Accelerometric signals in automatic balance assessment.

摘要: The paper presents the automatic computer-aided balance assessment system for supporting and monitoring diagnosis rehabilitation process of patients with limited mobility or disabled in home environment. has adopted seven Berg Balance Scale activities. approach is based on accelerometric signals acquired by inertial sensors worn patient. Several specific, mostly medium frequency features are introduced discussed. reduction feature vector been performed using multilevel Fisher's linear discriminant. classification employs multilayer perceptron artificial neural network. direct effectiveness ranges from 75% to 94% various