Proposição de um serious game para dispositivos móveis com um modelo inteligente para avaliar o conhecimento médico em suporte avançado de vida cardiovascular

Abstract

Introduction: Advanced Cardiovascular Life Support (ACLS) training is essential for equipping healthcare professionals who work in emergency settings. Traditionally, this training is conducted in person, featuring realistic simulations and assessments led by instructors. However, this approach presents challenges such as high costs, which hinder frequent reassessments, and the need for specific infrastructure. Objective: In light of these challenges, this study proposes the development of SAV-me, a graphic interactive game for mobile platforms designed to complement traditional ACLS training, assess users’ theoretical

knowledge, and provide feedback through an intelligent model. Materials and Methods: SAV- me was built according to the 2020 guidelines from the American Heart Association (AHA), a

reference widely adopted in Brazil and the Americas due to the international acceptance of these algorithms. The decision-making core employs a Knowledge-Based System (KBS) that compares each player’s action to the official algorithm. Throughout the game, feedback is provided—brief messages and sounds that confirm the action taken—and at the end, a report is generated. This report is a synthetic summary highlighting correct actions, mistakes, and omissions. Although it does not yet assess motor skills, SAV-me has proven to be an accessible and reliable solution for frequent refreshers of theoretical ACLS knowledge, complementing—

but not replacing—hands-on training. Results and Discussion: The results indicate that SAV- me is a promising tool for evaluating theoretical knowledge in ACLS. The implementation of

a KBS allowed for objective and structured assessment, ensuring standardization and reliability in performance evaluation. SAV-me intelligent model provides feedback during and after gameplay—a summary report of correct and incorrect responses and omissions. This mechanism accelerated self-correction, increased engagement, and kept training costs low thanks to mobile distribution. Conclusion: SAV-me achieved all its goals by operationalizing theoretical evaluation criteria in a KBS aligned with AHA algorithms. The automatic evaluation provides feedback highly consistent with that of specialists. The proposed assessment model not only ensures reliability of results but also contributes to a more autonomous, interactive, and motivating learning experience, making training more accessible and effective for healthcare professionals. Explicit rules ensure transparency and allow for quick updates when new guidelines are released. The game complements but does not replace in-person training; practices requiring motor skills assessment cannot be addressed in SAV-me. Distributed on mobile devices, it reduces costs and proves to be an excellent alternative to the logistical limitations of in-person courses.