Constellation, le dépôt institutionnel de l'Université du Québec à Chicoutimi

Résumé

Degradation of postural control observed with aging is responsible for balance problems in the elderly, especially during the activity of walking. This gradual loss of performance generates abnormal gait, and therefore increases the risk of falling. This risk worsens in people with neuronal pathologies like Parkinson and Ataxia diseases. Many clinical tests are used for fall assessment such as the Timed up and go (TUG) test. Recently, many works have improved this test by using instrumentation, especially body-worn sensors. However, during the instrumented TUG (iTUG) test, the type of ground can influence risk of falling. In this paper, we present a new methodology for fall risk assessment based on quantitative gait parameters measurement in order to improve instrumented TUG test. The proposed measurement unit is used on different types of ground, which is known to affect human gait. The methodology is closer to the real walking environment and therefore enhances ability to differentiate risks level. Our system assesses the risk of falling's level by quantitative measurement of intrinsic gait parameters using fuzzy logic. He is also able to measure environmental parameters such as temperature, humidity and atmospheric pressure for a better evaluation of the risk in activities of daily living (ADL).