Automated veterinary appointment documentation system and method thereof

Abstract

Disclosed herein are devices, systems, and methods for performing automated veterinary documentation, the method including: receiving an audio input representing speech of at least one user; transcribing the audio input using at least one speech recognition model to generate a transcription; and generating a summary output of the transcripton using at least one large language model, the summary output categorized into structured categories. The structured categories can be based on a SOAP (Subjective, Objective, Assessment, Plan) format.

Description

AUTOMATED VETERINARY APPOINTMENT DOCUMENTATION SYSTEM AND METHOD THEREOFFIELD

[0001] The invention pertains to the field of automated veterinary documentation systems, specifically focusing on integrating advanced technologies to optimize and streamline the process of documenting veterinary appointments (including, optionally, treatments and opinions).BACKGROUND

[0002] In veterinary practice, the documentation of patient appointments (i.e. where pets are examined and medical advice and / or treatments are provided) is a crucial aspect of providing high- quality care and maintaining accurate health records. Traditionally, veterinary professionals have relied on manual methods of documentation, such as handwritten notes, paper-based charts, and standalone software systems. However, these traditional approaches often present several challenges that can impact efficiency, accuracy, and overall patient care.

[0003] Manual documentation methods, such as handwriting notes or manually inputting data into software systems, are time-consuming and can divert valuable resources away from direct patient care. Handwritten notes and manual data entry increase the risk of errors, such as illegible handwriting or transcription mistakes, which can lead to inaccuracies in medical records and treatment plans. Further, paper-based records or standalone software systems may not be easily accessible remotely, making it challenging for veterinary professionals to access critical patient information outside of the clinic setting leading to inconsistent documentation practices across different veterinarians or clinics that can result in fragmented patient information, hindering continuity of care and collaboration among veterinary teams.

[0004] Given these challenges, there is a clear need for an automated veterinary documentation system that leverages modern technologies, such as speech recognition, and natural language processing to streamline and optimize the documentation process. Such a system can enhance accuracy, improve accessibility, ensure compliance with regulations, facilitate collaboration among veterinary teams, and ultimately support better patient care in the veterinary field.BRIEF SUMMARY

[0005] In accordance with an aspect, there is provided a system for performing automated veterinary documentation, wherein the system includes: at least one processor; and one or more non-transitory memory units communicatively coupled to the at least one processor and storing machine-readable instructions that, when executed, cause the at least one processor to perform at least the following steps: receive an audio input representing speech of at least one user; transcribe the audio input using at least one speech recognition model to generate a transcription; and generate a summary output of the transcripton using at least one large language model, the summary output categorized into structured categories.

[0006] In some embodiments, the structured categories are based on a SOAP (Subjective, Objective, Assessment, Plan) format.

[0007] In some embodiments, the structured categories are based on at least one adaptive template defining at least one data type and at least one format for the summary output.

[0008] In some embodiments, the steps further comprise documenting the summary output in a medical record using a medical record creation unit.

[0009] In some embodiments, the at least one speech recognition model comprises a dictation unit configured to transcribe the audio input into the transcription, the dictation unit configured based on veterinary terminology.

[0010] In some embodiments, generating the summary output comprises generating summaries of the transcription using a recap unit.

[0011] In some embodiments, generating the summary output comprises using a callback unit on the audio input, the audio input representing client communication.

[0012] In some embodiments, the system further includes performing speech-based dental charting by capturing the audio input, the audio input representing a narration of dental examination by a first user in real-time, and wherein the structured categories are structured dental chart entries.

[0013] In some embodiments, the system further includes performing an ultrasound assessment documentation by capturing the audio input, the audio input representing a narration of an ultrasound assessment by a first user in real-time, and wherein the at least one large language model is trained on ultrasound terminology.

[0014] In some embodiments, the system further includes performing medical assessment documentation by capturing the audio input, the audio input representing a narration by a first user in real-time of a medical assessment of at least one subject, and wherein the summary output is categorized into the structured categories for each subject, wherein each subject is defined in the audio input by a description.

[0015] In some embodiments, the summary output comprises predefined expressions inserted based on predefined conditions represented in the audio input.

[0016] In some embodiments, the processor is configured to accurately capture and interpret the spoken dialogue using speech recognition algorithms.

[0017] In some embodiments, the summary output represents treatment progress, medical assessment, surgical outcomes, diagnostic findings, follow-up recommendations, client updates about a subject, or any combination thereof of a subject.

[0018] In accordance with an aspect, there is provided a method for performing automated veterinary documentation, the method including: receiving an audio input representing speech of at least one user; transcribing the audio input using at least one speech recognition model to generate a transcription; and generating a summary output of the transcripton using at least one large language model, the summary output categorized into structured categories.

[0019] In some embodiments, the structured categories are based on a SOAP (Subjective, Objective, Assessment, Plan) format.

[0020] In some embodiments, the method includes generating the summary output to include at least one data type and at least one format, each defined by at least one adaptive template.

[0021] In some embodiments, the method includes performing speech-based dental charting by capturing the audio input, the audio input representing a narration of dental examination by a firstuser in real-time, and wherein the structured categories are structured dental chart entries.

