WO2025221968A1

WO2025221968A1 - Artificial intelligence patient actor system

Info

Publication number: WO2025221968A1
Application number: PCT/US2025/025111
Authority: WO
Inventors: Thomas THESEN; Simon Stone
Original assignee: Dartmouth College
Current assignee: Dartmouth College
Priority date: 2024-04-17
Filing date: 2025-04-17
Publication date: 2025-10-23
Anticipated expiration: 2026-10-17

Abstract

An artificial intelligence (AI) patient actor system and associated functionality can be used to simulate an interaction between a medical student and a patient for educational purposes. The system can include a database including a patient case file and an assessment rubric, a prompt refinement system that provides the patient case file and the assessment rubric to an artificial intelligence model (e.g., a large language model (LLM), etc.), and a student interface system that causes a student user interface for simulating the interaction between the medical student and the patient to be presented via a computing device.

Description

ARTIFICIAL INTELLIGENCE PATIENT ACTOR SYSTEM

CROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of and priority to U.S. Provisional Patent Application No. 63/635,377, filed April 17, 2024, the entirety of which is incorporated by reference herein.

BACKGROUND

[0002] Students including medical students need to prepare to practice professionally. Today, students typically need other humans to serve as “actors” when learning skills such as performing a medical evaluation of a patient, for example.

BRIEF DESCRIPTION OF THE DRAWINGS

(0003J FIG. 1 is a block diagram showing components of an example distributed computing environment, in accordance with some aspects of the disclosure.

[0004] FIG. 2 is a block diagram showing components of an example computing system, in accordance with some aspects of the disclosure.

[0005] FIG. 3 is a block diagram illustrating an example artificial intelligence (Al) patient actor system, in accordance with some aspects of the disclosure.

[0006] FIG. 4 is a flowchart of an example process for simulating an interaction between a medical student and a patient that can be performed using the system of FIG. 3, in accordance with some aspects of the disclosure.

[0007] FIG. 5 is an illustration showing an example student user interface that can be caused to be provided by the system of FIG. 3, in accordance with some aspects of the disclosure.

[0008] FIGS. 6A-6B are illustrations showing more example student user interfaces that can be caused to be provided by the system of FIG. 3, in accordance with some aspects of the disclosure.

DETAILED DESCRIPTION

[0009] Referring to FIG. 1, a non-limiting example of a distributed computing environment 100 is shown, in accordance with some aspects of the disclosure. In some examples, the distributed computing environment 100 may include one or more server(s) 102 (e.g., data servers, computing devices, computers, etc.), one or more client computing devices 106, and/or other components that may implement certain features described herein. Other devices, such as specialized sensor devices, etc., may interact with the client computing device(s) 106 and/or the server(s) 102. The server(s) 102, client computing device(s) 106, or any other devices may be configured to implement a client-server model or any other distributed computing architecture. In an illustrative and non-limiting example, the client devices 106 may include a first client device 106A and a second client device 106B. The first client device 106A may correspond to a first user in a class and the second client device 106B may correspond to a second user in the class or another class. In some examples, the client devices 106 can include a virtual reality headset or any suitable computing device with a display (e.g., smartphone, tablet, laptop computer, etc.).

|0010] In some examples, the server(s) 102, the client computing device(s) 106, and any other disclosed devices may be communicatively coupled via one or more communication network(s) 120. The communication network(s) 120 may be any type of communication networks supporting data communications. As non-limiting examples, network 120 may be a local area network (LAN; e.g., Ethernet, Token-Ring, etc.), a wide-area network (e.g., the Internet), an infrared or wireless network, a public switched telephone networks (PSTNs), a virtual network, etc. Network 120 may use any available protocols, such as, e.g., transmission control protocol/Internet protocol (TCP/IP), systems network architecture (SNA), Internet packet exchange (IPX), Secure Sockets Layer (SSL), Transport Layer Security (TLS), Hypertext Transfer Protocol (HTTP), Secure Hypertext Transfer Protocol (HTTPS), Institute of Electrical and Electronics (IEEE) 802.11 protocol suite or other wireless protocols, and the like.

[0011] The examples shown in FIG. 1 and/or FIG. 2 are respective examples of a distributed computing system and are not intended to be limiting. The subsystems and components within the server(s) 102 and the client computing device(s) 106 may be implemented in hardware, firmware, software, or combinations thereof. Various different subsystems and/or components 104 may be implemented on the server 102. Users operating the client computing device(s) 106 may initiate one or more client applications to use services provided by these subsystems and components. Various different system configurations are possible in different types of distributed computing environments and content distribution networks. Server 102 may be configured to run one or more server software applications or services, for example, web-based or cloud-based services, to support content distribution and interaction with client computing device(s) 106. Users operating client computing device(s) 106 may in turn utilize one or more client applications (e.g., virtual client applications) to interact with server 102 to utilize the services provided by these components. The client computing device(s) 106 may be configured to receive and execute client applications over the communication network(s) 120. Such client applications may be web browser-based applications and/or standalone software applications, such as mobile device applications. The client computing device(s) 106 may receive client applications from server 102 or from other application providers (e.g., public or private application stores).

[0012] As shown in FIG. 1, various security and integration components 108 may be used to manage communications over the communication network(s) 120 (e.g., a file-based integration scheme, a service-based integration scheme, etc.). In some examples, the security and integration components 108 may implement various security features for data transmission and storage, such as authenticating users or restricting access to unknown or unauthorized users. As non-limiting examples, the security and integration components 108 may include any dedicated hardware, specialized networking components, and/or software (e.g., web servers, authentication servers, firewalls, routers, gateways, load balancers, etc.) within one or more data centers in one or more physical location(s) and/or operated by one or more entities, and/or may be operated within a cloud infrastructure. In various implementations, the security and integration components 108 may transmit data between the various devices in the distribution computing environment 100 (e.g., in a content distribution system or network). In some examples, the security and integration components 108 may use secure data transmission protocols and/or encryption (e.g., File Transfer Protocol (FTP), Secure File Transfer Protocol (SFTP), and/or Pretty Good Privacy (PGP) encryption) for data transfers, etc.).

[0013] In some examples, the security and integration components 108 may implement one or more web services (e.g., cross-domain and/or cross-platform web services) within the distribution computing environment 100, and may be developed for enterprise use in accordance with various web service standards (e.g., the Web Service Interoperability (WS-I) guidelines). In an example, some web services may provide secure connections, authentication, and/or confidentiality throughout the network using technologies such as SSL, TLS, HTTP, HTTPS, WS-Security standard (providing secure SOAP messages using XML encryption), etc. In some examples, the security and integration components 108 may include specialized hardware, network appliances, and the like (e.g., hardware-accelerated SSL and HTTPS), possibly installed and configured between one or more server(s) 102 and other network components. In such examples, the security and integration components 108 may thus provide secure web services, thereby allowing any external devices to communicate directly with the specialized hardware, network appliances, etc.

