More Like this

1. Comparing alternative approaches to debriefing in a tool to support peer-led simulation-based training

   https://doi.org/10.1007/978-3-031-09680-8_8  

   Katz, S.; Albacete, P.; Gallagher, J.; Jordan, P.; Platt, T.; Silliman, S.; Yang, T. (January 2022, Intelligent Tutoring Systems: 18th International Conference, ITS 2022)

   This poster describes an early-stage project. It introduces MedDbriefer, a tablet-based tool that allows small groups of paramedic students to practice realistic prehospital emergency care scenarios. While two or more students collaborate as members of an emergency medical service (EMS) team, a peer uses the tablet’s checklists to record the team’s actions. The system then analyzes the event log to provide an automated debriefing on the team’s performance. Although debriefing is purported to be one of simulation-based training’s most critical components, there is little research to guide human and automated debriefing. We are imple-menting two approaches to automated debriefing and will compare their effective-ness in an upcoming randomized controlled trial. 

   more » « less
2. Implementing Distributed Feedback in a Tool that Supports Peer-to-Peer Simulation in Healthcare

   Katz, Sandra; Albacete, Patricia; Jordan, Pamela; Silliman, Scott; Wrzesniewski, Matthew (June 2024, Springer, Cham)

   Sifaleras, A; Lin, F (Ed.)

   MedDbriefer is a web based ITS designed to enable healthcare students to do clinical scenarios anytime, anywhere. While one student “voice treats” a scenario’s patient(s) as the leader of a mock Emergency Medical Services (EMS) team, a peer records the team’s actions by using the system’s checklists, on a tablet. When the scenario ends, MedDbriefer analyzes the event log and generates a debriefing. MedDbriefer also provides a platform for research on simulation-based training. This paper describes how the system’s debriefing engine could be extended to deliver feedback during a scenario, as well as afterwards. MedDbriefer could then be used to compare the effectiveness of different ways of timing feedback delivery in computer-based simulation systems. 

   more » « less
3. MultiHeart: Secure and Robust Heartbeat Pattern Recognition in Multimodal Cardiac Monitoring System

   https://doi.org/10.3390/electronics14153149  

   Ahmadi, Hossein; Zhang, Yan; Tran, Nghi H (August 2025, Electronics)

   The widespread adoption of heartbeat monitoring sensors has increased the demand for secure and trustworthy multimodal cardiac monitoring systems capable of accurate heartbeat pattern recognition. While existing systems offer convenience, they often suffer from critical limitations, such as variability in the number of available modalities and missing or noisy data during multimodal fusion, which may compromise both performance and data security. To address these challenges, we propose MultiHeart, which is a robust and secure multimodal interactive cardiac monitoring system designed to provide reliable heartbeat pattern recognition through the integration of diverse and trustworthy cardiac signals. MultiHeart features a novel multi-task learning architecture that includes a reconstruction module to handle missing or noisy modalities and a classification module dedicated to heartbeat pattern recognition. At its core, the system employs a multimodal autoencoder for feature extraction with shared latent representations used by lightweight decoders in the reconstruction module and by a classifier in the classification module. This design enables resilient multimodal fusion while supporting both data reconstruction and heartbeat pattern classification tasks. We implement MultiHeart and conduct comprehensive experiments to evaluate its performance. The system achieves 99.80% accuracy in heartbeat recognition, surpassing single-modal methods by 10% and outperforming existing multimodal approaches by 4%. Even under conditions of partial data input, MultiHeart maintains 94.64% accuracy, demonstrating strong robustness, high reliability, and its effectiveness as a secure solution for next-generation health-monitoring applications. 

   more » « less
4. Haptic paradigms for multimodal interactive simulations

   Tennison, J. L.; Greenberg, J.; Moore, E. B.; & Gorlewicz, J. L. (January 2021, Journal on technology and persons with disabilities)

   Touch is often omitted or viewed as unnecessary in digital learning. Lack of touch feedback limits the accessibility and multimodal capacity of digital educational content. Touchscreens with vibratory, haptic feedback are prevalent, yet this kind of feedback is often under-utilized. This work provides initial investigations into the design, development, and use of vibratory feedback within multimodal, interactive, educational simulations on touchscreen devices by learners with and without visual impairments. The objective of this work is to design and evaluate different haptic paradigms that could support interaction and learning in educational simulations. We investigated the implementation of four haptic paradigms in two physics simulations. Interviews were conducted with eight learners (five sighted learners; three learners with visual impairments) on one simulation and initial results are shared. We discuss the learner outcomes of each paradigm and how they impact design and development moving forward. 

   more » « less
5. Automated Assessment in Math Education: A Comparative Analysis of LLMs for Open-Ended Responses

   https://doi.org/10.5281/zenodo.12729932  

   Baral, Sami; Worden, Eamon; Lim, Wen-Chiang; Luo, Zhuang; Santorelli, Christopher; Gurung, Ashish (January 2024, International Educational Data Mining Society)

   Benjamin, Paaßen; Carrie, Demmans Epp (Ed.)

   The effectiveness of feedback in enhancing learning outcomes is well documented within Educational Data Mining (EDM). Various prior research have explored methodologies to enhance the effectiveness of feedback to students in various ways. Recent developments in Large Language Models (LLMs) have extended their utility in enhancing automated feedback systems. This study aims to explore the potential of LLMs in facilitating automated feedback in math education in the form of numeric assessment scores. We examine the effectiveness of LLMs in evaluating student responses and scoring the responses by comparing 3 different models: Llama, SBERT-Canberra, and GPT4 model. The evaluation requires the model to provide a quantitative score on the student's responses to open-ended math problems. We employ Mistral, a version of Llama catered to math, and fine-tune this model for evaluating student responses by leveraging a dataset of student responses and teacher-provided scores for middle-school math problems. A similar approach was taken for training the SBERT-Canberra model, while the GPT4 model used a zero-shot learning approach. We evaluate and compare the models' performance in scoring accuracy. This study aims to further the ongoing development of automated assessment and feedback systems and outline potential future directions for leveraging generative LLMs in building automated feedback systems. 

   more » « less