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 .
Claims 1-20 are pending in the Application and Claims 1, 11 and 20 are independent claims.
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 (i.e., changing from AIA to pre-AIA ) 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.
Claims 1, 6-7, 11, 16-17 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Al-Rikabi et al. Pat No. US 12223080 B1 (Al-Rikabi) in view of Oduro-Afriyie et al., “Enabling the informed patient paradigm with secure and personalized medical question answering,” In Proceedings of the 14th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics, 2023 Sep 3 (pp. 1-6). .
Regarding Claim 1. Al-Rikabi discloses a system, comprising:
one or more processors, coupled with memory (Al-Rikabi, col 42, ln 15-17, The CPUs 1904 can be standard programmable processors that perform arithmetic and logical operations necessary for the operation …), to:
receive, via a chatbot interface, a textual input related to an electronic report (Al-Rikabi, col 8, ln 45-49, The NER model input 208 takes the NLQ text and determines spans in the text (identified by offsets in the text) along with possible high-level types. The NER model 208 outputs labels that describe the expected semantic role of an entity mentioned in the NLQ; [i.e., NLQ as a “chatbot interface”; NLQ (Natural Language Query) is a chatbot which is an AI assistant that translates plain, conversational language into structured database queries (like SQL)]);
generate, in response to the textual input, using a large language model, an output comprising a set of keywords, a rephrased version of the textual input, and an intent sentence (Al-Rikabi, col 18, ln 60-col 19, ln 3, For the model-based approach, a text generation model may be used to curate synonym candidates for a target phrase (which may be performed using a ML model 116 of NLQ query search service 110 or a ML model 120 of data model service 118). For each target phrase, a context sentence may be generated (with the phrase-to-be-replaced identified by some textual marker, e.g., single quotes.) A general pretrained encoder-decoder model may be fine-tuned to specifically replace the identified phrase with an alternate phrase and collect that candidate noted by the same textual marker in the output. For example, using a model-based text generation model, e.g., one of ML models 116 or 120, an input may be: “paraphrase: in the context of health and healthcare, the health service area is Western New York </s> right carrot.” The output may be “in the context of health and healthcare, the service region is Western New York”... FIG. 6 schematically illustrates an example flow 600 for generating one or more synonyms for columns of dataset 122. At 602, a target phrase is determined. At 604, a domain specific phrase is generated to generate a prompt for a large language model to fill in);
filter, using the output from the large language model, an ontology stored in a database to generate a filtered ontology comprising one or more ontology elements, intents, and examples (Al-Rikabi, col 3, ln 30-38, For example, with column synonym generation, a knowledge graph (e.g., using ConceptNet) may be traversed for a target phrase, such as “Patient”, to generate candidate synonyms. A domain specific phrase may be generated to generate a prompt for a large language model to fill in. For example, using “Patient” the phrase generated might be “In the context of healthcare, Patient is <candidate>”. The hydrated phrase may be ranked via an ensemble of similarity models; [i.e., “traverse for a target phrase”” as “filter an ontology”]; Al-Rikabi, col 4, ln 4-32, The techniques and architecture may also provide row level security (RLS) for autocomplete during entry of NLQs and fuzzy matching in natural language question answering; Al-Rikabi, col 5, ln 20-30, In configurations, synthetic question generation may be implemented on previously unseen schema using a three part process: composable, synthetic templates that generate combinations of potential questions using schema and data elements, an ML model that filters and scores the combinations based on how sensible the combinations are (as a perplexity score from the fine-tuned large language model), and finally an answerability filter that re-runs the question through an existing inference pipeline to make sure that in fact it is possible to generate the intended answer from that question text; [i.e., the concept of “output of the LLM being used to filter an ontology” was taught in Al-Rikabi]);
display, via the chatbot interface, the list of actions (Al-Rikabi, para col 16, ln 17-22, Fig 1 and FIG. 11 schematically illustrates an example flow: 1100 for CA analysis at topic creation time that may be implemented by an algorithm. At 1102, select potential dimensions, e.g., columns. At 1104, scan visuals that are displayed by BI service 108 or NLQ query service 110 on the client device 114. For example, obtain all visuals, e.g., dashboards, reports; Al-Rikabi, para col 7, ln 58-61, In response to the NLQs 124, the BI service 108 performs searches the one or more datasets 122 for one or more possible answers. This can involve using one or more ML models 116 and/or 120 to analyze the NLQs 124. [i.e., Fig. 1: “NLQ query service 110” which is part of the “Business Intelligence (BI) service 108” as “chatbot interface”; Fig 1, user connected to client device 114 whose output NLQ 124 is input to the BI 102 to be analyzed (“list of actions”) by ML models 116 and/or 120]);
receive, via the chatbot interface, an indication to execute an action from the list of actions (Al-Rikabi, para 16, ln 17-22, the author may execute the IR to get the corresponding visual (e.g., dashboard) so that the author may better understand the answer. At 512, the author may accept or reject the some or all of the synthetic NLQs, or in some instances may rephrase some or all of the synthetic NLQs to be more natural); and
provide, responsive to the indication, instructions to execute the action associated with the electronic report to modify the electronic report (Al-Rikabi, col 41, ln 54-col 42, ln 8, Such requests can, for instance, be requests to add, delete, change or otherwise modify policy for a customer, service, or system, or for other administrative actions, such as providing an inventory of existing policies and the like… The computer architecture shown in FIG. 19 illustrates a server computer, workstation, desktop computer, laptop, tablet, network appliance, e-reader, smartphone, or other computing device, and can be utilized to execute any of the software components presented herein).
Al-Rikabi does not specifically disclose generate, using the large language model and the filtered ontology, a list of actions that are executable to modify the electronic report.