[0022] In some embodiments, the method includes performing an ultrasound assessment documentation by capturing the audio input, the audio input representing a narration of an ultrasound assessment by a first user in real-time, and wherein the at least one large language model is trained on ultrasound terminology.

[0023] In some embodiments, the method includes performing medical assessment documentation by capturing the audio input, the audio input representing a narration by a first user in real-time of a medical assessment of at least one subject, and wherein the summary output is categorized into the structured categories for each subject, wherein each subject is defined in the audio input by a description.

[0024] In some embodiments, the method includes inserting predefined expressions in the summary output based on predefined conditions represented in the audio input.

[0025] In some embodiments, an automated veterinary documentation system integrates several innovative units within a processor to optimize and streamline the process of documenting veterinary appointments. In one embodiment, the system utilizes advanced speech recognition technology to capture and categorize information provided by the veterinarian into structured categories (Subjective, Objective, Assessment, Plan - SOAP format). In one embodiment, the system employs advanced speech recognition algorithms to accurately capture and interpret spoken dialogue in real-time, converting spoken words into written text seamlessly. In one embodiment, the system automatically generates concise summaries based on input data from various sources within the veterinary practice, suitable for practice managers during team meetings or rounds. In one embodiment, the system facilitates comprehensive documentation of client communications by recording important details regarding treatment progress, surgical outcomes, and follow-up recommendations. In one embodiment, the system performs voice-activated dental charting, capturing and translating spoken observations into structured dental chart entries with visual mapping of dental findings. In one embodiment, the system employs machine learning techniques, including speech recognition, natural language processing, and automated summarization, to enhance workflow efficiency, accuracy, and documentation quality in veterinary practice. In one embodiment, the system's capabilities enable rapid and detaileddocumentation of veterinary appointments, supporting veterinarians in delivering optimal patient care and client communication. Embodiments can automate medical record documentation during (and otherwise, in connection with) patient appointments, enhancing efficiency and accuracy in veterinary practice.BRIEF DESCRIPTION OF THE DRAWINGS

[0026] A clear understanding may be had by reference to the appended drawings, which illustrate the method and system of embodiments, although it will be understood that such drawings depict embodiments and, therefore, are not to be considered as limiting its scope with regard to other embodiments. Accordingly:

[0027] FIG. 1 depicts an automated veterinary documentation system according to various embodiments.

[0028] FIG. 2 depicts an example user interface of the automated veterinary documentation system of FIG.1 according to various embodiments.

[0029] FIG. 3 depicts example user interfaces showing various functionalities of the automated veterinary documentation system of FIG.1 according to various embodiments.

[0030] FIG. 4 depicts an example user interface for a desktop application of the automated veterinary documentation system of FIG.1 according to various embodiments.

[0031] FIGS. 5 A and 5B illustrate workflows of obtaining the medical record of the system of FIG.l according to various embodiments.

[0032] FIG. 6 depicts an example user interface dashboard of a automated veterinary documentation system according to various embodiments.

[0033] FIG. 7 depicts an example user interface recording screen of a automated veterinary documentation system according to various embodiments.

[0034] FIG. 8 depicts an example user interface recording screen of a automated veterinary documentation system according to various embodiments.

[0035] FIG. 9 depicts an example user interface export screen of a automated veterinary documentation system according to various embodiments.

[0036] FIG. 10 depicts an example user interface settings screen of a automated veterinary documentation system according to various embodiments.

[0037] FIG. 11 depicts an example user interface team screen of a automated veterinary documentation system according to various embodiments.

[0038] FIG. 12 depicts an example user interface for a desktop application of the automated veterinary treatment documentation system according to various embodiments.

[0039] FIG. 13 depicts an example user interface for a desktop application of the automated veterinary treatment documentation system according to various embodiments.

[0040] In the drawings, like reference numerals refer to like parts throughout the several views of the drawings.DETAILED DESCRIPTION

[0041] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items. As used herein, the singular forms "a," "an," and "the" are intended to include the plural forms as well as the singular forms, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and / or groups thereof.

[0042] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one having ordinary skill in the art. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and the present disclosure and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0043] In describing embodiments, it will be understood that a number of techniques and steps are disclosed. Each of these has individual benefit and each can also be used in conjunction with one or more, or in some cases all, of the other disclosed techniques. Accordingly, for the sake of clarity, this description will refrain from repeating every possible combination of the individual steps in an unnecessary fashion.

[0044] Nevertheless, the specification and claims should be read with the understanding that such combinations are entirely within the scope of the claims. An automated veterinary treatment documentation system 100 and method thereof is discussed herein. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding. It will be evident, however, to one skilled in the art that various embodiments may be practiced without these specific details. The present disclosure is to be considered as an exemplification of embodiments and is not intended to limit the embodiments to the specific embodiments illustrated by the figures or description below. Embodiments will now be described by referencing the appended figures.