[0014] The distributed computing environment 100 may further include one or more data stores 110. In some examples, the one or more data stores 110 may include, and/or reside on, one or more back-end servers 112, operating in one or more data center(s) in one or more physical locations. In such examples, the one or more data stores 110 may communicate data between one or more devices, such as those connected via the one or more communication network(s) 120. In some cases, the one or more data stores 110 may reside on a non-transitory storage medium within one or more server(s) 102. In some examples, data stores 110 and back-end servers 112 may reside in a storage-area network (SAN). In addition, access to one or more data stores 110, in some examples, may be limited and/or denied based on the processes, user credentials, and/or devices attempting to interact with the one or more data stores 110.

[0015] Referring to FIG. 2, a block diagram of an example computing system 200 is shown, in accordance with some aspects of the disclosure. The computing system 200 (e.g., one or more connected computers) may correspond to any one or more of the computing devices or servers of the distributed computing environment 100, or any other computing devices described herein. In an example, the computing system 200 may represent an example of one or more server(s) 102 and/or of one or more server(s) 112 of the distributed computing environment 100. In another example, the computing system 200 may represent an example of the client computing device(s) 106 of the distributed computing environment 100. In some examples, the computing system 200 may represent a combination of one or more computing devices and/or servers of the distributed computing environment 100.

[0016] In some examples, the computing system 200 may include processing circuitry 204, such as one or more processing unit(s), processor(s), etc. In some examples, the processing circuitry 204 may communicate (e.g., interface) electronically with a number of peripheral subsystems via a bus subsystem 202. These peripheral subsystems may include, for example, a storage subsystem 210, an input/output (I/O) subsystem 226, and a communications subsystem 232.

[0017] In some examples, the processing circuitry 204 may be implemented as one or more integrated circuits (e.g., a micro-processor or microcontroller). In an example, the processing circuitry 204 may control the operation of the computing system 200. The processing circuitry 204 may include single core and/or multicore (e g., quad core, hexa-core, octo-core, ten-core, etc.) processors and processor caches (e.g., central processing units (CPUs), graphics processing units (GPUs), etc.). The processing circuitry 204 may execute a variety of resident software processes embodied in program code, and may maintain multiple concurrently executing programs or processes. In some examples, the processing circuitry 204 may include one or more specialized processors, (e.g., digital signal processors (DSPs), outboard, graphics application-specific, and/or other processors).

[0018] In some examples, the bus subsystem 202 provides a mechanism for intended communication between the various components and subsystems of computing system 200. Although the bus subsystem 202 is shown schematically as a single bus, other implementations of the bus subsystem may utilize multiple buses. In some examples, the bus subsystem 202 may include a memory bus, memory controller, peripheral bus, and/or local bus using any of a variety of bus architectures (e.g., Industry Standard Architecture (ISA), Micro Channel Architecture (MCA), Enhanced ISA (EISA), Video Electronics Standards Association (VESA), and/or Peripheral Component Interconnect (PCI) bus, possibly implemented as a Mezzanine bus manufactured to the IEEE P1386.1 standard, etc.).

[0019] In some examples, the I/O subsystem 226 may include one or more device controller(s) 228 for one or more user interface input devices and/or user interface output devices, possibly integrated with the computing system 200 (e.g., virtual reality headsets, integrated audio/video systems, and/or touchscreen displays), or may be separate peripheral devices which are attachable/detachable from the computing system 200. Input may include keyboard or mouse input, audio input (e.g., spoken commands), motion sensing, gesture recognition (e.g., eye gestures), etc. As non-limiting examples, input devices may include a keyboard, pointing devices (e.g., mouse, trackball, and associated input), touchpads, touch screens, scroll wheels, click wheels, dials, buttons, switches, keypad, audio input devices, voice command recognition systems, microphones, three dimensional (3D) mice, joysticks, pointing sticks, gamepads, graphic tablets, speakers, digital cameras, digital camcorders, portable media players, webcams, image scanners, fingerprint scanners, barcode readers, 3D scanners, 3D printers, laser rangefinders, eye gaze tracking devices, medical imaging input devices, MIDI keyboards, digital musical instruments, and the like. In general, use of the term “output device” is intended to include all possible types of devices and mechanisms for outputting information from computing system 200, such as to a user (e.g., via a display device) or any other computing system, such as a second computing system 200. In an example, output devices may include one or more display subsystems and/or display devices that visually convey text, graphics and audio/video information (e.g., cathode ray tube (CRT) displays, flat-panel devices, liquid crystal display (LCD) or plasma display devices, projection devices, touch screens, etc.), and/or may include one or more non-visual display subsystems and/or non-visual display devices, such as audio output devices, etc. As non-limiting examples, output devices may include, virtual reality headsets, indicator lights, monitors, printers, speakers, headphones, automotive navigation systems, plotters, voice output devices, modems, etc.

[0020] In some examples, the computing system 200 may include one or more storage subsystems 210, including hardware and software components used for storing data and program instructions, such as system memory 218 and computer-readable storage media 216. In some examples, the system memory 218 and/or the computer-readable storage media 216 may store and/or include program instructions that are loadable and executable on the processor(s) 204. In an example, the system memory 218 may load and/or execute an operating system 224, program data 222, server applications, application program(s) 220 (e.g., client applications), Internet browsers, mid-tier applications, etc. In some examples, the system memory 218 may further store data generated during execution of these instructions.

[0021] In some examples, the system memory 218 may be stored in volatile memory (e.g., random-access memory (RAM) 212, including static random-access memory (SRAM) or dynamic random-access memory (DRAM)). In an example, the RAM 212 may contain data and/or program modules that are immediately accessible to and/or operated and executed by the processing circuitry 204. In some examples, the system memory 218 may also be stored in non-volatile storage drives 214 (e.g., read-only memory (ROM), flash memory, etc ). In an example, a basic input/output system (BIOS), containing the basic routines that help to transfer information between elements within the computing system 200 (e.g., during start-up), may typically be stored in the non-volatile storage drives 214.

[0022] In some examples, the storage subsystem 210 may include one or more tangible computer-readable storage media 216 for storing the basic programming and data constructs that provide various functionality. In an example, the storage subsystem 210 may include software, programs, code modules, instructions, etc., that may be executed by the processing circuitry 204, in order to provide the functionality described herein. In some examples, data generated from the executed software, programs, code, modules, or instructions may be stored within a data storage repository within the storage subsystem 210. In some examples, the storage subsystem 210 may also include a computer-readable storage media reader connected to the computer-readable storage media 216.

