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
Claims 1-27 are pending. Claims 1, 16, 22, and 25 are independent and are all method Claims.
Claims 1-15 are pending and under examination of which Claim 1 is independent.
This Application was published as U.S. 20250131214.
Apparent priority: 19 October 2023.
Restriction/Election
Claims 16-27 were restricted out and Claims 1-15 elected without traverse during a telephone interview and subsequent telephonic response of 6/11/2026 and 6/12/2026.
Claims 1, 16, 22, and 15 are directed to
Restriction to one of the following inventions is required under 35 U.S.C. 121:
I. Claims 1-15 are directed to a method of placing a medical report in user-friendly non-technical language and to training of a model for doing so. (G06F40/10, 157, 166, 20, 237, 242, 247, 284, 289, 30, 40.)
II. Claims 17-21 and 22-24 are directed to a method of anonymizing patient data. (G06F40/295.)
III. Claims 25-27 are directed to a method of filling health insurance forms. (G06F40/174, 177, 18, 183, 186.)
1. A method comprising:
obtaining, at data processing hardware, a first input, the first input including medical information associated with a user, the medical information associated with the user including one or more medical terms;
generating, by the data processing hardware, a first output based on the first input using an algorithm, the first output including a first interpretation of the medical information associated with the user; and
providing, by the data processing hardware, the first output to the user;
wherein the first interpretation includes a first result of paraphrasing the one or more medical terms in the first input into broadly understood language.
16. A method, comprising:
obtaining, at data processing hardware, a consent to data sharing from a user;
de-identifying, by the data processing hardware, data of the user;
transmitting, by the data processing hardware, the de-identified data to a server; and
receiving, by the data processing hardware, a connection suggestion from the server,
wherein the data includes a first interpretation of medical information associated with the user,
wherein the medical information associated with the user includes one or more medical terms, and
wherein the first interpretation includes a first result of paraphrasing the one or more medical terms in the medical information associated with the user into broadly understood language.
22. A method, comprising:
obtaining, at data processing hardware, first de-identified data from a first software application and second de-identified data from a second software application;
storing, by the data processing hardware, the first de-identified data and the second de-identified data;
determining, by the data processing hardware, one or more commonalities based on one or more keywords found in the first de-identified data and the second de-identified data; and
in response to a determination that the first de-identified data and the second de-identified data share one or more commonalities, transmitting, by the data processing hardware, a connection offer to the first software application and the second software application.
25. A method comprising:
obtaining, at data processing hardware, a first question from a user, the first question about user's health insurance plan;
translating, by data processing hardware, the first question into language commonly used by health insurance industry using an algorithm;
analyzing, by the data processing hardware, health insurance document associated with the user's health insurance plan based on the first translated question; and
providing, by the data processing hardware, a first answer based on the analysis outcome.
The inventions need to be independent or distinct, each from the other. Here, the Inventions above are directed to related concept of medical reporting and medical information of a patient. The related inventions are distinct if: (1) the inventions as claimed are either not capable of use together or can have a materially different design, mode of operation, function, or effect; (2) the inventions do not overlap in scope, i.e., are mutually exclusive; and (3) the inventions as claimed are not obvious variants. See MPEP § 806.05(j). In the instant case, the inventions as claimed are distinct. Furthermore, the inventions as claimed do not encompass overlapping subject matter and there is nothing of record to show them to be obvious variants.
Restriction for examination purposes as indicated is proper because all the inventions listed in this action are independent or distinct for the reasons given above and there would be a serious search and/or examination burden if restriction were not required because one or more of the following reasons apply:
During a telephone conversation with Cecily O’Regan on 6/11 and 6/12 of 2026 a provisional election was made without traverse to prosecute the invention of group I claims 1-15. Affirmation of this election must be made by applicant in replying to this Office action. Claims 16-27 are withdrawn from further consideration by the examiner, 37 CFR 1.142(b), as being drawn to a non-elected invention.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim 12 is provisionally rejected under 35 U.S.C. 101 as claiming the same invention as Claim 11 of the instant Application. This is a provisional statutory double patenting rejection since the claims directed to the same invention have not in fact been patented.
The two Claims are identical with identical dependencies. Only a single claim to a single invention is permitted.
11. The method of claim 9, wherein:
the first reading level is associated with a first Flesch Reading Ease Score and the second reading level is associated with a second Flesch Reading Ease Score.
12. The method of claim 9, wherein:
the first reading level is associated with a first Flesch-Kincaid Grade Level and the second reading level is associated with a second Flesch-Kincaid Grade Level.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception.
