“…The first dissimilarity metric used was the Hausdorff distance [33]. Assuming two sets of points A and B, the Hausdorff distance h is described as:…”
Abstract-Dexterity and procedural knowledge are two critical skills surgeons need to master to perform accurate and safe surgical interventions. However, current training systems do not allow to provide an in-depth analysis of surgical gestures to precisely assess these skills. Our objective is to develop a method for the automatic and quantitative assessment of surgical gestures. To reach this goal, we propose a new unsupervised algorithm that can automatically segment kinematic data from robotic training sessions. Without relying on any prior information or model, this algorithm detects critical points in the kinematic data which define relevant spatio-temporal segments. Based on the association of these segments, we obtain an accurate recognition of the gestures involved in the surgical training task. We then perform an advanced analysis and assess our algorithm using datasets recorded during real expert training sessions. After comparing our approach with the manual annotations of the surgical gestures, we observe 97.4% accuracy for the learning purpose and an average matching score of 81.9% for the fullyautomated gesture recognition process. Our results show that trainees workflow can be followed and surgical gestures may be automatically evaluated according to an expert database. This approach tends towards improving training efficiency by minimizing the learning curve.
“…The first dissimilarity metric used was the Hausdorff distance [33]. Assuming two sets of points A and B, the Hausdorff distance h is described as:…”
Abstract-Dexterity and procedural knowledge are two critical skills surgeons need to master to perform accurate and safe surgical interventions. However, current training systems do not allow to provide an in-depth analysis of surgical gestures to precisely assess these skills. Our objective is to develop a method for the automatic and quantitative assessment of surgical gestures. To reach this goal, we propose a new unsupervised algorithm that can automatically segment kinematic data from robotic training sessions. Without relying on any prior information or model, this algorithm detects critical points in the kinematic data which define relevant spatio-temporal segments. Based on the association of these segments, we obtain an accurate recognition of the gestures involved in the surgical training task. We then perform an advanced analysis and assess our algorithm using datasets recorded during real expert training sessions. After comparing our approach with the manual annotations of the surgical gestures, we observe 97.4% accuracy for the learning purpose and an average matching score of 81.9% for the fullyautomated gesture recognition process. Our results show that trainees workflow can be followed and surgical gestures may be automatically evaluated according to an expert database. This approach tends towards improving training efficiency by minimizing the learning curve.
“…In this paper, we define network trajectories as a set of connected vertices: tx = [v1, ..., vn]. Figure 1 contains four network trajectories: tA = [16,17,18,19,20]; tB = [1,2,3,4,9,10]; tC = [16,17,18,19,20,15,10]; tD = [6,7,8,9,10].…”
“…In our pseudocode and implementation, we use Dijkstra's single-source shortest-paths algorithm [4] taking a source node v and a graph G as input: distM ap[] = Dijkstra(G, v). It returns the shortest-path distance to all other nodes in G. In Figure 1, an example dist (3,8) would be a distance of 3, traversing the shortestpath 3-2-7-8 or 3-4-9-8.…”
“…One popular metric is Hausdorff distance, a commonly used measure to compare similarity between two geometric objects (e.g., polygons, lines, sets of points) [11]. A number of methods have focused on applying Hausdorff distance to trajectories in geometric space [2,3,9,15].…”
Given a set of trajectories on a road network, the goal of the All-Pair Network Trajectory Similarity (APNTS) problem is to calculate the similarity between all trajectories using the Network Hausdorff Distance. This problem is important for a variety of societal applications, such as facilitating greener travel via bicycle corridor identification. The APNTS problem is challenging due to the high cost of computing the exact Network Hausdorff Distance between trajectories in spatial big datasets. Previous work on the APNTS problem takes over 16 hours of computation time on a real-world dataset of bicycle GPS trajectories in Minneapolis, MN. In contrast, this paper focuses on a scalable method for the APNTS problem using the idea of row-wise computation, resulting in a computation time of less than 6 minutes on the same datasets. We provide a case study for transportation services using a data-driven approach to identify primary bicycle corridors for public transportation by leveraging emerging GPS trajectory datasets.
“…This method can be used to measure the 'closeness' of two non-empty point sets which are subsets of a metric space. The method assigns a scalar score or distance to the two trajectories, which measures the similarity between the two trajectories (Chen et al, 2011). This distance is defined as: Table 2 shows the distances between the experimental trajectories and the simulated trajectories, for the four experimental subjects, at the 'POSITION' steps of the finite state machines (see Figure 7).…”
Digital human models can be used for biomechanical risk factors assessment of a workstation and work activity design for which there is no physical equipment that can be tested using actual human postures and forces. Yet, using digital human model software packages is usually complex and time-consuming. A challenging aim therefore consists in developing an easy-to-use digital human model capable of computing dynamic, realistic movements and internal characteristics in quasi-real time, based on a simple description of future work tasks, in order to achieve reliable ergonomics assessments of various work task scenarios. We developed such a dynamic digital human model, which is automatically controlled in force and acceleration and inspired by human motor control and based on robotics and physics simulation. In our simulation framework, the digital human model motion was controlled by real-world Newtonian physical and mechanical laws. We also simulated and assessed experimental insert-fitting activities according to the occupational repetitive actions (OCRA) ergonomic index. Simulation led to satisfactory results: experimental and simulated ergonomics evaluations were consistent, and both joint torques and digital human model movements were realistic and coherent with human-like behaviours and performances.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.