However, Oduro-Afriyie, in the same field of endeavor, discloses generate, using the large language model and the filtered ontology, a list of actions that are executable to modify the electronic report (Oduro-Afriyie, 1st – 2nd page, Section 1.1, 1st - 2nd para, In this paper, we introduce a knowledge graph (KG)-based question answering (QA) system, Medicient, designed to tackle these challenges by incorporating data from patient EHRs/PHLs, drug databases, and medical literature/research to provide personalized treatment recommendations for patients while safeguarding HIPPA mandated privacy… The proposed system leverages the power of large language models for question understanding and natural language (NL) response generation while concealing the EHR or PHL from the language model to preserve privacy; Oduro-Afriyie, 2nd page, Section 2.4, 1st para, Our proposed system aims to bridge this gap by incorporating the strengths of knowledge graphs and large language models while ensuring the protection of sensitive patient information; Oduro-Afriyie, 4th page, Section 4.1.2, 2nd col, Algorithm 1 Semantic Path Discovery (SPD) to answer an NL query Q from a KG; Oduro-Afriyie, 4th page, Section 4.1.3, 2nd col, As shown in Algorithm 1, the output of our SPD algorithm is an aggregation 𝐴 of entities and relations from the KG. These entities and relations represent the concepts and actions that can be used to answer the main query 𝑄; [i.e., “actions that can be used to answer the main query 𝑄” are generated by Algorithm 1 based on knowledge graph (KG) (ontology) traversing and LLM] ).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Oduro-Afriyie in the method of Al-Rikabi because this would enable developing a privacy-preserving KG-based QA system that leverages EHRs/PHLs, drug databases, general medical literature to equip clinicians with a powerful tool to navigate the vast landscape of medical information (Oduro-Afriyie, 2nd page, Section 1.1, 2nd para).
Regarding Claim 6. Al-Rikabi in view of Oduro-Afriyie disclose the system of claim 1, wherein the one or more processors (Al-Rikabi, para 42, ln 15-16, The CPUs 1904 can be standard programmable processors) are further configured to:
generate a search query based on the set of keywords, the rephrased version of the textual input, and the intent sentence (Al-Rikabi, para 33, ln 44-56, The one or more paraphrased synthetic NLQs may be generated based on global question patterns within the datasets, e.g., dataset 122, a user's topic metadata, and the user's IR. The IR may be generated as described with respect to FIG. 2 herein. In order to obtain a high quality paraphrased synthetic NLQ, the one or more paraphrased synthetic NLQs may be filtered. For example, the one or more paraphrased synthetic NLQs may be filtered with respect to fluency using a language model in natural language to filter the less fluent paraphrased synthetic NLQs. For cycle consistency, the paraphrased synthetic NLQs may be filtered based on which paraphrased synthetic NLQs are executable within the NLQ query service 110); and
provide the search query to a search engine to identify the one or more ontology elements, the intents, and the examples (Al-Rikabi, para 20, ln 23-35, For the graph-based approach, synonyms from one or more public knowledge graph sources may be directly collected. For a model-based approach, a text generation model may be used to curate synonym candidates for a target phrase (which may be performed using a ML model 116 of NLQ query search service 110 or a ML model 120 of data model service 118). For each target phrase, a context sentence may be generated (with the phrase-to-be-replaced identified by some textual marker, e.g., single quotes.) A general pretrained encoder-decoder model may be fine-tuned to specifically replace the identified phrase with an alternate phrase and collect that candidate noted by the same textual marker in the output; Al-Rikabi, para 25, ln 36-45, In configurations, a type of NLQ is a “Why” question, e.g., “why is revenue down in North America Q1 2022. In order to answer such questions, it is necessary to predict which factors are most likely to contribute to this change in metric so that various dimensions, e.g., columns of a dataset, of potential contribution can actually be analyzed to find insights. In particular, to answer “why” questions it is necessary to not only understand the intent of the user's question (what metric to analyze, what time period), but it is also necessary to automatically predict which factors to use; [“NLQ query search service …to identify…” and also “the NLQ… not only understand the intent of the user's question (what metric to analyze, what time period), but it is also necessary to automatically predict which factors to use” as “the search query to a search engine to identify the one or more ontology elements, the intents, and the examples”]).
Regarding Claim 7. Al-Rikabi in view of Oduro-Afriyie disclose the system of claim 1, wherein the one or more processors (Al-Rikabi, para 42, ln 15-16, The CPUs 1904 can be standard programmable processors) are further configured to:
generate an input context based on the filtered ontology (Al-Rikabi, para 34, ln 14-47, At, 1406, generating one or more synonyms for each name… For the graph-based approach, synonyms from one or more public knowledge graph sources may be directly collected. For a model-based approach, a text generation model may be used to curate synonym candidates for a target phrase (which may be performed using a ML model 116 of NLQ query search service 110 or a ML model 120 of data model service 118). For each target phrase, a context sentence may be generated (with the phrase-to-be-replaced identified by some textual marker, e.g., single quotes.) A general pretrained encoder-decoder model may be fine-tuned to specifically replace the identified phrase with an alternate phrase and collect that candidate noted by the same textual marker in the output. The candidate synonyms may be filtered using one of ML models 116 or 120 to include only those candidate synonyms that are most semantically relevant to the original target phrase); and
provide the input context to the large language model to cause the large language model to generate the list of actions (At, 1406, generating one or more synonyms for each name. For example, generating one or more synonyms for columns of dataset 122 determining a target phrase. The target phrase is used to generate candidate synonyms. For example, a knowledge graph (e.g., using ConceptNet) may be traversed for a target phrase, such as “Patient”, to generate candidate synonyms. A domain specific phrase may be generated to generate a prompt for a large language model to fill in… a text generation model may be used to curate synonym candidates for a target phrase (which may be performed using a ML model 116 of NLQ query search service 110 or a ML model 120 of data model service 118). For each target phrase, a context sentence may be generated (with the phrase-to-be-replaced identified by some textual marker, e.g., single quotes.)).
