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 Amendment
The amendments filed January 2nd 2026 have been entered. Claims 1-20 remain pending. Applicant’s amendments to the claims have overcome the 35 USC 112(b) rejections previously set forth in the non-final office action mailed October 1st 2025. Applicant’s amendments to the claims have overcome the 35 USC 103 rejections previously set forth, however new rejections have been issued, as necessitated by amendment.
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, 3-4, 8, 10-11, 15, 17-18 are rejected under 35 U.S.C. 103 as being unpatentable over Stille, C. M., Bekolay, T., Blouw, P., & Kröger, B. J. (2019). Natural language processing in large-scale neural models for medical screenings. Frontiers in Robotics and AI, 6. https://doi.org/10.3389/frobt.2019.00062 (Hereinafter "Stille") in view of Mefford, J. A., Zhao, Z., Heilier, L., Xu, M., Zhou, G., Mace, R., Sloane, K. L., Sheppard, S. M., & Glenn, S. (2023). Varied performance of picture description task as a screening tool across MCI subtypes. PLOS Digital Health, 2(3). https://doi.org/10.1371/journal.pdig.0000197 (Hereinafter "Mefford"), Bean (US 2024/0378398 A1) and Y. Feng et al., "PromptMagician: Interactive Prompt Engineering for Text-to-Image Creation," in IEEE Transactions on Visualization and Computer Graphics, vol. 30, no. 1, pp. 295-305, Jan. 2024, doi: 10.1109/TVCG.2023.3327168 (hereinafter “Feng”).
Regarding claim 1, Stille teaches wherein the image is described by the AI (abstract; results subsection “picture induced word naming”); using AI to generate written descriptions of the generated image simulating a cohort of healthy individuals and a cohort of individuals with a predetermined condition ("for both screenings the same model is simulated with and without neural deficits" abstract; results subsection “picture induced word naming”);
extracting diagnostic linguistic features from written descriptions of a cohort of healthy individuals and a cohort of individuals with the predetermined condition (pg. 2 left column lines 7-9; pg. 3 left column lines 35-40; pg. 3 right column lines 7-10; pg. 15 lines 11-13 of discussion section);
comparing the extracted diagnostic linguistic features of the written descriptions for the cohort of healthy individuals and the cohort of individuals with the predetermined condition (lines 3-8 of introduction, pg. 15 right column lines 35-38); and
using an image in a picture description task (figure 2)
Stille describes a system which simulates the language and speech of individuals with and without neural deficits when conducting medical screenings for conditions such as Alzheimer’s and dementia. One of the medical screening tests used in the experiment is a “picture induced word naming” task. This is analogous to the picture description task of the claimed invention. Stille extracts linguistic features as well as other information representative of neural activities and uses these in the modeling of cognitive process through the semantic pointer architecture. Stille also describes comparing the performance of its system to previous experiments, receiving similar results to the previous studies. This comparison between studies involves the comparison of linguistic features.
Stille fails to teach
using AI to generate a story based on at least one of a user input prompt and a predetermined input;
using AI to generate an image based on the generated story, wherein generating the image comprises:
translating an analytic target into a text prompt, which then guides the AI in generating the image, wherein the resulting description is compared with an original text prompt for quality control;
using the generated image in a picture description task when the compared extracted features of the written descriptions of the cohort of healthy individuals and the cohort of individuals with the predetermined condition exhibit at least a predetermined threshold of difference;
However, Bean teaches using AI to generate a story (Figure 3 #304, paragraph [0038]) based on at least one of a user input prompt and a predetermined input (Figure 3 #300 & #302, paragraphs [0036]-[0038],);
using AI to generate an image based on the generated story (Figure 3 #310, paragraphs [0007], [0042]) wherein generating the image comprises: translating an analytic target into a text prompt (Fig. 3, paragraphs [0037]-[0039] – generated story is a text prompt, and is generated to reflect the identified/classified object of the image which can be considered an analytical target), and
a text prompt, which guides the AI in generating the image (Fig. 3, paragraph [0007] – the story inputted into the image generation is analogous to a prompt); Bean describes an AI system that takes images of a user’s environment as input, which is a user input prompt in the form of an image. Bean then generates a story about the user’s environment, and then generates images based on this story. Bean is considered analogous to the claimed invention as it is in the same field of generative artificial intelligence. Therefore it would have been obvious to one of ordinary skill in the art, to replace the word naming dataset of Stille with the AI generated images of Bean to improve the time it takes to create a word naming dataset.
Bean fails to teach wherein generating the image comprises: the resulting description is compared with an original text prompt for quality control; using the generated image in a picture description task when the compared extracted features of the written descriptions of the cohort of healthy individuals and the cohort of individuals with the predetermined condition exhibit at least a predetermined threshold of difference.