Step 1: The independent Claims are directed to statutory categories:
Claim 1 is a method claim and directed to the process category of patentable subject matter.
Step 2A, Prong One: Does the Claim recite a Judicially Recognized Exception? Abstract Idea? Are these Claims nevertheless considered Abstract as a Mathematical Concept (mathematical relationships, mathematical formulas or equations, mathematical calculations), Mental Process (concepts performed in the human mind (including an observation, evaluation, judgment, opinion), or Certain Methods of Organizing Human Activity (1-fundamental economic principles or practices (including hedging, insurance, mitigating risk), 2-commercial or legal interactions (including agreements in the form of contracts; legal obligations; advertising, marketing or sales activities or behaviors; business relations), 3- managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions) and fall under the judicial exception to patentable subject matter?)
The rejected Claims recite Mental Processes or Methods of Organizing Human Activity.
Step 2A, Prong Two: Additional Elements that Integrate the Judicial Exception into a Practical Application? Identifying whether there are any additional elements recited in the claim beyond the judicial exception(s), and evaluating those additional elements to determine whether they integrate the exception into a practical application of the exception. “Integration into a practical application” requires an additional element(s) or a combination of additional elements in the claim to apply, rely on, or use the judicial exception in a manner that imposes a meaningful limit on the judicial exception, such that the claim is more than a drafting effort designed to monopolize the exception. Uses the considerations laid out by the Supreme Court and the Federal Circuit to evaluate whether the judicial exception is integrated into a practical application.
The rejected Claims do not include additional limitations that point to integration of the abstract idea into a practical application and are therefore directed to the abstract idea.
Claim 1 is a generic automation of a mental process of reading a medical report of a patient that includes medical terms and paraphrasing it for the patient in a language that is more understandable for a lay person.
1. A method comprising:
obtaining, at data processing hardware, a first input, the first input including medical information associated with a user, the medical information associated with the user including one or more medical terms; [Nurse pulls the file of a patient.]
generating, by the data processing hardware, a first output based on the first input using an algorithm, the first output including a first interpretation of the medical information associated with the user; and [Nurse interprets the results of the patient’s blood test.]
providing, by the data processing hardware, the first output to the user; [Nurse provides the results to the patient in person or over the phone or shows the result to the patient on a chart.]
wherein the first interpretation includes a first result of paraphrasing the one or more medical terms in the first input into broadly understood language. [Nurse tries to use terminology that is plain and understandable to the user.]
Step 2B: Search for Inventive Concept: Additional Elements Do not amount to Significantly More: The limitations of data processing hardware or server that are included in the Claims are well-understood, routine, and conventional machine components that are being used for their well-understood, routine, conventional and rather generic functions. Additionally, these limitations are expressed parenthetically and lack nexus to the Claim language and as such are a separable and divisible mention to a machine. Accordingly, they are not sufficient to cause the Claim as a whole to amount to significantly more than the underlying abstract idea.
The Dependent Claims do not add limitations that could integrate the abstract idea into a practical technological application or could help the Claim as a whole to amount to significantly more than the Abstract idea identified for the Independent Claim:
2. The method of claim 1, wherein the medical information associated with the user is a transcription based on spoken words from a healthcare provider. [A medical record may first be spoken. This Claim is still a human mental activity.]
3. The method of claim 1, wherein the medical information of the user includes an electronic health record (EHR) of the user or a written communication from a healthcare provider. [Neither option helps elevate the abstract idea to a practical application or significantly more. The medical information is often from a healthcare provider.]
4. The method of claim 1, wherein providing the first output to the user includes displaying the first interpretation to the user. [The simplified report can be shown to the patient.]
5. The method of claim 1, wherein providing the first output to the user includes:
converting the first interpretation into a first synthesized voice; and [This claim includes a potential technological component of a speech synthesizer which is not sufficient to make the claim non-abstract because it is an add-on unrelated to the inventive core and a post processing event with a werc element.]
playing the first synthesized voice.
6. The method of claim 1, wherein the algorithm includes one or more algorithms, the one or more algorithms including natural language processing (NLP). [This claim includes a potential technological component of NLP processing which is not sufficient to make the claim non-abstract because it is an add-on unrelated to the inventive core and a post processing event with a werc element.]
