Artificial intelligence-based system and method for training and education

Abstract

To streamline and automate content creation, and ensure process compliance, memory retention, and improvement in the quality of education and training.SOLUTION: A system for training and education includes a server that collects inputs from users through user devices and receives these inputs through a communication network. The server includes a memory and a processor that executes instructions from the memory. A query processing and scripting generation module uses artificial intelligence to generate a draft script from information acquired from databases. A compliance check module verifies the draft script on the basis of predefined parameters, and generates structured data. An audio / video search engine utilizes artificial intelligence to generate diverse audio / video contents on the basis of structured data. A review module evaluates the content's relevance, creates concise audio / video segments from the pertinent material, and filters out irrelevant content.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a training and education platform, and more particularly to an artificial intelligence-based system and method for assisting in content creation and its process for training and education, for example, for experts in the medical industry.

Background Art

[0002] Many factors are involved in decision-making regarding training and education, including the characteristics of trainers and trainees, educational levels, personal beliefs, perceptions, behaviors, and criteria and guidelines that define educational paths. Furthermore, guidelines set by educational and training systems and related parties also increase the complexity of decision-making regarding content selection. Since it is necessary to consider these multiple parameters, the decision-making process becomes extremely complex. Such multi-dimensional requirements are further complicated by issues such as the complexity of content search, filtering of relevant materials, and the cost burden, especially in highly skilled fields such as medicine.

[0003] Furthermore, training for professional occupations in the medical field, for example, faces the following problems: a mismatch in skills between medical practitioners and those on the side of transmitting important information, a lack of knowledge among the parties involved, a lack of regular skill improvement, a lack of training in the ability to provide individualized support, etc. This results in problems such as burnout syndrome, an increase in the turnover rate, inefficiency, excessive costs, redundancy, etc., and ultimately leads to a loss of profits.

[0004] With the rapid increase in the adoption of digital marketing strategies, pharmaceutical brands have access to large amounts of data utilizing artificial intelligence (AI) in order to build engagement with healthcare professionals (HCPs). AI uses machine learning, natural language processing (NLP), and deep learning to analyze HCP and customer data and help increase HCP engagement through personalization. Also, AI can derive insights and assist in creating more effective content. Furthermore, AI can also help to derive insights and create more accurate segments of target HCPs. This enables the formation of cluster groups based on specific characteristics, for example, to personalize communication. Additionally, it is also possible to analyze the behavior of the target audience in different industries and specialties using predictive models. There is also an opportunity to enhance engagement by leveraging chatbots, but currently, it is mainly limited to NLP.

[0005] Furthermore, healthcare professionals (HCPs) require continuous training and education to stay updated with the latest findings. While providing education is important, face-to-face interactions with HCPs are still cumbersome and inefficient. Also, the content of the training mainly depends on the patient groups that HCPs handle. AI may be useful in creating high-quality training videos, but there is a need to maintain continuous engagement and ensure relevant content.

[0006] Medical institutions are spending a large amount of money on many non-productive efforts and human customer engagement staff, which ultimately increases costs. In the past, attempts have been made to solve these problems. Patent document WO2019207456 A1 discloses a system that provides details of products and services of a life sciences company to multiple HCPs along with slide content through a bot-form salesperson. This system includes a plurality of computing devices, an authoring module, an information module, an HCP behavior processing module, a campaign module, an interactive communication module, and a communication network. Although these tools can cost-effectively provide product and service information to healthcare service providers and customers, due to the complexity of the content accessible to users and the rapidly increasing amount, the level of engagement is still not sufficient.

[0007] While these technologies are very useful, they are still in their infancy in terms of the range of content that these systems can provide and how end-users consume it. Therefore, there is a strong demand for the development of AI-based systems and methods that streamline and automate content creation, ensure process compliance, enhance memory retention, and improve quality in education and training. Summary of the Invention Problems to be Solved by the Invention

[0008] The present invention aims to address the problems of conventional computer-assisted training and education platforms for new hires and professionals in specific fields. This includes closing skill gaps, reducing turnover rates, cutting costs, improving overall efficiency, empowering learners, creating growth opportunities, and promoting a dynamic and personalized learning experience.

[0009] One objective of the present invention is to provide a system and method that facilitate, rationalize, and automate content creation, ensure process compliance, memory retention, and educational quality in the training and education of market experts within the industry.

[0010] Another objective is to utilize AI to curate content, leverage feedback, make data-driven recommendations, individually optimize the engagement between pharmaceutical companies / wholesalers / distributors and healthcare professionals (HCPs), and improve the continuous learning experience.

[0011] Yet another objective is to achieve the provision of a high affinity score, a high Net Promoter Score (NPS), a highly reliable content source, a user-centric system, and on-demand verified generative AI answers.

[0012] Another objective of the present invention is to enhance sales efficiency through integration with Sales Force Automation (SFA).

[0013] In addition, the present invention aims to establish an internal guidance system. SUMMARY OF THE INVENTION

[0014] It should be understood that the present disclosure is not limited to specific systems and methodologies, and there may be multiple embodiments not explicitly illustrated in the present disclosure. Also, the terms used herein are for the purpose of describing a particular version or embodiment and are not intended to limit the scope of the present disclosure.

[0015] Described herein are, for example, artificial intelligence (AI)-based systems and methods for the training and education of medical practitioners. In one embodiment, the system includes a plurality of user devices configured to receive input from one or more users. The server is configured to receive user input through a user interface via a communication network and obtain information from one or more databases, such as a medical research database related to the training and education of medical practitioners. This server includes a memory and a processor configured to execute instructions stored in the memory. Further, a query processing and script creation module is configured to process query input from a user, generate a draft script using AI based on information obtained from one or more databases, and save it. A compliance check module is configured to verify a draft script based on one or more predefined rules stored in the server and generate structured data from the verified script. An audio / video search engine is configured to generate a plurality of audio and / or video contents using AI based on structured data. The review module is configured to receive the generated audio / video content and determine its relevance based on predetermined parameters, generate an explanatory snippet from a compliance-checked script and an audio / video snippet based on the relevant content, exclude irrelevant content, and link the explanatory snippet and the corresponding audio / video snippet based on a matching coefficient. The output module displays the linked content on the user device.

