Input representations and classification strategies for automated human gait analysis

• Published in: Gait & posture Vol. 76; no. NA; pp. 198 - 203
• Main Authors: Slijepcevic, Djordje, Zeppelzauer, Matthias, Schwab, Caterine, Raberger, Anna-Maria, Breiteneder, Christian, Horsak, Brian
• Format: Journal Article
• Language: English
• Published: England Elsevier B.V 01.02.2020
• Subjects: Gait classification; Gait disorders; Ground reaction force; Machine learning; Support vector machine; Gait classification; Support vector machine; Ground reaction force; Machine learning; Gait disorders

•Aggregating data from several trials of one person is advantageous.•Majority voting on classifier's predictions and mean waveforms performed best.•Utilizing derived signal representations (e.g. signal differences) is advantageous.•Combining original input signals and derived representations is...