DESIGNING AI-ASSISTED HEALTH EDUCATION FOR OLDER ADULTS: A HUMAN-CENTERED FRAMEWORK BASED ON GERIATRIC ASSESSMENT
Abstract
This study addresses the gap in health literacy among geriatric populations, which is influenced by cognitive decline, visual impairment, and limited digital independence. Using a mixed-methods approach based on Comprehensive Geriatric Assessment (CGA) data, this research identifies key challenges faced by elderly users in accessing and understanding health information. The findings reveal a strong preference for simple, high-contrast visual communication supported by clear narrative structure and familiar contextual representation. Based on these insights, the study proposes a human-centered visual communication framework implemented through short-form educational video, designed to be accessible, repetitive, and cognitively adaptive. The production process incorporates an AI-assisted workflow, including prompt-based visual generation, character design, and multimodal integration of visual and audio elements. The use of 3D animated characters as instructional agents enhances engagement and facilitates comprehension through narrative and emotional cues. This study contributes to the development of an adaptive visual communication design framework for geriatric health education, highlighting the role of AI as an assistive tool within a human-in-the-loop design process. The findings provide practical implications for designing inclusive, accessible, and contextually relevant health communication media for elderly users and their caregivers.