A robot-assisted sliding palpation system uses a calibrated digital twin to generate labeled tactile sequences, reducing reliance on real-world data. The model of sensor-vessel contact is calibrated using Bayesian optimization against real palpation trajectories. A spatio-temporal graph neural network trained on simulated marker trajectories performs vessel classification and creates a localisation map.
This is relevant to modeling and management of systems where reducing data collection costs and improving accuracy of state estimation for biological or technical objects is key.
Translation and summary are machine-generated from the source. The full article is not republished; its rights belong to the source.
