Inertial Sensor Based Detection of Freezing of Gait for On-Demand Cueing in Parkinson’s Disease
| Autor: | Andrea A. Kühn, Christina Salchow-Hömmen, Thomas Seel, Nikolaus Wenger, H. Voigt, Ardit Dvorani, M.C.E. Jochner, Thomas Schauer, J. Meyer-Ohle, Constantin Wiesener |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
0209 industrial biotechnology
Inertial frame of reference Orientation (computer vision) business.industry Computer science 020208 electrical & electronic engineering Parkinsonian gait 02 engineering and technology Acceleration 020901 industrial engineering & automation Gait (human) Control and Systems Engineering Inertial measurement unit 0202 electrical engineering electronic engineering information engineering medicine Computer vision Artificial intelligence Sensitivity (control systems) medicine.symptom Falling (sensation) business |
| Zdroj: | IFAC-PapersOnLine. 53:16004-16009 |
| ISSN: | 2405-8963 |
| Popis: | Freezing of Gait (FoG) is one of the cardinal symptoms of Parkinson’s disease, which arises in the late stages of the disease. It affects the gait cycle and increases the risk of falling. FoG leads to heterogeneous gait cycles, which makes the detection of gait phases and events difficult. In this article, we introduce a new inertial measurement unit-based approach for detecting Parkinsonian gait phases based on the acceleration, velocity, rate of turn and orientation of the foot. Furthermore, we introduce a new gait evaluation measurement, the so-called GaitScore, for distinguishing between normal and FoG-affected motion phases and thus for detecting FoG episodes. Preliminary results show that the extreme values of the pitch angle during a motion phase provide valuable information for the detection of FoG. The proposed method can detect FoG episodes with a sensitivity of 97% and specificity of 87%. The reference data were generated by clinical experts who annotated FoG episodes in video data synchronized with the measurements of the inertial sensors. The detection of FoG in real-time enables on-demand cueing. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|