[0023] In some examples, the computer-readable storage media 216 may contain program code, or portions of program code. Together and optionally in combination with the system memory 218, the computer-readable storage media 216 may comprehensively represent remote, local, fixed, and/or removable storage devices plus storage media for temporarily and/or more permanently containing, storing, transmitting, and/or retrieving computer-readable information. In some examples, the computer-readable storage media 216 may include any appropriate media known or used in the art, including storage media and communication media, such as but not limited to, volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage and/or transmission of information. This can include tangible computer-readable storage media such as RAM, ROM, electronically erasable programmable ROM (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disk (DVD), or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or other tangible computer-readable media. This can also include nontangible computer-readable media, such as data signals, data transmissions, or any other medium which can be used to transmit the desired information, and which can be accessed by the computing system 200. In an illustrative and non-limiting example, the computer-readable storage media 216 may include a hard disk drive that reads from or writes to non-removable, nonvolatile magnetic media, a magnetic disk drive that reads from or writes to a removable, nonvolatile magnetic disk, and an optical disk drive that reads from or writes to a removable, nonvolatile optical disk such as a CD ROM, DVD, and Blu-Ray® disk, or other optical media.

[0024] In some examples, the computer-readable storage media 216 may include, but is not limited to, Zip® drives, flash memory cards, universal serial bus (USB) flash drives, secure digital (SD) cards, DVD disks, digital video tape, and the like. In some examples, the computer-readable storage media 216 may include, solid-state drives (SSD) based on non-volatile memory such as flash-memory based SSDs, enterprise flash drives, solid state ROM, and the like, SSDs based on volatile memory such as solid-state RAM, dynamic RAM, static RAM, DRAM-based SSDs, magneto-resistive RAM (MRAM) SSDs, and hybrid SSDs that use a combination of DRAM and flash memory-based SSDs. The disk drives and their associated computer-readable media may provide non-volatile storage of computer-readable instructions, data structures, program modules, and other data for the computing system 200.

[0025] In some examples, the communications subsystem 232 may provide a communication interface from the computing system 200 and external computing devices via one or more communication networks, including local area networks (LANs), wide area networks (WANs) (e.g., the Internet), and various wireless telecommunications networks. As illustrated in FIG. 2, the communications subsystem 232 may include, for example, one or more network interface controllers (NICs) 234, such as Ethernet cards, Asynchronous Transfer Mode NICs, Token Ring NICs, and the like, as well as one or more wireless communications interfaces 236, such as wireless network interface controllers (WNICs), wireless network adapters, and the like. Additionally, and/or alternatively, the communications subsystem 232 may include one or more modems (telephone, satellite, cable, ISDN), synchronous or asynchronous digital subscriber line (DSL) units, Fire Wire® interfaces, USB® interfaces, and the like. Communications subsystem 232 also may include radio frequency (RF) transceiver components for accessing wireless voice and/or data networks (e.g., using cellular telephone technology, advanced data network technology, such as 3G, 4G, 5G or EDGE (enhanced data rates for global evolution), Wi-Fi (IEEE 802.11 family standards, or other mobile communication technologies, or any combination thereof), global positioning system (GPS) receiver components, and/or other components.

[0026] In some examples, the communications subsystem 232 may also receive input communication in the form of structured and/or unstructured data feeds, event streams, event updates, and the like, on behalf of one or more users who may use or access the computing system 200. In an example, the communications subsystem 232 may be configured to receive data feeds in real-time from users of social networks and/or other communication services, web feeds such as Rich Site Summary (RSS) feeds, and/or real-time updates from one or more third party information sources (e.g., data aggregators). Additionally, the communications subsystem 232 may be configured to receive data in the form of continuous data streams, which may include event streams of real-time events and/or event updates (e.g., sensor data applications, financial tickers, network performance measuring tools, clickstream analysis tools, automobile traffic monitoring, etc.). In some examples, the communications subsystem 232 may output such structured and/or unstructured data feeds, event streams, event updates, and the like to one or more data stores that may be in communication with one or more streaming data source computing systems (e.g., one or more data source computers, etc.) coupled to the computing system 200. The various physical components of the communications subsystem 232 may be detachable components coupled to the computing system 200 via a computer network (e.g., a communication network 120), a FireWire® bus, or the like, and/or may be physically integrated onto a motherboard of the computing system 200. In some examples, the communications subsystem 232 may be implemented in whole or in part by software.

[0027] Due to the ever-changing nature of computers and networks, the description of the computing system 200 depicted in the figure is intended only as a specific example. Many other configurations having more or fewer components than the system depicted in the figure are possible. For example, customized hardware might also be used and/or particular elements might be implemented in hardware, firmware, software, or a combination. Further, connection to other computing devices, such as network input/output devices, may be employed. Based on the disclosure and teachings provided herein, a person of ordinary skill in the art will appreciate other ways and/or methods to implement the various aspects of the disclosure.

(0028] Referring to FIG. 3, a block diagram illustrating an example artificial intelligence (Al) patient actor system 300 is shown, in accordance with some aspects of the disclosure. The example artificial intelligence patient actor system 300 can be implemented using a variety of different hardware, software, firmware, and networking configurations, such as, for example, configuration that are similar to those detailed above with respect to the distributed computing environment 100 and/or the computing system 200. The artificial intelligence patient actor system 300 can be used for simulating an interaction between a medical student and a patient for educational purposes. However, the artificial intelligence patient actor system 300 can also be adapted for use in a variety of educational contexts beyond medical education. For example, the artificial intelligence patient actor system 300 can be adapted for use in contexts such as mock legal trials, mental health counseling, business negotiations (e.g., as part of a business school), human resource discussions (e.g., performance reviews, disciplinary, performance improvements, etc.), specialty training (e.g., studying for residency board exams, etc.), and other types of conversational training purposes.

[0029| The artificial intelligence patient actor system 300 generally uses artificial intelligence to help medical students that need practice examining patients. Accordingly, medical students can use the artificial intelligence patient actor system 300 to develop critical skills including, for example, medical interviewing skills, examination skills, communication skills, and differential diagnosis skills. The artificial intelligence patient actor system 300 can provide a realistic and safe learning environment for medical students to develop these skills while providing consistent performance evaluations and formative feedback. The artificial intelligence patient actor system 300 can also be provided in a cost-effective and scalable manner, thereby providing opportunities for improved equity across differently resourced medical schools. The artificial intelligence patient actor system 300 can use artificial intelligence solely as a conversation agent, while casespecific medical knowledge can be pre-determined by subject matter experts and used by the artificial intelligence patient actor system 300 to constrain artificial intelligence. As a result, the artificial intelligence patient actor system 300 can reduce the potential for erroneous medical knowledge generation (e.g., “hallucinations”) by artificial intelligence and/or reduce the potential for potential expression of biases by artificial intelligence. The design of the artificial intelligence patient actor system 300 also allows medical educators to retain greater control over educational content, thereby facilitating a more intentional and focused learning experience for students.

(0030] Currently, without the existence of a system like the artificial intelligence patient actor system 300, students typically need other humans to serve as “actors” when learning skills such as performing a medical evaluation of a patient, for example. The actors may be paid actors that read from pre-determined scripts, for example. Additionally, coordination of this type of practice can require alignment of time commitments from the medical student, the associated teacher, and the actor. While the growing prevalence of artificial intelligence technologies can facilitate the creation of new systems for helping students develop various skills, there are several technical challenges that exist in terms of interfacing with artificial intelligence technologies (e.g., large language models (LLMs), etc.) for this purpose.