[0016] In another embodiment of the present invention, it is disclosed that the relevance of the audio / video content is determined by predetermined parameters such as an affinity score, and the output is ranked based on the opinions, comments, transcripts, published papers, etc. of famous individuals / scientists / practitioners.

[0017] In yet another embodiment of the present invention, it is disclosed that the system learns about the user through assessment, selects a rank based on the user's proficiency, classifies the user as a beginner, practitioner, or expert according to the content or experience, and establishes a positioning within the system.

[0018] In another embodiment of the present invention, it is disclosed that the compliance check module is further configured to perform scrutiny and verification for compliance of the audio / video content.

[0019] In a preferred embodiment of the present invention, it is disclosed that the review module provides a review and retake function that enables iterative improvement of the audio / video content and the compliance-checked script, and provides snippets of relevant content.

[0020] In yet another embodiment of the present invention, it is disclosed that the review module is further configured to perform a process of rerecording a part of the relevant audio / video information and selectively modifying sections based on inputs from at least one stakeholder.

[0021] In another embodiment of the present invention, the users include healthcare professionals (HCPs), medical representatives (MRs), medical science liaisons (MSLs), etc. Further, it is disclosed that doctors, pharmacists, caregivers, and other relevant persons in medical institutions may be included.

[0022] In a preferred embodiment of the present invention, the predefined parameters include prescription preferences, opinions on brands or ingredients, behavioral characteristics, tendencies, medication compliance, behavioral patterns, etc.

[0023] In yet another embodiment of the present invention, it is disclosed that the server is configured to store and access information, media content, questionnaires (pre- and post-evaluations), reference documents, medical journals, marketing materials, medical communication materials, as well as training and educational content, etc., obtained from various information sources related to products and services.

[0024] In a preferred embodiment of the present invention, it is disclosed that the communication network includes either wired or wireless means and is used for data communication between the user device and the server.

[0025] In one embodiment of the present invention, it is further disclosed that the review module is configured to link the targeted description with the corresponding part of the compliance-checked script.

[0026] In yet another embodiment of the present invention, it is disclosed that the draft script can be edited by the user and can provide personalized feedback.

[0027] Also, the user query is processed by at least two generative AI technologies, generating two or more draft scripts, which are compared based on pre-stored parameters such as user recognition, persona, knowledge level, accuracy, relevance, comprehensiveness, clarity, context, etc. The draft scripts are updated simultaneously.

[0028] In a preferred embodiment of the present invention, it is disclosed that instructions are stored in a non-transitory computer-readable medium. When these instructions are executed by one or more processors of a computing system, they cause the following method to be executed. Input is received from one or more user devices. The user's query is processed as input, and a draft script is generated by artificial intelligence using information retrieved from one or more databases, and the resulting data is saved. Subsequently, the draft script is verified based on pre-defined rules / parameters stored on the server. Structured data is generated from the compliance-checked script. Furthermore, based on the structured data obtained using artificial intelligence, a plurality of audio / video contents are generated. The relevance of the generated audio / video content is determined using predetermined parameters. This process includes creating a naget (summary) of the description from the compliance-checked script, generating a naget of the audio / video content as well, and excluding less relevant content. The description naget is linked to the corresponding audio / video naget based on a match coefficient. The linked content is displayed on the user device.

[0029] In another preferred embodiment of the present invention, a non-transitory computer-readable medium discloses the following method. The method includes receiving, by one or more processors, a search query from a digital assistant application running on a remote client device.

[0030] In yet another embodiment of the present invention, a non-transitory computer-readable medium discloses that structured data is generated. This structured data includes the creation of multi-category classification data based on scientific or clinical documents, diagnoses, treatments, and drug names, and includes a plurality of metadata categories.

[0031] Furthermore, in this embodiment of the present invention, user-originated feedback on a naget leads to the modification of the remaining nagets in the naget list.

[0032] In a preferred embodiment of the present invention, a computer-implemented method for promoting personalized content creation and compliance for training and education purposes is disclosed. Various inputs from one or more user devices are received as queries. At least one user query is processed, and as a response to the query, a draft script is generated using artificial intelligence based on information obtained from one or more databases. The draft script is verified based on one or more predefined rules stored on a server, and structured data is generated from the compliance-checked script. Based on the structured data obtained using artificial intelligence, a plurality of audio / video contents are generated. The relevance of the generated audio / video contents is determined based on predefined parameters. The review module is configured to receive the generated audio / video contents and determine their relevance based on predefined parameters, generating a list of descriptive nagets derived from the compliance-checked script and a list of audio / video nagets based on the related contents, and filtering out irrelevant contents. The descriptive nagets are linked to the corresponding audio / video content nagets based on a match factor. The output module displays the linked contents on the user device.

[0033] In one embodiment of the present invention, a computer-implemented method is disclosed, and the "judgment" in the method is shown to include review and retake (re-recording) functions that enable iterative improvement of audio / video content and a compliance-checked script. As a result, nagets are provided.

[0034] In a preferred embodiment of the present invention, a computer-implemented system is disclosed, in which related content is further processed, and based on input from stakeholders, a part of the related audio / video information is partially retaken, and a selectively modified section is generated.

[0035] Various objectives, features, aspects, and advantages related to the subject matter of the present invention will become clearer through the following detailed description of the preferred embodiments and the accompanying drawings. In the drawings, the same reference numerals indicate similar components.

Brief Description of the Drawings

[0036] Non-limiting examples of the present disclosure will be described in the following description with reference to the accompanying drawings.

[0037]

Figure 1

Figure 2

Figure 3

Figure 4

[0038] Some embodiments of the present invention will be described in detail below as illustrative of all features. "comprising( ~ including)", "having( ~ having)", "containing( ~ including)", "including( ~ including)" and other forms thereof have equivalent meanings and are used as open expressions, not intending that the listed items are limiting. That is, the items following these words do not mean an exhaustive listing of the items, nor are they limited to the described items.

[0039] Also, the singular forms "a", "an", "the" used in this specification and the appended claims are to be construed as including the plural unless the context specifically dictates otherwise. In practicing the present invention, systems and methods similar or equivalent to those described herein can be used, but preferred systems and methods will be described below.

