DETAILED ACTION
This action is in response to the amendment filed on 07/28/2025.
Response to Amendment
Applicant’s amendment filed on 07/28/2025 has been entered. Claims 1, 16 and 17 have been amended. Claims 8, 10 – 15 and 18 have been canceled. Claims 22 and 23 have been added. Claims 1 -7, 9, 16, 17 and 19 – 23 are still pending in this application, with claims 1, 16, 22 and 23 being independent.
Allowable Subject Matter
Claims 16 (with dependent claims 17, 19 – 21), 22 and 23 are allowed.
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.
Claim(s) 1 and 9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Thawkar et al. (“XrayGPT: Chest Radiographs Summarization using Large Medical Vision-Language Models)(“Thawkar”) in view of Tanwani (US 2023/0386646) and further in view of Li et al. (“LLaA-Med: Training a Large Language -and-Vision Assistant for Biomedicine in One Day”) (“Li”).
For claim 1, Thawkar discloses a system for providing natural language decision support for pathology, the system comprising: a vision encoder (MedicalVLM visual encoder, Fig.1) that receives a pathology image (chest X-Rays reveal indicators of disease, Figure 3 - Figure 6)(3.1 Model Architecture) and generates a representation of the pathology image (raw embeddings mapped to an output dimension of 512 as representation, 3.1 Model Architecture); a multimodal projector (Linear Transformer, Fig.1) that generates a first set of tokens from the representation of the pathology image (To bridge the gap between image-level features and the language decoder’s embedding space, we employ a trainable linear transformation layer, denoted as t. This layer projects the image-level features, represented by Vp, into corresponding language embedding tokens, denoted as Lv:, 3.1 Model Architecture); and a large language model (MedicalLLM, Figure 1, 1. Introduction and 3 Method) that is trained on an instruction dataset (domain-specific instruction set comprising instructions used as prompts, e.g. In both stage-1 and stage-2 of training, we utilize predetermined prompts in the given format: ###Doctor: <Img><ImageFeature></Img><Instructions>###Assistant: Here, refers to a randomly selected instruction from our pre-established set of instructions, which includes different forms of instructions like "Describe the given chest x-ray image in detail." or "Are there any potential complications or risks associated with the observed abnormalities in this chest x-ray image? or the x-ray is normal." or "Is the overall impression provided by this chest x-ray image normal or abnormal?, 1. Introduction and 5. Experiments, 5.1 Implementation Details ) complied plurality of pathology-related sources (MiMiC-CXR and OpenI, 4. Curating high quality data, Datasets), a given training sample within the instruction dataset comprising a set of pathology images and text describing or answering specific queries pertaining to the images (the chest radiographs comprise associated reports which describe the radiographs, 4. Curating high-quality data, Datasets), the large language model receiving a second set of tokens associated with a prompt (We have two text queries in our overall architecture. The first query, denoted as ###Assistant, serves the purpose of determining the system role, which in our case is defined as "You are a helpful healthcare virtual assistant." The second text query, ###Doctor, corresponds to the prompt itself. To ensure consistency, both queries undergo tokenization, resulting in dimensions represented by Lt . Finally, Lv is concatenated with Lt and fed into the medical LLM, Fine-tuned Vicuna, 3. Method, 3.1 Model Architecture) and determining a response from the first set of tokens and the second set of tokens (Finally, Lv is concatenated with Lt and fed into the medical LLM, Fine-tuned Vicuna, which generates the summary of the chest x-ray, 3. Method, 3.1 Model Architecture; Figure 2 – Figure 5). Yet, Thawkar fails to teach that the system comprises: a processor and a non-transitory computer readable medium storing instructions executable by the processor, the machine-executable instructions comprising the vision encoder, multimodal projector and large language model; and the multimodal projection model is trained on a set of image-caption pairs while a set of weights associated with the large language model are maintained unchanged during a first stage of training of the system and the multimodal projector and the large language model are trained together on the instruction dataset during a second stage of training of the system.
