Context and Motion Primitive Aggregate Surgical Set
Contains code and instructions for labeling videos.
Contains kinematic and video data organized by task, subject, and trial. Includes the Suturing (S), Needle Passing (NP), and Knot Tying (KT) tasks from the JIGSAWS dataset [3], Peg Transfer (PT) from the DESK dataset [4], and Pea on a Peg (PoaP) and Post and Sleeve (PaS) from the ROSMA dataset [5].
Motion Primitive | Suturing | Needle Passing | Knot Tying | Peg Transfer | Pea on a Peg | Post and Sleeve | COMPASS |
---|---|---|---|---|---|---|---|
Grasp | 471 | 373 | 283 | 323 | 577 | 824 | 2851 |
Release | 441 | 365 | 247 | 313 | 556 | 776 | 2698 |
Touch | 518 | 330 | 135 | 539 | 1782 | 1598 | 4902 |
Untouch | 314 | 206 | 111 | 364 | 1261 | 1131 | 3387 |
Pull | 194 | 114 | 235 | 0 | 525 | 0 | 1068 |
Push | 179 | 119 | 0 | 0 | 2 | 0 | 300 |
Contains code and instructions for running ML models; and eventually our trained models.
Contains scripts for converting context labels to motion primitives.
[1] Hutchinson, K., Reyes, I., Li, Z., & Alemzadeh, H. (2023). COMPASS: a formal framework and aggregate dataset for generalized surgical procedure modeling. International Journal of Computer Assisted Radiology and Surgery, 1-12. [arXiv, pdf]
[2] Hutchinson, K., Reyes, I., Li, Z., & Alemzadeh, H. (2023). Evaluating the Task Generalization of Temporal Convolutional Networks for Surgical Gesture and Motion Recognition using Kinematic Data. IEEE Robotics and Automation Letters. [arXiv]
[1] Gao, Y., Vedula, S. S., Reiley, C. E., Ahmidi, N., Varadarajan, B., Lin, H. C., ... & Hager, G. D. (2014, September). Jhu-isi gesture and skill assessment working set (jigsaws): A surgical activity dataset for human motion modeling. In MICCAI workshop: M2cai (Vol. 3, No. 3).
[2] Gonzalez, G. T., Kaur, U., Rahma, M., Venkatesh, V., Sanchez, N., Hager, G., ... & Wachs, J. (2020). From the DESK (Dexterous Surgical Skill) to the Battlefield--A Robotics Exploratory Study. arXiv preprint arXiv:2011.15100.
[3] Rivas-Blanco, I., Pérez-del-Pulgar, C. J., Mariani, A., Quaglia, C., Tortora, G., Menciassi, A., & Muñoz, V. F. (2021). A surgical dataset from the da Vinci Research Kit for task automation and recognition. arXiv preprint arXiv:2102.03643.