[0040] Particularly when the training for pharmaceutical company personnel and liaison officers is insufficient or suboptimal, it may lead to consequences such as a decrease in sales, a decline in the company's reputation, legal problems, penalties, and loss of credibility. In training targeted at professionals in the market (e.g., medical practitioners), the lack of relevant content and personalization creates a divergence from industry experts such as medical practitioners, hinders the optimal utilization of the final product, and has an adverse impact on revenue.

[0041] The present invention discloses an innovative platform to address the significant challenges faced by marketing departments of pharmaceutical companies, medical device companies, wholesalers, etc. The platform addresses issues such as the skill mismatch of market professionals with industry needs, regular and personalized white paper commissioning (cult brand) competency training, high burnout / attrition rates, inefficiencies / excessive costs / redundancies.

[0042] When the platform curates content, utilizes feedback, and makes data-driven recommendations, it personalizes the engagement between market professionals such as medical information representatives (MRs) and industry professionals such as healthcare providers (HCPs), and promotes an improved process of continuous learning based on AI and user-specific interventions.

[0043] This methodology enhances the roles of MRs and marketing specialists, leading to improved patient outcomes by enabling higher-level opportunities and quality care through continuous education and skill building.

[0044] The disclosed systems and methods provide content that can be quickly acquired in short, digestible chunks while summarizing the latest research, assisted by GenAI (generative AI), and simulate question-and-answer (Q&A) sessions with industry professionals. Specifically for healthcare providers (HCPs), it provides on-site support and access to complex real-world cases and the insights of knowledge leaders, enabling accelerated product understanding and information-based decision-making.

[0045] Figure 1 shows a network diagram 100 for implementing an AI-based system 104 (hereinafter referred to as "system 104") for training and education for professionals such as medical market professionals. As shown, a plurality of user devices 102-1, 102-2,..., 102n (hereinafter collectively referred to as "102") can communicate with a server 108 via a network 106, and the server 108 is configured to host the system 104. In another embodiment, the user device 102 may execute a digital assistant application (not shown), and the application may be a computer-based software application or a web application. For example, the digital assistant application may be integrated with a customer relationship management (CRM) application or an electronic medical record (EMR) application, or may be provided in the form of flashcards or a stand-alone web application. The digital assistant application can communicate with or form part of system 104. For example, user device 102 is used by a doctor 102-1, a marketer 102-2, a patient 102-3, etc., who wish to receive or send information to obtain training materials regarding a certain medical topic or health condition. Such a distributed configuration of server 108 can be provided in the form of a distributed computing network in which multiple devices are interconnected via network 106 and managed by ecosystem members. All or part of system 104 is executed by this network. Inputs from stakeholders may include public content and comments such as recommendations and public feedback provided by companies, HCPs (healthcare providers), industry experts, etc. Therefore, this solution can be integrated into the medical field, the CRM system of a pharmaceutical company, and other locations.

[0046] Server 108 may be a remote server or a cloud server, and is configured to access multiple databases 110 to obtain the latest industry-specific information and data in real time. Communication network 106 may be either a wired or wireless system and is used for data transmission. Network 106 enables seamless communication and data exchange between user 102 and server 108. Server 108 is adapted to store a training dataset 112 for system 104 to automatically learn using machine learning.

[0047] System 104 is configured to process and analyze user queries received as input and ensure efficient interaction and information flow. The user queries are processed by System 104 and at least two draft scripts are generated using at least two artificial intelligence technologies (e.g., ChatGPT, Gemini, etc.). In a preferred embodiment, System 104 is configured to compare each generated script based on pre-defined rules / parameters and present the optimal and personalized draft script for the user's query. This comparison includes, but is not limited to, the user's cognition / persona, knowledge criteria, content relevance, comprehensiveness, clarity and context, and reflection of the latest information at the same time, which System 104 has grasped from past interactions with the user.

[0048] System 104 processes the query to generate one or more draft scripts, and then executes a compliance mechanism to convert the draft script into a compliance-checked script. In one embodiment, the draft can be manually edited by a privileged user, e.g., content creator user 102-2, both before and after the content is published, to add personalization. In the operation of a preferred embodiment, System 104 provides a draft script based on the action recognition, persona, and knowledge criteria of the user or target audience learned from the interaction between user 102 and System 104. Subsequently, the compliance mechanism is applied and the draft script is converted into a compliance-checked script based on pre-defined and pre-saved parameters such as validity, accuracy, comprehensiveness, clarity, and currency as content approval by industry experts. During the compliance check process, the latest industry information is compared with past data, and the updated content at the same time is incorporated into the content, ensuring consistency and enhancing the accuracy and completeness of the content, which is essential for effective decision-making by medical professionals. In one embodiment, a timestamp for use in training and educational purposes is applied to the draft script, the compliance-checked script, or the final content, either before or after publication. The timestamp is stored in the system 104 and used to check the update status of the content by comparison with scripts to be generated in the future. Furthermore, human information sources may also be included in the provision of the compliance-checked script. For example, in the medical industry, it includes content from medical professionals, doctors, and medical transcription. Anonymized data (e.g., transcription data from doctors who have opted in for learning promotion) will be used in the future as intelligence for learning and content generation.

[0049] In one embodiment, the content may be approved by at least one healthcare professional (HCP) or an expert in the field, weighted based on knowledge, experience, expertise, affinity, and relevant parameters. From the compliance-checked script, the system 104 generates structured data, and using the structured data, audio / video content is generated by another AI-based audio / video generation engine. In one embodiment, the generation of structured data includes, for example, the generation of classification system (taxonomy) data in multiple categories based on scientific or clinical documents, diagnoses, treatments, and drug names, and includes metadata in multiple categories.

[0050] The audio / video content generated based on the structured data in this way is again subject to a compliance check from perspectives such as relevance, effectiveness, and legal guidelines. Furthermore, filtering is performed based on a defined set of rules (such as content differences, compliance, length, etc.), and a list of nuggets (fragmentary information) of relevant audio / video content is provided. The obtained audio / video content is ordered based on its consistency (match factor) with the content of the compliance-checked script and is distributed as subtitles to the scenes of the audio / video linked to the relevant descriptive nuggets. The content generated in this way is provided to the user as fragmentary (piecemeal) learning material along with the description, maximizing the user's engagement with the teaching materials.