However, Tanwani discloses a system and method for automatically generating medical reports (Abstract), comprising the following: a processor (Fig.3, 310; [0068]); and a non-transitory computer readable medium (Fig.3, 320) storing instructions executable by the processor ([0069] [0071]), the machine-executable instructions comprising components (image encoder, text encoder and language model, Fig.2, 220, 230 and 250) which perform medical report generation based on a medical image and question ([0036 – 0041] [0071] [0072]).
Additionally, Li discloses a method for training a large language and vision assistant for biomedicine (Abstract), comprising the following: multimodal projection model (projection matrix) is trained on a set of image-caption pairs while a set of weights associated with the large language model are maintained unchanged during a first stage of training of the system (4. Adapting Multimodal Conversational Models to the Biomedical Domain, Stage 1 : Biomedical Concept Feature Alignment, pg. 4 and 5); and the multimodal projector and the large language model are trained together on the instruction dataset during a second stage of training of the system ( Adapting Multimodal Conversational Models to the Biomedical Domain, Stage 2 : End-to End Instruction Tuning, pg. 5).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve Thawkar’s invention in the same way that Tanwani’s invention has been improved to achieve the following predictable results for the purpose of efficiently and effectively automating the time-consuming and cumbersome task of analyzing and interpretating medical images (Thawkar, Abstract) (Tanwani, [0003] [0004]): the system further comprises a processor and a non-transitory computer readable medium storing instructions executable by the processor, the machine-executable instructions comprising components including a vision encoder, multimodal projector and large language model.
Additionally, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve the invention disclosed by the combination of Thawkar and Tanwani in the same way that Li’s invention has been improved to achieve the following predictable results for the purpose of efficiently and effectively automating the time-consuming and cumbersome task of analyzing and interpretating medical images (Thawkar, Abstract) (Tanwani, [0003] [0004]): the multimodal projection model is trained on a set of image-caption pairs while a set of weights associated with the large language model are maintained unchanged during a first stage of training of the system; and the multimodal projector and the large language model are trained together on the instruction dataset during a second stage of training of the system.
For claim 9, Thawkar further discloses wherein the instruction dataset (pre-established set of instructions and datasets including MIMIC-CXR and OpenI) is selected as to exclude any text below a threshold length (Thawkar, To generate concise and coherent medical summaries from the unstructured reports, we perform the following pre-processing steps for both datasets: (1) Removal of incomplete reports lacking finding or impression sections. (2) Elimination of reports that have finding sections containing less than 10 words. (3) Exclusion of reports with impression sections containing less than 2 words. … In stage-1 training, the model is designed to gain understanding of how Xray image features and corresponding reports are interconnected by analysing a large set of image-text pairs … We use high quality interactive report summary as described in sec. 4 of MIMIC-CXR (Johnson et al., 2019) train set with 213,514 image text pairs for training…. In stage-2 training, the pretrained stage-1 model is enforced to gain radiology specific summary of how Xray image features by examining set of highly curated image-text pairs from OpenI dataset … In both stage-1 and stage-2 of training, we utilize predetermined prompts in the given format: ###Doctor: <Img><ImageFeature></Img><Instruction>###Assistant: Here, refers to a randomly selected instruction from our pre-established set of instructions, which includes different forms of instructions like "Describe the given chest x-ray image in detail." or "Are there any potential complications or risks associated with the observed abnormalities in this chest x-ray image? or the x-ray is normal." or "Is the overall impression provided by this chest x-ray image normal or abnormal? Answer based on the observed findings.", 4.Curating high-quality data and 5.Experiments, 5.1 Implementation Details).
Claim(s) 2 is/are rejected under 35 U.S.C. 103 as being unpatentable over Thawkar et al. (“XrayGPT: Chest Radiographs Summarization using Large Medical Vision-Language Models)(“Thawkar”) in view of Tanwani (US 2023/0386646), and further in view of Li et al. (“LLaA-Med: Training a Large Language -and-Vision Assistant for Biomedicine in One Day”) (“Li”) and further in view of Zhou et al. (US 2022/0130499) (“Zhou”).