[0031] For example, challenges exist with respect to designing a system that keeps artificial intelligence “in character” by preventing the artificial intelligence from switching roles. In the particular context of simulating an interaction between a medical student and a patient for educational purposes, it can be difficult to prevent artificial intelligence from staying in character as a patient experiencing symptoms but unsure of a diagnosis. Without the appropriate guidance, the artificial intelligence can unintentionally reveal the correct diagnosis to the medical student. Technical challenges also exist in terms of designing a system that provides meaningful feedback to the medical student without requiring constant review and/or input from the associated teacher. As detailed herein, the specific design of the artificial intelligence patient actor system 300 and the associated functionality can be used to provide solutions to address these technical problems.

[0032] The artificial intelligence patient actor system 300 is shown to include a student interface system 310, a teacher interface system 320, a prompt refinement system 330, and a database 340 that stores one or more patient case files 342 and one or more assessment rubrics 344. Further, as shown, the artificial intelligence patient actor system 300 can communicate with a student device 350 and cause a student user interface 352 to be presented via the student device 350. The artificial intelligence patient actor system 300 can also communicate with a teacher device 360 and cause a teacher user interface 362 to be presented via the teach device 410. Moreover, the artificial intelligence patient actor system 300 can interface with an artificial intelligence model 372 via an application programming interface (API) 500. In some implementations, however, it should be noted that the artificial intelligence patient actor system 300 can interface with one or more separate artificial intelligence models through one or more separate application programming interfaces.

(0033] The student device 350 can be any suitable type of computing device. For example, the student device 350 can be a smartphone, a tablet, a laptop, a wearable device, a virtual reality (VR) device, and/or any other suitable type of computing device. The student device 350 can generally be used by a medical student to access the artificial intelligence patient actor system 300. As such, the medical student can use the student device 350 to develop skills with respect to performing medical examinations through the artificial intelligence patient actor system 300. In particular, the medical student can perform an artificial intelligence based simulated medical examination of a patient by interacting with the student user interface 352. The student user interface 352 can be any suitable type of user interface such as, for example, a web-based interface (e.g., as accessed through a web browser executing on the student device 350), an application interface (e.g., as accessed by launching a mobile application or a desktop application on the student device 350), or any other type of user interface.

|0034| The teacher device 360 can likewise be any suitable type of computing device. For example, the teacher device 360 can be a smartphone, a tablet, a laptop, a wearable device, a virtual reality (VR) device, and/or any other suitable type of computing device. The teacher device 360 can generally be used by a teacher associated with one or more medical students (e.g., a professor at a medical school) to access the artificial intelligence patient actor system 300. As such, the teacher can use the teacher device 360 to evaluate the performance of medical students, submit one or more of the patient case files 342 for storing in the database 340, submit one or more of the assessment rubrics 344 for storing in the database 340, and/or other actions associated with the artificial intelligence patient actor system 300. In particular, the teacher can perform these actions by interacting with the teacher user interface 362 via the teacher device 360. The teacher user interface 362 can again be any suitable type of user interface such as, for example, a web-based interface, an application interface, or any other type of user interface.

[0035] The artificial intelligence model 372 can be any suitable type of artificial intelligence model such as, for example, a large language model (LLM) or another suitable type of model. The application programming interface 370 can also be any suitable type of application programming interface. For example, the artificial intelligence model 372 can be the GPT-4 (Generative Pretrained Transformer 4) Turbo model created by OpenAI, Inc., and the application programming interface 370 can be an API provided by OpenAI, Inc. for interfacing with the GPT-4 Turbo model. As noted, however, the artificial intelligence patient actor system 300 can interface with one or more separate artificial intelligence models through one or more separate application programming interfaces. For example, the artificial intelligence patient actor system 300 can also interface with the Whisper machine learning model created by OpenAI, Inc. through an associated API and/or another speech-to-text artificial intelligence model and associated API (e g., a speech-to-text API as created by Google, LLC) to enable acceptance of voice inputs and/or the provision of voice outputs through the student user interface 352 and/or the teacher user interface 362. For example, the application programming interface 370 can manage aspects of the retrieval, processing, and dispatching of information exchanged between the artificial intelligence model 372 and the artificial intelligence patient actor system 300 during simulated student-patient interactions.

[0036] The student interface system 310 can generally cause the student user interface 352 to be presented on the student device 350. For example, the student interface system 310 can generate and send data to the student device 350 that causes the student device 350 to present the student user interface 352 to a medical student via a display of the student device 350. More specifically, the student interface system 310 can interface with frontend software tools such as, for example, Streamlit and/or other similar frontend software tools to facilitate the provision of the student user interface 352 on the student device 350. The teacher interface system 320 can generally cause the teacher user interface 362 to be presented on the teacher device 360. For example, teacher interface system 320 can generate and send data to the student device 350 that causes the teacher device 360 to present the teacher user interface 362 to a teacher that is associated with one or more medical students via a display of the teacher device 360. Additionally, the teacher interface system 320 can interface with frontend software tools such as, for example, Streamlit and/or other frontend software tools to facilitate the provision of the teacher user interface 362 on teacher device 360.

[0037| Referring to FIG. 5, an illustration showing an example of the student user interface 352 is shown, in accordance with some aspects of the disclosure. Specifically, as shown in FIG. 5, the student user interface 352 includes various user interface elements associated with a simulated interaction between a medical student and a patient. The student user interface 352 is shown to include an example input 532, where the input 532 is a question asked by the medical student to the patient. The artificial intelligence patient actor system 300 can provide the input 532 to the artificial intelligence model 372, for example. The student user interface 352 is shown to include an example output 534, where the output 534 is a response provided by the patient to the question asked by the medical student. The artificial intelligence patient actor system 300 can receive the output 534 from the artificial intelligence model 372 responsive to providing the input 532 to the artificial intelligence model 372, for example. [0038] The student user interface 352 as shown in FIG. 5 also includes selectable user interface elements that the medical student can select to order an examination or a diagnostic test for the patient. Specifically, the student user interface 352 is shown to include a physical examination button 542 that is selectable by the medical student to order a physical examination for the patient. The student user interface 352 is also shown to include a neurological examination button 544 that is selectable by the medical student to order a neurological examination for the patient. The student user interface 352 is also shown to include a diagnostic tests button 546 that is selectable by the medical student to order diagnostic tests for the patient (e.g., magnetic resonance imaging (MRI) tests, etc.). Responsive to receiving a selection of the physical examination button 542, the neurological examination button 544, or the diagnostic tests button 546, the artificial intelligence patient actor system 300 can provide an input to the artificial intelligence model 372 and receive an output back from the artificial intelligence model 372 including simulated results of the ordered examination or a diagnostic test (e.g., based on constraints defined by the patient case file 342).