[0051] During operation, user 102 needs to register with system 104 to start an educational or training process. At this time, in addition to basic information, the user provides information regarding their educational / vocational history and field of expertise (if applicable). Based on the information received at registration and the specific wishes / requirements of user 102, the system presents the user with various course options and recommendations. These options and recommendations include, for example, a specific field, course level, duration, and the number of hours of study possible per week. Depending on the complexity of the course selected by the user and the desired learning duration, system 104 can design a personalized course tailored to the user. In one embodiment, system 104 evaluates the user's behavior pattern and prepares course materials based on it. In another embodiment, system 104 presents a specific number of options and mandatory / regular courses based on the initial information provided by the user in relation to the course, enabling the user to select the theme that they are most interested in or in need of at that time. System 104 provides the optimal information nuggets to maximize the user's engagement with the course based on the said selection and other information provided by user 102 (such as the course duration), as well as the information learned by system 104 from user 102. These information nuggets are part of a draft script, and compliance checks, i.e., checks based on relevance, accuracy, comprehensiveness, and currency of the information, are performed, and a compliant script is provided that presents the information nuggets suitable for the user. The nuggets are divided to maximize the user's attention. In one embodiment, the system 104 enables a user, or a user checking the compliance of the draft script, to provide feedback on a particular nugget, and based on that feedback, a list of nuggets or modifications corresponding to the content of the subsequently generated nuggets is made.

[0052] In one embodiment, the system 104 enables the user to personalize the nuggets according to their skill level. Contextual information flows between interconnected nuggets, and the system 104 adapts to the user's personalization choices and provides an average number of nuggets for the selected course.

[0053] In one embodiment, the system 104 is configured to curate (select and edit) user - centric content with high affinity scores, high net promoter scores, reliability, and click - through rates, and is capable of providing answers by verified generative AI as needed. This involves calculating affinity scores for various attributes, with particular emphasis on recent interactions and the user's preferences, giving high weights to real - time behavior and high engagement. At an initial stage, an individual affinity score based on specific variables is defined for the "neighborhood" of each participant. Next, these multivariate affinity score profiles are utilized by diagnostic verification and classification algorithms. Affinity - based classification involves splitting the data into training samples and test samples, performing cross - validation on the training data, and then comparing with weighted k - nearest neighbor (KNN) classification.

[0054] There are several purposes for the implementation of this affinity score calculation. For example, it includes assisting in treating various patients based on deep insights into disease symptoms, as well as enhancing sales by educating or training patients. It may also include healthcare providers (HCPs) who incorporate data from individuals with chronic diseases and treatment-resistant conditions, as well as healthy control groups. The individualized affinity score facilitates training or education for the target individual based on the creation of relevant content and enables the generation of audio or video "nuggets". These nuggets correspond to existing influencers that exist between the provider and the trainee, and form important links in the learning process. For example, there may be four influencers identified between the provider and the trainee. Furthermore, the influencers themselves can also obtain the knowledge of the course through the trainees, and a learning flow that chains from individual to individual is constructed. By achieving high accuracy in training, nested cross-validation, and prediction steps, the affinity score-based classification outperforms the K-Nearest Neighbor (KNN) classification in both the training dataset and the test dataset.

[0055] Specifically, the "affinity score" captures patterns related to treatment responsiveness, treatment resistance, or sales improvement, and indicates the degree of affinity an individual has for multiple diagnostic groups across various variables. This individualized approach assists clinicians in considering comorbidities, severity, disabilities, prognosis, and adjusting intervention strategies accordingly. The present invention integrates the affinity score into diagnostic verification, classification, and prediction algorithms, and functions as a clinical decision support system aimed at establishing diagnoses and predicting prognoses. Finally, using k-means clustering, it is evaluated whether individuals with treatment resistance can be separated from healthy controls and those with chronic diseases based on the multivariate affinity score profile.

[0056] The present invention provides on-demand personalized rapid reskilling solutions via a training and education platform. This system enables a rapid learning curve in the medical field by providing rapid skilling content that summarizes the latest research findings in a concise and understandable format. With the support of generative artificial intelligence (Generative AI), simulated question-and-answer sessions with healthcare professionals (HCPs) are conducted to ensure rapid understanding and retention of information, contributing to the acceleration of the learning curve.

[0057] Furthermore, this system incorporates a "Just-In-Time Customer Centricity" function specifically designed for the medical field and other education-related applications, enabling medical science liaisons (MSRs) to quickly provide the preparatory content required immediately before meeting with healthcare professionals (HCPs). This content is provided in "bite-sized" form, condensed into key points, enhancing the engagement of HCPs and improving satisfaction with more reliable and accurate information.

[0058] This system streamlines the content creation process by creating AI-generated visual materials from the final approved script in just a few minutes. The selection of narration, background music, and emphasis of subtitles are also simplified, and AI support functions are available that can significantly save time and cost for content creators and agencies. The present invention enables the generation of rapid skilling content in the medical field, providing numerous functions and advantages. It simplifies content creation by leveraging AI-generated visual materials and can be implemented within minutes from the final approved script. This system can simplify the selection of narrators and background music and efficiently perform subtitle highlighting processing. The availability of AI assistance realizes valuable time and cost savings for medical writers, content creators, and agencies.

[0059] Another notable function realized by this system is the peer connection function, especially in the medical field. This peer connection builds collaborations among medical information representatives (MSR) / medical science liaison (MSL) within the company and with a panel of "friendly" healthcare professionals (HCP), providing on-demand and reliable domain knowledge approved by experts. This enables the acquisition of insights from key opinion leaders (KOL) and promotes the deepening of customer understanding.

[0060] This system promotes peer-to-peer support among healthcare professionals (HCP) within medical institutions and supports HCP decision-making through on-site medical support and access to complex actual cases and insights from KOL (key opinion leaders). This effort aims to accelerate the improvement of drug awareness and enable information-based decision-making by HCP.

[0061] This system provides a pre-survey function to customize courses according to the skill levels of learners and optimize learning efficiency by skipping unnecessary content. The performance of medical information representatives (MSR) and medical science liaison (MSL) is monitored through voting, quizzes, and badges, promoting active participation. Using AI / ML algorithms, ranks of doctors, nurses, MSR / MSL are determined based on quiz results, classified as beginners or experts, and individual certifications are awarded to each. By personalizing content nudges using pre-surveys, the learning experience is personalized and the required training time is significantly reduced.

