INTEGRATING SIMULATION TECHNOLOGIES AND HYBRID ARTIFICIAL INTELLIGENCE IN TEACHING BIOPHYSICS TO MEDICAL STUDENTS

Abstract

This article proposes an instructional model for integrating simulation technologies and hybrid artificial intelligence (AI) into biophysics teaching for medical students. Hybrid AI is operationally defined as the coordinated use of explicit biophysical rules and computational checks with a neural language component that provides adaptive, source-grounded feedback under teacher supervision. The approach distinguishes scientific simulation from AI-generated explanation: the simulator produces model-based evidence, whereas AI supports interpretation without replacing the student's reasoning. A conceptual instructional-design methodology links learning outcomes, simulation tasks, verification rules, scaffolding, and individual assessment. Six modules are considered: measurement and uncertainty, hemodynamics, membrane transport, bioelectricity, biomedical optics, and ultrasound. The proposed architecture comprises verified content, simulation-based inquiry, symbolic validation, adaptive dialogue, and evidence-based assessment. A 90-minute lesson sequence, a module integration matrix, performance criteria, and implementation safeguards are specified. A hemodynamics example demonstrates how prediction, parameter manipulation, quantitative verification, and critical discussion can be connected within one learning cycle. The central design requirement is that students independently formulate predictions, evaluate numerical evidence, identify model limitations, and defend their conclusions. The contribution is a testable methodological framework rather than an experimentally validated intervention. Its educational value should be investigated through authentic measures of conceptual understanding, quantitative reasoning, delayed retention, and transfer to physical laboratory tasks.