[0045] Disclosed herein is an automated veterinary treatment documentation system 100 and methods. In some embodiments, during appointments, the system 100 automates record-keeping with advanced speech recognition, categorizing data into Subjective, Objective, Assessment, Plan formats using large language models trained on vast amounts of text data, including medical literature, clinical notes, patient records, and other sources of medical information. In some embodiments, the system 100 includes real-time speech recognition that accurately converts dialogue into text, enabling seamless documentation. In some embodiments, the system 100 generates concise summaries from practice data, aiding managers in meetings. In some embodiments, the system 100 performs speech-based dental charting that translates observations into structured entries with visual mapping. In some embodiments, the system 100 employs machine learning techniques, including speech recognition and natural language processing to enhance workflow and documentation quality. In some embodiments, the system 100 supports veterinarians in delivering optimal care and communication within veterinary practice, streamlining processes for improved efficiency and comprehensive record-keeping.

[0046] In some embodiments, there is provided a method for performing automated veterinarydocumentation, the method including: receiving an audio input representing speech of at least one user; transcribing the audio input using at least one speech recognition model to generate a transcription; and generating a summary output of the transcription using at least one large language model, the summary output categorized into structured categories. The structured categories can be according to a SOAP format as further described herein. In some embodiments, speech to text converters (e.g., via API calls; fine tuned speech recognition models) receive the audio input and generate a textual transcription of that audio input. The textual output is stored by the automated veterinary documentation system 100 such as as in a local or remote database(s), and the textual output is received by a large language model(s) such as models that have been trained on and / or prompted with veterinary terms, medical information, and animal subjects and / or fine-tuned based on training data. The large language model(s) are configured to generate the summary ouput in structured categories. In some embodiments, one or more Al prompts are generated based on the textual output and the one or more Al prompts are used by the large language model(s) to generate the summary output in structured categories.

[0047] FIG. 1 depicts an automated veterinary documentation system 100 according to various embodiments. The example system 100 includes a first interactive device 102 associated with a first user 104 (e.g. veterinarian or veterinarian technician) who interacts with a second user 108 (e.g. pet owner), a processor 110 and a network 112. The processor 110 is communicatively connected with the first interactive device 102 associated with the first user 104 through the network 112. The first interactive device 102 may be a smartphone, laptop, desktop computer or similar device. The first user 104 is a veterinarian. The second user 108 is a client associated with a pet.

[0048] The processor 110 incorporates a database 114 alongside various specialized units aimed at optimizing and streamlining the process of documenting veterinary treatments. The database 114 serves as a central repository for storing critical veterinary data, including patient records, treatment histories, medication details, diagnostic reports, and client information. This database 114 is essential for maintaining comprehensive and organized records of veterinary treatments. The processor 110 includes various units that leverages advanced algorithms and predefined templates to generate accurate and detailed treatment documentation automatically. In some embodiments, the processor 110 is configured to execute instructions in non-transitory memory toconfigure a medical record creation unit 116; a dictation unit 128; a recap unit 130; a callback unit 132; a dental record creation unit 134; a multi-pet creation unit 136; a pack mode unit 138; an ultrasound unit 140; and / or an adaptive template unit 142. In some embodiments, processor 110 stores data in and retrieves data from one or more databases 114, which may be stored locally or remotely over a network, such as a local network or Internet network.

[0049] In some embodiments, automated veterinary documentation system 100 is configured to receive audio input following user request for actuation of audio capture. FIG. 7 and FIG. 8 depicts an example user interface recording screen of a automated veterinary documentation system 100 according to various embodiments. A user can select a large “record” button or other user interface element to initiate the audio capture by the system 100. Such audio capture can provide audio input for each of a medical record creation unit 116; a dictation unit 128; a recap unit 130; a callback unit 132; a dental record creation unit 134; a multi-pet creation unit 136; a pack mode unit 138; and / or an ultrasound unit 140.

[0050] The processor 110 includes a medical record creation unit 116. The medical record creation unit 116 is designed to streamline and automate the process of medical record documentation during patient appointments. The unit 116 operates seamlessly to capture, categorize, and summarize information provided by the veterinarian user 104 and the client user 108 during consultations. As the veterinarian conducts the appointment in the patient's room, the unit 116 utilizes advanced speech recognition technology to capture spoken dialogue accurately and in realtime. The transcribed content is automatically organized into structured categories using a categorization unit 118 based on the SOAP (Subjective, Objective, Assessment, Plan) format using large language models that are trained on vast amounts of text data, including medical literature, clinical notes, patient records, and other sources of medical information. The user can view the data after being organized into the structured categories, thereby allowing the user to easily see a structured output based on their audio input. In some embodiments, the large language model is GPT-4 (Generative Pre-trained Transformer 4) based on a transformer architecture, which uses self-attention mechanisms to process sequential data, such as text. The categorization unit 118 includes a subjective category sub-unit 120 configured to document veterinarian's observations, patient history, and information provided by the pet owner regarding the pet's condition and symptoms. The categorization unit 118 includes an objective category subunit 122 configured todocument objective findings such as physical examination results, diagnostic test outcomes, and vital signs. The categorization unit 118 includes an assessment subunit 124 configured to document the veterinarian's assessment and diagnosis based on the gathered information. The categorization unit 118 includes a plan subunit 126 configured to document detailed treatment plans, prescriptions, follow-up recommendations, and patient care instructions. In an example embodiment, the processor 110 is configured to processing a 30-minute veterinary appointment in approximately 5 minutes or less (e.g. 2 minutes). This rapid turnaround time significantly enhances workflow efficiency and allows veterinarians to focus more on patient care and less on administrative tasks. The medical record creation unit 116 may be customized to align with specific veterinary practice requirements, accommodating variations in terminology, templates, and reporting preferences.