[0039] The student user interface 352 as shown in FIG. 5 also includes a selectable user interface element that the medical student can select to end the patient interaction and receive feedback on the performance of the medical student with respect to the interaction (an end interaction button 520 as shown). The student user interface 352 as shown in FIG. 5 also includes a settings menu 510 including a variety of selectable fields to configure various parameters associated with the simulated interaction between the medical student and the patient. Via the settings menu 510, the medical student can toggle the input setting between text-only and speech plus text, toggle the output setting between text and speech, toggle a mode selection, select a case (e g., associated with one of the patient case files 342), and select a language, for example. The student user interface 352 as shown in FIG. 5 also includes an input field 550 through which the medical student can enter text inputs used in the simulated interaction between the medical student and the patient.

[0040] Referring to FIGS. 6A-6B, illustrations showing additional examples of the student user interface 352 are shown, in accordance with some aspects of the disclosure. Specifically, FIGS. 6A-6B show two examples of feedback that can be provided to a medical student by the artificial intelligence patient actor system 300 via the student user interface 352 upon completion of a simulated interaction between the medical student and a patient (e.g., upon receiving a selection of the end interaction button 520 via the student user interface 352). As shown in FIG. 6A, the student user interface 352 can include an input field 610 through which the medical student can enter a text input indicative of a predicted diagnosis for the patient that is based on the simulated interaction between the medical student and a patient (e.g., based on symptoms experienced by the patient, based on one or more examinations and/or diagnostic tests, etc.). Also shown in FIG. 6A, the student user interface 352 can display a correct diagnosis 620 to the medical student (e.g., a correct diagnosis as defined in the patient case file 342).

[0041] Further shown in FIG. 6 A as well as in FIG. 6B, the student user interface 352 can present a performance evaluation (e.g., based on an output that is received from the artificial intelligence model 372) to the medical student using textual and/or numerical feedback indicators. The performance evaluation presented via the student user interface 352 can be based on the predicted diagnosis that is submitted by the medical student via the input field 610, for example, as well as the associated assessment rubric 344. The numerical feedback can include scores for different categories (e.g., introduction, open ended inquiry, focused inquiry, etc.) defined in the associated assessment rubric 344, for example. The textual feedback can include any suitable text for helping the medical student improve examination skills/ For example, the textual feedback can indicate why the medical student received certain scores, provide suggestions for additional questions that should or could have been asked by the medical student, provide potential differential diagnosis for the patient’s symptoms, and/or any other suitable type of feedback that may help the medical student learn from the simulated interaction with the patient.

[0042| The prompt refinement system 330 can generally refine inputs received from the student device 350 based on interactions between the medical student and the student user interface 352 such that the inputs can be appropriately provided to the artificial intelligence model 372. The prompt refinement system 330 can tune prompts provided to the artificial intelligence model 372 by the artificial intelligence patient actor system 300 such that the outputs received by the artificial intelligence patient actor system 300 responsive to the prompts are suitable for presenting to the medical student via the student user interface 352 on the student device 350. For example, the prompt refinement system 330 can interface with backend software management tools such as, for example, LangChain and/or other backend software tools to refine prompts before providing them as input to the artificial intelligence model 372. The prompt refinement system 330 can ensure that the artificial intelligence patient actor system 300 presents various information to the artificial intelligence model 372 in a format that is efficient for interaction and context augmentation. For example, the prompt refinement system 330 can implement a chain-of-thought (CoT) prompting approach to interacting with the artificial intelligence model 372.

100431 The database 340 can be implemented using any suitable type and/or types of databases. The artificial intelligence patient actor system 300 can use the patient case files 342 and/or the assessment rubrics 344 in a retrieval -augmented generation (RAG) process to constrain outputs provided by the artificial intelligence model 372, for example. The patient case files 342 generally serve as ground truth for the artificial intelligence model 372 during simulated student-patient interactions and constrain the artificial intelligence model 372 to parameters defined in the patient case files 342. The parameters defined in the patient case files 342 can include a correct diagnosis for the patient and one or more symptoms being experienced by the patient, for example. The parameters defined in the patient case files 342 can also include other suitable information such as, for example, patient demographics (e.g., gender, age, name, occupation, etc.), a presenting complaint that may or overlap with the one or more symptoms, history of present illness, past medical history, medications, allergies, family and/or social history, review of relevant symptoms, examination results (e g., physical, neurological, etc.), diagnostic studies and/or results (e.g., MRI results, blood tests, etc ), one or more differential diagnoses, and/or any other suitable parameters that may be associated with a particular case. The patient case files 342 can include, for example, real deidentified patient images (e.g., MRI images, computed tomography (CT) scans, etc.) and/or images generated by artificial intelligence. For example, the artificial intelligence patient actor system 300 can query an external system (e.g., via one or more APIs) to receive specific types of images generated by artificial intelligence for use in one or more of the patient case files 342.

[0044] The assessment rubrics 344 generally constrain the artificial intelligence model 372 in generating performance evaluations for medical students based on simulated interactions with patients. The assessment rubrics 344 in some examples can be associated with one or more of the patient case files 342 in the database 340. The assessment rubrics 344 can use the conversational log and predicted diagnosis submitted by a medical student to provide immediate and personalized feedback to the medical student using the artificial intelligence model 372. The assessment rubrics 344 can include different rubric categories such as, for example, introduction and building rapport, open-ended inquiry, focused inquiry and probing questions, differential diagnosis development, communication skills and professionalism, and/or motivation and counseling. The assessment rubrics 344 can also define numerical scores for different categories and indicate how medical students can achieve the numerical scores. The assessments rubrics 344 generally can be used to constrain the artificial intelligence model 372 in generating performance evaluations for medical students based on simulated interactions with patients such that the feedback provided via the student user interface 352 is useful in skill development.

[0045] As noted, the artificial intelligence patient actor system 300 as well as its components as illustrated in FIG. 3 can be implemented using a variety of different hardware, software, firmware, and/or networking configurations, such as, for example, configuration that are similar to those detailed above with respect to the distributed computing environment 100 and the computing system 200. Moreover, the artificial intelligence patient actor system 300 can include more, fewer, and/or alternative arrangements of the components as illustrated in FIG. 3. For example, the student interface system 310, the teacher interface system 320, the prompt refinement system 330, and/or the database 340 can be provided as the same component or can be provided as separate components depending on the implementation of the artificial intelligence patient actor system 300. In some examples, the artificial intelligence patient actor system 300 can be implemented as a web application.