For claim 2, the combination of Thawkar, Tanwani and Li further discloses: a user interface that receives the prompt from the user as natural language text (Thawkar, Figure 2 – Figure 5)(Tanwani, Fig.3, 330; [0007] [0071]); and a tokenizer that generates the second set of tokens (Thawkar, 3.1 Model Architecture)(Tanwani, Fig.2, 230; [0036] [0047]). Yet, the combination of Thawkar and Tanwani fails to teach that the user interface displays the response to the user at an associated display.
However, Zhou discloses a system and method for performing medical visual question answering (Abstract), wherein a response to a user query regarding a medical image is displayed to the user on a display at the user’s computing device ([0035] [0037 – 0042] [0049] [0050] [0075] [0077]).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve the invention disclosed by the combination of Thawkar, Tanwani and Li in the same way Zhou’s invention has been improved to achieve the predictable results of the user interface (input/output devices) further displaying the response to the user at an associated display for the purpose of efficiently and effectively automating the time-consuming and cumbersome task of analyzing and interpretating medical images (Thawkar, Abstract) (Tanwani, [0003] [0004]).
Claim(s) 3 and 4 is/are rejected under 35 U.S.C. 103 as being unpatentable over Thawkar et al. (“XrayGPT: Chest Radiographs Summarization using Large Medical Vision-Language Models)(“Thawkar”) in view of Tanwani (US 2023/0386646), and further in view of Li et al. (“LLaA-Med: Training a Large Language -and-Vision Assistant for Biomedicine in One Day”) (“Li”), and further in view of Zhou et al. (US 2022/0130499) (“Zhou”) and further in view of Conjeti (US 2023/0076903).
For claim 3, the combination of Thawkar, Tanwani, Li and Zhou fails to teach, wherein the response is a first response of a plurality of responses and the user interface allows the user to select a response from the plurality of responses, the image and the selected response being added to the instruction data set as a training sample.
However, Conjeti discloses a method for the purpose of providing cue-based medical reporting assistance (Abstract), comprising the following: providing a plurality of diagnostic cues (responses, e.g. questions and/or notes, provided by a processing algorithm) ([0023 -0025] [0031] [0092 - 0096]; Table 1) to a user interface ([0029]) based on medical imaging data and textual input ([0032 – 0039] [0040] [0056] [0058 – 0063] [0070] [0074] [0076]); selecting, by a user, a diagnostic cue from the plurality of cues ([0042] [0077]), the image and the selected response being added to a data set as a training sample (In some examples, the training phase 2005 may be re-executed from time to time, based on ground-truth labels obtained from or during the inference phase 2010. This is a technique that can be labeled life-long-learning (LLL) … Optionally, at box 4050, LLL could be implemented, as part of the training phase 2005. This can be based on further user feedback regarding a quality of the one or more cues that are posed to the user at box 4035. A “like” or “dislike” button may be used. A “1 to 5 star rating” or the like may be used. Thereby, ground-truth labels for the training can be derived . The ground-truth labels are the cues corresponding to the input medical image data. Re-executing training using the derived ground truth labels and associated input medical image data implies that the selected response, e.g. cue, and image have been added to the training data, [0045] [0047] [0078] [0097 – 0100]).
, [0045] [0047] [0078] [0097 – 0100]).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve the invention disclosed by the combination of Thawkar, Tanwani, Li and Zhou in the same way that Conjeti’s invention has been improved to achieve the following predictable results for the purpose of efficiently and effectively automating the time-consuming and cumbersome task of analyzing and interpretating medical images using optimized machine learning models (Thawkar, Abstract) (Tanwani, [0003] [0004] (Conjeti, [0079]): further providing a plurality of responses, wherein the response is a first response of a plurality of responses; and further allowing, by the user interface the user to select a response from the plurality of responses, the image and the selected response being added to a data set, e.g. instruction data set, as a training sample.
