1303 showing synthetic image 1302 scaled, aligned, masked and overlain onto geospatial image 1301 merged into synthetic image overlay area 1330”; see also abstract), including assign pixels of the synthetic feature a different pixel value than other pixels in the feature image (Estrada, pixels that are not the synthetic feature are transparent, para 74: “using a masking function 1185 to set the background of the synthetic image to transparent such that existing imagery is not occluded”; see 1302 in FIG 13, attached below). It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to have utilized the method of assigned pixels, taught by Estrada above, with the method of Comaniciu in order to seamlessly blend the synthetic feature with the features of the repurposed image (Estrada, the overlay does not obscure the existing image, para 9: “masks the background colors surrounding the synthetic image to become transparent such that overlay onto the real image does not obscure existing images to create a manipulated synthetic image”).
PNG
media_image7.png
437
623
media_image7.png
Greyscale
Further, OpenCV dicloses a method for creating a synthetic image by adding the assigned different pixel values to corresponding pixel values in the at least one image that does not include the visual feature (OpenCV, “Image Blending” section: “This is also image addition…Images are added as per the equation below:
g(x) = (1-\alpha) f_(0)(x) + \alpha f_(1)(x)”; adding corresponding pixel values from the middle image to the leftmost image in the image attached below) to add the synthetic feature in the at least one image that does not include the visual feature (OpenCV, rightmost image in the image attached below).
.
PNG
media_image8.png
259
835
media_image8.png
Greyscale
Comaniciu discloses the addition of a feature in an existing image (Comaniciu, adding a synthetic tumor to scan data representing healthy tissue in para 71 and 74), but does not disclose specific instructions for combining the two images (Comaniciu, para 71: “tumor (e.g., based on population analysis) is automatically selected and virtually simulated at that location with various perturbations”). OpenCV teaches adding a feature to an existing image with the known technique of adding/summing the pixel values of two images to combine them. A person having ordinary skill in the art, before the effective filing date of the claimed invention, could have applied the known technique, as taught by OpenCV, in the same way to the system and generated training images of Comaniciu in view of Estrada and achieved predictable results of combining two images to add a feature to an existing image using a basic image pixel addition operation.
Regarding claim 3 (dependent on claim 1), Comaniciu in view of Estrada and OpenCV teaches wherein the processor is further configured to generate the feature image with the synthetic feature (Comaniciu, para 74: “Each of the synthetic tumors with or without scan data representing healthy tissue is used to generate an image or images”) based on a mathematical model (Comaniciu, para 54: “The computational model is defined by parameters that control characteristics of the synthetic tumor”; para 58: "In one embodiment, the mathematical modeling of cancer uses a multi-scale model of tumor physiology").
Regarding claim 4 (dependent on claim 3), Comaniciu in view of Estrada and OpenCV teaches wherein the mathematical model includes at least one user adjustable parameter (Comaniciu, para 64: "In act 40, the medical system (e.g., processor) varies parameters of the computational tumor model. The tumor model is defined by one or more parameters, such as any of the modeling parameters in the equations or used as input features in the computational model"; para 67: "The variation may be controlled or user set").
Regarding claim 5 (dependent on claim 4), Comaniciu in view of Estrada and OpenCV teaches wherein the processor if further configured to set the at least one adjustable parameter based on human input (Comaniciu, para 67: "The variation may be controlled or user set").
Regarding claim 8 (dependent on claim 1), Comaniciu in view of Estrada and OpenCV teaches wherein the processor is further configured to create at least one additional repurposed image (Comaniciu, multiple images may be generated, para 74: “generate an image or images”) by visually manipulating the synthetic feature (Comaniciu, para 66: “For example, thousands of stochastic perturbations to the synthetic tumor model produce thousands of corresponding tumor models. Since the tumor models have known conditions (e.g., known phenotype information to be used as the ground truth), a rich dataset on which to train the machine learning model is provided”).
Regarding claim 9 (dependent on claim 1), Comaniciu in view of Estrada and OpenCV teaches where the processor is further configured to validate the artificial intelligence module based on at least one image from the at least one data repository that includes the visual feature (Comaniciu, para 21: “During the testing or application phase, the models are then applied to unseen data for early diagnosis and/or virtual biopsy”; testing images include the visual feature - see synthetic samples using the models in para 20).
Regarding claim 10 (dependent on claim 1), Comaniciu in view of Estrada and OpenCV teaches where the artificial intelligence module includes a deep learning artificial intelligence algorithm (Comaniciu, para 86: “deep-learnt classifier…such as using a neural network”).