7. The method of claim 1, further comprising:
obtaining, at the data processing hardware, a second input, the second input including medical information not associated with the user, the medical information not associated with user including one or more medical terms; [This is just the medical report of a second patient.]
generating, by the data processing hardware, a second output based on the second input using the algorithm, the second output including a second interpretation of the medical information not associated with the user; and [Simplifying the report of the second patient.]
providing, by the data processing hardware, the second output to the user, wherein the second interpretation includes a second result of paraphrasing the one or more medical terms in the second input into broadly understood language. [This is a repeat of Claim 1 for a different report.]
8. The method of claim 7, wherein the second input includes a medical publication. [A medial article is simplified and provided to the user/patient.]
9. The method of claim 1, wherein the algorithm is trained by: [This Claim is directed to a very generic and general training process for an algorithm which falls under a mathematical concept.]
performing one or more training iterations, the one or more training iterations including:
obtaining, at the data processing hardware, a training input, the training input including training medical information;
generating, by the data processing hardware, a plurality of training outputs based on the training input using the algorithm, the plurality of training outputs including a second interpretation of the training medical information targeted at a first reading level and a third interpretation of the training medical information targeted at a second reading level;
providing, by the data processing hardware, the plurality of training outputs to the user;
receiving, by the data processing hardware, feedback from the user on the training outputs; and
training, by the data processing hardware, the algorithm based on the feedback received from the user.
10. The method of claim 9, wherein the training medical information is the medical information associated with the user. [The Algorithm is trained on the patient data.]
11. The method of claim 9, wherein:
the first reading level is associated with a first Flesch Reading Ease Score and the second reading level is associated with a second Flesch Reading Ease Score. [These are reading scores and are not technological.]
12. The method of claim 9, wherein:
the first reading level is associated with a first Flesch-Kincaid Grade Level and the second reading level is associated with a second Flesch-Kincaid Grade Level. [Repeat of Claim 11.]
13. The method of claim 9, wherein the receiving the feedback from the user on the training outputs includes receiving an indication of a preferred output by the user. [User selects the correct choice. Preparing data.]
14. The method of claim 9, wherein the training the algorithm based on the feedback includes adjusting the algorithm to target a reading level corresponding to the preferred output. [User select the appropriate outputs.]
15. The method of claim 1, further comprising:
generating, by the data processing hardware, one or more suggested questions that provide additional valuable information for the user; and [Adding questions.]
incorporating, by the data processing hardware, the one or more suggested questions into the first output. [Adding questions to the material provided to the user.]
Claim Rejections - 35 USC § 102
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.
Claims 1, 3-4, 6-8 and 15 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Said (U.S. 20200234826).
PNG
media_image1.png
702
964
media_image1.png
Greyscale
PNG
media_image2.png
564
1022
media_image2.png
Greyscale
PNG
media_image3.png
686
984
media_image3.png
Greyscale
PNG
media_image4.png
540
646
media_image4.png
Greyscale
Regarding Claim 1, Said teaches:
1. A method comprising: [Said is directed to the same concept: “… Treatment recommendations and options are displayed to the patient in ordinary language to facilitate proper understanding. Translation of treatment options to ordinary language can be performed automatically.” Abstract. As to the hardware: “[0233] The present document also relates to an apparatus for performing the operations herein. This apparatus may be specially constructed for the required purposes, or it may comprise a general-purpose computing device selectively activated or reconfigured by a computer program stored in the computing device. Such a computer program may be stored in a computer readable storage medium, such as, but is not limited to, any type of disk including floppy disks, optical disks, CD-ROMs, DVD-ROMs, magnetic-optical disks, read-only memories (ROMs), random access memories (RAMs), EPROMs, EEPROMs, flash memory, solid state drives, magnetic or optical cards, application specific integrated circuits (ASICs), or any type of media suitable for storing electronic instructions, and each coupled to a computer system bus. Further, the computing devices referred to herein may include a single processor or may be architectures employing multiple processor designs for increased computing capability.”]
obtaining, at data processing hardware, a first input, the first input including medical information associated with a user, the medical information associated with the user including one or more medical terms; [Said, Figure 2, obtaining medical information includes direct question answering by the patient or retrieving of medical records: “[0137] If, in step 204, the user indicates that he or she would like to answer medical questions instead of authorizing access, the system collects 207 user-identifying information and answers to medical questions …” “[0142] … Treatment options include but are not limited to: approved drugs, interventions such as surgery or radiation therapy, and clinical trials.” “[0136] Examples of patient data include but are not limited to electronic medical records, genomic information, claims records, geolocation, diagnostic tests, medical benefits, insurance plan information, and the like….” Medical records include medical terms as shown in Figures of Said and Figure 12 for example: “consider endocrine therapy 1202.”]