[0046] Moreover, as noted, the artificial intelligence patient actor system 300 can be adapted for use in different contexts beyond doctor-patient interviews and other types of medical applications. These contexts can include, but are not limited to, simulated legal trials, mental health counseling, business negotiations, human resource discussions, and different types of specialty training (e.g., studying for residency board exams, etc.) contexts. In these scenarios, the patient case files 342 can be adapted as appropriate to include any suitable information used to constrain outputs provided by the artificial intelligence model 372 based on parameters defined in the associated case file. The assessment rubrics 344 can likewise be adapted depending on the application to constrain performance evaluations generated by the artificial intelligence model 372. Depending on the context, the user of the user device 350 may not necessarily be a student, and the student interface system 310 can be adapted to cause presentation of the user interface 352 on the user device 350 for a user that is not necessarily a student (e.g., a practitioner, a trainee, etc.). Similarly, depending on the context, the user of the teacher device 360 may not necessarily be a teacher, and the teacher interface system 320 can be adapted to cause presentation of the user interface 362 on the teacher device 360 for a user that is not necessarily a teacher (e.g., an evaluator, a supervisor, etc.).

|0047| Further, the artificial intelligence patient actor system 300 can be adapted to allow multiple users working at the same time to practice different types of skills (e.g., working as part of a medical team, team-based negotiations, etc.). For example, the artificial intelligence patient actor system 300 can be adapted to facilitate training for scenarios where a primary care physician (PCP) refers a patient to a medical oncologist, a surgical oncologist, and/or a radiation oncologist, etc., to work together and create a treatment plan for a patient. Multiple users can interface with the patient actor system 300 at the same time such that additional user devices beyond the user device 350 and the teacher device 360 can simultaneously interface with the patient actor system 300 by accessing separate user interfaces on the separate user devices. Accordingly, the artificial intelligence patient actor system 300 can facilitate training for things like team-based negotiations, multiple physicians working together on a treatment plan for a patient, and other functionality in this manner.

[0048] Moreover, the artificial intelligence patient actor system 300 can interact with additional APIs such as, for example, one or more emotional voice APIs such that the outputs provided via the student user interface 352 can incorporate emotion and/or tone of speech detection functionality. The patient case files 342 and other similar types of constraining case files for different applications using the artificial intelligence patient actor system 300 can also include parameters to constrain the artificial intelligence model 372 to different types of personalities (e.g., exhibited by the simulated patient, etc ). For example, the personalities can include variable levels of agreeableness, neuroticism, belief in the medical system, and/or other variable types of personality traits to improve the educational experience.

[0049] Referring to FIG. 4, a flowchart illustrating an example process 400 for simulating an interaction between a medical student and a patient is shown, in accordance with some aspects of the disclosure. The process 400 can be performed by the artificial intelligence patient actor system 300 as detailed above, for example. The process 400 can generally be used to for simulate an interaction between a medical student and a patient for educational purposes. The incorporation of the patient case files 342 and the assessment rubrics 344 ensure that the artificial intelligence model 372 provides coherent, contextually appropriate, and informative responses. Additionally, the process 400 only uses artificial intelligence as a conversational agent, and does not rely on the artificial intelligence to generate any medical knowledge. The incorporation of the patient case files 342 in particular allows clinical information used in the simulated student-patient interactions to be vetted by subject matter experts. The incorporation of the patient case files 342 also allows educators to control the difficulty and the complexity of simulated student-patient interactions. As such, the process 400 allows educators to remain “in the loop” with respect to simulated studentpatient interactions.

[0050| At 402, the process 400 includes receiving a patient case file and an assessment rubric. For example, the patient case file can be one of the patient case files 342 and the assessment rubric can be one of the assessment rubrics 344. The patient case file and the assessment rubric can be received at 402 by the artificial intelligence patient actor system 300 from the teacher device 360. For example, the artificial intelligence patient actor system 300 can cause the teacher user interface 362 to be presented via the teacher device 360, and the teacher user interface 362 can then allow a teacher (e.g., a professor at a medical school) to submit the patient case file and the assessment rubric received at 402. At 404, the process 400 then includes storing the patient case file and the assessment rubric received at 402 in a database. For example, the artificial intelligence patient actor system 300 can store the patient case file and the assessment rubric received at 402 in the database 340. l0051 | At 406, the process 400 includes causing a student user interface for simulating the interaction between the medical student and the patient to be presented via a computing device. For example, at 406, the artificial intelligence patient actor system 300 can cause the student user interface 352 to be presented on the student device 350. The student user interface 352 can include a variety of different user interface elements such as detailed above with respect to FIG. 5, for example. At 408, the process 400 includes providing an initial prompt to an artificial intelligence model, where the initial prompt includes a patient case file that constrains the artificial intelligence model to parameters defined in the patient case file. For example, the artificial intelligence patient actor system 300 can provide a prompt to the artificial intelligence model 372 that includes the patient case file received at 402. The artificial intelligence patient actor system 300 can select a particular one of the patient case files 342 to provide with the initial prompt at 408 based on an input that is provided by the medical student via the student user interface 352 (e.g., a selection of a particular case via the settings menu 510). The artificial intelligence patient actor system 300 can provide the initial prompt to the artificial intelligence model 372 at 408 using the application programming interface 370, for example.

[0052] The initial prompt provided by the artificial intelligence patient actor system 300 to the artificial intelligence model 372 at 408 can include a set of instructions that provides context to the artificial intelligence model 372 and specifies the roles and rules for the user prompts (e.g., the inputs provided by the medical student via the student user interface 352) and/or specifies the chatbot responses (e.g., the outputs provided by the artificial intelligence model 372 responsive to the inputs provided by the medical student via the student user interface 352) within a particular patient-doctor interaction scenario. The patient case file can be a clinical case description provided by a human medical expert including patient signs, symptoms, and psychosocial history, for example. The initial prompt and associated inputs provided to the artificial intelligence model 372 at 408 serve as constraints to responses provided by the artificial intelligence model 372 and additional context for the artificial intelligence model 372 to ground its responses in. By constraining the artificial intelligence model 372 in this manner, the artificial intelligence patient actor system 300 can prevent the artificial intelligence model 372 from hallucinating or otherwise straying from the specific educational context intended for the educational interaction.

[0053] At 410, the process 400 includes receiving an input from the medical student associated with the interaction between the medical student and the patient from the computing device. For example, the artificial intelligence patient actor system 300 can receive an input that is provided by the medical student via the student user interface 352 from the student device 350. The medical student can provide the input (e.g., the input 532) received at 410 in various ways. For example, the medical student can enter text in the input field 550 or the input field 610 as detailed above, or the medical student can provide a voice input. The input received at 410 can also indicate that the medical student wants to order an examination or a diagnostic test for the patient (e.g., based on a selection of the physical examination button 542, the neurological examination button 544, or the diagnostic tests button 546), for example.

[0054] At 412, the process 400 includes providing the input from the medical student as input to the artificial intelligence model. For example, the artificial intelligence patient actor system 300 can provide the input that is received at 410 to the artificial intelligence model 372 at 412 using the application programming interface 370. The prompt refinement system 330 can refine the input that is received at 410 such that the artificial intelligence patient actor system 300 provides the input that is received at 410 to the artificial intelligence model 372 at 412 in a format that is efficient for interaction and context augmentation.

