A Footwear-Based Methodology for Fall Detection

摘要: Automatic fall detection is an active research area since several years. Basically, this motivated by the impact that falls have, in terms of mortality, morbidity, and social costs, which make them comparable to road traffic injuries. The early a can be critical reduce mortality rate limit associated health consequences. Technological solutions designed automatically detect notify may classified into wearable non-wearable. Among former ones, use specific devices worn subject very common assumption, but it fails address user’s acceptability issues. In fact, position sensor or its visibility perceived as stigma with primary function detection. To such issue, paper presents methodology for relies on pair smart shoes, equipped force sensors tri-axial accelerometer, able supervising system. instrumented footwear enables analysis subject’s motion foot orientation, recognizing abnormal configurations. developed algorithm not computationally intensive, therefore, easily executed board device. Laboratory tests provided satisfactory performances correct classification: 544 136 activities daily living, performed 17 healthy subjects, 97.1% accuracy has been achieved. Further experiments involving two elderly users demonstrate effectiveness proposed method real-life scenario.