BGU-SKILLS is a comprehensive dataset of long-term surgical training collected using the da Vinci Research Kit (dVRK), featuring synchronized kinematic and video data. The dataset comprises recordings from 18 surgical residents who, once per month over a six-month period, performed three standardized robot-assisted surgical tasks—Ring Tower Transfer, Knot-Tying, and Suturing—before, during, and after a 26-hour hospital shift. It is designed to support research in surgical robotics, motor learning, and skill assessment by capturing the progression of surgical expertise over time.
For each session, the dataset provides kinematic data, video recordings, manual annotations, and questionnaire-based activity and sleep information.
A detailed PDF file describes the dataset structure, and data are organized by task-specific folders.
For further details, please refer to the associated publication.
For any questions or additional information, please contact Yarden Sharon (shayar@post.bgu.ac.il).
We recommend using the "Tree" view to explore the data by folder.