[0062] In the medical field, through the "Engage" function, exciting polls and post-surveys are conducted to create a sense of community and relevance to the content, enhancing learner satisfaction. To promote the elimination of skill gaps and improve patient outcomes in MSR / MSL, the medical field incorporates learning using generative artificial intelligence (Generative AI). In particular, post-surveys are used for certification purposes and enable further personalization. To achieve even better results, other personal data is also utilized to deepen personalization.

[0063] In the medical field, AI-driven courses revolutionize the content creation process, saving time for content creators. By adjusting content based on pre-surveys and incorporating engaging polls and quizzes, the learning experience of MSR / MSL is improved, fostering a sense of community and satisfaction. Education using generative AI further strengthens the skills of MSR / MSL, ultimately improving patient treatment outcomes.

[0064] The present invention provides real-time insights for a "just-in-time" customer centricity and informed engagement approach. This function enables medical science liaisons (MSR) to effectively engage based on the information of current healthcare professionals (HCP). In the medical field, this system collects the latest research from information sources such as Pubmed, Umin, and social networking services (SNS), and provides a script that summarizes the key points. MSR can read these summaries, convert them into audio, or present them as short visual videos using Content Creator AI. This enables MSR to accurately engage with HCPs at an appropriate level.

[0065] The smart HCP tracking and updating function in medical institutions actively monitors the engagement status of HCPs. This includes changes in trends in personas and community segments, such as attendance records at public events, lecture activities, event visit histories, changes in areas of interest, and changes in relationships with colleagues. This information content is provided to medical information staff (MSR) in real time, enhancing the timing of engagement and the accuracy of content.

[0066] To improve the quality of conversations in the medical field, the present invention presents the main questions that MSR / MSL should ask during engagement with HCPs and provides essential guidelines. This guidance includes a simulation of a question-and-answer session that includes expected answers and predicted responses, assisting in the preparation of MSR / MSL and promoting deeper and more engaging conversations.

[0067] With the present invention, engagement with healthcare professionals (HCPs) is revolutionized, enabling more information-based and effective conversations, accelerating market penetration, and achieving unprecedented improvements in medical outcomes. Through the transformation of content generation by AI, the time and costs of content producers can be saved. Furthermore, education using generative AI improves the skills of MSR / MSL and also enhances patient treatment outcomes.

[0068] This system promotes peer-to-peer connections (a network among colleagues) and offers multiple functions and benefits, such as access to communities, access to medical knowledge, engagement with HCPs, and ensuring compliance with training. In the medical field, this system connects novice MSRs / MSLs with experienced colleagues and promotes knowledge sharing and skill support within the community.

[0069] Furthermore, medical institutions can achieve knowledge sharing and skill support by collaborating experienced colleagues with the "Friendly HCP (Healthcare Provider)" group. This platform enables practitioners and expert-level MSRs / MSLs to connect with HCPs, share timely expertise on drug diagnosis, prescription, treatment, evaluation, and care, and accelerates conversion (achieving results) by facilitating access to the "Friendly HCP group". For example, the Friendly HCP group on the service provider side can include registered industry experts within the industry who are candidates for recommending pre-release content, and the Friendly HCP group on the client side can include experts registered in the client's panel.

[0070] Furthermore, in the medical field, this system 104 monitors professionals such as the new generation, mid-career, and senior levels at individual content levels (beginner level, practitioner level, expert level). The content varies according to the targeted learners (such as doctors, nurses, MSR / MSL participants, etc.), thereby providing useful competency data, progress, and feedback to individuals, managers, executives, and the human resources department (HR). This innovative approach enables two-way engagement with HCPs, contributes to closing the skills gap, reducing costs, and eliminating inefficiencies. Such a dynamic learning environment creates an individually optimized learning experience, ultimately leading to improved patient treatment outcomes and the quality of medical care.

[0071] Furthermore, the present invention provides peer-to-peer point-of-care access among HCPs (healthcare professionals) and encompasses various functions and advantages such as an HCP network, MR / MS / MSL support, MSL collaboration, a community board, and a transcription of content that can serve as knowledge for future professionals. System 104 enables HCPs to access and collaborate with their fellow HCPs and supports them in quickly obtaining the necessary support at the point of care through advanced search. Within the medical platform, connected HCPs can support other HCPs during a call, which is realized through the involvement of MSLs (medical science liaisons). Also, in one embodiment, System 104 enables MSLs to invite other experienced HCPs during a call, realizing the timely provision of expertise and the rapid acquisition of real-world data. Furthermore, by using Generative AI, these conversations are further enhanced, contributing to better decision-making and improved patient outcomes.

[0072] In the implementation of System 104, individual evaluations based on past evaluation data and comprehension tests are conducted to assess the learning outcomes of users. The expansion of this system develops beyond individual-level evaluations into a comprehensive talent evaluation system. For example, detailed data on physicians are obtained through this learning system 104, and these data can later be linked to the marketing activities of pharmaceutical companies.

[0073] System (100) enables physicians, medical representatives (MRs), pharmacists, and users to register and log in through an authenticated application on their user terminals. In particular, signed-up or registered physicians and users can access information. Registered users (102) can communicate with experts in their respective fields, such as the medical representatives and other professionals they have selected. Through this communication, it becomes possible to obtain accurate and detailed knowledge about, for example, the latest diagnoses, treatments, cures, medication compliance, and medical devices, and a wide range of personalization of information according to the needs of the users is realized.

[0074] The systems and methods according to the present invention facilitate discussions on complex patient cases through meetings and webinars, accelerate awareness improvement regarding pharmaceuticals, and provide useful insights, voting, and survey data for collecting market intelligence. Further, it automatically transcribes and anonymizes communication content, collects actual case data, and enables obtaining supplementary information and insights from KOLs (Key Opinion Leaders) important for pharmaceutical intelligence reports. This promotes a quicker and deeper understanding of pharmaceuticals and enables informed decision-making. The comprehensive functions of this system eliminate skill gaps, reduce costs, and eliminate inefficiencies.

