Interpretability of Input Representations for Gait Classification in Patients after Total Hip Arthroplasty

• Published in: Sensors (Basel, Switzerland) Vol. 20; no. 16; p. 4385
• Main Authors: Dindorf, Carlo, Teufl, Wolfgang, Taetz, Bertram, Bleser, Gabriele, Fröhlich, Michael
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 06.08.2020; MDPI
• Subjects: Accuracy; Algorithms; Arthritis; Arthroplasty, Replacement, Hip; Artificial Intelligence; Biomechanical Phenomena; Biomechanics; Classification; explainable artificial intelligence; Gait; Humans; inertial measurement unit; Joint replacement surgery; Joint surgery; Kinematics; Knee Joint; Machine Learning; Osteoarthritis; total hip replacement; explainable artificial intelligence; inertial measurement unit; total hip replacement; biomechanics; machine learning; gait

Many machine learning models show black box characteristics and, therefore, a lack of transparency, interpretability, and trustworthiness. This strongly limits their practical application in clinical contexts. For overcoming these limitations, Explainable Artificial Intelligence (XAI) has shown prom...