[0051] The processor 110 includes a dictation unit 128. The dictation unit 128 employs advanced speech recognition algorithms to accurately capture and interpret spoken dialogue. This technology ensures high accuracy in transcribing veterinary professionals' verbal communication.

[0052] As the veterinarian speaks during consultations or examinations, the dictation unit 128 is configured to convert spoken words into written text. This transcription allows for efficient documentation without interrupting the natural flow of communication. The dictation unit 128 supports customizable vocabularies tailored to veterinary terminology, ensuring accurate transcription of specialized terms, pet breeds, medical conditions, and treatment protocols commonly used in veterinary practice.

[0053] The processor 110 includes a recap unit 130 that utilizes advanced algorithms to automatically generate generic summaries based on input data from various sources within the veterinary practice. This data may include patient records, appointment schedules, treatment logs, and operational metrics. The recap unit 130 is configured to generate concise and informative summaries suitable for practice managers during team meetings or rounds within veterinary clinics. Practice managers may customize the parameters and scope of summaries generated by the unit 130 to align with specific meeting agendas or areas of focus. In an embodiment, the recap unit 130 unit incorporates end-user prompting features. This functionality enables practice managers to request specific types of summaries or insights based on their immediate needs during meetingsor rounds.

[0054] In one embodiment, the processor 110 includes a callback unit 132 that facilitates the veterinarians to record their end of the conversation when providing updates to the second user 108 who is a client associated with a pet. The callback unit 132 allows veterinarians to capture important details regarding treatment progress, surgical outcomes, diagnostic findings, and followup recommendations. Alternatively, veterinarians may seek client consent to record the entire callback conversation. By putting the client on speakerphone and obtaining consent, the callback unit 132 records the full interaction, ensuring comprehensive documentation of client communications. The system 100 100 may be used as an ancillary tool to an existing documentation system 100 of the veterinarian.

[0055] In one example embodiment, the processor 110 includes a dental record creation unit 134 configured to perform speech-based dental charting. The veterinarians perform dental exams by verbally narrating their observations tooth by tooth. The unit 134 is configured to capture the narration of the veterinarian and translates spoken descriptions into structured dental chart entries. After the veterinarian describes each tooth during the exam, the unit 134 automatically populates a digital dental chart with corresponding findings for each tooth. This includes notations on tooth condition, abnormalities, dental disease, and treatment recommendations. In one embodiment, the unit 134 maps the recorded dental findings to a visual diagram of the animal's dental anatomy. Each entry in the dental chart is visually linked to the respective tooth in the diagram, providing a comprehensive and intuitive representation of the dental exam results. Upon completion of the dental exam, veterinarians may instantly print the populated dental chart with mapped tooth findings. This feature enables quick and efficient documentation for patient records and client communication.

[0056] In some embodiments, the processor 110 includes a multi-pet creation unit 136 configured to perform speech to text generation according to medical record unit 116, except for more than one patient at a time, with the output generated and categorized based on the patient it pertains to. The patient can be identified by name in the audio input. The output can be generated in a single SOAP output, with the data generated organized by the patient it pertains to in each category. Alternatively, the output can be generated in separate medical record or SOAP outputs or files.This can allow two or more pets to have their medical exam performed in the same room at a time. Audio is transcribed and inputted into a large language model(s) or other Al to output summaries and categorize the outputted text into SOAP categories.

[0057] In some embodiments, the processor 110 includes a pack mode unit 138 configured to perform speech to text generation according to medical creation unit 116, except for more than one patient at a time, with the output generated and categorized based on the patient it pertains to, the patients being identified in the audio input by descriptors (e.g., “kitten with pink eyes”). The output can be generated in a single SOAP output, with the data generated organized by the patient it pertains to in each category. Alternatively, the output can be generated in separate medical record or SOAP outputs or files. This can allow two or more pets to have their medical exam performed in the same room at a time. Audio is transcribed and inputted into a large language model(s) or other Al to output summaries and the outputted text is categorized into SOAP categories. In some embodiments, the large language model is configured to determine more than one subject (patient), which may be represented in the audio input as names or by description. The large language model can be configured by an input prompt that configures the large language model to determine and categorize input for more than one subject instead of simply one subject.

[0058] In some embodiments, the processor 110 includes an ultrasound unit 140 configured to perform speech to text generation for an ultrasound such as an abdominal ultrasound. Audio is transcribed and inputted into a large language model(s) or other Al to output summaries tailored to ultrasound assessment data and categorize the outputted text. The outputted text can be in a free structured note format, instead of a SOAP format.