[0075] In FIG. 2, a block diagram of system 104 is shown as one embodiment. As shown, system 104 includes one or more processors 202, a memory 204, at least one user interface 206, data 208, a query processing and script creation module 210, an audio / video engine 212, a review module 214, and a compliance check module 216.

[0076] In one embodiment, system 104 receives a query from a user from the network 106 via at least one user terminal, such as 102-1. The user query is received at server 108 that hosts system 104. The query processing and script generation module 210 obtains information using artificial intelligence from one or more knowledge databases (such as 110-1) based on the user query and generates a draft script. In one embodiment, the query processing and script generation module 210 can perform preprocessing, for example, using natural language processing techniques, before obtaining information from one or more databases 110 for the user query.

[0077] Furthermore, the compliance check module 216 is configured to verify the draft script and provide a compliant script for its query. The draft script may include a part of structured data, unstructured data (such as real-world data, scientific document data, and domain-specific expertise data). The compliance check module 216 can generate structured data or metadata such as expert-defined rules based on the expertise data and the unstructured data generated by one or more AI engines, and save them as part of the data 208.

[0078] The structured data can include layers of information specialized for a particular disease or drug. In one embodiment, the real-world data is converted into a common format for machine learning and processed. The real-world data, scientific document data, and expert opinion data are processed by a machine learning classifier to generate taxonomic metadata. For example, the unstructured data may be tagged with medical-related topics by a machine learning classifier. In one embodiment, the data is mined to generate multi-category classification data including relevant metadata.

[0079] The verification process is performed based on one or more pre - stored rules and / or parameters (e.g., part of data 208) or parameters generated from unstructured data. In one embodiment, the compliance check module 216 considers multiple draft scripts generated using one or more AI algorithms in response to different but related user queries regarding the same topic, and generates a compliant script based on predefined rules. The module 216 can access a database of participating users, such as medical providers and market experts, along with rules such as rules for introducing a specific doctor based on symptoms or medical conditions, rules defined by experts, Medicaid or private insurance benefit rules, affinity scores, user - level scores, etc. In one embodiment, the compliance check module 216 can receive input from at least one user to complete the compliance process and personalize the script for the user query. Further, the compliant script can also take into account, for example, real - time news feeds regarding the topic, real - time opinions of experts collected through interviews, and survey results of professionals at different levels within the industry.

[0080] Furthermore, the audio / video engine 212 is configured to generate audio / video content based on structured data using artificial intelligence. In one embodiment, the audio / video engine 212 provides audio - video content related to the structured data received from the compliance check module 216 through an AI - based text - to - video generator. Here too, machine learning algorithms such as natural language processing and autoregressive transformers may be used in the generation of audio / video content. The audio / video engine 212 can also learn from the existing dataset 112.

[0081] Furthermore, the review module 214 is configured to receive the generated audio / video content and evaluate its relevance. Rules and / or parameters for evaluating relevance may be stored as part of the data 208 or may be generated from structured data. The rules / parameters regarding the approval of content information include multiple criteria, including evaluations by nurses, doctors, hospital panels, corporate physicians, medical writers, and experts. These relevant parties evaluate the relevance, accuracy, and fact-checking of the content, and sub-classify the relevant content based on four main perspectives: the relevance, timeliness, and comprehensiveness of the audio / video content information.

[0082] The review module 214 separates relevant audio / video content from the content generated by the audio / video engine 212 and generates it as a nugget (clip) that condenses the audio / video content. On the other hand, irrelevant content is filtered. In one embodiment, the review module 214 also plays the role of extracting relevant sections and nuggets from the compliance-checked scripts.

[0083] In one embodiment, the review module 214 is configured to provide a review and retake function that enables iterative improvement of the audio / video content and the compliance-checked script, and to generate nuggets of relevant content. In yet another embodiment, the relevance of the curated content is evaluated based on parameters such as affinity scores, accuracy based on peer-reviewed papers, rankings of opinions of famous people on the topic, and recommendations by famous people. In one embodiment, the system 104 collects user information through evaluation methods such as online questionnaires and surveys, assigns a rank according to the user's skill level, and classifies them, for example, as beginners, experts, and experts, and defines levels for educational and training purposes.

[0084] More specifically, the predetermined parameters include behavioral characteristics, which include perception, prescriptive actions, prescriptive preferences, opinions on brands or drugs, characteristics, tendencies, medication compliance, and compliance.

[0085] The present disclosure further discloses that the server 108 is configured to store and access various data, which includes information, media content, survey items including pre- and post-questionnaires, reference documents, scientific papers, as well as training and educational content. These data are obtained from various sources related to the underlying products and services.

[0086] Finally, the linked content is displayed on the user device 102-1 via the user interface 206. This content has been refined and curated through various processing stages of the system and is presented in a form that can be utilized by the user.

[0087] The architecture of the system 104 indicates that various modules within the server are integrated to manage user input, process queries, retrieve information from the database, and present refined personalized content to the end user. Through this complex network of modules and functions, the system aims to rationalize the creation, processing, and provision of relevant content in a structured user-friendly manner.

[0088] To enhance the user experience and efficiency, the system 104 demonstrates a synergy where multiple components work together to consistently process the flow of information and content in the context of education or training and provide accurate and targeted materials to the final user.

[0089] Figure 3 is a flowchart showing a method for promoting content creation and compliance for training and education. In step 302, communication and data transfer are established between the user and the server, and the user's query is received as input. In step 304, the user's query is processed, and a draft script is generated based on information obtained from the knowledge database using AI.

[0090] In step 306, the draft script is verified based on one or more predefined rules / parameters stored on the server, and structured data is generated from the compliance-checked script. Further, the system 104 incorporates behavior learning for new users and generates content information.

[0091] In step 308, multiple audio / video contents are generated using artificial intelligence based on the structured data. In step 310, the relevance of the generated audio / video contents is evaluated based on predefined parameters, a condensed nugget of relevant audio / video content is created, and irrelevant segments are excluded. These condensed nuggets are linked to the compliance-checked script and displayed on the user device in step 312.

