Download Full Text (198.1 MB)
Daniel Felix Ritchie School of Engineering and Computer Science, Center for Orthopaedic Biomechanics
Joint kinematics, Lower limb, Biomechanics
Biomechanical Engineering | Biomechanics | Engineering
This dataset is derived from a publicly available dataset generated by Camargo et al. The dataset aimed to convert the exciting dataset to a format which is used for training and evaluation of machine learning models, specifically transformer-based model, proposed by the Sharifi et al study.
The dataset comprising comprehensive data, joint kinematics along with IMU data, from 19 healthy subjects, during various activities: LW (self-selected speeds), RA, RD, SA, and SD. Each subject was equipped with four six-axis IMUs attached to the foot, shank, thigh, and torso, as well as 32 motion capture markers. The dataset offers valuable insights into lower limb biomechanics, specifically focusing on the lower limb IMU data and sagittal plane joint kinematics at the hip, knee, and ankle.
This page holds the model. Download the data set at https://digitalcommons.du.edu/biomat/1
The dataset is stored in a pickle file format, containing four sections represented as keys in the BioMAT dictionary:
- Format: List of dataframes
- Number of dataframes: 2578
- Columns in each dataframe: ['Header', 'Label', 'subject', 'trialType', 'trialNum', 'trialStarts', 'trialEnds', 'leadingLegStart', 'leadingLegStop', 'turn', 'speed', 'transLegAscent', 'transLegDescent', 'rampInclide', 'stairHeight']
- Description: Each dataframe provides trial-specific information related to the trials conducted.
- Format: List of tuples
- Description: Contains information about the dataset itself.
- Information included:
- 'dataset_name': 'camargo'
- 'activity_list': ['levelground', 'ramp', 'stair']
- 'subject_list': ["AB06", "AB07", "AB08", ..., "AB25"]
- 'datatype_list': ['imu', 'ik']
- Format: List of 2578 dataframes
- Columns in each dataframe: ['Header', 'foot_Accel_X', 'foot_Accel_Y', 'foot_Accel_Z', 'foot_Gyro_X', 'foot_Gyro_Y', 'foot_Gyro_Z', 'shank_Accel_X', 'shank_Accel_Y', 'shank_Accel_Z', 'shank_Gyro_X', 'shank_Gyro_Y', 'shank_Gyro_Z', 'thigh_Accel_X', 'thigh_Accel_Y', 'thigh_Accel_Z', 'thigh_Gyro_X', 'thigh_Gyro_Y', 'thigh_Gyro_Z', 'trunk_Accel_X', 'trunk_Accel_Y', 'trunk_Accel_Z', 'trunk_Gyro_X', 'trunk_Gyro_Y', 'trunk_Gyro_Z']
- Description: Each dataframe represents IMU data for a specific trial and includes time-related columns, acceleration measurements (X, Y, Z) for foot, shank, thigh, and trunk, as well as gyroscope measurements (X, Y, Z) for foot, shank, thigh, and trunk.
- Format: List of 2578 dataframes
- Columns in each dataframe: ['Header', 'pelvis_tilt', 'pelvis_list', 'pelvis_rotation', 'pelvis_tx', 'pelvis_ty', 'pelvis_tz', 'hip_flexion_r', 'hip_adduction_r', 'hip_rotation_r', 'knee_angle_r', 'ankle_angle_r', 'subtalar_angle_r', 'mtp_angle_r', 'hip_flexion_l', 'hip_adduction_l', 'hip_rotation_l', 'knee_angle_l', 'ankle_angle_l', 'subtalar_angle_l', 'mtp_angle_l', 'lumbar_extension', 'lumbar_bending', 'lumbar_rotation']
- Description: Each dataframe represents joint kinematic measurements for a specific trial and includes time-related columns, along with measurements for pelvis tilt, pelvis list, pelvis rotation, pelvis translations (X, Y, Z), joint angles (hip flexion, hip adduction, hip rotation, knee angle, ankle angle, subtalar angle, mtp angle), and lumbar measurements (lumbar extension, lumbar bending, lumbar rotation).
Dataset Use Case
This dataset has been used in training and evaluation of multiple deep neural networks architecture such as BioMAT, BiLSTM, and CNNLSTM. The source code related the usage of this dataset is availabe here: https://github.com/MohsenSharifi1991/BioMAT
Copyright Statement / License for Reuse
This work is licensed under a Creative Commons Attribution 4.0 International License.
The dataset is available for public use and can be accessed through the provided link https://digitalcommons.du.edu/biomat/ . Researchers, data analysts, and machine learning practitioners can utilize this dataset for academic, research, or commercial purposes, under the terms of the Creative Commons Attribution 4.0 International License.
If you use this dataset in your research or publications, we kindly request that you cite it using the following reference:
Sharifi-Renani, M.; Mahoor, M.H.; Clary, C.W. BioMAT: An Open-Source Biomechanics Multi-Activity Transformer for Joint Kinematic Predictions Using Wearable Sensors. Sensors 2023, 23, x.
Camargo, Jonathan, et al. "A comprehensive, open-source dataset of lower limb biomechanics in multiple conditions of stairs, ramps, and level-ground ambulation and transitions." Journal of Biomechanics 119 (2021): 110320. https://doi.org/10.1016/j.jbiomech.2021.110320