[0059] In some embodiments, the processor 110 includes an adaptive template unit 142 configured to configure output generated by the system 100 (e.g., the SOAP output) into a template format specified by a user. The user can input data (e.g., by text or audio) to request particular types of data to be included in the output and / or position the output pertaining to the particular types of data in a specified location in the generated output. The user can also request particular types of data to be omitted from the output even if captured in the audio. The user’ s specifications including these requests for customization of the output can be saved as a template style and used as a format for subsequent output generated by the system 100, such as when the user requests that the outputfor a particular audio capture to have the template applied to the output. The system 100 is then configured to generate the output in the format requested according to the specifications saved in the template. Templates can also be shared via teams mode. Templates can be applied to output generated by medical record unit 116, dictation unit 128, recap unit 130, callback unit 132, and dental record creation unit 134, each of which, in some embodiments, can generate and categorize output in a SOAP format, which can be configured according to specifications requested in a particular template applied. In some embodiments, each of these can generate output in a freeform format, instead of a SOAP format. Examples of user specifications provided as input that can be saved in a template to be applied to configure output generated by the system 100 include inputs, such as inputs that specify tone, format, dictation or terminology or phrasing or units, content, organization. As examples, the following inputs can be accepted and applied by adaptive template unit 142 to configure the output generated.

[0060] Regarding tone, these can be instructions that set the style and “sound” of the output — how concise or detailed it is, the perspective (active voice), and how findings are summarized. Example inputs include the following phrases: Be concise. Be detailed. Be verbose. Summarize findings. Write in the active voice. Do not use pronouns. Format.

[0061] Regarding format, these can be instructions on how information is arranged and presented, covering layout (bullets, paragraphs) and emphasis. Example inputs include the following phrases: Write abnormals in all capitals. Write in paragraph form. Write in point form OR write in point form (use dashes). Maintain formatting. Only list the objectives mentioned. List only abnormal objectives. Bring abnormal findings to the top of the objective category.

[0062] Regarding dicatation style, these can be instructions governing the specific words or terminology to use, such as level of technicality, abbreviations, phrasing and even specified measurement units. Example inputs include the following phrases: Use the following Legend: Nrm = normal, Abn = abnormal, N / A = not applicable, N / E = not examined, Occ = occasional. Write in medical terminology. Write in layman's terms. Use (f / c) units of measurement. Use veterinary abbreviations.

[0063] Additional writing style instruction can be provided. Example inputs include the following phrases: Write follow-up recommendations based on the appointment. Write differentials basedon the appointment. Create a detailed plan based on the information provided. Create a discharge summary with these headings: What to Do. What to Watch For. When to Follow-Up.

[0064] Regarding content and organization, instructions can be guidelines on what to include or exclude based on specifics of each appointment in your note, how to categorize findings, and how to reflect treatments or recommendations. Example inputs include the following phrases: Include if the owner has declined treatments. If the patient is vomiting, list the following differentials: Pancreatitis, Foreign body and gastroenteritis. Include masses in the body system 100 they are found in. Include masses under the skin / integument category. Include Anal gland expression under rectal / under other / at the end of the objective section. Include all potential courses of treatments discussed with costs, even if declined (note if declined). If vaccines noted, note locations: e.g. Rabies RH SQ 3y, DHPP / Lepto LH SQ 3y / ly, Lepto LH SQ ly (if alone), Bord orally. No diagnosis in Subjective. Do not write subjective information in objective. Only include exam findings in objectives (no assumption of condition). Do not assume any follow-ups unless explicitly stated in the transcript. Do not list recommendations unless discussed in the appointment. Only use info in the transcript. If there are no findings, return “not mentioned”. Do not repeat information.

[0065] Various templates can include a SOAP format (as described herein), templates specified in freeform data input, and / or templates specified automatically (where the system 100 can detect a template format from an uploaded document that uses or contains the template). A user can customize each template, save each template, and specify which template to be used for a particular audio capture session or as outputted by a particular unit described herein. For templates specified by freeform data input, a user can type in data input that outlines the type of data to be captured, with the configuration of the data inputted also outlining the format that the data is to be outputted in. In some embodiments, the templates can be configured by a user by providing data input representing commands or requests that specify what types of data is to be included in the output and in what format. Templates created by the user can be saved as different types of templates applicable for different use cases, such as ER rounds, case summary, structured callback, and structured surgery. By request, a user can specify which template to apply to a particular output.

[0066] A referral code functionality may be provided to reward and incentivize the customer basewithin the veterinary community. In one embodiment, the system 100 includes personalized record-keeping capabilities to tailor documentation to individual patient needs and preferences. In one embodiment, the system 100 features a sync recording menu that allows the user to save recordings offline when internet connectivity is limited, enabling seamless synchronization with a cloud server at a later time. This functionality ensures reliable and uninterrupted documentation capture, even in challenging network conditions, enhancing efficiency and data integrity in veterinary practice management. In some embodiments, an audio recording generated from an audio capture is saved in fractional portions, such as in one minute pieces of the entire audio recording, and each fractional portion can be successively stored such as in a mobile device’s storage or at a remote storage. Storage of fractional portions can reduce file corruption, such as if the storage process is interrupted partway through saving the entire audio recording. The audio recording or the portions thereof that have been stored can later be transmitted to a remote storage such as when internet connectivity is enabled. In one embodiment, the users have unlimited access to all recording functionalities for free, except medical records which are capped based on usage. In one embodiment, upon sign-up, users receive 50 complimentary medical records, and thereafter can purchase a subscription based plan for unlimited medical record capacity. Other subscription plans providing various types of access to functionality and levels of functionality are permissible in various embodiments. In some embodiments, the users have access to callbacks, dictation, recaps, detal records, and abdominal ultrasounds for free. In some embodiments, a limited or full version of all features described herein are provided for free, except adaptive templates.