[0055] At 414, the process 400 includes receiving an output from the artificial intelligence model that is associated with the patient responsive to providing the input from the medical student to the artificial intelligence model. For example, the artificial intelligence patient actor system 300 can receive a response associated with the patient from the artificial intelligence model 372 at 414 through the application programming interface 370. Since the artificial intelligence patient actor system 300 constrains the artificial intelligence model 372 based on the patient case file provided in the initial prompt at 408, the response received from the artificial intelligence model 372 is likely to be coherent, contextually appropriate, and informative. The response received from the artificial intelligence model 372 can include the results of an examination or a diagnostic test, a conversational response that the simulated patient provides to the medical student, or any other kind of response that may be associated with the patient.

[0056] At 416, the process 400 includes causing the output from the artificial intelligence model to be presented via the student user interface on the computing device. For example, the artificial intelligence patient actor system 300 can cause the response received at 414 to be presented via the student user interface 352 on the student device 350. The artificial intelligence patient actor system 300 can cause the response to be presented via the student user interface 352 on the student device 350 in various ways at 416, such as, for example, as text or as audio in the form of speech. Again, since the artificial intelligence patient actor system 300 constrains the artificial intelligence model 372 in accordance with the patient case file that is provided in the initial prompt at 408, the response presented to the medical student is likely to be coherent, contextually appropriate, and informative such that it helps facilitate learning for the medical student.

[0057] At 418, the process 400 includes determining that the interaction between the medical student and the patient is complete. For example, the artificial intelligence patient actor system 300 can determine that the interaction is complete responsive to receiving an indication that the medical student has selected the end interaction button 520 via the student user interface 352 from the student device 350. The artificial intelligence patient actor system 300 can also receive an indication from the artificial intelligence model 372 through the application programming interface 370 indicating that the interaction is complete (e.g., because the details in the patient case file have been exhausted, because the medical student or the patient indicated that the interaction has been completed, etc ). Other possible approaches can be implemented using the artificial intelligence patient actor system 300 to determine that the interaction between the medical student and the patient is complete at 418.

[0058] At 420, the process 400 includes providing an assessment rubric to the artificial intelligence model that constrains the artificial intelligence model in generating a performance evaluation for the medical student. For example, the artificial intelligence patient actor system 300 can provide the assessment rubric received at 402 to the to the artificial intelligence model 372 via the application programming interface 370. The assessment rubric provided at 420 can guide the artificial intelligence model 372 to generate the performance evaluation by using the conversational log and/or the predicted diagnosis submitted by the medical student (e.g., via the input field 610) to provide immediate and personalized feedback to the medical student. The assessment rubric provided at 420 constrains the artificial intelligence model 372 in generating the performance evaluation for the medical student such that the feedback provided by the artificial intelligence model 372 is useful to the medical student in terms skill development. The artificial intelligence patient actor system 300 can provide the assessment rubric to the artificial intelligence model 372 with the initial prompt at 408 in some examples.

[0059] At 422, the process 400 includes receiving the performance evaluation for the medical student from the artificial intelligence model. For example, the artificial intelligence patient actor system 300 can receive the performance evaluation from the artificial intelligence model 372 via the application programming interface 370. Again, since the since the artificial intelligence patient actor system 300 constrains the artificial intelligence model 372 in accordance with the assessment rubric, the performance evaluation generated by the artificial intelligence model 372 is likely to be useful to the medical student in terms skill development. At 424, the process 400 includes causing the performance evaluation for the medical student to be presented via the student user interface on the computing device. For example, the artificial intelligence patient actor system 300 can cause the performance evaluation received at 422 to be presented via the student user interface 352 on the student device 350. The artificial intelligence patient actor system 300 can cause the performance evaluation to be presented via the student user interface 352 on the student device 350 in various ways at 424, such as, for example, as text or as audio in the form of speech.

100601 The process 400 as detailed above can be adapted for use in various educational contexts beyond medical students and patients. For example, the process 400 can be adapted for use in contexts such as mock legal trials, mental health counseling, business negotiations (e.g., as part of a business school), human resource discussions and applications (e.g., performance reviews, disciplinary, performance improvements, etc.), specialty training (e.g., studying for residency board exams, etc.), and other types of conversational training purposes. In such examples, the interaction may not necessarily be between a medical student and a patient. Instead, the interaction can more generally be an educational interaction associated with a student (e.g., a law student, a business student, a nursing student, etc.) that is facilitated as part of the process 400. The “student” in these examples may not necessarily be a student that is enrolled at an educational institution (e.g., a graduate program at a university, etc.). Instead, the “student” can be any person that wishes to learn through the educational interaction facilitated via the process 400. Additionally, the “case files” used to constraint the artificial intelligence model 372 in these scenarios can include any suitable types of constraints for placing on the artificial intelligence model 372 for a given educational interaction to prevent the artificial intelligence model 372 from hallucinating or otherwise straying from the specific educational context intended for the educational interaction.

[00611 It should be noted that, while the steps of the process 400, 600, 700, and 800 are shown in a particular order in FIG. 4, in some implementations, the processes 400 may not include all steps shown, may include additional steps, and/or may include the shown steps in a different order. Further, the steps of the process 400 can be combined in different ways in certain implementations.

|0062| Other examples and uses of the disclosed technology will be apparent to those having ordinary skill in the art upon consideration of the disclosure. The specification and examples given should be considered as examples only, and it is contemplated that the appended claims will cover any other such implementation or modifications as fall within the true scope of the disclosure. .

Claims

1. A system for simulating an interaction between a medical student and a patient, the system comprising: a database comprising a patient case file and an assessment rubric; a prompt refinement system that provides the patient case file and the assessment rubric to an artificial intelligence model; and a student interface system that causes a student user interface for simulating the interaction between the medical student and the patient to be presented via a computing device; wherein the student interface system receives inputs from the medical student and provides outputs associated with the patient to the medical student via the student user interface; and wherein the prompt refinement system provides the inputs from the medical student to the artificial intelligence model and receives the outputs associated with the patient from the artificial intelligence model.

2. The system of claim 1, comprising a teacher interface system that causes a teacher user interface for evaluating performance of the medical student with respect to the interaction between the medical student and the patient to be presented via a second computing device.

3. The system of claim 2, wherein the teacher user interface allows the teacher to submit the patient case file or the assessment rubric for storing in the database.

4. The system of claim 1, wherein the patient case file constrains the artificial intelligence model to parameters defined in the patient case file.

5. The system of claim 4, wherein the parameters defined in the patient case file comprise a symptom experienced by the patient and a correct diagnosis for the patient.

6. The system of claim 1, wherein the assessment rubric constrains the artificial intelligence model in generating a performance evaluation for the medical student upon completion of the interaction.

7. The system of claim 1 , wherein the artificial intelligence model comprises a large language model (LLM).

8. The system of claim 1, wherein the student user interface allows the medical student to order an examination or a diagnostic test for the patient.

