Design and development of an interactive and customizable digital patient avatar to suit the needs of medical simulation operators

Nathan English

doi:10.17918/00010892

While current technology for developing and delivering medical simulations for nurses in training is ever expanding and improving, current platforms are still limited in their ability to allow the Subject Matter Experts (SMEs) to create or customize existing scenarios while presenting the same content across different modalities to reach their global audience. In order for simulation training to be more widely applicable and accessible world-wide, scenarios require a large range of patient types (e.g. gender, ethnicity, body dimensions, medical conditions) to mirror the diversity of real patients across the globe. In addition, they must be able to meet their learners in a variety of learning settings, including in a simulation lab, in a classroom, or even at home, and must support a range of delivery and interaction modes, including web-based point-and-click delivery, tablet or mobile delivery, computer-based delivery, or augmented reality-based delivery. However, many current simulation platforms experience time, cost, and hardware limitations that prevent certain modes of delivery, interaction options, and custom content creation from being available to simulation operators. One major shortcoming of current technologies is the limited ability for simulation operators (subject matter experts) to dynamically change the visual and behavioral parameters of simulated patients and their interaction capabilities in a flexible, reusable, and cost-effective way without limitations imposed by the simulation developer. Provided simulated patients often have limited conversational ability, limited animations, and limited ability for modification. Secondly, medical simulations are often developed for and limited to one mode of delivery, making them potentially hardware-prohibitive for lower fidelity simulation systems and limiting the range of interaction modes that students can utilize without purchasing extra content for different delivery modalities and learning environments. This research aimed to address these shortcomings by developing a highly customizable, reactive, and reusable 3D patient avatar integrated in the flexible Cerebrum[TM] medical simulation software platform, such that simulation operators have access to a wide range of interactive patient avatars to tailor simulation scenarios without prohibitive costs or turnaround times. Through augmenting a pre-existing patient avatar, characteristics such as appearance, age, gender, conditions, and diagnoses were made changeable through plain-text keywords and variables, and Large-Language-Model powered interpretations of these keywords ensures that patient behaviors, emotions, and speech can be tailored in real-time to fit any scenario description provided by the operator. The digital patient avatar is suited for use with augmented reality headsets, such that it can be used in both lower-end digital simulations in any location or in combination with physical training manikin labs for a holistic digital-physical mixed-reality simulated patient experience. This research provides foundations for a viable option for healthcare professionals that require highly flexible, cost-effective, and realistically reactive 3D simulated patients to tailor simulation scenarios. The digital patient avatar is highly adaptable to fit different simulation scenarios and deployment methods without sacrificing fidelity, reactivity, or customizability, and is open for further augmentations to improve visual realism and interactivity.

Design and development of an interactive and customizable digital patient avatar to suit the needs of medical simulation operators

Files and links (1)

Abstract

Metrics

Details

Design and development of an interactive and customizable digital patient avatar to suit the needs of medical simulation operators

Files and links (1)

Abstract

Metrics

Details

Drexel University Social media