[0067] FIG. 2 depicts an example user interface of the automated veterinary treatment documentation system 100 of FIG.1 according to various embodiments. The user interface 200 includes all the recordings categorized and labelled according to the type of recording that includes the medical record, dictation, recap, callback and dental record. When a record is finished, it goes into the inbox and is labelled according to the type of recording.

[0068] FIG. 3 depicts example user interfaces showing various functionalities of the automated veterinary treatment documentation system 100 of FIG.1 according to various embodiments. The user interface 300 includes drafts functionality that includes incomplete or ongoing records that are being worked on. The interface 300 includes processing functionality that includes records currently being transcribed or processed. The interface 300 includes available for review tab thatincludes completed records ready for review and editing. The interface 300 includes ready to export functionality that includes records finalized and ready for export to other management software. The interface 300 includes exported functionality that includes records that have been successfully exported to external system 100s. FIG. 9 depicts an example user interface export screen of a automated veterinary documentation system 100 according to various embodiments. The interface 300 includes an “add” functionality that facilitates multi-device collaboration. In one embodiment, veterinarians and the user are enabled to contribute to the same patient record in realtime. The “add” functionality facilitates seamless transitions and combines notes created by different users, ensuring a cohesive and comprehensive patient record without duplication or gaps. The “add” functionality stitches together audio recordings from different devices, creating a unified and synchronized record of patient interactions and appointments. The “add” functionality allows a user to select a user interface element such as a button and actuate the stitching of new audio with existing audio stored by the system 100. The “add” functionality facilitates categorising different types of additions to the recording, such as continuing appointments, dictations, callbacks, and more. This categorization enhances accessibility of recorded data. In some embodiments, a user can specify the type of audio addition, such as audio from a continued appointment, medical record creation, dictation, callback, or dental record creation. The corresponding unit is actuated to capture the audio and generate the output as described in the foregoing and add it to the selected existing audio and output generated.

[0069] The interface 300 includes a “select and replace” functionality that enhances text editing capabilities for veterinary professionals, facilitating efficient and consistent modifications to patient records and documentation. The user can highlight a specific word within the document by clicking or tapping on it. After selecting a word, the user can replace all occurrences of that word throughout the document with a new term or phrase. The “select and replace” functionality offers options to find and replace individual words or phrases, enabling quick and comprehensive text modifications.

[0070] The interface 300 includes an “autosuggest” functionality that leverages a programmed list of veterinary terminology and / or artificial intelligence to provide intelligent word suggestions based on veterinary terminology, enhancing efficiency and accuracy in document creation. The “autosuggest” functionality suggests words and phrases relevant to veterinary medicine as userstype, helping to expedite document creation and ensure consistency in terminology. The Autosuggest feature continuously learns from user input and evolves its suggestions based on patterns and context, improving its accuracy and relevance over time. In some embodiments, the “autosuggesf ’ functionality generates the words and / or phrases based on context data such as the medical record of the user whose file is presently open or other data associated with that user (e.g., output generated from a unit described herein).

[0071] The interface 300 includes an “export flow” functionality that provides clients with a range of options to export notes from the system 100, offering convenient and customizable methods for sharing and storing patient records. Users can export notes or other records (e.g., SOAP data) via email in different formats, including as a shareable link or embedded text within the email body. In some embodiments, after a user completes a note (e.g., generates data output by the system 100 after capturing audio), the system 100 presents a user interface element such as a button that the user can click to quickly and easy export the note (e.g., generated output that may or may not have manual user changes) to a designated destination (e.g., a selected user account accessible via a remote computer such as a smartphone or personal computer) via a designated mechanism (e.g., email).

[0072] The interface 300 includes a “Show data” functionality that automatically extracts relevant data points from meeting transcripts, including topics, actions, decisions, and follow-up items. The Show Data tab presents extracted data in a structured format, enabling users to quickly navigate and access specific information without manually reviewing lengthy transcripts.

[0073] In one embodiment, the interface 300 includes a settings functionality that allows the user to define custom default words or phrases to be automatically inserted in specific scenarios, enhancing consistency and efficiency in veterinary documentation. FIG. 10 depicts an example user interface settings screen of a automated veterinary documentation system 100 according to various embodiments. The users can specify default expressions (words or phrases) to be used automatically based on predefined conditions or scenarios. The settings functionality facilitates defining conditions where the custom default word will be applied, such as for benign positive cases (e.g., where the audio input represents data that should not be recorded such as noting that a certain body part of an animal during a physical exam is unproblematic or “good”) or whenomitting specific body parts (e.g., where the audio input does not contain any data pertaining to a specific body part but that the output generated should still note as having been checked and noted as unproblematic during a physical exam) in notes. The defined default word or phrase will be automatically inserted into the document or generated as output whenever the specified condition is met during note-taking. When audio input at a different time represents data that should be recorded, the system 100 is configured to generate the corresponding output based on that audio data or insert the same into the document instead of the default words or phrases.