Regarding Claim 11. Al-Rikabi discloses a method, comprising:
receiving, via a chatbot interface, a textual input related to an electronic report (Al-Rikabi, col 8, ln 45-49, The NER model input 208 takes the NLQ text and determines spans in the text (identified by offsets in the text) along with possible high-level types. The NER model 208 outputs labels that describe the expected semantic role of an entity mentioned in the NLQ; [i.e., NLQ as a “chatbot interface”; NLQ (Natural Language Query) is a chatbot which is an AI assistant that translates plain, conversational language into structured database queries (like SQL)]);
generating, in response to the textual input, using a large language model, an output comprising a set of keywords, a rephrased version of the textual input, and an intent sentence (Al-Rikabi, col 18, ln 60-col 19, ln 3, For the model-based approach, a text generation model may be used to curate synonym candidates for a target phrase (which may be performed using a ML model 116 of NLQ query search service 110 or a ML model 120 of data model service 118). For each target phrase, a context sentence may be generated (with the phrase-to-be-replaced identified by some textual marker, e.g., single quotes.) A general pretrained encoder-decoder model may be fine-tuned to specifically replace the identified phrase with an alternate phrase and collect that candidate noted by the same textual marker in the output. For example, using a model-based text generation model, e.g., one of ML models 116 or 120, an input may be: “paraphrase: in the context of health and healthcare, the health service area is Western New York </s> right carrot.” The output may be “in the context of health and healthcare, the service region is Western New York”... FIG. 6 schematically illustrates an example flow 600 for generating one or more synonyms for columns of dataset 122. At 602, a target phrase is determined. At 604, a domain specific phrase is generated to generate a prompt for a large language model to fill in);
filtering, using the output from the large language model, an ontology stored in a database to generate a filtered ontology comprising one or more ontology elements, intents, and examples (Al-Rikabi, col 3, ln 30-38, For example, with column synonym generation, a knowledge graph (e.g., using ConceptNet) may be traversed for a target phrase, such as “Patient”, to generate candidate synonyms. A domain specific phrase may be generated to generate a prompt for a large language model to fill in. For example, using “Patient” the phrase generated might be “In the context of healthcare, Patient is <candidate>”. The hydrated phrase may be ranked via an ensemble of similarity models; [i.e., “traverse for a target phrase”” as “filter an ontology”]; Al-Rikabi, col 4, ln 4-32, The techniques and architecture may also provide row level security (RLS) for autocomplete during entry of NLQs and fuzzy matching in natural language question answering; Al-Rikabi, col 5, ln 20-30, In configurations, synthetic question generation may be implemented on previously unseen schema using a three part process: composable, synthetic templates that generate combinations of potential questions using schema and data elements, an ML model that filters and scores the combinations based on how sensible the combinations are (as a perplexity score from the fine-tuned large language model), and finally an answerability filter that re-runs the question through an existing inference pipeline to make sure that in fact it is possible to generate the intended answer from that question text; [i.e., the concept of “output of the LLM being used to filter an ontology” was taught in Al-Rikabi]);
displaying, via the chatbot interface, the list of actions (Al-Rikabi, para col 16, ln 17-22, Fig 1 and FIG. 11 schematically illustrates an example flow: 1100 for CA analysis at topic creation time that may be implemented by an algorithm. At 1102, select potential dimensions, e.g., columns. At 1104, scan visuals that are displayed by BI service 108 or NLQ query service 110 on the client device 114. For example, obtain all visuals, e.g., dashboards, reports; Al-Rikabi, para col 7, ln 58-61, In response to the NLQs 124, the BI service 108 performs searches the one or more datasets 122 for one or more possible answers. This can involve using one or more ML models 116 and/or 120 to analyze the NLQs 124. [i.e., Fig. 1: “NLQ query service 110” which is part of the “Business Intelligence (BI) service 108” as “chatbot interface”; Fig 1, user connected to client device 114 whose output NLQ 124 is input to the BI 102 to be analyzed (“list of actions”) by ML models 116 and/or 120]);
receiving, via the chatbot interface, an indication to execute an action from the list of actions (Al-Rikabi, para 16, ln 17-22, the author may execute the IR to get the corresponding visual (e.g., dashboard) so that the author may better understand the answer. At 512, the author may accept or reject the some or all of the synthetic NLQs, or in some instances may rephrase some or all of the synthetic NLQs to be more natural); and
providing, responsive to the indication, instructions to execute the action associated with the electronic report to modify the electronic report (Al-Rikabi, col 41, ln 54-col 42, ln 8, Such requests can, for instance, be requests to add, delete, change or otherwise modify policy for a customer, service, or system, or for other administrative actions, such as providing an inventory of existing policies and the like… The computer architecture shown in FIG. 19 illustrates a server computer, workstation, desktop computer, laptop, tablet, network appliance, e-reader, smartphone, or other computing device, and can be utilized to execute any of the software components presented herein).
Al-Rikabi does not specifically disclose generating, using the large language model and the filtered ontology, a list of actions that are executable to modify the electronic report;
However, Oduro-Afriyie, in the same field of endeavor, discloses generating, using the large language model and the filtered ontology, a list of actions that are executable to modify the electronic report (Oduro-Afriyie, 1st – 2nd page, Section 1.1, 1st - 2nd para, In this paper, we introduce a knowledge graph (KG)-based question answering (QA) system, Medicient, designed to tackle these challenges by incorporating data from patient EHRs/PHLs, drug databases, and medical literature/research to provide personalized treatment recommendations for patients while safeguarding HIPPA mandated privacy… The proposed system leverages the power of large language models for question understanding and natural language (NL) response generation while concealing the EHR or PHL from the language model to preserve privacy; Oduro-Afriyie, 2nd page, Section 2.4, 1st para, Our proposed system aims to bridge this gap by incorporating the strengths of knowledge graphs and large language models while ensuring the protection of sensitive patient information; Oduro-Afriyie, 4th page, Section 4.1.2, 2nd col, Algorithm 1 Semantic Path Discovery (SPD) to answer an NL query Q from a KG; Oduro-Afriyie, 4th page, Section 4.1.3, 2nd col, As shown in Algorithm 1, the output of our SPD algorithm is an aggregation 𝐴 of entities and relations from the KG. These entities and relations represent the concepts and actions that can be used to answer the main query 𝑄; [i.e., “actions that can be used to answer the main query 𝑄” are generated by Algorithm 1 based on knowledge graph (KG) (ontology) traversing and LLM]).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Oduro-Afriyie in the method of Al-Rikabi because this would enable developing a privacy-preserving KG-based QA system that leverages EHRs/PHLs, drug databases, general medical literature to equip clinicians with a powerful tool to navigate the vast landscape of medical information (Oduro-Afriyie, 2nd page, Section 1.1, 2nd para).