[0092] In operation, user 102 is required to register with system 104 for educational or training purposes by providing information regarding basic information as well as academic / occupational history and expertise (if applicable). Based on the information provided at the time of registration and the specific wishes / requirements of user 102, the system presents the user with various course options and recommendations. These options and recommendations include, for example, specific fields, course levels, course durations, time available per week, etc. Depending on the complexity and time constraints of the course selected by the user, system 104 can provide a personalized course tailored to the user. In an embodiment, system 104 can evaluate the user's behavior pattern and accordingly prepare course materials. In another embodiment, based on the initial course-related information provided by the user, system 104 presents a certain number of elective courses or mandatory / regular courses, from which the user can select the topic most suitable for their current interests or needs at that time. System 104 provides an information nugget that is optimal for maximizing user engagement based on the said selection and other information provided by user 102 (such as course duration), as well as what system 104 has learned from user 102. These information nuggets are part of a draft script and are compliance-checked from the perspectives of relevance, accuracy, comprehensiveness, and novelty of information, and are provided as a compliant script in a nugget format suitable for the user. These nuggets are segmented to maximally attract the user's attention. In an embodiment, system 104 enables the user or a user who checks the compliance of the draft script to provide feedback on the nuggets, thereby allowing modifications to be made to the nugget list or the content of subsequent nuggets.

[0093] In one embodiment, system 104 is configured to allow a user to personalize the nagets according to their skill level and to enable the flow of context information between related nagets. System 104 adapts to the user's selection of personalization and provides an average number of nagets for the selected course.

[0094] The method also includes review and retake functions, enabling iterative improvement of both audio / video content and structured information, and presenting refined and highly relevant nagets. Ultimately, refined relevant content is displayed on the user's device. The growth of the user's knowledge and the curation of content are adjusted in line with their evolving proficiency. Monitoring the user across social media, publications, academic journals, and data collection helps to identify changing needs. Subsequently, the engine generates refined content and recommendations based on this. External factors are given a specific weighting, while behavioral patterns are typically given an 80% weighting.

[0095] The series of steps of the method emphasizes a systematic approach to content creation and refinement, leveraging multiple processing stages to provide high-quality and relevant audio / video content (shown in Figure 4). The process is iterative in nature and, with review and retake functions, aims to optimize the user experience in an educational or training environment, realizing a continuous improvement loop in the quality and relevance of the content provided.

[0096] Figure 4 shows a conditional selection flowchart regarding content creation and compliance for education and training. First, data transfer and communication occur between the user and the server. Then, the user query is processed as input. Based on the input query, a draft script is generated using information retrieved from one or more databases using artificial intelligence.

[0097] The relevance of the generated audio / video content is checked based on predetermined parameters. If it is determined to be relevant, a nugget of the relevant audio / video content is created and irrelevant segments are excluded. If it is not relevant, the process loops until information is retrieved from a different database using artificial intelligence and a new draft script is generated. When a nugget linked to the compliance-checked script is created, the final output is displayed on the user's device.

[0098] Embodiments of the present invention can be provided as a computer program product. This product can include a computer-readable medium that specifically embodies instructions for programming a computer (or other electronic device) to execute a process. Computer-readable media include, but are not limited to, hard disks, magnetic tapes, floppy disks, optical disks, CD-ROMs, magneto-optical disks, ROMs, RAMs, PROMs, EPROMs, EEPROMs, flash memories, magnetic cards or optical cards, and other media / machine-readable media suitable for storing electronic instructions (e.g., computer program code as software or firmware). Furthermore, embodiments of the present invention can also be downloaded as one or more computer program products, and the programs can be transferred from a remote computer to the requesting computer via a communication link (e.g., a modem or network connection) through data signals embodied in a carrier wave or other propagation medium.

[0099] In various embodiments of the present invention, a manufactured product (e.g., a computer program product) can be used by directly executing code from a computer-readable medium, or by copying the code from a computer-readable medium to another computer-readable medium (e.g., a hard disk, RAM, etc.), or by transferring the code over a network for remote execution. By combining one or more computer-readable media containing the code according to the present invention with appropriate standard computer hardware for executing the code, the various methods described herein can be implemented. An apparatus for implementing various embodiments of the present invention can include one or more computers (or one or more processors within a single computer, or one or more processor cores) and a storage system accessible to a computer program coded according to the various methods described herein. Also, the method steps of the present invention can be executed by a module, routine, subroutine, or subpart of a computer program product.

[0100] For purposes of brevity, the illustrated methodologies are shown and described as a series of blocks / steps, but it should be understood that the methodologies are not limited to the order of the blocks. A block may occur in a different order than other blocks, or in parallel. Further, not all of the illustrated blocks are necessary for the implementation of the exemplary methodologies. A block may be integrated or divided into multiple components. Additionally, additional or alternative methodologies may include additional blocks not shown.

[0101] In the foregoing description, specific terms have been used for the sake of brevity, clarity, and understanding, but these terms are used for descriptive purposes only and do not imply any unnecessary limitations beyond the requirements of the prior art. Accordingly, the present invention is not limited to the specific details, representative embodiments, and illustrated examples described. This application is intended to cover modifications, variations, and equivalents within the scope of the appended claims.

[0102] The methodologies and techniques described with respect to the above embodiments can be executed using a machine or other computing device, and by executing a set of instructions in this device, one or more of the above methodologies can be made to execute. In some embodiments, the machine can function as a stand-alone device. In another embodiment, the machine is connected to other machines (e.g., using a network) and can function as a server in a network arrangement, or as a client-user machine in a client-user network environment, or as a peer machine in a peer-to-peer (or distributed) network environment.

[0103] Furthermore, although the present invention and its advantages have been described in detail, it should be understood that various changes, substitutions, and modifications can be made herein without departing from the scope of the invention as defined by the appended claims. Moreover, the scope of the present application is not intended to be limited to the specific embodiments of the processes, machines, manufactures, compositions of matter, means, methods, and steps described herein. As can be readily understood from the disclosure herein, existing or future-developed processes, machines, manufactures, compositions of matter, means, methods, or steps designed to perform substantially the same function or achieve substantially the same result can be used as embodiments of the present invention. Accordingly, the appended claims are intended to include within their scope such processes, machines, manufactures, compositions of matter, means, methods, or steps as described above.

[0104] The description presented herein refers to various embodiments. Those skilled in the art to which this application pertains will understand that changes and modifications to the described structures and methods of operation can be implemented without substantially departing from the principles, spirit, and scope.

