Skills evaluation in minimally invasive surgery using force/torque signatures

Abstract

Preliminary data suggest that F/T magnitudes associated with the tool/tissue interactions provide an objective means of distinguishing novices from skilled surgeons. Clinical F/T analysis using the proposed technology and methodology may be helpful in training, developing surgical simulators, and measuring technical proficiency during laparoscopic surgery.

Towards automatic skill evaluation: Detection and segmentation of robot-assisted surgical motions

“…In the field of high-level surgical modeling, Rosen et al [14][15][16] demonstrated that statistical models derived from recorded force and motion data can be used to objectively classify surgical skill level as either novice or expert. The results show that the statistical distances between Hidden Markov Models (HMMs) representing varying levels of surgical skill were significantly different (a , 0.05).…”

An observation support system with an adaptive ontology-driven user interface for the modeling of complex behaviors during surgical interventions

“…The procedure of modeling surgical processes by observation, especially with the objective of performance assessment in the context of surgical training (Leong et al, 2007;Megali, Signigaglia, Tonet, & Dario, 2006;Richards, Rosen, Hannaford, Pellegrini, & Sinanan, 2000;Rosen, Brown, Chang, & Hannaford, 2006) or of comparing strategies for surgical treatment (den Boer et al, 1999;Strauss et al, 2006), generally focuses on two measurement strategies: high-resolution, sensor-based measurement of the performance of a limited number of surgical actions-such as, for instance, instrument movement trajectories while placing knots-or low-resolution and simpler observer-based measurements for surgical interventions or interventional phases (e.g., Archer & Macario, 2006;Schuster, Wicha, Fiege, & Goetz, 2007), without reference to specific surgical process steps. In previous work, we proposed an approach that allows for a medium level of granularity to be used for the decomposition of surgical process steps into categories, as described in the following section, in order to accommodate the complexity and diversity of information and the high variability of surgery.…”

Data-Derived Models for Segmentation with Application to Surgical Assessment and Training

“…Dosis et al [5] have used hidden Markov models to model hand manipulations and to classify simple surgical tasks. Richards et al [6] have demonstrated that force/torque signatures may be used in RMIS for two-way skill classification. Rosen et al [7] have used HMMs to model tool-tissue interactions in laparoscopic surgery; a seperate HMM for each skill level was trained using a pool of surgeons, and a statistical distance between these HMMs was shown to correlate well with the learning curve of these trainee surgeons.…”