Regarding Claim 16. Al-Rikabi in view of Oduro-Afriyie disclose the method of claim 11, further comprising:
generating a search query based on the set of keywords, the rephrased version of the textual input, and the intent sentence (Al-Rikabi, para 33, ln 44-56, The one or more paraphrased synthetic NLQs may be generated based on global question patterns within the datasets, e.g., dataset 122, a user's topic metadata, and the user's IR. The IR may be generated as described with respect to FIG. 2 herein. In order to obtain a high quality paraphrased synthetic NLQ, the one or more paraphrased synthetic NLQs may be filtered. For example, the one or more paraphrased synthetic NLQs may be filtered with respect to fluency using a language model in natural language to filter the less fluent paraphrased synthetic NLQs. For cycle consistency, the paraphrased synthetic NLQs may be filtered based on which paraphrased synthetic NLQs are executable within the NLQ query service 110); and
providing the search query to a search engine to identify the one or more ontology elements, the intents, and the examples (Al-Rikabi, para 20, ln 23-35, For the graph-based approach, synonyms from one or more public knowledge graph sources may be directly collected. For a model-based approach, a text generation model may be used to curate synonym candidates for a target phrase (which may be performed using a ML model 116 of NLQ query search service 110 or a ML model 120 of data model service 118). For each target phrase, a context sentence may be generated (with the phrase-to-be-replaced identified by some textual marker, e.g., single quotes.) A general pretrained encoder-decoder model may be fine-tuned to specifically replace the identified phrase with an alternate phrase and collect that candidate noted by the same textual marker in the output; Al-Rikabi, para 25, ln 36-45, In configurations, a type of NLQ is a “Why” question, e.g., “why is revenue down in North America Q1 2022. In order to answer such questions, it is necessary to predict which factors are most likely to contribute to this change in metric so that various dimensions, e.g., columns of a dataset, of potential contribution can actually be analyzed to find insights. In particular, to answer “why” questions it is necessary to not only understand the intent of the user's question (what metric to analyze, what time period), but it is also necessary to automatically predict which factors to use; [“NLQ query search service …to identify…” and also “the NLQ… not only understand the intent of the user's question (what metric to analyze, what time period), but it is also necessary to automatically predict which factors to use” as “the search query to a search engine to identify the one or more ontology elements, the intents, and the examples”]).
Regarding Claim 17. Al-Rikabi in view of Oduro-Afriyie disclose the method of claim 11, further comprising:
generating an input context based on the filtered ontology (Al-Rikabi, para 34, ln 14-47, At, 1406, generating one or more synonyms for each name… For the graph-based approach, synonyms from one or more public knowledge graph sources may be directly collected. For a model-based approach, a text generation model may be used to curate synonym candidates for a target phrase (which may be performed using a ML model 116 of NLQ query search service 110 or a ML model 120 of data model service 118). For each target phrase, a context sentence may be generated (with the phrase-to-be-replaced identified by some textual marker, e.g., single quotes.) A general pretrained encoder-decoder model may be fine-tuned to specifically replace the identified phrase with an alternate phrase and collect that candidate noted by the same textual marker in the output. The candidate synonyms may be filtered using one of ML models 116 or 120 to include only those candidate synonyms that are most semantically relevant to the original target phrase); and
providing the input context to the large language model to cause the large language model to generate the list of actions (At, 1406, generating one or more synonyms for each name. For example, generating one or more synonyms for columns of dataset 122 determining a target phrase. The target phrase is used to generate candidate synonyms. For example, a knowledge graph (e.g., using ConceptNet) may be traversed for a target phrase, such as “Patient”, to generate candidate synonyms. A domain specific phrase may be generated to generate a prompt for a large language model to fill in… a text generation model may be used to curate synonym candidates for a target phrase (which may be performed using a ML model 116 of NLQ query search service 110 or a ML model 120 of data model service 118). For each target phrase, a context sentence may be generated (with the phrase-to-be-replaced identified by some textual marker, e.g., single quotes.)).
Regarding Claim 20. A non-transitory computer readable medium including one or more instructions stored thereon and executable by a processor to:
receive, by the processor, via a chatbot interface, a textual input related to an electronic report (Al-Rikabi, col 8, ln 45-49, The NER model input 208 takes the NLQ text and determines spans in the text (identified by offsets in the text) along with possible high-level types. The NER model 208 outputs labels that describe the expected semantic role of an entity mentioned in the NLQ; [i.e., NLQ as a “chatbot interface”; NLQ (Natural Language Query) is a chatbot which is an AI assistant that translates plain, conversational language into structured database queries (like SQL)]);
generate, by the processor, in response to the textual input, using a large language model, an output comprising a set of keywords, a rephrased version of the textual input, and an intent sentence (Al-Rikabi, col 18, ln 60-col 19, ln 3, For the model-based approach, a text generation model may be used to curate synonym candidates for a target phrase (which may be performed using a ML model 116 of NLQ query search service 110 or a ML model 120 of data model service 118). For each target phrase, a context sentence may be generated (with the phrase-to-be-replaced identified by some textual marker, e.g., single quotes.) A general pretrained encoder-decoder model may be fine-tuned to specifically replace the identified phrase with an alternate phrase and collect that candidate noted by the same textual marker in the output. For example, using a model-based text generation model, e.g., one of ML models 116 or 120, an input may be: “paraphrase: in the context of health and healthcare, the health service area is Western New York </s> right carrot.” The output may be “in the context of health and healthcare, the service region is Western New York”... FIG. 6 schematically illustrates an example flow 600 for generating one or more synonyms for columns of dataset 122. At 602, a target phrase is determined. At 604, a domain specific phrase is generated to generate a prompt for a large language model to fill in);
