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 .
DETAILED ACTION
This action is issued in response to Application filed April 24, 2025.
Claims 1-20 are pending.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 2-4 and 6 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 2: the limitation 4 states “transforms each at least one section into a vector embedding” and limitation 7 states “one query is converted into at least one vector embedding”. It is unclear whether these vector embedding are the same or different. It is further unclear because limitation 8 states “the at least one vector embedding” and limitation 9 states “process uses the vector embedding”; wherein the examiner is unsure if the mentioned vector embedding for limitations 8 and 9 are referencing the vector embedding transformed from limitation 4 or the vector embedding converted from limitation 7. Due to the lack of clarity, the examiner will examiner the claims with the broadest reasonable interpretation. Corrections are required.
Claim 3: limitation 5, states “a storage hyponym process comprising a hyponym field… storing at least one hypernym”; and limitation 9 states “input into the storage hyponym process, which returns the at least one hypernym for the hyponym field”. The limitations are interchanging hyponym and hypernym improperly. This is believed to be a mistake, otherwise it is unclear why a storage hyponym process with a hyponym field is storing a hypernym and how that hypernym is being return for a hyponym field. Corrections are required.
Claim 4: limitation 8 states “the at least one vector embedding is sent to the at least one storage process”; however, it is unclear which vector embedding (i.e., first vector embedding, second vector embedding, or both) is sent to the storage process. Due to the lack of clarity, the examiner will examiner the claims with the broadest reasonable interpretation. Corrections are required.
Claim 6: contains the same lack of clarity issues as claim 4 above. Corrections are required.
Claim 6: recites the feature "the hyponym field" in limitation 6. There is no prior mention of a hyponym field within claim 6 nor within the relied upon Independent claim 5. There is insufficient antecedent basis for this limitation in the claim.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 3 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Colley (U.S. Patent Application No. 2022/0181029).
Regarding Claim 3, Colley discloses a system for storing and retrieving information from an electronic knowledge base, the system comprising:
a computer and an associated memory (par [0043], [0287], Colley – while a mobile device or tablet is referenced throughout, it is understood that the device may also include any device such as a personal computer, etc.,… it is understood that a personal computer has associated memory and par [0200] makes mention of a memory);
at least one electronic document (par [0045], [0050-0051], Colley – generating a report is performed by capturing or uploading a report and validating any fields… electronically capturing a document, wherein the report corresponds to the electronic document);
at least one process for splitting the at least one electronic document into at least one section (par [0064], [0069], Colley – a report includes a first section and a second section, wherein the different sections identify separate features);
at least one storage process (par [0066], Colley – a document such as a genetic testing report may extract some or all medical information and then the report and extracted information may be stored in a database or other data repository);
a storage hyponym process comprising a hyponym field for determining and storing at least one hypernym of at least one term in the at least one section (par [0191], [0197], Colley – there’s a database for viewing links; wherein links between two concepts may represent specific known relationships between those two concepts. For example, “Tylenol” may be linked to “acetaminophen” by a “trade name” marker, and may be linked to “Tylenol 50 mg” by a “dosage of” marker. There may also be markers to identify taxonomic “is a” relationships between concepts. “Is a” markers provide relationships between over some clinical dictionaries (such as SNOMEDCT_US, Campbell W S, Pederson J, etc.) to establish relationships between each database with the others. For example, we can follow “is a” relationships from “Tylenol”, “Tylenol 50 mg”, or “acetaminophen” to the concept for a generic drug. Such a relationship may not be available for another concept, for example, a match to the dictionary for UMLS to “the patient” or “patient” may not have a relationship to a medication dictionary due to the conceptually distinct natures of each entity. Relationships may be found between drugs that have the same ingredients or are used to treat the same illnesses);
a retrieval hyponym process (par [0191], [0197], [0203], Colley);
at least one query (par [0046], Colley – interface receives text input… par [0099], Colley – system is triggered in response to specific queries); and
a query filter construction process (par [0046], Colley - search indicator that, upon selection by the user, receives text input such as a patient's name, unique identifier, or diagnosis, that permits the user to filter the patients by the search criteria of the text input to search for a specific patient);
wherein the at least one section is input into the storage hyponym process, which returns the at least one hypernym for the hyponym field (par [0191], [0197], [0203], Colley - there’s a database for viewing links; wherein links between two concepts may represent specific known relationships between those two concepts. For example, “Tylenol” may be linked to “acetaminophen” by a “trade name” marker, and may be linked to “Tylenol 50 mg” by a “dosage of” marker. There may also be markers to identify taxonomic “is a” relationships between concepts);
wherein the at least one query is input into the retrieval hyponym process, which returns at least one term from the at least one query in the hyponym field (par [0203], Colley – query is input and may return CUIs related to concepts which are lined as “is a” descendants relationship);
wherein the query filter construction process constructs a query filter that comprises prefiltering on the at least one query in the hyponym field and its associated value (par [0191], [0197], [0203], Colley);
wherein the query filter is sent to the at least one storage process (par [0207-0208], [0231-0232], Colley – narrowing the search to specify that “Tylenol” is a brand name of the generic brand “acetaminophen” and processed through the database associated with the Abstraction Engine toolbox); and
wherein a response is received from the at least one storage process (par [0114], Colley – generate and serve a response to the system based on receiving a request… par [0234], Colley – response is received from toolbox service… also see par [0099], [0219]).
