Dataset of human walking kinematics, knee external moments, and knee compressive contact forces from 5 open datasets

This dataset contains the following time series data during the stance phase of walking: Lumbar, pelvis, hip, knee, and ankle joint angles (kinematics) Knee external flexion-extension and abduction-adduction moments Medial and lateral tibiofemoral compressive contact forces and their summed "total contact force" Patellofemoral compressive contact force Subjects' body mass, body height, age, and sex Five original datasets (described below) were processed with musculoskeletal modeling and simulation to produce this combined dataset. In total, the data comes from 208 subjects and 5009 stance phases of walking. The same data is available in both CSV and MATLAB formats. In the CSV files, "padded" means the time series are sampled at 100 Hz and padded with trailing zeros to 250 elements per time series; "normalized" means that the time series are sampled to 101 data points (without trailing zeros). In the MATLAB file, both types of time series are contained in one file. The table below presents a description of all columns in the data: Column label Description dataset the original source dataset, named by the first author except for "KGD" (Kuopio Gait Dataset) subject_name identifier of the subject in the original dataset trial_name identifier of the trial in the original dataset row index of the row in the table (metadata not related to biomechanics) age age of the subject in years sex M for male, F for female body_height body height of the subject in meters body_mass body mass of the subject in kilograms (colloquially "weight") leg the leg (R for right, L for left) that was undergoing stance when the time series were extracted TYPE_timeseries_jointangles_NAME_LEG time series of kinematics, where TYPE is "padded" or "normalized" (see explanation above) and NAME is the same as the name of the generalized coordinate in the utilized musculoskeletal model; LEG, when applicable, is "primary" when the time series is from the leg undergoing the stance phase and "secondary" when the time series is from the leg undergoing the swing phase TYPE_timeseries_jointmoments_ekfm external knee flexion moment time series in the stance leg TYPE_timeseries_jointmoments_ekam external knee adduction moment time series in the stance leg TYPE_timeseries_contactforces_kcf_summed total tibiofemoral compressive contact force time series in the stance leg; note that this is not simply medial+lateral "summed" although the name so implies TYPE_timeseries_contactforces_kcf_medial medial tibiofemoral compressive contact force time series in the stance leg TYPE_timeseries_contactforces_kcf_lateral lateral tibiofemoral compressive contact force time series in the stance leg TYPE_timeseries_contactforces_kcf_patellofemoral patellomoferal compressive contact force time series in the stance leg This dataset comes from the following five original datasets, all licensed under CC BY 4.0: Camargo et al., "A comprehensive, open-source dataset of lower limb biomechanics in multiple conditions of stairs, ramps, and level-ground ambulation and transitions", https://doi.org/10.1016/j.jbiomech.2021.110320 Latest version of the dataset is available at: https://www.epic.gatech.edu/opensource-biomechanics-camargo-et-al/ Fukuchi et al., "A public dataset of running biomechanics and the effects of running speed on lower extremity kinematics and kinetics", https://doi.org/10.7717/peerj.3298 Dataset at: https://doi.org/10.6084/m9.figshare.4543435 Horst, Fabian; Lapuschkin, Sebastian; Samek, Wojciech; Müller, Klaus-Robert; Schöllhorn, Wolfgang I. (2019), “A public dataset of overground walking kinetics and full-body kinematics in healthy adult individuals”, Mendeley Data, V3, doi: 10.17632/svx74xcrjr.3 Schreiber, C., Moissenet, F. A multimodal dataset of human gait at different walking speeds established on injury-free adult participants. Sci Data 6, 111 (2019). https://doi.org/10.1038/s41597-019-0124-4 Dataset at: https://doi.org/10.6084/m9.figshare.7734767 Lavikainen et al., "Gait data from 51 healthy participants with motion capture, inertial measurement units, and computer vision", Ann Biomed Eng, https://doi.org/10.1016/j.dib.2024.110841 Dataset at: https://doi.org/10.5281/zenodo.10559504 If you use this dataset, kindly cite the original datasets as well. Python scripts using this data for training knee contact force predicting models is available here: https://github.com/jerela/kinematics2kinetics