generating, by the data processing hardware, a first output based on the first input using an algorithm, the first output including a first interpretation of the medical information associated with the user; and [Said, Figure 8 shows one example of output to the user based on interpretation of input data pertaining to the user. Figure 11, 1110. “[0040] Referring now to FIG. 11, there is shown a block diagram depicting an overall schematic architecture for a system 1100 that provides personalized treatment recommendations to patients, according to one embodiment. Personalization engine 1101 takes input from various sources, including services 1102 and 1105, to generate user output 1110, which may include personalized treatment, clinical trials, news, and/or patient-reported outcomes (PRO)….”]
providing, by the data processing hardware, the first output to the user; [Said, Figure 1, 104: “[0072] In at least one embodiment, patient mobile app 110 runs on device(s) 104 associated with patients and/or caregivers. Patient mobile app 110 interfaces with patient API 114 as needed, so as to provide a user-friendly interface that allows access to tools for patients and/or caregivers to develop individual patient profiles. The system then uses these patient profiles to provide specifically-tailored content and experiences appropriate for a particular user….” ]
wherein the first interpretation includes a first result of paraphrasing the one or more medical terms in the first input into broadly understood language. [Said: “[0142] The system then provides 215 all treatment path permutations and associated content to the user. This may include displaying treatment options, recommendations, and/or other information via patient mobile app 110. In at least one embodiment, treatment options are displayed in ordinary language to facilitate proper understanding. In at least one embodiment, the translation of treatment options to ordinary language can be performed automatically. This can be done, for example, by looking up the various options in a patient-friendly translation dictionary whereby every medical concept is translated into patient-friendly simplified language. Treatment options include but are not limited to: approved drugs, interventions such as surgery or radiation therapy, and clinical trials.”]
Regarding Claim 3, Said teaches:
3. The method of claim 1, wherein the medical information of the user includes an electronic health record (EHR) of the user or a written communication from a healthcare provider. [Said, Figure 1, “EHR systems 102.” “[0117] In other embodiments, integration can be performed using other approaches. These include, for example, HL7 interfacing, custom flat-file extracts out of EHR systems 102 such as EPIC, and/or custom document submission feeds into EHR systems 102….” See also Figures 14A and 15 pertaining to HER.]
Regarding Claim 4, Said teaches:
4. The method of claim 1, wherein providing the first output to the user includes displaying the first interpretation to the user. [Said, see Figure 8 and Figure 11, “user output/visualization 1110.”]
Regarding Claim 6, Said teaches:
6. The method of claim 1, wherein the algorithm includes one or more algorithms, the one or more algorithms including natural language processing (NLP). [Said: “[0195] In at least one embodiment, a treatment method similar to that depicted in FIG. 2 is used to select which trials to display. The system looks up clinical attributes from patient records and/or answers to questions, and then uses these attributes to perform a multi-layered search of clinical trials; the search can include inclusion and/or exclusion criteria. In at least one embodiment, the method uses natural language processing to determine which keywords match which part of the trials. Natural language processing can also be used to analyze medical records and health information, so that the information can be integrated with clinical trials, thereby providing improved personalization.”]
Regarding Claim 7, Said teaches:
7. The method of claim 1, further comprising:
obtaining, at the data processing hardware, a second input, the second input including medical information not associated with the user, the medical information not associated with user including one or more medical terms; [Said, Figure 8, e.g., the “Drug Information,” “Patients’ Outcomes,” and the rest of information on this screen include medical terms and are not associated with the user rather associated with broad spectrum studies of a drug, its costs, and outcomes. At least the name of the drug teaches the medical terms whereas many of the other terms shown in Figure 8 such as “chemotherapy regimen” or “breast cancer” are also medical terms. See also Figure 6A which begins with information regarding the user but ends with information regarding “Clinical Trials” and Figure 6B which is entirely information not about a particular patient and provided in plain language.]
generating, by the data processing hardware, a second output based on the second input using the algorithm, the second output including a second interpretation of the medical information not associated with the user; and [Said, Figure 8, the middle panel is not associated with the particular user and rather with a general study.]
providing, by the data processing hardware, the second output to the user, wherein the second interpretation includes a second result of paraphrasing the one or more medical terms in the second input into broadly understood language. [Said, Figure 7B shows a presentation to the user regarding the data of studies conducted on other people. This is a continuation of Figure 7A which also includes information regarding Doxorubicin which is a medical term in plain language: “A common chemotherapy medication.”]