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.
Claim(s) 1 is/are rejected under 35 U.S.C. 103 as being unpatentable over Colley (U.S. Patent Application No. 2022/0181029) in view of Paterson (U.S. Patent Application No. 2018/0067932).
Regarding Claim 1, Colley discloses a system for storing and retrieving information from an electronic knowledge base, the system comprising:
a computer and an associated memory (par [0043], [0287], Colley – while a mobile device or tablet is referenced throughout, it is understood that the device may also include any device such as a personal computer, etc.,… it is understood that a personal computer has associated memory and par [0200] makes mention of a memory);
at least one electronic document (par [0045], [0050-0051], Colley – generating a report is performed by capturing or uploading a report and validating any fields… electronically capturing a document, wherein the report corresponds to the electronic document);
at least one process for splitting the at least one electronic document into at least one section (par [0064], [0069], Colley – a report includes a first section and a second section, wherein the different sections identify separate features);
at least one storage process (par [0066], Colley – a document such as a genetic testing report may extract some or all medical information and then the report and extracted information may be stored in a database or other data repository);
a storage entity count process for determining a total number of unique references to at least one entity type in each at least one section (par [0175], Colley - The number of candidate concepts which may be extracted may be needlessly large. For example, in patient file with thousands of documents, a concept candidate for breast cancer may occur hundreds of times, a concept for lung cancer may occur tens of times, and a concept for liver cancer may only occur once. It may be useful to filter/rank the mentions of each concept candidate to reduce repetition in the following stages in the pipeline. For concept candidates which may be consolidated (such as mentions of breast cancer for diagnosis) the concept candidate may be reduced to a single concept with a count field in the hundreds. Furthermore, if concept candidates are competing for the same field, the concept candidate may be coupled with a reliability index based upon the frequency of the concept candidate occurring in relationship to the others (such as 200 mentions of breast cancer, 13 mentions of lung cancer, and 1 mention of liver cancer may be processed to a 200/214 reliability index that the patient has breast cancer). The highest ranked competing concept candidate may be preserved along with a reliability index, or a consolidated report of the most frequent competing concept candidates may be preserved along with their count values and/or reliability index);
a retrieval entity count process (par [0175], Colley);
at least one query (par [0046], Colley – interface receives text input… par [0099], Colley – system is triggered in response to specific queries); and
a query filter construction process (par [0046], Colley - search indicator that, upon selection by the user, receives text input such as a patient's name, unique identifier, or diagnosis, that permits the user to filter the patients by the search criteria of the text input to search for a specific patient);
wherein the at least one process for splitting the document creates at least one section (par [0064], [0069], Colley – a report includes a first section and a second section, wherein the different sections identify separate features);
wherein the query filter is sent to the at least one storage process (par [0207-0208], [0231-0232], Colley – narrowing the search to specify that “Tylenol” is a brand name of the generic brand “acetaminophen” and processed through the database associated with the Abstraction Engine toolbox); and
wherein a response is received from the at least one storage process (par [0114], Colley – generate and serve a response to the system based on receiving a request… par [0234], Colley – response is received from toolbox service… also see par [0099], [0219]).
While Colley discloses all of the claimed subject matter as stated above. However, Colley is not as detailed with respect to where the at least one section is input into the storage entity count process, which returns at least one entity type field along with a count value for the at least one entity type field; wherein the entity type field and its count value are sent to the storage process for storage; wherein the at least one query is input into the retrieval entity count process, which returns at least one query entity type field along with a count value for the at least one query entity type field; wherein the query filter construction process constructs a query filter that comprises filtering on the at least one query entity type field and its associated count value.