[0074] In some embodiments, the interface 300 includes a team mode setting functionality that facilitates seamless collaboration among individuals by enabling shared access to information and notes within a single account. In some embodiments, the team mode setting functionality enables teams mode to grant multiple individuals access to the same Scribenote account, fostering collaboration and information sharing. Changes made by team members are synchronized in realtime across devices, ensuring that everyone has access to the latest information and updates. In some embodiments, teams mode does not permit multiple user access to a single account providing access to the data, medical records, output generated, adaptive templates, and the like, but instead permits users to collaborated on the same each via their own user accounts that provide access to the same.

[0075] In some embodiments, a team can be managed via an account maintained by the system 100. A user account provides access to a team invite functionality that allows the user accessing the user account to invite other user accounts to a team associated with each user account. The user accounts invited may not yet exist at the time of invite. The invite can be sent to email addresses and contain an invitation to create an account and accept the team invitation or, if the user already has an account, simply accept the team invitation. A user account managing a team can specify user permissions and billings associated with their team members and access to data associated with or created by a user accessing the system 100 via the user account (e.g., output generated by the system 100). In some embodiments, multiple user accounts can be linked together, and files, recordings, outputs generated, data, and templates generated or stored by one account can be viewed and used by other user accounts in the team. FIG. 11 depicts an example user interface team screen of a automated veterinary documentation system 100 according to various embodiments.

[0076] FIG. 4 depicts an example user interface for a desktop application of the automated veterinary treatment documentation system 100 of FIG.1 according to various embodiments. The desktop application operates as a convenient widget directly accessible on the user's device. The desktop widget can be overlay ed over a user’s screen to allow the widget to be presented on the screen even while another application that appears behind the widget, at least for a portion of the other application, is selected. This widget interface streamlines the process of exporting completed medical records by seamlessly integrating with external programs. The user may simply click on the widget, select the desired medical record, and efficiently transfer the information into their target document or application through drag-and-drop functionality or copy-and-paste actions. The desktop application optimizes workflow by minimizing the need to switch between applications. The desktop application may employ machine learning and automation to eliminate manual importing steps, allowing for smoother and more intelligent integration of data into external systems. In some embodiments, the widget contains user interface elements corresponding to each of “Subjective”, “Objective”, “Assessment”, “Plan”, and / or “Full SOAP”, selection and dragging to another data input application (e.g., practice management software) of which copies and pastes the corresponding data associated with the particular user record into the other data input application.

[0077] FIG. 12 and FIG. 13 show an example user interfaces for a desktop application of the automated veterinary treatment documentation system 100, according to some embodiments. FIG. 13 shows previews of multiple summary outputs generated by the automated veterinary documentation system 100 based on audio input.

[0078] In some embodiments, the system 100 includes an “import” functionality where a user can request (e.g., by selecting a user interface element) importing of a record (or other data generated by the system 100) to another application, and the system 100 generates an output that maps to the corresponding other application and allows the other application to receive the outputted data according to the type and configuration of data input expected by that applicaton.

[0079] In some embodiments, the system 100 includes a client summary feature configured to generate a written output summarizing the data updates to the SOAP data for a particular medical record. The written output can be provided to the patient to summarize the medical exam.

[0080] In some embodiments, a writing assistant functionality is provided, which can be used by any of the units described herein to determine a format, style, content, data type, or other particular of output generated by system 100. For example, the output generated by system 100 by a large language model can be configured by the writing assistant according to particulars specified by a user, and a final output can be generated. The written output can be outputted in a specified language. The written output provides a summary, such as including of the medical exam with recommendations provided by the medical examiner, in a concise manner. The written output can be outputted in a specified format, such as in point form or paragraph form. The user can specify the language and / or format by selecting a user interface element of the system 100. The written output can be outputted with or without medical terminology, such as in a layperson summary. The written output can be outputted with or without abbreviations. These output specifications can be selected and chosen by a user, with the outputted summary generated upon request.

[0081] In one example user interface for a mobile application of the automated veterinary treatment documentation system 100, the mobile application focuses on delivering a more streamlined and concise user experience while retaining core functionalities. The mobile applications available on both iOS and Android platforms offer enhanced synchronization capabilities, allowing users to upload recordings seamlessly even when offline or experiencing connectivity issues. The mobile application enables users to manage patient-related notes and quick notes within a unified interface. FIG. 6 depicts an example user interface dashboard of an automated veterinary documentation system 100 according to various embodiments.

[0082] In some embodiments, the system 100 stores an association between a particular patient and a medical record or other output generated by system 100 based on the audio. For example, a user can mark or label a medical record as pertaining to a particular patient and organize, view, and / or access all data, including all medical records, associated with that particular patient.

[0083] In some embodiments, system 100 is configured to generate output from audio input captured by generating a textual transcript of the audio input using speech to text methods and use Al such as large language models to output data based on the textual transcript. In some embodiments, the outputted data is summaries structured according to the template applied (e.g., SOAP or a user specified template). The large language models are trained to determine andclassify text input according to data types desired in the output based on the template applied. For example, if data about a patient’s eyes is to be included in the output as deteremined by the template to be applied to the output, the large language model is configured to generate data that summarizes the transcript data pertaining to the patient’s eyes, such as remarks that a veterinary doctor provided during a medical exam in a medical record mode implemented by medical record creation unit 116.