Regarding Claim 8, Said teaches:
8. The method of claim 7, wherein the second input includes a medical publication. [Said, Figure 6B; 6005 presents a “research study” or “clinical trial” which comes in a publication form: “[0202] FIG. 6B depicts screen 6005, which provides information about a clinical trial. The user can tap on buttons 6005A through 6005E to see different types of information; in at least one embodiment, such information is pulled live via APIs and/or from existing stored databases. Button 6005A activates screen 6006, which provides general information about the clinical trial. Button 6005B activates map view 6007, which displays locations for the clinical trial….”]
Regarding Claim 15, Said teaches:
15. The method of claim 1, further comprising:
generating, by the data processing hardware, one or more suggested questions that provide additional valuable information for the user; and [Said, “[0140] The system then supplements 213 any existing medical answers with new medical question answers, such as those collected in step 207.” “[0036] FIGS. 17A and 17B depict various examples of screens of a user interface for completing a Patient Record Outcomes Questionnaire, according to one embodiment.” “[0137] If, in step 204, the user indicates that he or she would like to answer medical questions instead of authorizing access, the system collects 207 user-identifying information and answers to medical questions, for example by prompting the user via patient mobile app 110 and receiving input from the user. The collected information is then encrypted and transmitted 208 to a server for secure storage.”]
incorporating, by the data processing hardware, the one or more suggested questions into the first output. [Said, the answers are incorporated into the output: “[0120] In various embodiments, information can be personalized based on patient information that is automatically retrieved from authorized patient records, and/or user answers to questions, for example via a questionnaire. In at least one embodiment, personalization is performed by automatically looking up specific treatments based on patient attributes.” “[0139] In step 110, the system determines whether the user has authorized patient record access. If so, patient records are retrieved 211 and then processed 212 to extract answers to medical questions.”]
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Said in view of Strader (U.S. 20220391083).
Regarding Claim 2, transcription of the medical professional’s statements is common but Said does not mention that.
Strader teaches:
2. The method of claim 1, wherein the medical information associated with the user is a transcription based on spoken words from a healthcare provider. [Strader: “A method and workstation for generating a transcript of a conversation between a patient and a healthcare practitioner is disclosed. A workstation is provided with a tool for rendering of an audio recording of the conversation and generating a display of a transcript of the audio recording using a speech-to-text engine, thereby enabling inspection of the accuracy of conversion of speech to text. A tool is provided for scrolling through the transcript and rendering the portion of the audio according to the position of the scrolling. There is a highlighting in the transcript of words or phrases spoken by the patient relating to symptoms, medications or other medically relevant concepts. Additionally, there is provided a set of transcript supplement tools enabling editing of specific portions of the transcript based on the content of the corresponding portion of audio recording.” Abstract.]
Said and Strader pertain to preparation of medical reports and it would have been obvious to combine the audio transcription of Strader with the system of Said as a logical method of generating patient file that can be used by Said. This combination falls under combining prior art elements according to known methods to yield predictable results or use of known technique to improve similar devices (methods, or products) in the same way. See MPEP 2141, KSR, 550 U.S. at 418, 82 USPQ2d at 1396.
Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Said in view of Menhard (U.S. 20040181412).
Regarding Claim 5, Said teaches: “[0055] In the context of the description herein, a “computing device” or “client device” may be any device capable of digital data processing, including (but not limited to) one that is able to receive any combination of text-based input, touchscreen input, voice-based (speech) input, and/or the like, and is further able to generate any combination of visual output, text-based output, haptic output, and/or audio output (including voice-based (speech) output). …”
Said does not expressly teach presentation by synthesized voice.
Menhard teaches:
5. The method of claim 1, wherein providing the first output to the user includes:
converting the first interpretation into a first synthesized voice; and [Menhard, “…The CAD report is processed to produce a speech synthesized CAD report in accordance with the at least one level of information. The digital image is simultaneously displayed with the delivery of the speech synthesized CAD report whereby the user can examine the digital image while simultaneously listening to the CAD report….” Figure 1, 104.]
playing the first synthesized voice. [Menhard, Figure 1, 106.]
Said and Menhard pertain to reporting of medical information and it would have been obvious to combine the speech synthesis of Menhard with the system of Said which provides for the capability in its device either for the reason set forth in Menhard such that a medical image can be viewed while its associated medical report is being read aloud or for a patient who is driving or visually impaired. This combination falls under combining prior art elements according to known methods to yield predictable results or use of known technique to improve similar devices (methods, or products) in the same way. See MPEP 2141, KSR, 550 U.S. at 418, 82 USPQ2d at 1396.
