A Supervised Autonomous Resection and Retraction Framework for Transurethral Enucleation of the Prostatic Median Lobe

2025-11-11

AI Summary

This study addresses transurethral median lobe enucleation, aiming to enhance semi-autonomy in image-guided robotic surgery and establish a foundation for fully automated minimally invasive prostatectomy. We propose a synergistic framework integrating model-driven cutting planning with learning-driven retraction control—combining CT image segmentation, biomechanics-informed trajectory generation, and a Push-Conditional Variational Autoencoder (PushCVAE) to enable three-stage median lobe resection using a dual-arm concentric-tube robot. Leveraging Level-3 supervised autonomy, we validate the approach on a hydrogel prostate phantom, achieving 97.1% target volume resection and sub-millimeter localization accuracy. The core contribution lies in the deep integration of interpretable physical models with data-driven soft-tissue interaction modeling, significantly improving robustness and generalizability of vision–force coordinated manipulation in complex intracavitary surgical environments.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Vision for Robotics & Autonomous Driving
• Natural Language Processing: Safety and Robustness

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Concentric tube robots (CTRs) offer dexterous motion at millimeter scales, enabling minimally invasive procedures through natural orifices. This work presents a coordinated model-based resection planner and learning-based retraction network that work together to enable semi-autonomous tissue resection using a dual-arm transurethral concentric tube robot (the Virtuoso). The resection planner operates directly on segmented CT volumes of prostate phantoms, automatically generating tool trajectories for a three-phase median lobe resection workflow: left/median trough resection, right/median trough resection, and median blunt dissection. The retraction network, PushCVAE, trained on surgeon demonstrations, generates retractions according to the procedural phase. The procedure is executed under Level-3 (supervised) autonomy on a prostate phantom composed of hydrogel materials that replicate the mechanical and cutting properties of tissue. As a feasibility study, we demonstrate that our combined autonomous system achieves a 97.1% resection of the targeted volume of the median lobe. Our study establishes a foundation for image-guided autonomy in transurethral robotic surgery and represents a first step toward fully automated minimally-invasive prostate enucleation.

Problem

• Developing autonomous resection and retraction for transurethral prostate surgery
• Creating coordinated planning and learning systems for robotic tissue removal
• Achieving supervised autonomy in minimally invasive prostate enucleation procedures

Innovation

• Model-based resection planner for tool trajectory generation
• Learning-based retraction network trained on surgeon demonstrations
• Supervised autonomous execution using dual-arm concentric tube robot