ProCut: Probabilistic Cutting Topology for Autonomous Electrosurgical Tissue Dissection

Xiao Liang; Fei Liu; Florian Richter; Genie Rubaiyat; Michael Yip

cs.ROOct 5, 2026

ProCut: Probabilistic Cutting Topology for Autonomous Electrosurgical Tissue Dissection

Authors: Xiao Liang, Fei Liu, Florian Richter, Genie Rubaiyat, Michael Yip

Abstract

Accurately modeling and tracking the deformation of soft tissue is critical for a wide range of interventional and surgical procedures. However, current methods struggle in scenarios involving topological changes, such as cutting and dissection, due to the inherent non-linearity and discontinuity introduced by explicit changes in connectivity. In this work, we present a novel, fully differentiable framework that enables robust estimation and modeling of topological changes during deformable tracking. Our method introduces a continuous, sigmoid-based formulation to smooth the otherwise discrete event of tissue cutting, making it amenable to gradient-based optimization within a differentiable Position-Based Dynamics (PBD) simulation. To account for uncertainty and improve robustness in the presence of noisy visual data, we incorporate Stein Variational Gradient Descent (SVGD) for particle-based probabilistic inference, generating multiple hypotheses for topological state estimation. Building on this foundation, we develop an autonomous dissection algorithm for thin-shell tissues that leverages topological updates to guide closed-loop cutting trajectory control. We evaluate our approach in both simulated and real-world electrosurgical environments, demonstrating significant improvements in topological estimation accuracy and dissection precision over existing methods. Our results highlight the potential of this framework to advance automation in soft-tissue surgical procedures by enabling reliable perception and control in the presence of complex structural changes.

Figures & tables

Fig. 1: Overview of our autonomous tissue dissection system. (Top) The robot performs electrosurgical dissection. (Bottom left) Our method estimates the cutting topology using probabilistic, gradient-based inference on a canonical reference frame. (Bottom right) The estimated topology enables accurate tracking of tissue deformation, providing critical feedback for precise dissection control.

Fig. 2: Key representations that enable differentiable optimization of the cutting topology. (a). A mesh $M_{t}$ representing tissue that is manipulated and cut by a robot. (b). In the tissue’s canonical mesh representation $\hat{M}$ , a cutting curve $S$ affects the constraint stiffness of its surrounding mesh edges. Edges that are closer to the curve will have smaller stiffness to achieve similar effect as if the edges are cut.

Fig. 3: Comparison of topology estimation results from three methods. WB roughly predicts the location and shape but lacks accuracy due to wrong optical flow prediction. HB is highly sensitive to observation noise, leading to many outliers. In contrast, our method achieves accurate topology estimation and enables the most realistic tissue state tracking.

Fig. 4: Ablation results of comparing variants of our method initialized with different noise level. Data of Init indicates initalization errors. Gradient-descent method ( GD-LQ ) is sensitive to initialization. Estimation with linear representation ( SVGD-L ) leads to lower accuracy, whereas quadratic representation alone ( SVGD-Q ) is accurate but not robust to larger initialization error. Combining SVGD with a linear to quadratic refinement strategy ( SVGD-LQ ) achieves good accuracy and robustness.

Fig. 5: Comparison of results from open-loop and the proposed method in simulation. (a) $\sim$ (c) cases have target cutting patterns of line, arc, and snake shape shown as the boundaries of two colors. The 1st row shows dissection error over time of 10 repetitions. The open loop method doesn’t execute trajectory accurately and couldn’t separate completely due to errors. The proposed method is able to follow desired pattern accurately.

Fig. 6: This figures shows an example flow of robot autonomous dissection using the proposed method. The robot utilizes the proposed method’s tissue tracking capability to perform electrosurgical dissection. It inevitably makes error due to motion noise and calibration error. Our topology estimation method identifies the dissection error and updates the dissection plan. This process can be repeated until the dissection trajectory is met or when the estimated error is small enough. In this procedure, tissue retraction is tele-operated.

Explore similar work

Jul 15, 2026cs.RO

Deformable State Estimation for Autonomous Surgical Tissue Retraction Under Partial Observability

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.

Everest Yang, Skye Thompson, George D. Konidaris

Brown University, Providence, RI, USA

State Estimation Deformable Linear Objects Deformable Cnn Category-Level Object Pose Estimation Partial Observability

Ask PDF

Jul 21, 2026cs.RO

Bayesian Retraction Optimization for Tissue Attachment Mapping in Surgical Dissection

With growing surgeon shortages, automating surgical sub-tasks such as tissue dissection offers a promising step toward reducing workload and expanding patient access. Prior work has relied on hand-crafted incision policies that cannot quantify uncertainty or has relied on simulation-based methods that require strong modeling assumptions. We instead view tissue attachment identification as an inherently probabilistic problem and propose a Bayesian approach that avoids explicit tissue modeling. Our method uses a Sequential Bayesian Hilbert Map (SBHM) to represent the likelihood that each tissue point is attached to the underlying resection surface. An ensemble of learned classifiers predicts attachment likelihoods from spatial data acquired during robotic tissue retraction, with each classifier serving as a noisy information source to update the SBHM. To plan the next retraction, we devise Bayesian Retraction Optimization (BRO) to select the most informative action under safety constraints. As the SBHM refines over time, regions with high attachment likelihood are selectively incised. We validate our method in simulation across diverse tissue geometries and acquisition strategies, and demonstrate zero-shot transfer to real robotic dissection experiments.

Shing-Hei Ho, Bao Thach, Toan Vo +2

The Robotics Center and the Kahlert School of Computing at the University of Utah, Salt Lake City, UT, USA · School of Computational Science and Engineering, Georgia Institute of Technology, GA, USA · Department of Electrical and Computer Engineering and Department of Computer Science, Vanderbilt University, TN, USA

Surgical Robots Tissue

Ask PDF

Sep 28, 2026cs.RO

AGRO-SUVIDE: Agentic Robotics for Surgical Viscoelastic Debridement

Augmented dexterity has the potential to reduce the fatigue experienced by surgeons during repetitive surgical tasks. In this paper, we propose the first AGentic RObotics framework for SUrgical VIscoelastic DEbridement (AGRO-SUVIDE), the repeated removal of small fragments attached to a viscoelastic substrate. Leveraging the self-improving and coding capability of agents, AGRO-SUVIDE adopts a modular framework. Specifically, the demonstration analysis module automatically identifies recurring skills from a single expert demonstration, using both visual and kinematic information. The construction module then builds each skill, either as a procedural model-based skill the agent codes against a scaffolded library or as a model-free policy-based skill. At runtime, the monitoring module composes the skills into a loop-style graph sized to the number of fragments it observes, then verifies pre- and post-conditions of each skill to decide whether to advance or retry. We evaluate AGRO-SUVIDE through 340 physical trials on the da Vinci Research Kit (dVRK). AGRO-SUVIDE achieves an average single-fragment removal success rate of 85%, completing consecutive three-fragment removal at 60% and at 95% with one human intervention. It further generalizes to unseen five-fragment scenarios with an average success rate of 80% for single-fragment removal. Project page: https://surgical-robotics.github.io/AGRO-SUVIDE/

Shutong Jin, Ziyang Chen, Preethi Satish +4

Department of Electrical Engineering and Computer Sciences, University of California, Berkeley, CA 94720, USA. · Department of Robotics, Perception and Learning, KTH Royal Institute of Technology, Stockholm, SE 10044, Sweden. · Intuitive Surgical, Sunnyvale, CA 94086, USA.

Surgical Robots Surgery Task Success Rate

Ask PDF