filter, by the processor, using the output from the large language model, an ontology stored in a database to generate a filtered ontology comprising one or more ontology elements, intents, and examples (Al-Rikabi, col 3, ln 30-38, For example, with column synonym generation, a knowledge graph (e.g., using ConceptNet) may be traversed for a target phrase, such as “Patient”, to generate candidate synonyms. A domain specific phrase may be generated to generate a prompt for a large language model to fill in. For example, using “Patient” the phrase generated might be “In the context of healthcare, Patient is <candidate>”. The hydrated phrase may be ranked via an ensemble of similarity models; [i.e., “traverse for a target phrase”” as “filter an ontology”]; Al-Rikabi, col 4, ln 4-32, The techniques and architecture may also provide row level security (RLS) for autocomplete during entry of NLQs and fuzzy matching in natural language question answering; Al-Rikabi, col 5, ln 20-30, In configurations, synthetic question generation may be implemented on previously unseen schema using a three part process: composable, synthetic templates that generate combinations of potential questions using schema and data elements, an ML model that filters and scores the combinations based on how sensible the combinations are (as a perplexity score from the fine-tuned large language model), and finally an answerability filter that re-runs the question through an existing inference pipeline to make sure that in fact it is possible to generate the intended answer from that question text; [i.e., the concept of “output of the LLM being used to filter an ontology” was taught in Al-Rikabi]);
display, by the processor, via the chatbot interface, the list of actions (Al-Rikabi, para col 16, ln 17-22, Fig 1 and FIG. 11 schematically illustrates an example flow: 1100 for CA analysis at topic creation time that may be implemented by an algorithm. At 1102, select potential dimensions, e.g., columns. At 1104, scan visuals that are displayed by BI service 108 or NLQ query service 110 on the client device 114. For example, obtain all visuals, e.g., dashboards, reports; Al-Rikabi, para col 7, ln 58-61, In response to the NLQs 124, the BI service 108 performs searches the one or more datasets 122 for one or more possible answers. This can involve using one or more ML models 116 and/or 120 to analyze the NLQs 124. [i.e., Fig. 1: “NLQ query service 110” which is part of the “Business Intelligence (BI) service 108” as “chatbot interface”; Fig 1, user connected to client device 114 whose output NLQ 124 is input to the BI 102 to be analyzed (“list of actions”) by ML models 116 and/or 120]);
receive, by the processor, via the chatbot interface, an indication to execute an action from the list of actions (Al-Rikabi, para 16, ln 17-22, the author may execute the IR to get the corresponding visual (e.g., dashboard) so that the author may better understand the answer. At 512, the author may accept or reject the some or all of the synthetic NLQs, or in some instances may rephrase some or all of the synthetic NLQs to be more natural); and
provide, by the processor, responsive to the indication, instructions to execute the action associated with the electronic report to modify the electronic report (Al-Rikabi, col 41, ln 54-col 42, ln 8, Such requests can, for instance, be requests to add, delete, change or otherwise modify policy for a customer, service, or system, or for other administrative actions, such as providing an inventory of existing policies and the like… The computer architecture shown in FIG. 19 illustrates a server computer, workstation, desktop computer, laptop, tablet, network appliance, e-reader, smartphone, or other computing device, and can be utilized to execute any of the software components presented herein).
Al-Rikabi does not specifically disclose generate, by the processor, using the large language model and the filtered ontology, a list of actions that are executable to modify the electronic report;
However, Oduro-Afriyie, in the same field of endeavor, discloses generate, by the processor, using the large language model and the filtered ontology, a list of actions that are executable to modify the electronic report (Oduro-Afriyie, 1st – 2nd page, Section 1.1, 1st - 2nd para, In this paper, we introduce a knowledge graph (KG)-based question answering (QA) system, Medicient, designed to tackle these challenges by incorporating data from patient EHRs/PHLs, drug databases, and medical literature/research to provide personalized treatment recommendations for patients while safeguarding HIPPA mandated privacy… The proposed system leverages the power of large language models for question understanding and natural language (NL) response generation while concealing the EHR or PHL from the language model to preserve privacy; Oduro-Afriyie, 2nd page, Section 2.4, 1st para, Our proposed system aims to bridge this gap by incorporating the strengths of knowledge graphs and large language models while ensuring the protection of sensitive patient information; Oduro-Afriyie, 4th page, Section 4.1.2, 2nd col, Algorithm 1 Semantic Path Discovery (SPD) to answer an NL query Q from a KG; Oduro-Afriyie, 4th page, Section 4.1.3, 2nd col, As shown in Algorithm 1, the output of our SPD algorithm is an aggregation 𝐴 of entities and relations from the KG. These entities and relations represent the concepts and actions that can be used to answer the main query 𝑄; [i.e., “actions that can be used to answer the main query 𝑄” are generated by Algorithm 1 based on knowledge graph (KG) (ontology) traversing and LLM]).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Oduro-Afriyie in the method of Al-Rikabi because this would enable developing a privacy-preserving KG-based QA system that leverages EHRs/PHLs, drug databases, general medical literature to equip clinicians with a powerful tool to navigate the vast landscape of medical information (Oduro-Afriyie, 2nd page, Section 1.1, 2nd para).
Claims 2 and 12 are rejected under 35 U.S.C. 103 as being unpatentable over Al-Rikabi in view of in view of Oduro-Afriyie, and further in view of Bandhu et al. Pat App No. US 20210173865 A1 (Bandhu).
Regarding Claim 2. Al-Rikabi in view of Oduro-Afriyie disclose the system of claim 1.
Al-Rikabi in view of Oduro-Afriyie do not specifically disclose wherein the chatbot interface is decoupled from a reporting interface displaying the electronic report
However, Bandhu, in the same field of endeavor, discloses wherein the chatbot interface is decoupled from a reporting interface displaying the electronic report (Bandhu. para 0072-0079, The voice assistant device 302, 342 opens the content item locally (e.g., displays the document, plays a media content item). [0072] “open the [content item] in a new tab”: similar to “open the content item”, but causes voice assistant tool device 302, 342 to display the opened document in a new tab in the user interface… Voice assistant tool service 310, 350 sends back a response indicating the CMS is disconnected, which the voice assistant tool device 302, 342 outputs to the user.; [i.e., “voice assistant” as a “chatbot”; “displays the document” as “displaying the electronic report”; “Voice assistant tool service 310, 350 sends back a response indicating the CMS is disconnected” as “the chatbot interface is decoupled”]).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Bandhu in the method of Al-Rikabi in view of Oduro-Afriyie because this would enable the voice assistant tool device to effectively disconnect the end-user application being used to interact with the CMS (Bandhu, para 0079).
Regarding Claim 12. Al-Rikabi in view of Oduro-Afriyie disclose the method of claim 11.
Al-Rikabi in view of Oduro-Afriyie do not specifically disclose wherein the chatbot interface is decoupled from a reporting interface displaying the electronic report.