Claims 9-14 are rejected under 35 U.S.C. 103 as being unpatentable over Said in view of Kargiannakis (U.S. 20200265185) and further in view of Strader.
Regarding Claim 9, Said does not discuss the training of the model.
Karginannakis teaches:
9. The method of claim 1, wherein the algorithm is trained by: performing one or more training iterations, the one or more training iterations including:
obtaining, at the data processing hardware, a training input, the training input including training medical information; [Karginannakis, Figure 1, the “content conversion system 100” which converts a content into different reading levels has access to various types of training data. Figure 5 shows the pipeline of input of “content acquisition 1100” to the “learning data repository and manager 1108.” “[0084] Content conversion system 100 interfaces with external data 160 which may include an external data repository and store partner data. External data 160 may include data such as training data, provided by an external source, and accessed by external data retrieval software 370, described in further detail below.” Figure 3, “Learning Data Store 398.” “[0104] Machine learning software 360 determines recommendations for content conversion to be performed by transformation software 330, as well as develop training sets of data to train machine learning models to process data using programming rules and code that can dynamically update over time. In some embodiments, machine learning software 360 is configured to learn from transformations made, for example, by transformation software 330, which may facilitate transformation software 330 performing in a more automated and more accurate way in future uses. Training data and machine learning models may be stored in learning data store 398.” “[0105] As illustrated, machine learning software 360 may include a learning data repository and manager 1108 for storing and managing training data collected by content conversion system 100.” “[0107] As illustrated, external data retrieval software 370 may include external data repositories and partner data 1109 for receiving data, such as training data, from external or partner sources instead of through content conversion system 100 directly.” ]
generating, by the data processing hardware, a plurality of training outputs based on the training input using the algorithm, the plurality of training outputs including a second interpretation of the training medical information targeted at a first reading level and a third interpretation of the training medical information targeted at a second reading level; [Karginannakis, Figure 1, the “content conversion system 100” is trained to convert an input “content 130” from one reading level to another reading level and therefore the “training outputs” which would be the outputs of the model “[0062] Content conversion system 100 may measure the readability levels of individual pieces of training data gathered from operation of content conversion system 100. Content conversion system 100 may also track each individual end-user (that is, a reader of converted content), for example, user 110 or one of other users 170, to compile a detailed profile of their individual readability levels across all the various dimensions mentioned above.” “[0112] … Once content 130 is acquired, content acquisition 1100 may request that user 110 input a target readability level (TRL) for content 130, as the desired readability level for content 130 following conversion, and a target comprehensibility level (TCL) for content 130, as the desired comprehensibility level for content 130 following conversion.”]
providing, by the data processing hardware, the plurality of training outputs to the user; [Karginannakis, Figures 3 and 4, “content presenter and feedback gatherer 1103.” “[0109] As illustrated, output software 380 may include a content presenter and feedback gatherer 1103 for formatting transformed text in preparation for presentation to a user such as user 110 as well as for soliciting and receiving feedback from users on transformations, final content delivery 1105 for delivering content to a user such as user 110 for external purposes, and application embedder 1107 for expressing transformations within other (e.g., external) applications in which digital content is being created, edit, or curated.” “[0150] Conversion controller 1102 may output raw converted content (both finalized and potential) to content presenter and feedback gatherer 1103.”]
receiving, by the data processing hardware, feedback from the user on the training outputs; and [Karginannakis, Figures 3 and 4, “content presenter and feedback gatherer 1103.” “[0111[ … Content presenter and feedback gatherer 1103 also receives end-user and customer feedback and communicates with application embedder 1107 and final content delivery 1105, as well as end-user/customer profiling and requirements manager 1106….” “[0204] Content presenter and feedback gatherer 1103 may be configured to give the end-user/customer, such as user 110, the opportunity to make judgments on whether the current state of conversion meets their requirements. User 110 can choose to comment on the state, change their overall requirements, and/or return the content for further conversion. Also, user 110 can provide more micro inputs on individual segments that have been converted—even to the point of changing the conversion details. If user 110 makes any direct changes to content, this information is fed into the learning data repository and manager 1108 which may improve the automation of the overall system.”]
training, by the data processing hardware, the algorithm based on the feedback received from the user. [Karginannakis, “[0062] Content conversion system 100 may measure the readability levels of individual pieces of training data gathered from operation of content conversion system 100. Content conversion system 100 may also track each individual end-user (that is, a reader of converted content), for example, user 110 or one of other users 170, to compile a detailed profile of their individual readability levels across all the various dimensions mentioned above.” See [0204] above. The user feedback is collected and fed into “learning data” to further train the content conversion system 100.]