Claims

1. A computer-implemented system (100) for training and education, the system (100) comprising: - One or more user devices (102-1, 102-2,... 102-n) configured to receive input from one or more users; - A server (108) configured to receive the user input via a communication network (106) and obtain information from one or more databases (110), wherein the server (108) is: - A memory (204), and - A processor (202) configured to execute instructions stored in the memory (204), and further comprising: - A query processing and script generation module (210) configured to process a user query received via a user interface as user input, generate a draft script based on information obtained from the one or more databases (110) using artificial intelligence, and store the same; - A compliance confirmation module (216) configured to verify the draft script based on one or more predetermined parameters stored in the server (108) and generate structured data from the compliance-confirmed script; - An audio / video search engine (212) configured to generate a plurality of audio / video contents based on the structured data by artificial intelligence; - A review module (214) configured to receive the generated audio / video contents to determine relevance based on predetermined parameters, wherein the review module (214) creates a list of audio / video content nuggets relevant to the descriptive nuggets from the compliance-confirmed script and removes irrelevant content; - Linking the descriptive nuggets with corresponding audio / video content nuggets based on match factors; - An output module for displaying the linked content on a user device.

2. The system (100) according to claim 1, wherein the relevance of the audio / video content is determined by a predetermined parameter such as an affinity score, and the output is ranked based on the opinions of well-known individuals.

3. ​ In the system (100) according to claim 1, the system (100) learns information about a user through assessment, selects the rank of the user based on the proficiency level of the user, thereby classifying the user as a beginner, a practitioner, or an expert, and establishing its presence within the system (100).

4. In the system (100) according to claim 1, the compliance confirmation module (216) is further configured to perform scrutiny and verification for regulatory compliance on audio / video content.

5. In the system (100) according to claim 1, the review module (214) is configured to facilitate a review and retake function that enables iterative improvement of audio / video content and a compliance-confirmed script, and to provide nuggets of related content.

6. In the system (100) according to claim 1, the review module (214) is further configured to partially retake related content and selectively modify a part of the related audio / video information based on input from at least one stakeholder.

7. In the system (100) according to claim 1, the users include healthcare professionals (HCPs), medical information personnel (MRs), and sales representatives. More specifically, the healthcare professionals (HCPs) include physicians, pharmacists, caregivers, and representatives of various medical institutions.

8. In the system (100) according to claim 1, the predetermined parameters include behavioral characteristics composed of cognition, behavior, prescription selection, preference for prescriptions, opinions regarding brands or ingredients, characteristics, tendencies, medication adherence, and compliance.

9. In the system (100) according to claim 1, the server (108) is further configured to store and access a wide range of data including information, media content, pre / post-questionnaire items, literature, medical journals, marketing and medical communication materials, and training and educational content, and the data is obtained from various information sources.

10. In the system (100) according to claim 1, the review module (214) is further configured to link nagets and relevant portions of the compliance-verified script.

11. In the system (100) according to claim 1, the draft script is edited by a user (102) and individualized feedback is provided.

12. In the system (100) according to claim 1, a user query is processed by at least two generative AI processing techniques, at least two draft scripts are generated, and the system (104) is configured to compare the generated scripts based on one or more of the pre-stored parameters, specifically the user's recognition and persona, knowledge criteria, accuracy, relevance, comprehensiveness, clarity and context, and the simultaneous updatability of the script.

13. A computer-implemented method for facilitating content creation and compliance for training and education, the method comprising the following steps: - Receiving input from one or more user devices (102-1, 102-2,... 102-n); - Processing a user query as user input, generating a draft script by artificial intelligence using information obtained from one or more databases (110), and storing it; - Verifying the draft script based on one or more pre-defined rules / parameters stored on the server and generating structured data from the compliance-verified script; - Generating a plurality of audio / video contents by artificial intelligence based on the structured data; - Judging the relevance of the generated audio / video contents based on pre-defined parameters, creating an explanatory naget and an audio / video naget from the compliance-verified script, and filtering out irrelevant contents; - Linking the explanatory naget and the corresponding audio / video naget based on a matching factor; - Displaying the linked contents on the user device.

14. The method according to claim 13, wherein the determination of the relevance includes a review and retake function that enables iterative improvement of the audio / video content and the compliance-verified script.

15. The method according to claim 13, wherein the relevant content is further characterized in that partial retakes and selective correction processes are performed based on the input of at least one stakeholder.

16. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause a computing system to perform the following method, the method comprising: - Receiving input from one or more user devices (102-1, 102-2,... 102-n); - Processing the user query as user input, generating a draft script using artificial intelligence based on information obtained from one or more databases (110), and storing it; - Verifying the draft script based on one or more predefined rules / parameters stored in the server (108) and generating structured data from the compliance-verified script; - Generating a plurality of audio / video contents based on the structured data using artificial intelligence; - Determining the relevance of the generated audio / video content based on predefined parameters, creating an explanatory naget and an audio / video naget from the compliance-verified script, and filtering out irrelevant content; - Linking the explanatory naget and the corresponding audio / video naget based on a match factor; - Displaying the linked content on the user device.

17. The non-transitory computer-readable medium according to claim 16, wherein the method further includes receiving a search query by one or more processors from a digital assistant application running on a remote client device.

18. The non-transitory computer-readable medium according to claim 16, The generation of the structured data includes generating taxonomic data in multiple categories based on scientific or clinical documents, diagnoses, treatments, and drug names, and further includes metadata in multiple categories. A medium characterized by this.

19. The method according to claim 13, characterized in that user feedback on a certain naget causes corrections to the remaining nagets in a predetermined naget list.

20. The method according to claim 13, characterized in that the verification step includes assigning a timestamp to the linked content.

21. A method and system for delivering personalized multimedia content to a target recipient via a desired communication channel, identifying the recipient's preferences and context data, generating or selecting disease-specific or drug-specific video or audio content, transmitting the multimedia content through email, SMS, or other desired communication platforms, characterized in that the recipient can access the content asynchronously, and as a scenario, when a patient receives an educational video tailored to their disease, or when a doctor receives an explanation or answer to a question based on a request for new drug information during commuting, as well as when receiving the latest information in audio format, etc. A method and system including this.

Images