However, Bandhu, in the same field of endeavor, discloses wherein the chatbot interface is decoupled from a reporting interface displaying the electronic report (Bandhu. para 0072-0079, The voice assistant device 302, 342 opens the content item locally (e.g., displays the document, plays a media content item). [0072] “open the [content item] in a new tab”: similar to “open the content item”, but causes voice assistant tool device 302, 342 to display the opened document in a new tab in the user interface… Voice assistant tool service 310, 350 sends back a response indicating the CMS is disconnected, which the voice assistant tool device 302, 342 outputs to the user. The voice assistant tool device 302, 342 may also close the end-user application being used to interact with CMS 330; [i.e., “voice assistant” as a “chatbot”; “displays the document” as “displaying the electronic report”; “Voice assistant tool service 310, 350 sends back a response indicating the CMS is disconnected” as “the chatbot interface is decoupled”]).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Bandhu in the method of Al-Rikabi in view of Oduro-Afriyie because this would enable the voice assistant tool device to effectively disconnect the end-user application being used to interact with the CMS (Bandhu, para 0079).
Claims 3-4 and 13-14 are rejected under 35 U.S.C. 103 as being unpatentable over Al-Rikabi in view of in view of Oduro-Afriyie, and further in view of DeRosa-Grund et al. Pat App No. US 20220270725 A1 (DeRosa-Grund ).
Regarding Claim 3. Al-Rikabi in view of Oduro-Afriyie discloses the system of claim 1.
Al-Rikabi in view of Oduro-Afriyie do not specifically disclose wherein in response to receiving the textual input, the one or more processors are further configured to determine a user identifier and a state of the electronic report.
However, DeRosa-Grund , in the same field of endeavor, discloses wherein in response to receiving the textual input, the one or more processors are further configured to determine a user identifier and a state of the electronic report (DeRosa-Grund, para 0130-0132, Accordingly, at the testing facility, once the user has logged onto and/or validated their identity with the MS module, the module enables the user to access and/or display the unique identifier for the test on their device and/or transmit the identifier to facility (e.g. via email, text or other media). In some embodiments, the user validates their identity on the MS module on their device using their biometric data (e.g. voice, fingerprint, face, etc.) via the biometric lock module described above. For example, the module is able to access a stored image of their driver's license (or valid school and/or parental ID) and compare it to a live captured facial image to verify the user's identity as the proper test taker. Alternatively or in addition, the user is able to validate their identity using their username and password and/or other identifying data submitted during the registration process… If accepted, after receiving the results (e.g. as captured from an image of the testing material), the MS module is able to transmit the test results/medical status along with the identifier of the user (e.g. alphanumeric identifier assigned during registration) to the platform 200 which records the results/medical status on the MIDC 202 of the associated user/user account as a transaction on PPI block or PPI update block in the same manner as described above. ).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of DeRosa-Grund in the method of Al-Rikabi in view of Oduro-Afriyie because this would enable the user to validate their identity using their username and password and/or other identifying data submitted during the registration process. (DeRosa-Grund, para 0130).
Regarding Claim 4. Al-Rikabi in view of Oduro-Afriyie and DeRosa-Grund disclose the system of claim 3.
Furthermore, DeRosa-Grund teaches:
wherein the state of the electronic report corresponds to an initial state of the electronic report prior to any modifications being made to the electronic report (para 0074, during login and registration establishing initial PPI and/or any time a user wishes to add new PPI, a copy of data needed for account recovery is added to their MIDC 202 using a key deterministically derived from an enrolled recovery method ).
Regarding Claim 13. Al-Rikabi in view of Oduro-Afriyie disclose the method of claim 11.
Al-Rikabi in view of Oduro-Afriyie do not specifically disclose wherein in response to receiving the textual input, further comprising determining a user identifier and a state of the electronic report
However, DeRosa-Grund, in the same field of endeavor, discloses wherein in response to receiving the textual input, further comprising determining a user identifier and a state of the electronic report (DeRosa-Grund, para 0130-0132, Accordingly, at the testing facility, once the user has logged onto and/or validated their identity with the MS module, the module enables the user to access and/or display the unique identifier for the test on their device and/or transmit the identifier to facility (e.g. via email, text or other media). In some embodiments, the user validates their identity on the MS module on their device using their biometric data (e.g. voice, fingerprint, face, etc.) via the biometric lock module described above. For example, the module is able to access a stored image of their driver's license (or valid school and/or parental ID) and compare it to a live captured facial image to verify the user's identity as the proper test taker. Alternatively or in addition, the user is able to validate their identity using their username and password and/or other identifying data submitted during the registration process… If accepted, after receiving the results (e.g. as captured from an image of the testing material), the MS module is able to transmit the test results/medical status along with the identifier of the user (e.g. alphanumeric identifier assigned during registration) to the platform 200 which records the results/medical status on the MIDC 202 of the associated user/user account as a transaction on PPI block or PPI update block in the same manner as described above. ).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of DeRosa-Grund in the method of Al-Rikabi in view of Oduro-Afriyie because this would enable the user to validate their identity using their username and password and/or other identifying data submitted during the registration process. (DeRosa-Grund, para 0130 ).
Regarding Claim 14. Al-Rikabi in view of Oduro-Afriyie and DeRosa-Grund disclose the method of claim 13.
Furthermore, DeRosa-Grund teaches:
wherein the state of the electronic report corresponds to an initial state of the electronic report prior to any modifications being made to the electronic report (DeRosa-Grund, para 0074, during login and registration establishing initial PPI and/or any time a user wishes to add new PPI, a copy of data needed for account recovery is added to their MIDC 202 using a key deterministically derived from an enrolled recovery method ; [i.e., “PPI” (Private Personal Information) as “electronic document/report”]).
Claims 5 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Al-Rikabi in view of in view of Oduro-Afriyie, further in view of DeRosa-Grund et al. Pat App No. US 20220270725 A1 (DeRosa-Grund ), and further in view of Chang et al. pat App No. US 20240370736 A1 (Chang).
Regarding Claim 5. Al-Rikabi in view of Oduro-Afriyie and DeRosa-Grund disclose the system of claim 3.
Al-Rikabi in view of Oduro-Afriyie and DeRosa-Grund do not specifically disclose wherein the one or more processors are further configured to provide the textual input, the user identifier, and the state of the electronic report to the large language model to cause the large language model to generate the output.
