Search papers, labs, and topics across Lattice.
This paper addresses the challenge of surgical tissue retraction under conditions of partial and noisy perception by developing a learned state estimator that reconstructs the full deformable mesh state from limited observations. The approach integrates a multilayer perceptron with a low-dimensional PCA latent representation, enhanced by geometry-aware regularization to ensure smooth and physically plausible deformations. Evaluation in a 2D deformable sheet simulation reveals that the estimator achieves 98.1% of oracle performance in multi-step retraction planning, highlighting its effectiveness in real-world surgical scenarios.
Achieving 98.1% of oracle performance in surgical tissue retraction planning with only 40 noisy observations could revolutionize autonomous surgical systems.
Surgical tissue retraction requires effective manipulation planning under partial and noisy perception. We study state estimation for deformable tissue retraction, where only sparse observations of the tissue surface are available at decision time. We propose a learned state estimator that reconstructs the full deformable mesh state from 40 noisy vertex observations. The estimator combines a multilayer perceptron with a low-dimensional PCA latent representation and is trained using geometry-aware regularization that encourages smooth and physically plausible deformations. We evaluate the approach in a 2D deformable sheet simulation using single-step and multi-step retraction planning. Results show that the learned estimator achieves 98.1% of oracle performance in multi-step retraction while supporting efficient inference. These results demonstrate that learned, geometry-regularized state estimation can support effective deformable manipulation under realistic perception constraints.