However, Mefford teaches using the generated image in a picture description task when the compared extracted features of the written descriptions of the cohort of healthy individuals and the cohort of individuals with the predetermined condition exhibit at least a predetermined threshold of difference (page 8 - "Classification performance of a test score using measures other than AUROC generally require specification of a threshold on values of the test score such that individuals with scores below the threshold are classified to one group and those with scores above the threshold are classified to another group."). Mefford describes classification of a response for mild cognitive impairment (MCI) based on a predetermined threshold. Mefford is considered analogous to the claimed invention as it is in the same field of computer aided medical diagnostics. Therefore it would have been obvious to one of ordinary skill in the art, before the effective filing date, to combine the classification MCI from Mefford with the simulated groups of Stille and the generated images of Bean to ensure the simulation of groups and their interaction with the images of Bean is accurate and applicable to a real world scenario.
Stille in view of Mefford and Bean fail to teach wherein generating the image comprises: the resulting description is compared with an original text prompt for quality control.
However, Feng teaches wherein generating the image comprises: the resulting description is compared with an original text prompt for quality control (figs. 2 & 3, sections 3.2 & 4). Feng describes a prompt engineering system for image generation. This prompt engineering system takes an image generation prompt from a user, generates a set of images and prompts with more specific keywords which illicit certain aspects of that image. This process is repeated and the generated prompts and images are refined. In this process, the various prompts are compared to each other, and the original prompt, to improve image quality. Feng is considered analogous to the claimed invention as it is in the same field of artificial intelligence and image generation. Therefore it would have been obvious to one of ordinary skill in the art, before the effective filing date, to combine the prompt refinement and improvement techniques of Feng with Stille in view of Mefford and Bean in order to improve generated image and prompt quality.
Regarding claim 3, Stille in view of Mefford, Bean and Feng teaches the method of claim 1, Mefford further teaches wherein the using the generated image in a picture description task includes outputting a diagnosis or non-diagnosis for the predetermined medical condition (Mefford, last 2 paragraphs of page 4, paragraph 3 of page 5, table 1 & 2).
Regarding claim 4, Stille in view of Mefford, Bean and Feng teaches the method of claim 1, Stille further teaches generating simulated written descriptions for at least one individual from at least one of the cohort of healthy individuals and the cohort of individuals with the predetermined condition (Stille, section “simulations” paragraphs 1 & 2, results subsection “picture induced word naming”, "for both screenings the same model is simulated with and without neural deficits" abstract).
Product claim(s) 8, 10-11 is/are drawn to the method claimed in claim(s) 1, 3-4. Therefore, the product claim(s) 8, 10-11 correspond(s) to the method claim(s) 1, 3-4, and is/are rejected for the same reasons of obviousness as used above.
Product claim(s) 15, 17-18 are drawn to the method claimed in claim(s) 1, 3-4. Therefore, the product claim(s) 15, 17-18 correspond(s) to the method claim(s) 1, 3-4, and is/are rejected for the same reasons of obviousness as used above.
Claims 2, 9, 16 are rejected under 35 U.S.C. 103 as being unpatentable over view of Mefford, Bean and Feng as applied to claim 1 above, and further in view of Hamamoto, R., Suvarna, K., Yamada, M., Kobayashi, K., Shinkai, N., Miyake, M., Takahashi, M., Jinnai, S., Shimoyama, R., Sakai, A., Takasawa, K., Bolatkan, A., Shozu, K., Dozen, A., Machino, H., Takahashi, S., Asada, K., Komatsu, M., Sese, J., & Kaneko, S. (2020). Application of artificial intelligence technology in oncology: Towards the establishment of Precision Medicine. Cancers, 12(12), 3532. https://doi.org/10.3390/cancers12123532 (hereinafter "Hamamoto").
Regarding claim 2, Stille in view of Mefford, Bean and Feng teaches the method of claim 1. Stille in view of Mefford, Bean and Feng fails to teach wherein the using the generated image in a picture description task includes approving the generated image for diagnostic use of the predetermined condition
However, Hamamoto teaches wherein the using the generated image in a picture description task includes approving the generated image for diagnostic use of the predetermined condition (Table 1, #2, #5, section 7). Hamamoto describes numerous FDA approved AI assisted medical diagnosis tools such as “An iPad-based memory-assessment tool for older adults” and “A non-invasive test using AI for diagnosis and treatment of ADHD in children”. Hamamoto is considered analogous to the claimed invention as it is in the same field of AI assisted medical technology. Therefore it would have been obvious to one of ordinary skill in the art to approve generated image for diagnostic use just as Hamamoto teaches numerous FDA approved diagnosis tools that use AI. The motivation to combine would be that approval is necessary for safe, ethical and possibly legal medical testing.