For claim 4, the combination of Thawkar, Tanwani, Li and Zhou fails to teach wherein the response is a first response of a plurality of responses and the user interface allows the user to rate a response as one of helpful and unhelpful, the image and the selected response being added to the instruction data set as a training sample when the response is rated as helpful.
However, Conjeti discloses a method for the purpose of providing cue-based medical reporting assistance (Abstract), comprising the following: providing a plurality of diagnostic cues (responses, e.g. questions and/or notes, provided by a processing algorithm) ([0023 -0025] [0031] [0092 - 0096]; Table 1) to a user interface ([0029]) based on medical imaging data and textual input ([0032 – 0039] [0040] [0056] [0058 – 0063] [0070] [0074] [0076]); rating, by a user, a diagnostic cue from the plurality of cues ([0042] [0077]) as one of helpful (“like” or “5 star”) and unhelpful (“dislike” or “1 star”, the image and the selected response being added to a data set as a training sample when the response is helpful (In some examples, the training phase 2005 may be re-executed from time to time, based on ground-truth labels obtained from or during the inference phase 2010. This is a technique that can be labeled life-long-learning (LLL) … Optionally, at box 4050, LLL could be implemented, as part of the training phase 2005. This can be based on further user feedback regarding a quality of the one or more cues that are posed to the user at box 4035. A “like” or “dislike” button may be used. A “1 to 5 star rating” or the like may be used. Thereby, ground-truth labels for the training can be derived. The ground-truth labels are the cues corresponding to the input medical image data. Re-executing training using the derived ground truth labels and associated input medical image data implies that the selected response, e.g. cue, and image have been added to the training data, [0045] [0047] [0078] [0097 – 0100]).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve the invention disclosed by the combination of Thawkar, Tanwani, Li and Zhou in the same way that Conjeti’s invention has been improved to achieve the following predictable results for the purpose of efficiently and effectively automating the time-consuming and cumbersome task of analyzing and interpretating medical images using optimized machine learning models (Thawkar, Abstract) (Tanwani, [0003] [0004] (Conjeti, [0079]): further providing a plurality of responses, wherein the response is a first response of a plurality of responses; and further allowing, via the user interface, the user to rate a response as one of helpful and unhelpful, the image and the selected response being added to a data set, e.g. instruction data set, as a training sample.
Claim(s) 5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Thawkar et al. (“XrayGPT: Chest Radiographs Summarization using Large Medical Vision-Language Models)(“Thawkar”) in view of Tanwani (US 2023/0386646), and further in view of Li et al. (“LLaA-Med: Training a Large Language -and-Vision Assistant for Biomedicine in One Day”) (“Li”), and further in view of Paulett et al. (US 2024/0347156) (“Paulett”) and further in view of Natarajan et al. (US 2024/0428937) (“Natarajan”).
For claim 5, the combination of Thawkar, Tanwani and Li fails to teach, wherein one of the second set of tokens and the prompt are stored on the non-transitory computer readable medium, and the second set of tokens is provided to the large language model without input by a user.
However, Paulett discloses a system and method for radiology reporting (Abstract), comprising the following: receiving inputs which are used to determine a radiology report, wherein the inputs comprise a set of images([0045] [0046]); and automatically generating a radiology report based on no manual inputs (zero-click) ([0090 - 0092]).