Said and Kargiannakis pertain to modifying content to improve readability and it would have been obvious to augment the system of Said which does not address the training of its model with the system of Karginnakis which includes the training features for a more complete system. This combination falls under combining prior art elements according to known methods to yield predictable results or use of known technique to improve similar devices (methods, or products) in the same way. See MPEP 2141, KSR, 550 U.S. at 418, 82 USPQ2d at 1396.
PNG
media_image5.png
616
570
media_image5.png
Greyscale
PNG
media_image6.png
744
542
media_image6.png
Greyscale
PNG
media_image7.png
658
1026
media_image7.png
Greyscale
PNG
media_image8.png
606
954
media_image8.png
Greyscale
Regarding medical data, Karginannakis teaches “[0523] … Leanne is a paralegal by training and has found the medical language used to explain her pregnancy and forthcoming delivery very overwhelming. While she and her wife, Mary, have taken to the internet to search some of the terms in the documents their OB-GYN and family doctor provided, they only found equally as confusing reports online. …” Medical terminology simplification is also taught by this reference. “[0532] … [0532] … Luckily, Yvette has integrated content conversion system 100 into her chatbot technology. Now the chatbot will respond and mirror the type of language used to ask it questions. If someone uses a lot of slang and local colloquialisms, the chatbot will mirror that language and adjust the medical information accordingly….”
In order to perform the above tasks the model has to be trained on medial data. However, that is not express in the reference.
Strader teaches:
9. The method of claim 1, wherein the algorithm is trained by: performing one or more training iterations, the one or more training iterations including:
obtaining, at the data processing hardware, a training input, the training input including training medical information; [Strader, Figures 1 and 2 show the training of the “machine learning system 104/218” based on the audio input from a conversation between a doctor 202 and a patient 204 which includes medical terms identified at “entity recognition for medically relevant words/phrases 112.”]
generating, by the data processing hardware, a plurality of training outputs based on the training input using the algorithm, the plurality of training outputs including a second interpretation of the training medical information targeted at a first reading level and a third interpretation of the training medical information targeted at a second reading level;
providing, by the data processing hardware, the plurality of training outputs to the user; [Strader, Figure 5 teaches providing the output of the model to the user for correction.
receiving, by the data processing hardware, feedback from the user on the training outputs; and [Strader, Figures 1 and 2, the “transcript … 114” is subjected to editing based on feedback: “[0005] … The transcript (and note) are editable, with version control to approve, reject or provide feedback on generated suggestions for editing the transcript. Alternative words are suggested for speech in the transcript that does not match vocabulary in the automated speech recognition, or for muffled and partially audible voice input.” See also Figure 5: “[0058] … The method may further include the step of automatically generating suggestions of alternative words or phrases for words or phrases in the transcript and tools to approve, reject or provide feedback on the generated suggestions to thereby edit the transcript, e.g., as described in FIG. 5 above….” “7. The method of claim 6, wherein the set of note supplement tools include at least one of: a) a display of smart suggestions for words or phrases and a tool for editing, approving, rejecting or providing feedback on the smart suggestions; b) a display of suggested corrected medical terminology; and c) a display of an indication of confidence level in suggested words or phrases.”]
training, by the data processing hardware, the algorithm based on the feedback received from the user. [Strader, Figure 1, the transcript 114 that is subjected to the feedback and edits of Figure 5 is fed back in a loop to the “machine learning 104.” This step is not expressly discussed.]
PNG
media_image9.png
374
468
media_image9.png
Greyscale
PNG
media_image10.png
350
496
media_image10.png
Greyscale
Said/Kargiannakis and Strader pertain to modifying content to improve readability and it would have been obvious to augment the system of combination with the medical data of Strader in order to have a model that is trained on medical data as well. This combination falls under simple substitution of one element with a similar one. See MPEP 2141, KSR, 550 U.S. at 418, 82 USPQ2d at 1396.
Regarding Claim 10, Said/Kargiannakis were not cited for training with medical data.
Strader teaches:
10. The method of claim 9, wherein the training medical information is the medical information associated with the user. [Strader, Figure 2, the input data from conversation with patient 204 and therefore the system will get trained on the patient data.]
