Estimating individual minimum calibration for deep-learning with predictive performance recovery: An example case of gait surface classification from wearable sensor gait data

• Published in: Journal of biomechanics Vol. 154; p. 111606
• Main Authors: Lam, Guillaume, Rish, Irina, Dixon, Philippe C.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.06.2023; Elsevier Limited
• Subjects: Adult; Biomechanics; Calibration; Clinical datasets; Clinical trials; Data points; Datasets; Deep Learning; Gait; Humans; Inertial platforms; Inertial sensing devices; Inter/intra-subject; Learning algorithms; Machine learning; Performance prediction; Random/Record-wise split; Subject-wise split; Test sets; Training; Walking; Wearable computers; Wearable Electronic Devices; Biomechanics; Clinical datasets; Gait; Inter/intra-subject; Machine learning; Calibration; Subject-wise split; Random/Record-wise split

Clinical datasets often comprise multiple data points or trials sampled from a single participant. When these datasets are used to train machine learning models, the method used to extract train and test sets must be carefully chosen. Using the standard machine learning approach (random-wise split),...