On the other hand, Paterson discloses returns at least one entity type field along with a count value for the at least one entity type field (par [0117], [0119], [0136], Paterson – retrieving a plurality of document and entity/category types based on the query, along with a count… a count can also be stored and retrieved… also see par [0112], Paterson);
wherein the entity type field and its count value are sent to the storage process for storage (par [0064], [0086], [0136], Paterson - provide remote or cloud based virtual filing cabinets that allow individuals to store documents in association with different categories and user entities. The embodiments described herein may facilitate viewing and filtering digital documents by document associations with business or personal entity identifiers, entity types, jurisdictions, as well as displaying a real-time count of the documents contained in each folder and sub-folder… a count can also be stored in the database for the particular document entity);
wherein the at least one query is input into the retrieval entity count process, which returns at least one query entity type field along with a count value for the at least one query entity type field (par [0136], [0140], Paterson - An entity filter can be used by a user to select those entity identifiers for which documents are to be displayed. A user may interact with an entity filter, for example using user interface. The entity filter allows the user to select one or more entity identifiers that can be used to query the non-relational database to identifier entity associations, category associations, active icon statuses, and document counts);
wherein the query filter construction process constructs a query filter that comprises filtering on the at least one query entity type field and its associated count value (par [0136], [0140], Paterson).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Paterson’s teachings of real-time document filtering into the Colley system. A skilled artisan would have been motivated to combine in order to ensure the requested data capable of being rapidly updated and provide efficient management of the system.
Claim(s) 2 is/are rejected under 35 U.S.C. 103 as being unpatentable over Colley in view of Paterson, further in view of Tunstall (U.S. Patent Application No. 2023/0259705).
Regarding Claim 2, the combination of Colley in view of Paterson, disclose the system of claim 1, further comprising:
a vector generation process (par [0168], Colley - at each level of granularity, the whole document classifier may be interrupted, for example, if a sufficient level of certainty has been reached or processing was intended to terminate at that level. For example, if a document has been determined to have a high incidence of accuracy because a table on page 3 of a document may always return the correct gender for the patient, then the algorithm may identify that high accuracy has been provided for the document based on the one sentence of that document and stop processing a gender classification at the sentence level vector for that patient. Furthermore, a patient level vector may not be generated if a document level vector has reached a certain threshold of certainty (such as 95%), or if, for example, only one document is being processed);
a vector database that supports metadata filtering, wherein the at least one vector database comprises at least one of the at least one storage process (par [0167], Colley - The rule sets may include a vector of, for example, three hundred words and their respective weights, and each rule set may be applied over all words in a sentence to generate weights for every sentence. For example, a sentence “The patient was given prostate exam after he complained about having difficulty urinating in the mornings” may be given a high weight for gender as male because of words “prostate exam” and “he”. After each word of each sentence is processed, each respective sentence may be assigned a sentence vector (such as 10% female, 90% male), then each sentence in a document may be processed to assign a document vector, and finally, each document in a patient's EMR or EHR may be processed to assign a patient vector); and
a point ID generation process (par [0180], Colley - Fuzzy matching is structured around the text concepts included in the above enumerated list or the UMLS, including metadata fields CUI (the UMLS unique ID) and AUI (dictionary-specific unique ID), so that an exhaustive search may be performed for all medical concepts. The dictionary search engine may also return metadata about the specific entry detected (such as universal ID assigned in the above enumerated list or the UMLS));
wherein the point ID generation process generates a unique ID (par [0180], Colley - Fuzzy matching is structured around the text concepts included in the above enumerated list or the UMLS, including metadata fields CUI (the UMLS unique ID) and AUI (dictionary-specific unique ID), so that an exhaustive search may be performed for all medical concepts. The dictionary search engine may also return metadata about the specific entry detected (such as universal ID assigned in the above enumerated list or the UMLS));
wherein the unique ID, and the entity type field and its count value are sent to the storage process for storage (par [0180], Colley; par [0064], [0086], [0136], Paterson - provide remote or cloud based virtual filing cabinets that allow individuals to store documents in association with different categories and user entities. The embodiments described herein may facilitate viewing and filtering digital documents by document associations with business or personal entity identifiers, entity types, jurisdictions, as well as displaying a real-time count of the documents contained in each folder and sub-folder… a count can also be stored in the database for the particular document entity).