Rationale for combination as provided for Claim 9. The feature of training with medical data was imported from Strader and the details of training come from the same reference under the same rationale.
Regarding Claim 11, Said does not discuss the training of the model.
Karginannakis teaches:
11. The method of claim 9, wherein:
the first reading level is associated with a first Flesch Reading Ease Score and the second reading level is associated with a second Flesch Reading Ease Score. [Karkiannakis teaches that a particular content can be converted into any number of reading levels which teaches the first and second reading levels of this Claim: “[0051] Content conversion system 100 may leverage both the reading/writing skills and reading challenges of a broad variety of users (as well as several existing linguistic resources) to build machine learning models to convert any content into any reading level, comprehensibility level or style.” The particular type of score of this Claim is also taught by Karginannakis: “[0259] A Flesch Reading Ease score of 90-100 can indicate content readable by a fifth grader, while Flesch Reading Ease scores between 0-30 indicate readability by college graduates.”]
Rationale for combination as provided for Claim 9. The feature of reading levels was imported from Karginannakis and the details of training come from the same reference under the same rationale.
12. The method of claim 9, wherein:
the first reading level is associated with a first Flesch-Kincaid Grade Level and the second reading level is associated with a second Flesch-Kincaid Grade Level. [This claim is identical to Claim 11.]
Regarding Claim 13, Said does not discuss the training of the model.
Karginnakis teaches:
13. The method of claim 9, wherein the receiving the feedback from the user on the training outputs includes receiving an indication of a preferred output by the user. [Karginnakis, the corrections at the 1103 module and step indicate the preference of the user. “[0045] … Thus, the user's interactions with content conversion system 100 may be tracked, for example, to track a user's preferences, readability level and comprehensibility level over time.” “[0530] … his allows the business units to send pre-simplified drafts to the communications & marketing team to review. It also automatically applies corporate dictionary, style sheets, style guide principles so the documents are streamlined with the organizational style, tone, and preferred language….”]
Rationale for combination as provided for Claim 9. The feature of training was imported from Karginnakis and the details of training come from the same reference under the same rationale.
(See also Strader, in several places teaches that the correction and editing that teach the feedback of this Claim. “[0044] … Furthermore, the audio recording tools on the workstation display include pause, rewind, play, fast forward, etc., so that the user can start and stop the recording to listen to sensitive or important patient information and confirm that the transcript, highlighted words or phrases, and insertion of words or phrases into the note are correct.” “[0046] FIG. 5 illustrates a further example of editing of a transcript 500 and a note. In FIG. 5, the patient has spoken the phrase “I am. Lipodrene.” (502) in response to a question about medication they are taking. The phrase “Lipodrene” is not recognized as a name of a medication and the NER model of FIG. 1 generates two smart suggestions: Amlodipine and Lipozene (504 and 506) which are placed adjacent to the suspect term “Lipodrene.” The user can accept either of these suggestions by clicking on them, or reject them by activating the X icon 508….” “[0067] … The transcription supplement tools enable editing of these portions of the transcript, such as by displaying suggested alternative phrases, displaying corrected medical terminology, displaying suggestions for incomplete words, etc., and tools for accepting, rejecting or editing the transcript and the generated suggestions…..” “7. The method of claim 6, wherein the set of note supplement tools include at least one of: a) a display of smart suggestions for words or phrases and a tool for editing, approving, rejecting or providing feedback on the smart suggestions; b) a display of suggested corrected medical terminology; and c) a display of an indication of confidence level in suggested words or phrases.”)
Regarding Claim 14, Said does not discuss the training of the model.
Kargiannakis teaches:
14. The method of claim 9, wherein the training the algorithm based on the feedback includes adjusting the algorithm to target a reading level corresponding to the preferred output. [Karginannakis is training its model to perform various NLP tasks such as transcription (speech to text) , “[0513] Applications of systems described herein, including content conversion system 100, include embedding into web browsers such that a webpage can be converted to a different reading level, training chat bots to modulate their language based on with whom they are speaking, and integration with speech technologies (e.g., speech assistants, speech-to-text, audio information, text-to-speech, etc.), amongst other applications.”]
Rationale for combination as provided for Claim 9. The feature of reading levels was imported from Karginannakis and the details of training come from the same reference under the same rationale.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to FARIBA SIRJANI whose telephone number is (571)270-1499. The examiner can normally be reached 9 to 5, M-F.
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, Pierre Desir can be reached at 571-272-7799. 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.
/Fariba Sirjani/
Primary Examiner, Art Unit 2659
Prosecution Insights