Moreover, Natarajan discloses a system and method for the purpose of processing data using a machine learning model (Abstract), wherein input (image data) and one of a prestored prompt data (exemplar prompt data values for input to one or more tokenizing or embedding layers) and prestored tokens (already embedded hard prompts) are provided to machine learned model to generate an task specific output ([0035] [0036] [0039] [0040] [0042] [0043] [0045] [0051] [0052] [0055] [0061- 0066]).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve the invention disclosed by the combination of Thawkar, Tanwani and Li in the same way that Paulett’s invention has been improved to achieve the following, predictable results for the purpose of or the purpose of efficiently and effectively automating the time-consuming and cumbersome task of analyzing and interpretating medical images (Thawkar, Abstract) (Tanwani, [0003] [0004]): the response (answer generated from radiology images, Thawkar, Abstract) generated by a large language model based on image data is further generated automatically without user input.
Additionally, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve the invention disclosed by the combination of Thawkar, Tanwani, Li and Paulett in the same way that Natarajan’s invention has been improved to achieve the following, predictable results for the purpose of efficiently and effectively automating the time-consuming and cumbersome task of analyzing and interpretating medical images using optimized machine learning models (Thawkar, Abstract) (Tanwani, [0003] [0004]): one of the prompt and second set of tokens (embedding values of the prompt) are stored on the non-transitory computer readable medium; and the prompt or second set of tokens is provided to the machine learning model, e.g. large language model, at any time, including automatically without user input after receiving image data.
Claim(s) 6 is/are rejected under 35 U.S.C. 103 as being unpatentable over Thawkar et al. (“XrayGPT: Chest Radiographs Summarization using Large Medical Vision-Language Models)(“Thawkar”) in view of Tanwani (US 2023/0386646), and further in view of Li et al. (“LLaA-Med: Training a Large Language -and-Vision Assistant for Biomedicine in One Day”) (“Li”), and further in view of Wang et al. (“Automated Radiographic Report Generation Purely on Transformer: A Multicriteria Supervised Approach”) and further in view of Tian et al. (“A Two-Stage Deep Learning Strategy for Pneumothorax Classification”).
For claim 6, the combination of Thawkar, Tanwani and Li fails to teach the system of claim 1, further comprising an image segmenter that selects a region of interest within an image, the pathology images being a plurality of tiles generated from the region of interest and the first set of tokens representing the content and position of the pathology images.
However, Wang discloses a system and method for the purpose of performing radiographic report generation (Abstract), comprising the following: generating a plurality of patches from a medical image (chest X-ray) (Fig.1; III Proposed Method, A. Structures of the Proposed Model, 1) Vision Transformer Encoder, pg. 2806); generating output of the vision encoder, wherein the output represents the content and position of the patches (the output of the vision encoder is generated base one embeddings comprising content of the patches and position information of the patches, Fig.1; III Proposed Method, A. Structures of the Proposed Model, 1) Vision Transformer Encoder, pg. 2806).
Additionally, Tian further discloses a system and method for the purpose of diagnosing disease (Abstract), comprising the following preprocessing steps: segmenting a medical image by selecting region of interest within the image (III. Materials and Methods, A. Materials); and dividing the ROI into patches to perform a task (III. Materials and Methods, A. Materials).
Therefore, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve the invention disclosed by the combination of Thawkar, Tanwani and Li in the same way that Wang’s invention has been improved to achieve the following, predictable results for the purpose of efficiently and effectively automating the time-consuming and cumbersome task of analyzing and interpretating medical images using optimized machine learning models and data pre-preprocessing (Thawkar, Abstract) (Tanwani, [0003] [0004]): the system further comprises image pre-processing functionally, e.g. image segmenter; the image segmenter further divides an image into pathology images being a plurality of tiles (patches), wherein the first set of tokens generated by the vision encoder further represent content and position of the pathology images (patches).
Additionally it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve the invention disclosed by the combination of Thawkar, Tanwani, Li and Wang in the same way that Tian’s invention has been improved to achieve the following, predictable results for the purpose of efficiently and effectively automating the time-consuming and cumbersome task of analyzing and interpretating medical images using optimized machine learning models and data pre-preprocessing (Thawkar, Abstract) (Tanwani, [0003] [0004]): the image segmenter further pre-processes the image by selecting a region of interest, wherein the pathology images are generated from the region of interest.