Regarding claim 11, all claim limitations are met and rendered obvious by Comaniciu in view of Estrada and OpenCV because the method steps of claim 11 are the same as claim 1.
Regarding claim 13 (dependent on claim 11), Comaniciu in view of Estrada and OpenCV teaches further comprising generating the synthetic feature based on a mathematical model (Comaniciu, para 54: “The computational model is defined by parameters that control characteristics of the synthetic tumor”; para 58: "In one embodiment, the mathematical modeling of cancer uses a multi-scale model of tumor physiology"), wherein the mathematical model includes at least one user adjustable parameter (Comaniciu, para 64: "In act 40, the medical system (e.g., processor) varies parameters of the computational tumor model. The tumor model is defined by one or more parameters, such as any of the modeling parameters in the equations or used as input features in the computational model"; para 67: "The variation may be controlled or user set").
Regarding claim 15 (dependent on claim 11), Comaniciu in view of Estrada and OpenCV teaches further comprising: creating at least one additional repurposed image (Comaniciu, multiple images may be generated, para 74: “generate an image or images”) by visually manipulating the synthetic feature (Comaniciu, para 66: "For example, thousands of stochastic perturbations to the synthetic tumor model produce thousands of corresponding tumor models. Since the tumor models have known conditions (e.g., known phenotype information to be used as the ground truth), a rich dataset on which to train the machine learning model is provided").
Regarding claim 16, Comaniciu teaches a non-transitory computer-readable storage medium storing computer executable instructions (Comaniciu, para 104: “The instructions for implementing the processes, methods and/or techniques discussed herein are provided on computer-readable storage media or memories, such as a cache, buffer, RAM, removable media, hard drive or other computer readable storage media”). All further claim limitations are met and rendered obvious by Comaniciu in view of Estrada and OpenCV because the remaining limitations and executed steps of claim 16 are the same as claim 1.
Regarding claim 18 (dependent on claim 16), Comaniciu in view of Estrada and OpenCV teaches wherein the computer executable instructions further cause the processor to generate the synthetic feature based on a mathematical model (Comaniciu, para 54: “The computational model is defined by parameters that control characteristics of the synthetic tumor”; para 58: "In one embodiment, the mathematical modeling of cancer uses a multi-scale model of tumor physiology"), wherein the mathematical model includes at least one user adjustable parameter (Comaniciu, para 64: "In act 40, the medical system (e.g., processor) varies parameters of the computational tumor model. The tumor model is defined by one or more parameters, such as any of the modeling parameters in the equations or used as input features in the computational model"; para 67: "The variation may be controlled or user set").
Regarding claim 20 (dependent on claim 16), Comaniciu in view of Estrada and OpenCV teaches wherein the computer executable instructions further cause the processor to: create at least one additional repurposed image (Comaniciu, multiple images may be generated, para 74: “generate an image or images”) by visually manipulating the synthetic feature (Comaniciu, para 66: "For example, thousands of stochastic perturbations to the synthetic tumor model produce thousands of corresponding tumor models. Since the tumor models have known conditions (e.g., known phenotype information to be used as the ground truth), a rich dataset on which to train the machine learning model is provided").
Regarding claim 21 (dependent on claim 1), Comaniciu in view of Estrada and OpenCV teaches wherein the processor is configured to create the synthetic training image by a summation of pixel values of the at least one image that does not include the visual feature with corresponding pixel values of the feature image (Taught by the Image Blending operation of OpenCV in the claim 1 rejection).
Regarding claim 22 (dependent on claim 11), all claim limitations are met and rendered obvious by Comaniciu in view of Estrada and OpenCV because the limitations of claim 22 are the same as claim 21.
Regarding claim 23 (dependent on claim 16), all claim limitations are met and rendered obvious by Comaniciu in view of Estrada and OpenCV because the limitations of claim 23 are the same as claim 21.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
U.S. Patent No. 2020/0356790 A1 (para 28)
U.S. Patent No. 2021/0201474 A1 (FIG. 11)
U.S. Patent No. 11,055,569 B2
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 EMMA E DRYDEN whose telephone number is (571)272-1179. The examiner can normally be reached M-F 9-5 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, ANDREW BEE can be reached at (571) 270-5183. 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.
/EMMA E DRYDEN/Examiner, Art Unit 2677
/ANDREW W BEE/Supervisory Patent Examiner, Art Unit 2677
Prosecution Insights