[0084] FIGs. 5 A and 5B illustrate a workflow of obtaining the medical record of the system 100 of FIG.1 according to various embodiments. At 602, the workflow includes initiating a recording of the spoken dialogue of a veterinarian during a consultation for a pet. The 604, the workflow includes transcribing the spoken dialogue of the veterinarian during consultation using a speech recognition model. At 606, the workflow includes categorizing the transcribed content into structured categories based on the SOAP (Subjective, Objective, Assessment, Plan) format using large language models that are trained on vast amounts of text data, including medical literature, clinical notes, patient records, and other sources of medical information. At 608, the workflow includes exporting the medical records to external systems.

[0085] The systems and methods may be implemented as cloud-based systems and methods which operate as a cloud by communicating using one or more communication networks (e.g., the Internet). In some embodiments, these services may be offered as web based or cloud services or under Software as a Service (SaaS) model to the users of systems and methods (i.e. services and / or applications) described herein. The services and / or applications may be provided via a communication network, such as the Internet.

[0086] Although embodiments have been illustrated and described herein with reference to preferred embodiments and specific examples thereof, it will be readily apparent to those of ordinary skill in the art that other embodiments and examples may perform similar functions and / or achieve like results. All such equivalent embodiments and examples are within the scope of various embodiments.

Claims

CLAIMS1. A system for performing automated veterinary documentation, wherein the system comprises: at least one processor; and one or more non-transitory memory units communicatively coupled to the at least one processor and storing machine-readable instructions that, when executed, cause the at least one processor to perform at least the following steps: receive an audio input representing speech of at least one user; transcribe the audio input using at least one speech recognition model to generate a transcription; and generate a summary output of the transcripton using at least one large language model, the summary output categorized into structured categories.

2. The system of claim 1 , wherein the structured categories are based on a SOAP (Subj ective, Objective, Assessment, Plan) format.

3. The system of claim 1 , wherein the structured categories are based on at least one adaptive template defining at least one data type and at least one format for the summary output.

4. The system of claim 1, wherein the steps further comprise documenting the summary output in a medical record using a medical record creation unit.

5. The system of claim 1, wherein the at least one speech recognition model comprises a dictation unit configured to transcribe the audio input into the transcription, the dictation unit configured based on veterinary terminology.

6. The system of claim 1, wherein generating the summary output comprises generating summaries of the transcription using a recap unit.

7. The system of claim 1 , wherein generating the summary output comprises using a callback unit on the audio input, the audio input representing client communication.

8. The system of claim 1, further comprising performing speech-based dental charting by capturing the audio input, the audio input representing a narration of dental examination by a first user in real-time, and wherein the structured categories are structured dental chart entries.

9. The system of claim 1, further comprising performing an ultrasound assessment documentation by capturing the audio input, the audio input representing a narration of an ultrasound assessment by a first user in real-time, and wherein the at least one large language model is trained on ultrasound terminology.

10. The system of claim 1, further comprising performing medical assessment documentation by capturing the audio input, the audio input representing a narration by a first user in real-time of a medical assessment of at least one subject, and wherein the summary output is categorized into the structured categories for each subject, wherein each subject is defined in the audio input by a description.

11. The system of claim 1, wherein the summary output comprises predefined expressions inserted based on predefined conditions represented in the audio input.

12. The system of claim 1, wherein the processor is configured to accurately capture and interpret the spoken dialogue using speech recognition algorithms.

13. The system of claim 1, wherein the summary output represents treatment progress, medical assessment, surgical outcomes, diagnostic findings, follow-up recommendations, client updates about a subject, or any combination thereof of a subject.

14. A method for performing automated veterinary documentation, the method comprising: receiving an audio input representing speech of at least one user; transcribing the audio input using at least one speech recognition model to generate a transcription; and generating a summary output of the transcripton using at least one large language model, the summary output categorized into structured categories.

15. The method of claim 14, wherein the structured categories are based on a SOAP (Subjective, Objective, Assessment, Plan) format.

16. The method of claim 14, further comprising generating the summary output to include at least one data type and at least one format, each defined by at least one adaptive template.

17. The method of claim 14, further comprising performing speech-based dental charting by capturing the audio input, the audio input representing a narration of dental examination by a first user in real-time, and wherein the structured categories are structured dental chart entries.

18. The method of claim 14, further comprising performing an ultrasound assessment documentation by capturing the audio input, the audio input representing a narration of an ultrasound assessment by a first user in real-time, and wherein the at least one large language model is trained on ultrasound terminology.

19. The method of claim 14, further comprising performing medical assessment documentation by capturing the audio input, the audio input representing a narration by a first user in real-time of a medical assessment of at least one subject, and wherein the summary output is categorized into the structured categories for each subject, wherein each subject is defined in the audio input by a description.

20. The method of claim 14, further comprising inserting predefined expressions in the summary output based on predefined conditions represented in the audio input.

Images