Claim(s) 7 is/are rejected under 35 U.S.C. 103 as being unpatentable over Thawkar et al. (“XrayGPT: Chest Radiographs Summarization using Large Medical Vision-Language Models)(“Thawkar”) in view of Tanwani (US 2023/0386646), and further in view of Li et al. (“LLaA-Med: Training a Large Language -and-Vision Assistant for Biomedicine in One Day”) (“Li”), and further in view of Liu et al. (US 2024/0177838) (“Liu”) and further in view of Zhang et al. (“Large-Scale Domain-Specific Pretraining for Biomedical Vision-Language Processing”) (“Zhang”).
For claim 7, the combination of Thawkar, Tanwani and Li further disclose wherein, plurality of pathology-related sources includes pathology case reports (Thawkar, MIMIC-CXR and OpenI, 4.Curating high-quality data, Datasets). Yet, the combination of Thawkar, Tanwani and Li fails to teach that the plurality of pathology-related sources includes at least two of captions of medical images, educational articles, pathology case reports, and extracted regions from whole slide imaging.
However, Liu discloses a system and method for processing electronic images using deep foundation models (Abstract), wherein pathology related sources which can be used to train a foundation model include whole slide images, extracted regions of whole slide images (In a first step, the embeddings of image tiles, which may be derived from Whole Slide Images (WSIs), may be stored in a vector database. This database may be designed for fast querying, and/or may enable efficient retrieval of similar image tiles based on their embeddings. To optimize storage, the system may identify and/or save only a few representative tiles from each WSI. This approach may reduce the storage footprint while maintaining the ability to retrieve a comprehensive range of similar tiles) and X-ray images ([0028] [0033] [0084]).
Moreover, Zhang discloses a system and method for pretraining for biomedical vision and language processing (Abstract), wherein pathology related sources which can be used for training (pretraining) include captions of medical images and educational articles (PubMed Central contains 4.4 million publicly available full-text articles (as of June 15 2022). We downloaded and extracted compressed directories with complete article packages1. Each article is represented as a package of XML, PDF, media, and supplementary materials. We extracted figure files and the matching captions, along with PMID and PMCID of the provenance articles. This yields a dataset PMC-15M with 15 million figure-caption pairs from over 3 million article … . Images in PMC-15M are extremely diverse, ranging from generic biomedical illustration (e.g., statistical figures, graphs, charts, and tables and forms) to radiography (e.g., magnetic resonance, computerized tomography, and X-ray) to microscopy (e.g., transmission microscopy, and electron microscopy), among others.)
Therefore, it would been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve the invention disclosed by the combination of Thawkar, Tanwani and Li in the same way that Liu’s invention has been improved to achieve the predictable results of the pathology related sources further including extracted regions from whole slide imaging for the purpose of efficiently and effectively automating the time-consuming and cumbersome task of analyzing and interpretating medical images using optimized machine learning models and data pre-preprocessing (Thawkar, Abstract) (Tanwani, [0003] [0004]).
Moreover, it would have been obvious to one of ordinary skill in the art at the time of applicant’s filing to improve the invention disclosed by the combination of Thawkar, Tanwani, Li and Liu in the same way that Zhang’s invention has been improved to achieve the predictable results of the pathology related sources further including captions of medical images for the purpose of efficiently and effectively automating the time-consuming and cumbersome task of analyzing and interpretating medical images using optimized machine learning models and data pre-preprocessing (Thawkar, Abstract) (Tanwani, [0003] [0004]):
Response to Arguments
Applicant’s arguments with respect to claim(s) 1- 7 and 9 have been considered but are moot in view of the new ground(s) of rejection.
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 SONIA L GAY whose telephone number is (571)270-1951. The examiner can normally be reached Monday-Friday 9-5 ET.
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, Daniel Washburn can be reached on 571-272-5551. 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.
/SONIA L GAY/Primary Examiner, Art Unit 2657