Transfemoral Amputee Stumble Induction Data for Five Subjects

A multivariate gait database for five transfemoral amputee subjects walking and experiencing induced stumbles. Data includes 6DoF IMU readings at 60 Hz and sEMG signals at 480 Hz from three thigh muscles, recorded while participants used three different prosthetic knee systems. The dataset was authored by Lucas Galey and last updated on 2026-05-14.

Use Cases

• Developing machine learning models for stumble detection based on the binary 'stumble' marker.
• Analyzing gait stability under different prosthetic knee systems (MPK, LIMBS M3, G-Knee).
• Correlating muscle activity (sEMG) with kinematic data (IMU) during induced stumbles.
• Comparing stumble responses across three induction methods (bungee, obstacle, uneven).

Strengths

• Multimodal data synchronized via shared timestamps, combining IMU and sEMG signals.
• Includes data from five subjects across three different prosthetic knee systems.
• Stumble events are induced via three distinct methods, providing varied scenarios.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

figshare

Collection Method

Experimental data collection using IMU sensors and surface electromyography (sEMG) during walking and induced stumble trials.

Freshness

Last updated 2026-05-14 00:25:18; freshness should be verified.