However, Chang, in the same field of endeavor, discloses wherein the one or more processors are further configured to provide the textual input, the user identifier, and the state of the electronic report to the large language model to cause the large language model to generate the output (Chang, para 0003, The LLM may generate better or worse results depending how a prompt is formulated; Chang, para 0063, all of a multi modal user input, a zero-shot learning response to a multi modal prompt, and/or other information encoded in a webpage (e.g. a collection of resources to be rendered by the browser and associated plug-ins, including execution of scripts, such as JavaScript.…the user may enter or select data, such as clickable or touchable display regions or display regions for text input… For example, context information comprising one or more images may be uploaded, in combination with one or more entered text commands. Such inputs may prompt… JavaScript™ object notation (JSON) or extensible markup language (XML) … . An identifier of the type of user device, either mobile or non-mobile, for example, may be encoded in the request for the webpage by the web browser (e.g., as a user agent type in an HTTP header associated with a GET request) ).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Chang in the method of Al-Rikabi in view of Oduro-Afriyie because this would enable an LLM to generate zero-shot predictions when a prompt is well-formulated and the LLM that has formulated prompts can generate better results (Chang, para 0003).
Regarding Claim 15. Al-Rikabi in view of Oduro-Afriyie and DeRosa-Grund disclose the method of claim 13.
Al-Rikabi in view of Oduro-Afriyie and DeRosa-Grund do not specifically disclose wherein the one or more processors are further configured to provide the textual input, the user identifier, and the state of the electronic report to the large language model to cause the large language model to generate the output.
However, Chang, in the same field of endeavor, discloses providing the textual input, the user identifier, and the state of the electronic report to the large language model to cause the large language model to generate the output (Chang, para 0003, The LLM may generate better or worse results depending how a prompt is formulated; Chang, para 0063, all of a multi modal user input, a zero-shot learning response to a multi modal prompt, and/or other information encoded in a webpage (e.g. a collection of resources to be rendered by the browser and associated plug-ins, including execution of scripts, such as JavaScript.…the user may enter or select data, such as clickable or touchable display regions or display regions for text input… For example, context information comprising one or more images may be uploaded, in combination with one or more entered text commands. Such inputs may prompt… JavaScript™ object notation (JSON) or extensible markup language (XML) … . An identifier of the type of user device, either mobile or non-mobile, for example, may be encoded in the request for the webpage by the web browser (e.g., as a user agent type in an HTTP header associated with a GET request) ).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Chang in the method of Al-Rikabi in view of Oduro-Afriyie because this would enable an LLM to generate zero-shot predictions when a prompt is well-formulated and the LLM that has formulated prompts can generate better results (Chang, para 0003).
Claims 8 and 18are rejected under 35 U.S.C. 103 as being unpatentable over Al-Rikabi in view of in view of Oduro-Afriyie, and further in view of Grigoriev et al. Pat App No. CN 102972003 A (Grigoriev).
Regarding Claim 8. Al-Rikabi in view of Oduro-Afriyie discloses the system of claim 1.
Al-Rikabi in view of Oduro-Afriyie do not specifically disclose wherein the ontology comprises a resource description framework including a plurality of nodes, wherein each node of the plurality of nodes comprises a plurality of attributes.
However, Grigoriev, in the same field of endeavor, discloses wherein the ontology comprises a resource description framework including a plurality of nodes, wherein each node of the plurality of nodes comprises a plurality of attributes (Grigoriev, para 0045-0060, It is to be noted, be not all resources all be network " searchable ", for example, the bound book in personnel, equipment, facility, fund, the library etc. For example, abstract concept can be resource, for example the type of the operator in the mathematical equation and operand, relation (for example, " father and mother " or " employee ") or numerical value (for example, 0,1 and infinite).Be assumed to be concept sign is provided, express concept by information presentation format (for example, resource description framework (RDF) tlv triple) or structure (for example, RDF figure), perhaps express its sign by given box-like Uniform Resource Identifier (URI), then concept also can be resource… As used herein, term " tlv triple " refers to adopt subject-predicate of RDF-object to express. Subject represents resource and is RDF Uniform Resource Identifier (URI) reference or blank node, predicate is RDF URI reference, characteristic or the each side of its expression resource and express subject and object between relation, object is RDF URI with reference to, literal or blank node… Fig. 1 is the schematic diagram according to the general passive authorization invocation sequence of an embodiment. The method of passive mandate is generally used for service and the application such as message transmission, social networks. Initiator's request access resource is such as photograph album, have attribute, message threads etc... in the intelligent space, the owner can use one or more nodes (e.g., movement telephone, computer or similar terminal) performs a task (such as accessing caller RDF graph)).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Grigoriev in the method of Al-Rikabi in view of Oduro-Afriyie because this would enable providing authorization relating to access to the semantic network resource (Grigoriev, Abstract and para 0002).
Regarding Claim 18. Al-Rikabi in view of Oduro-Afriyie disclose the method of claim 11, wherein the ontology comprises a resource description framework including a plurality of nodes.
Al-Rikabi in view of Oduro-Afriyie do not specifically disclose wherein each node of the plurality of nodes comprises a plurality of attributes.
However, Grigoriev, in the same field of endeavor, discloses wherein each node of the plurality of nodes comprises a plurality of attributes (Grigoriev, para 0045-0060, It is to be noted, be not all resources all be network " searchable ", for example, the bound book in personnel, equipment, facility, fund, the library etc.For example, abstract concept can be resource, for example the type of the operator in the mathematical equation and operand, relation (for example, " father and mother " or " employee ") or numerical value (for example, 0,1 and infinite).Be assumed to be concept sign is provided, express concept by information presentation format (for example, resource description framework (RDF) tlv triple) or structure (for example, RDF figure), perhaps express its sign by given box-like Uniform Resource Identifier (URI), then concept also can be resource… As used herein, term " tlv triple " refers to adopt subject-predicate of RDF-object to express. Subject represents resource and is RDF Uniform Resource Identifier (URI) reference or blank node, predicate is RDF URI reference, characteristic or the each side of its expression resource and express subject and object between relation, object is RDF URI with reference to, literal or blank node… Fig. 1 is the schematic diagram according to the general passive authorization invocation sequence of an embodiment. The method of passive mandate is generally used for service and the application such as message transmission, social networks. Initiator's request access resource is such as photograph album, have attribute, message threads etc. ... in the intelligent space, the owner can use one or more nodes (e.g., movement telephone, computer or similar terminal) performs a task (such as accessing caller RDF graph)).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Grigoriev in the method of Al-Rikabi in view of Oduro-Afriyie because this would enable providing authorization relating to access to the semantic network resource (Grigoriev, Abstract and para 0002).
