A three-dimensional gait dataset from typically developing children aged 3–13 years in Shanghai, China

Direct identifiers, videos, school information, exact dates of birth, contact details, and clinical diagnoses are not included. Age is provided as age groups.Access requests may be submitted through the ScienceDB repository page. Approved users must use the dataset in accordance with the data use conditions specified in the ScienceDB access request form and data use agreement, which are designed to protect participants and support responsible data reuse.Repository structureDataset/├── raw_c3d.zip├── subject_metadata.csv├── file_manifest.csv├── data_dictionary.txt├── marker_set_description.txt└── README.txtFile descriptionraw_c3d.zipRaw C3D files for each participant.

Users should extract raw_c3d.zip before processing the C3D files. These typically include one static calibration trial (for example, S001-static-trial01.c3d) and 5 to 7 walking trials (for example, S001-walk-trial01.c3d to S001-walk-trial07.c3d).subject_metadata.csvAnonymized participant-level metadata.file_manifest.csvSummary of distributed files and file contents.data_dictionary.txtDefinitions of metadata fields.README.txtDataset overview and usage notes.ParticipantsThe dataset includes 431 typically developing children (224 males, 207 females) aged 3–13 years.For descriptive presentation in the associated article, participants were stratified into four age groups:•  3–5 years •  6–8 years •  9–10 years •  11–13 years Acquisition settingsMobile settingParticipants aged 3–5 years were assessed in kindergartens using a mobile motion capture system with 12 infrared cameras at 100 Hz. Laboratory settingParticipants aged 6–13 years were assessed in a gait analysis laboratory using 19 infrared cameras at 100 - 120 Hz. The same marker model and general processing workflow were applied except that 40 subjects collected at early period have no tibial tuberosity markers.Experimental protocolA modified Plug-in Gait lower-body model was used.

Additional markers were placed at the tibial tuberosity and the first and fifth metatarsal heads.Markers were placed on the upper body, pelvis, lower limbs, and feet during static calibration. A static calibration trial was acquired in a standardized T-pose for 3–5 s. The medial femoral condyle and medial malleolus markers were removed before dynamic trials.Participants walked barefoot at a self-selected speed along a 10-m walkway.

Only the middle portion of each walking trial was used for analysis. A trial was considered valid when marker trajectories were complete and walking was continuous and natural. At least seven valid walking trials were collected for each participant.System calibration was performed according to the manufacturer’s standard procedure using L-frame and T-wand calibration tools.

The acquisition environment was recalibrated after every five participants.In 40 participants, the tibial tuberosity marker was unavailable during acquisition. This is indicated in the metadata by the TT_available field.Data contentsC3D filesEach C3D file contains raw data from a single static calibration trial or walking trial.For all participants:•  marker trajectoriesNot included in C3D:•  gait event labels •  Metadata fieldssubject_metadata.csv contains:•  Subject_ID •  Sex •  Age_group •  Height •  Weight •  Collection_setting•  TT_available •  CameraSRMetadata policyAge is provided in categorized form, whereas height and weight are provided as exact continuous values.

Participant metadata are included to support data interpretation and data selection for secondary analysis.Processed individual-level outputs and embedded gait event labels are not distributed.File namingExamples:•  S001-static-trial01.c3d •  S001-walk-trial01.c3d •  S001-walk-trial02.c3d Format:•  S001 = anonymized participant ID •  static = static calibration trial •  trial01 = trial number Recommended softwareC3D files can be opened and processed using software that supports the C3D format, including:•  Visual3D •  OpenSim •  Mokka •  BTK-compatible tools Notes for users•  Kinematic data are available for all participants. •  The dataset includes two acquisition settings; this should be considered in secondary analyses. •  Gait events are not embedded in the distributed C3D files and must be identified by users during reprocessing.•  Age is provided as an age-group category, whereas height and weight are provided as exact continuous values. •  The TT_available field should be considered in analyses sensitive to shank marker configuration. The dataset is designed for reprocessing and secondary analysis from raw C3D files. 

Citation

Please cite the associated data descriptor when using this dataset.DOI: 10.57760/sciencedb.40305