Product claim 9 is drawn to the method claimed in claim 2. Therefore, the product claim 9 corresponds to the method claim 2, and is rejected for the same reasons of obviousness as used above.
Product claim 16 is drawn to the method claimed in claim 2. Therefore, the product claim 16 corresponds to the method claim 2, and is rejected for the same reasons of obviousness as used above.
Claim(s) 5-6 are rejected under 35 U.S.C. 103 as being unpatentable over view of Mefford, Bean and Feng as applied to claims 1-4 above, and further in view of Noever, David, and Samantha Elizabeth Miller Noever. "Visual AI and Linguistic Intelligence Through Steerability and Composability." arXiv preprint arXiv:2312.12383 (2023). (hereinafter "Noever").
Regarding claim 5, Stille in view of Mefford, Bean and Feng teaches the method of claim 1. Noever further teaches generating a written description prompt related to the generated image (Noever, Page 15, Appendix A – “Write a detailed prompt for dalle-2 to describe what it is”). Noever describes various methods for the use of generative artificial intelligence. In one of these use cases Noever demonstrates prompting an AI to create images of increasing complexity, in this process the user prompts the AI to create a prompt which to describes the image and can be used as a prompt for another AI agent. Noever is considered analogous to the claimed invention as it is in the same field of generative artificial intelligence. Therefore it would have been obvious to one of ordinary skill in the art to combine the teachings of Noever with the teachings of Stille in view of Mefford, Bean and Feng in order mimic human-like thought process.
Regarding claim 6, Stille in view of Mefford, Bean, Feng and in further view of Noever teaches The method of claim 5, Stille further teaches wherein at least one of the generated story, the generated image, and the generated written description prompt are designed to elicit written descriptions from the cohort of healthy individuals and the cohort of individuals with the predetermined condition with compared extracted diagnostic linguistic features that exhibit at least the predetermined threshold of difference (Stille, Fig 2 description; “simulations” section; page 5, left side, lines 1-12). Stille within the system of simulating neural deficiencies, describes test images from the WWT dataset such as the one in figure 2, which is “supposed to elicit the word wheelbarrow” and is a dataset for differentiating those with neural deficiencies such as Alzheimer’s by assessing various language skills.
Product claims 12-13 are drawn to the method claimed in claims 5-6. Therefore, the product claims 12-13 correspond to the method claims 5-6, and is/are rejected for the same reasons of obviousness as used above.
Product claims 18-20 are drawn to the method claimed in claims 5-6. Therefore, the product claims 18-20 correspond to the method claims 5-6, and are rejected for the same reasons of obviousness as used above.
Claim(s) 7 is rejected under 35 U.S.C. 103 as being unpatentable over view of Mefford, Bean, Feng and Noever as applied to claims 5-6 above, and further in view of Benedetto (US 2024/0299851 A1).
Regarding claim 7, Stille in view of Mefford, Bean, Feng and in further view of Noever teaches The method of claim 5,
Stille in view of Mefford, Bean and Feng fails to teach tuning at least one of the predetermined input, the generated story, the generated image, the generated written description prompt, and the generated simulated written descriptions when the compared extracted features of the written descriptions of the cohort of healthy individuals and the cohort of individuals with the predetermined condition do not exhibit at least the predetermined threshold of difference.
However Bendetto teaches tuning at least one of the predetermined input, the generated story, the generated image, the generated written description prompt, and the generated simulated written descriptions when the compared extracted features of the written descriptions of the cohort of healthy individuals and the cohort of individuals with the predetermined condition do not exhibit at least the predetermined threshold of difference (paragraphs [0020],[0076],[0078]). Bendetto describes tuning a generated story or image based on defined constraints. These constraints and the tuning process enables mitigation and prevention of undesirable results. Bendetto is considered analogous to the claimed invention as it is in the same field of generative artificial intelligence. Therefore it would have been obvious to one of ordinary skill in the art to combine the tuning of results present in Bendetto with the teachings of view of Mefford, Bean and Feng to mitigate and present undesirable results such as similar responses from healthy individuals and individuals with a predetermined condition.
Product claim(s) 14 is drawn to the method claimed in claim 7. Therefore, the product claim 14 corresponds to the method claim 7, and is rejected for the same reasons of obviousness as used above.
Response to Arguments
Applicant’s arguments with respect to claim(s) 1 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 amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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 Aidan W McCoy whose telephone number is (571)272-5935. The examiner can normally be reached 8:00 AM-5:00 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, Tammy Goddard can be reached at (571)272-7773. 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.
/AIDAN W MCCOY/Examiner, Art Unit 2611
/TAMMY PAIGE GODDARD/Supervisory Patent Examiner, Art Unit 2611