While Colley and Paterson disclose all of the claimed subject matter as stated above However, the references are not as detailed with respect to wherein the vector generation process transforms each at least one section into a vector embedding; wherein the at least one query is converted into at least one vector embedding; wherein the at least one vector embedding is sent to the at least one storage process; and wherein the at least one storage process uses the vector embedding in the construction of the response.
On the other hand, Tunstall discloses wherein the vector generation process transforms each at least one section into a vector embedding (par [0446], [0475], Tunstall - an encoder which turns a source sentence into an internal vector or sequence of vectors that encodes the source sentence); wherein the at least one query is converted into at least one vector embedding (par [0990, Tunstall); wherein the at least one vector embedding is sent to the at least one storage process (par [0446], [0475], Tunstall); and wherein the at least one storage process uses the vector embedding in the construction of the response (par [1553-1562], Tunstall).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Tunstall’s teachings into the Colley and Paterson system. A skilled artisan would have been motivated to combine in order to improve output within a language model environment.
Claim(s) 4-9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Colley (U.S. Patent Application No. 2022/0181029) in view of Tunstall (U.S. Patent Application No. 2023/0259705).
Regarding Claim 5, Colley discloses a system for storing and retrieving information from an electronic knowledge base, the system comprising:
a computer and an associated memory (par [0043], [0287], Colley – while a mobile device or tablet is referenced throughout, it is understood that the device may also include any device such as a personal computer, etc.,… it is understood that a personal computer has associated memory and par [0200] makes mention of a memory);
at least one electronic document (par [0045], [0050-0051], Colley – generating a report is performed by capturing or uploading a report and validating any fields… electronically capturing a document, wherein the report corresponds to the electronic document);
at least one process for splitting the at least one electronic document into at least one section (par [0064], [0069], Colley – a report includes a first section and a second section, wherein the different sections identify separate features);
at least one storage process (par [0066], Colley – a document such as a genetic testing report may extract some or all medical information and then the report and extracted information may be stored in a database or other data repository);
at least one retrieval process wherein the retrieval process selects at least one section from the at least one storage process based on at least one query (par [0196-0197], Colley – a query is generated to search a database and selection criteria is determined to identify what is needed based on the query).
While Colley discloses all of the claimed subject matter as stated above. However, Colley is not as detailed with respect to a relevant facts extraction process, which uses an LLM to extract extracted facts from at least one of the at least one section that is provided to the relevant facts extraction process, wherein the extracted facts are relevant to the at least one query; wherein the at least one query is input into the at least one retrieval process to produce at least one retrieval output; and wherein at least one of the at least one retrieval output from the at least one retrieval process is provided as input to the relevant facts extraction process to produce at least one relevant facts output.
On the other hand, Tunstall discloses a relevant facts extraction process, which uses an LLM to extract extracted facts from at least one of the at least one section that is provided to the relevant facts extraction process, wherein the extracted facts are relevant to the at least one query (par [0105-0112], Tunstall - receiving a text input and providing it to the LLM, wherein an output is produced… par [1079]); wherein the at least one query is input into the at least one retrieval process to produce at least one retrieval output (par [0104-0108], Tunstall – receiving a text input and providing it to the LLM, wherein an output is produced); and wherein at least one of the at least one retrieval output from the at least one retrieval process is provided as input to the relevant facts extraction process to produce at least one relevant facts output (par [0107-0108], Tunstall - to generate output based on the input to the LLM includes extracting the assertions in text generated by the LLM… par [0112], Tunstall – fact checking the output from the LLM).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Tunstall’s teachings into the Colley system. A skilled artisan would have been motivated to combine in order to improve output within a language model environment.