Claims 9 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Al-Rikabi in view of in view of Oduro-Afriyie, further in view of Grigoriev et al. Pat App No. CN 102972003 A (Grigoriev), and further in view of Hunt Pat No. US 12008652 B1 (Hunt).
Regarding Claim 9. Al-Rikabi in view of Oduro-Afriyie and Grigoriev disclose the system of claim 8.
Al-Rikabi in view of Oduro-Afriyie and Grigoriev do not specifically disclose wherein the node is associated with a probability score based on a relative frequency of one or more attribute combinations in historical data maintained in the database.
However, Hunt, in the same field of endeavor, discloses wherein the node is associated with a probability score based on a relative frequency of one or more attribute combinations in historical data maintained in the database (Hunt, col 1, ln 39- col 2, ln 38, graphical probability map representing a graphical performance prediction relative to a point in time of the instrument data based on the historical value statistics… determining a list of dates indicating locations where the dimension occurs within the instrument data; determining the historical value statistics for the set of time intervals relative to the occurrences of the dimension includes generating a collection of lists based on dates in the list of dates… determining the historical value statistics for the set of time intervals relative to the occurrences of the dimension includes determining a relative frequency of the occurrences of the dimension, sorting the occurrences by frequency, grouping the occurrences having corresponding frequencies into groupings of zero or more instances; Hunt, col 6, ln 29- 31, The storage device 110 can store financial instrument data which may include historical and/or real-time data for one or more financial instruments…; Hunt, col 8, ln 10-col 9, ln 21,The memory 154 is also capable of storing other instructions and data, including, for example, hardware drivers, operating systems, other software applications, databases, etc… the data store 160 may be incorporated with the memory 154 or may be distinct therefrom. In some implementations, the data store 160 may store data in association with a database management system (DBMS) operable on the computing system 150 ).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Grigoriev in the method of Al-Rikabi in view of Oduro-Afriyie because this would enable determining historical value statistics for a set of time intervals relative to the occurrences of the dimension and automatically generating one or more regions for a computer display indicating a graphical probability map representing a graphical performance prediction (Hunt, Abstract).
Regarding Claim 19. Al-Rikabi in view of Oduro-Afriyie and Grigoriev disclose the method of claim 18.
Al-Rikabi in view of Oduro-Afriyie and Grigoriev do not specifically disclose wherein the node is associated with a probability score based on a relative frequency of one or more attribute combinations in historical data maintained in the database.
However, Hunt, in the same field of endeavor, discloses wherein the node is associated with a probability score based on a relative frequency of one or more attribute combinations in historical data maintained in the database (Hunt, col 1, ln 39- col 2, ln 38, graphical probability map representing a graphical performance prediction relative to a point in time of the instrument data based on the historical value statistics… determining a list of dates indicating locations where the dimension occurs within the instrument data; determining the historical value statistics for the set of time intervals relative to the occurrences of the dimension includes generating a collection of lists based on dates in the list of dates… determining the historical value statistics for the set of time intervals relative to the occurrences of the dimension includes determining a relative frequency of the occurrences of the dimension, sorting the occurrences by frequency, grouping the occurrences having corresponding frequencies into groupings of zero or more instances; Hunt, col 6, ln 29- 31, The storage device 110 can store financial instrument data which may include historical and/or real-time data for one or more financial instruments…; Hunt, col 8, ln 10-col 9, ln 21,The memory 154 is also capable of storing other instructions and data, including, for example, hardware drivers, operating systems, other software applications, databases, etc… the data store 160 may be incorporated with the memory 154 or may be distinct therefrom. In some implementations, the data store 160 may store data in association with a database management system (DBMS) operable on the computing system 150).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Grigoriev in the method of Al-Rikabi in view of Oduro-Afriyie because this would enable determining historical value statistics for a set of time intervals relative to the occurrences of the dimension and automatically generating one or more regions for a computer display indicating a graphical probability map representing a graphical performance prediction (Hunt, Abstract).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Al-Rikabi in view of in view of Oduro-Afriyie, and further in view of Zhang et al. Pat App No. CN 115732051 A (Zhang).
Regarding Claim 10. Al-Rikabi in view of Oduro-Afriyie disclose the system of claim 1.
Al-Rikabi in view of Oduro-Afriyie do not specifically disclose wherein each action from the list of actions corresponds to a respective field of the electronic report.
However, Zhang, in the same field of endeavor, discloses wherein each action from the list of actions corresponds to a respective field of the electronic report (Zhang, page 13, para 11- page 14, para 1, For the field of the single selection and multiple selection, it can display option configuration item. As shown in FIG. 5, for the field of the single selection and multiple selection, the corresponding option configuration item comprises (value domain 1, value domain 2, value domain 3). A user may perform a single selection or a plurality of selections in the above option configuration item. In some embodiments of the generation method of the electronic case report shown in FIG. 2 and FIG. 3, the target form data acquisition mode is a structured extraction, and the method further comprises: establishing the association relationship between the established electronic case report and the preset database).
Therefore, it would have been obvious for one having ordinary skill in the art before the effective filing date of the claimed invention to incorporate the method of Zhang in the method of Al-Rikabi in view of Oduro-Afriyie because this would enable improving the efficiency of creating diversified electronic case reports (Zhang, Abstract).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MULUGETA T. DUGDA whose telephone number is (703)756-1106. The examiner can normally be reached Mon - Fri, 4:30am - 7:00pm.
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, Paras D. Shah can be reached at 571-270-1650. 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.
/MULUGETA TUJI DUGDA/Examiner, Art Unit 2653
/Paras D Shah/Supervisory Patent Examiner, Art Unit 2653
06/13/2026
Prosecution Insights