Sensoring and features extraction for the detection of Freeze of Gait in Parkinson disease

Freezing of Gait (FoG) in Parkinson disease (PD) is a sudden episode characterized by brief failure to walk. This study aims to build a preliminary prototype that is able to acquire signals from different sensors, and detect changes during FoG. This paper gives the basic concepts to build an algorithm. The different kinds of sensors used in this study are two acceleration sensors, two telemeter sensors and a goniometer. The acquired signals were compared to normal gait modes and analyzed in order to extract both time and frequency domain features that can separate the FoG class from the other modes. After selectingoptimal features, a windowing technique is applied to confirm the significance of these features when FoG episode occurs.Results prove the feasibility to detect FoG usingnew types of sensors and new extracted features.