Regarding Claim 6, the combination of Colley in view of Tunstall, disclose the system of claim 5, further comprising:
a vector generation process (par [0168], Colley - at each level of granularity, the whole document classifier may be interrupted, for example, if a sufficient level of certainty has been reached or processing was intended to terminate at that level. For example, if a document has been determined to have a high incidence of accuracy because a table on page 3 of a document may always return the correct gender for the patient, then the algorithm may identify that high accuracy has been provided for the document based on the one sentence of that document and stop processing a gender classification at the sentence level vector for that patient. Furthermore, a patient level vector may not be generated if a document level vector has reached a certain threshold of certainty (such as 95%), or if, for example, only one document is being processed);
a vector database that supports metadata filtering, wherein the at least one vector database comprises at least one of the at least one storage process (par [0167], Colley - The rule sets may include a vector of, for example, three hundred words and their respective weights, and each rule set may be applied over all words in a sentence to generate weights for every sentence. For example, a sentence “The patient was given prostate exam after he complained about having difficulty urinating in the mornings” may be given a high weight for gender as male because of words “prostate exam” and “he”. After each word of each sentence is processed, each respective sentence may be assigned a sentence vector (such as 10% female, 90% male), then each sentence in a document may be processed to assign a document vector, and finally, each document in a patient's EMR or EHR may be processed to assign a patient vector); and
a point ID generation process (par [0180], Colley - Fuzzy matching is structured around the text concepts included in the above enumerated list or the UMLS, including metadata fields CUI (the UMLS unique ID) and AUI (dictionary-specific unique ID), so that an exhaustive search may be performed for all medical concepts. The dictionary search engine may also return metadata about the specific entry detected (such as universal ID assigned in the above enumerated list or the UMLS));
wherein the vector generation process transforms each at least one section into a vector embedding (par [0446], [0475], Tunstall - an encoder which turns a source sentence into an internal vector or sequence of vectors that encodes the source sentence);
wherein the point ID generation process generates a unique ID (par [0180], Colley - Fuzzy matching is structured around the text concepts included in the above enumerated list or the UMLS, including metadata fields CUI (the UMLS unique ID) and AUI (dictionary-specific unique ID), so that an exhaustive search may be performed for all medical concepts. The dictionary search engine may also return metadata about the specific entry detected (such as universal ID assigned in the above enumerated list or the UMLS));
wherein the unique ID, the vector embedding, and the hyponym field and its values are sent to the storage process for storage (par [0180], Colley; par [0446], [0475], Tunstall - an encoder which turns a source sentence into an internal vector or sequence of vectors that encodes the source sentence);
wherein the at least one query is converted into at least one second vector embedding (par [0990, Tunstall);
wherein the at least one vector embedding is sent to the at least one storage process (par [0446], [0475], Tunstall); and
wherein the at least one storage process uses the at least one first vector embedding in the at least one second vector embedding in the construction of the response (par [1553-1562], Tunstall).
Regarding Claim 7, the combination of Colley in view of Tunstall, disclose the system of claim 6, wherein the relevant facts extraction process utilizes at least one categorical validation subprocess to remove facts that do not pass at least one validation criterion (par [0440], [0478], Tunstall – invalid passages/fact are removed).
Regarding Claim 4, the combination of Colley in view of Tunstall, disclose the system of claim 3, further comprising
a vector generation process (par [0168], Colley - at each level of granularity, the whole document classifier may be interrupted, for example, if a sufficient level of certainty has been reached or processing was intended to terminate at that level. For example, if a document has been determined to have a high incidence of accuracy because a table on page 3 of a document may always return the correct gender for the patient, then the algorithm may identify that high accuracy has been provided for the document based on the one sentence of that document and stop processing a gender classification at the sentence level vector for that patient. Furthermore, a patient level vector may not be generated if a document level vector has reached a certain threshold of certainty (such as 95%), or if, for example, only one document is being processed); a
vector database that supports metadata filtering, wherein the at least one vector database comprises at least one of the at least one storage process (par [0167], Colley - The rule sets may include a vector of, for example, three hundred words and their respective weights, and each rule set may be applied over all words in a sentence to generate weights for every sentence. For example, a sentence “The patient was given prostate exam after he complained about having difficulty urinating in the mornings” may be given a high weight for gender as male because of words “prostate exam” and “he”. After each word of each sentence is processed, each respective sentence may be assigned a sentence vector (such as 10% female, 90% male), then each sentence in a document may be processed to assign a document vector, and finally, each document in a patient's EMR or EHR may be processed to assign a patient vector); and
a point ID generation process (par [0180], Colley - Fuzzy matching is structured around the text concepts included in the above enumerated list or the UMLS, including metadata fields CUI (the UMLS unique ID) and AUI (dictionary-specific unique ID), so that an exhaustive search may be performed for all medical concepts. The dictionary search engine may also return metadata about the specific entry detected (such as universal ID assigned in the above enumerated list or the UMLS));
wherein the vector generation process transforms the at least one section into at least one first vector embedding (par [0446], [0475], Tunstall - an encoder which turns a source sentence into an internal vector or sequence of vectors that encodes the source sentence);
wherein the point ID generation process generates a unique ID (par [0180], Colley - Fuzzy matching is structured around the text concepts included in the above enumerated list or the UMLS, including metadata fields CUI (the UMLS unique ID) and AUI (dictionary-specific unique ID), so that an exhaustive search may be performed for all medical concepts. The dictionary search engine may also return metadata about the specific entry detected (such as universal ID assigned in the above enumerated list or the UMLS));
wherein the unique ID, the vector embedding, and the hyponym field and its values are sent to the at least one storage process for storage (par [0180], Colley; par [0191], [0197], Tunstall – for hyponym relation… par [0446], [0475], Tunstall - an encoder which turns a source sentence into an internal vector or sequence of vectors that encodes the source sentence);
wherein the at least one query is converted into at least one second vector embedding (par [0099], Tunstall);
wherein the at least one vector embedding is sent to the at least one storage process (par [0446], [0475], Tunstall); and
wherein the at least one storage process uses the at least one first vector embedding and the at least one second vector embedding in the construction of the response (par [1553-1562], Tunstall).
