A Framework for User-Centered Adoption of AI Fall Detection Systems in Healthcare

Abstract

Falls are one of the main causes of injury and death among older adults, making them a serious healthcare challenge. AI-based fall detection systems using sensors and machine learning have the potential to prevent injuries by sending quick alerts to caregivers and healthcare providers. However, these systems are not widely adopted because of problems such as difficult interfaces, too many false alarms and low user trust. This paper presents a framework for improving the adoption of AI fall detection systems in healthcare through a user-centered approach. The framework combines usability, user experience (UX) and technology adoption models (TAM and UTAUT) to explain how users interact with these systems and what factors influence their acceptance. The proposed framework highlights the needs of three main groups, namely elderly users, family caregivers and healthcare professionals, and explains how design improvements can reduce barriers while increasing trust, satisfaction and long-term use. The study proposes a mixed-method framework combining interviews, observations, surveys, and usability testing to guide the evaluation and refinement of AI fall detection systems in future research. Prototypes will be created and refined using user feedback to test how design changes affect usability and adoption. Unlike previous studies that separately address usability or adoption, this paper integrates user-centered design principles with TAM and UTAUT models to form a unified framework that guides the development of AI-based fall detection systems.

Ähnliche Arbeiten