9. A computer-implemented method for simulating an interaction between a medical student and a patient, the method comprising: causing a student user interface for simulating the interaction between the medical student and the patient to be presented via a computing device; providing an initial prompt to an artificial intelligence model, the initial prompt comprising a patient case file that constrains the artificial intelligence model to parameters defined in the patient case file; receiving an input from the medical student associated with the interaction between the medical student and the patient from the computing device; providing the input from the medical student to the artificial intelligence model; receiving an output from the artificial intelligence model that is associated with the patient responsive to providing the input from the medical student to the artificial intelligence model; causing the output from the artificial intelligence model to be presented via the student user interface on the computing device; determining that the interaction between the medical student and the patient is complete; providing an assessment rubric to the artificial intelligence model that constrains the artificial intelligence model in generating a performance evaluation for the medical student; receiving the performance evaluation for the medical student from the artificial intelligence model; and causing the performance evaluation for the medical student to be presented via the student user interface on the computing device.

10. The method of claim 9, wherein the student user interface allows the medical student to order an examination or a diagnostic test for the patient.

11. The method of claim 9, comprising causing a teacher user interface for reviewing the performance evaluation for the medical student to be presented via a second computing device.

12. The method of claim 11, wherein the teacher user interface allows a teacher to submit the patient case file or the assessment rubric.

13. The method of claim 9, wherein the parameters defined in the patient case file comprise a symptom experienced by the patient and a correct diagnosis for the patient.

14. The method of claim 9, wherein the artificial intelligence model comprises a large language model (LLM).

15. A non-transitory computer-readable storage medium having instructions stored thereon that, when executed by processing circuitry, cause the processing circuitry to: cause a student user interface for simulating the interaction between the medical student and the patient to be presented via a computing device; provide an initial prompt to an artificial intelligence model, the initial prompt comprising a patient case file that constrains the artificial intelligence model to parameters defined in the patient case file; receive an input from the medical student associated with the interaction between the medical student and the patient from the computing device; provide the input from the medical student to the artificial intelligence model; receive an output from the artificial intelligence model that is associated with the patient responsive to providing the input from the medical student to the artificial intelligence model; cause the output from the artificial intelligence model to be presented via the student user interface on the computing device; determine that the interaction between the medical student and the patient is complete; provide an assessment rubric to the artificial intelligence model that constrains the artificial intelligence model in generating a performance evaluation for the medical student; receive the performance evaluation for the medical student from the artificial intelligence model; and cause the performance evaluation for the medical student to be presented via the student user interface on the computing device.

16. The computer-readable medium of claim 15, wherein the student user interface allows the medical student to order an examination or a diagnostic test for the patient.

17. The computer-readable medium of claim 15, wherein the instructions, when executed by the processing circuitry, cause the processing circuitry to cause a teacher user interface for reviewing the performance evaluation to be presented via a second computing device.

18. The computer-readable medium of claim 17, wherein the teacher user interface allows a teacher to submit the patient case file or the assessment rubric.

19. The computer-readable medium of claim 15, wherein the parameters defined in the patient case file comprise a symptom experienced by the patient and a correct diagnosis for the patient.

20. The computer-readable medium of claim 15, wherein the artificial intelligence model comprises a large language model (LLM).

21. A system for simulating an educational interaction associated with a student, the system comprising: a database comprising a case file and an assessment rubric; a prompt refinement system that provides the case file and the assessment rubric to an artificial intelligence model; and a student interface system that causes a student user interface for simulating the educational interaction associated with the student to be presented via a computing device; wherein the student interface system receives inputs from the student and provides outputs associated with the educational interaction to the student via the student user interface; and wherein the prompt refinement system provides the inputs from the student to the artificial intelligence model and receives the outputs associated with the educational interaction from the artificial intelligence model.

22. The system of claim 21, comprising a teacher interface system that causes a teacher user interface for evaluating performance of the student with respect to the educational interaction to be presented via a second computing device.

23. The system of claim 22, wherein the teacher user interface allows the teacher to submit the case file or the assessment rubric for storing in the database.

24. The system of claim 21, wherein the case file constrains the artificial intelligence model to parameters defined in the case file.

25. The system of claim 21, wherein the assessment rubric constrains the artificial intelligence model in generating a performance evaluation for the student upon completion of the educational interaction.

26. The system of claim 21, wherein the artificial intelligence model comprises a large language model (LLM).

27. A computer-implemented method for simulating an educational interaction associated with a student, the method comprising: causing a student user interface for simulating the educational interaction associated with the student to be presented via a computing device; providing an initial prompt to an artificial intelligence model, the initial prompt comprising a case file that constrains the artificial intelligence model to parameters defined in the case file; receiving an input from the student that is associated with the educational interaction from the computing device; providing the input from the student to the artificial intelligence model; responsive to providing the input from the student to the artificial intelligence model, receiving an output from the artificial intelligence model that is associated with the educational interaction; causing the output from the artificial intelligence model to be presented via the student user interface on the computing device; determining that the educational interaction associated with the student is complete; providing an assessment rubric to the artificial intelligence model that constrains the artificial intelligence model in generating a performance evaluation for the student; receiving the performance evaluation for the student from the artificial intelligence model; and causing the performance evaluation for the student to be presented via the student user interface on the computing device.

28. The method of claim 27, comprising causing a teacher user interface for reviewing the performance evaluation for the student to be presented via a second computing device.

29. The method of claim 28, wherein the teacher user interface allows a teacher to submit the case file and the assessment rubric.

30. The method of claim 27, wherein the artificial intelligence model comprises a large language model (LLM).

31. A non-transitory computer-readable storage medium having instructions stored thereon that, when executed by processing circuitry, cause the processing circuitry to: cause a student user interface for simulating the educational interaction associated with the student to be presented via a computing device; provide an initial prompt to an artificial intelligence model, the initial prompt comprising a case file that constrains the artificial intelligence model to parameters defined in the case file; receive an input from the student that is associated with the educational interaction from the computing device; provide the input from the student to the artificial intelligence model; responsive to providing the input from the student to the artificial intelligence model, receive an output from the artificial intelligence model that is associated with the educational interaction; cause the output from the artificial intelligence model to be presented via the student user interface on the computing device; determine that the educational interaction associated with the student is complete; provide an assessment rubric to the artificial intelligence model that constrains the artificial intelligence model in generating a performance evaluation for the student; receive the performance evaluation for the student from the artificial intelligence model; and cause the performance evaluation for the student to be presented via the student user interface on the computing device.

32. The computer-readable medium of claim 31, wherein the instructions, when executed by the processing circuitry, cause the processing circuitry to cause a teacher user interface for reviewing the performance evaluation for the student to be presented via a second computing device.

33. The computer-readable medium of claim 32, wherein the teacher user interface allows a teacher to submit the case fde and the assessment rubric.

34. The computer-readable medium of claim 31, wherein the artificial intelligence model comprises a large language model (LLM).