Regarding Claim 8, the combination of Colley in view of Tunstall, disclose a system for storing and retrieving information from an electronic knowledge base, the system comprising:
a computer and an associated memory (par [0043], [0287], Colley – while a mobile device or tablet is referenced throughout, it is understood that the device may also include any device such as a personal computer, etc.,… it is understood that a personal computer has associated memory and par [0200] makes mention of a memory);
at least one electronic document (par [0045], [0050-0051], Colley – generating a report is performed by capturing or uploading a report and validating any fields… electronically capturing a document, wherein the report corresponds to the electronic document);
at least one process for splitting the at least one electronic document into at least one section (par [0064], [0069], Colley – a report includes a first section and a second section, wherein the different sections identify separate features);
at least one storage process (par [0066], Colley – a document such as a genetic testing report may extract some or all medical information and then the report and extracted information may be stored in a database or other data repository);
a storage keywords process (par [0079], Colley);
a retrieval keywords process (par [0079], Colley – MLA may search for one or more keywords or phrases);
a retrieval keyword expansion process (par [0093-0094], Colley);
at least one query (par [0046], Colley – interface receives text input… par [0099], Colley – system is triggered in response to specific queries); and
a query filter construction process (par [0046], Colley - search indicator that, upon selection by the user, receives text input such as a patient's name, unique identifier, or diagnosis, that permits the user to filter the patients by the search criteria of the text input to search for a specific patient);
wherein the at least one process for splitting the document creates the at least one section (par [0064], [0069], Colley – a report includes a first section and a second section, wherein the different sections identify separate features);
wherein the at least one section is input into the storage keywords process, which returns at least one keywords field along with at least one first keyword for the at least one keywords field (par [0093-0094], [0120], [0123], Colley);
wherein the at least one keywords field and its associated at least one first keyword are sent to the storage process for storage (par [0119-0120], [0207], Colley);
wherein the at least one query is input into the retrieval keywords process, which returns at least one second keyword (par [0093-0094], [0120], [0123], Colley);
wherein the at least one second keyword is input into the retrieval keyword expansion process, which returns multiple keywords based on the input of the at least one second keyword (par [0093-0094], [0120], [0123], Colley; par [0638], Tunstall);
wherein the query filter construction process constructs a query filter that comprises filtering on the at least one keywords field and the multiple keywords returned from the retrieval keyword expansion process (par [0093-0094], [0120], [0123], Colley);
wherein the query filter is sent to the at least one storage process (par [0207-0208], [0231-0232], Colley – narrowing the search to specify that “Tylenol” is a brand name of the generic brand “acetaminophen” and processed through the database associated with the Abstraction Engine toolbox); and
wherein a response is received from the at least one storage process (par [0114], Colley – generate and serve a response to the system based on receiving a request… par [0234], Colley – response is received from toolbox service… also see par [0099], [0219]).
Regarding Claim 9, the combination of Colley in view of Tunstall, disclose the system of claim 8, wherein the retrieval keyword expansion process expands the at least one second keyword based on a keyword type (par [0638], Tunstall).
Points of Contact
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CHELCIE L DAYE whose telephone number is (571) 272-3891. The examiner can normally be reached on Monday-Friday 7:30-4:00pm. 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, Apu Mofiz can be reached on 571-272-4080. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
Chelcie Daye
Patent Examiner
Technology Center 2100
March 9, 2026
/CHELCIE L DAYE/Primary Examiner, Art Unit 2161