Open Kinematic Dataset of Walking at Different Speeds in Type 2 Diabetes and Control Participants

The data is stored as a hierarchical JSON file for ease of parsing, analysis, and expansion. --- 📐 Structure The data is organized in four hierarchical levels: Subject ID → Speed → Trial Number → Kinematic Variables (markers 3D position) Where: -- **Subject ID**: A numeric string (e.g., `"7"`), - **Speed**: Integer string representing the gait speedin hm/h (e.g., `"18"` means 1.8 km/h). Speed values can be easily converted to m/s by dividing for 36 (e.g. 18/36 = 0.5 m/s). - **Trial Number**: A string ID for each 30-second trial (e.g., `"1"`, `"2"`, etc.) -Each trial entry contains a dictionary of 56 sampled variables over time: Frame and Subframe are index vars - The following variables are grouped by three, and named as right or left (R, L) + the name of the body marker + the axis (e.g. RLOBx, RLOBy, RLOBz, for a total of 9 anatomical positions and 18 markers).

LOB = Temporal lobe; SH = Acromion; ELB = Elbow; WR = Ulnar stiloyd process; GT = Great Trochanter; KN = Knee (lateral); AN = Malleolus lateral; HEE = Heel; MT = 5th Metatarsal. The 3D space was calibrated following the right hand rule, with the axis "X" = mediolateral; "Y" = anteroposterior; "Z" = vertical. Values are in mm. 📁 Subjects_groups_db (CSV File) This table contains the participants' data --- 🔎 Example { { "7": { "45": { "1": { "Frame": [1, 2, 3, ...], "Subframe": [0, 0, 0, ...], "RLOBx": [61.44, 61.92, 60.88, ...], "RLOBy": [56.09, 56.09, 54.58, ...], ... } } } }