Objective: The aim of this study is to objectively detect falls and their direction in Parkinson’s disease (PD) patients during everyday activities.

Background: Falling is one of the debilitating features in PD. However, patients may forget the circumstances of falls. There have been no satisfying methods to identify and quantify falls outside of the clinic.

Methods: Patients were selected from among 36 patients who participated in our previous prospective study on FOG and falls. We developed a motion recorder (body-fixed 3D accelerometer) with a long-lasting battery. The movements of healthy volunteers and recurrent PD fallers with severe FOG were recorded using the waist-mounted device in the outpatient clinic and during their everyday activities. A newly developed algorithm for determining falling directions was applied to the control subjects mimicking falls and the frequent fallers. We also recorded a rapid oscillation due to knee trembling before falls. Video recording was performed in some cases.

Results: Characteristic patterns of acceleration signals were recorded for freezing and falls. Knee trembling was recorded as a rapid oscillation of acceleration. Imitating falls by healthy volunteers and actual falls of PD patients in everyday life were detected as abrupt trunk angle changes. In addition, falling direction (forward, lateral, and backward) was determined by our measurements with confirmation by video recordings.

Conclusions: Motion recording using our wearable 3D sensor is useful for detecting falls in PD patients. Determining falling directions may contribute to the pathophysiology of recurrent falls.

« Back to 2018 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/determination-of-falling-direction-in-parkinsons-disease-patients-using-a-wearable-sensor/