← Back ICRA 2026

STITCH 2.0: Extending Augmented Suturing with EKF Needle Estimation and Thread Management

Kush Hari, Ziyang Chen, Hansoul Kim, Ken Goldberg

PDF

AI summary

Key figure (auto-extracted from paper)

STITCH 2.0 significantly improves autonomous surgical suturing accuracy and speed over its predecessor, achieving up to 100% wound closure with minimal human intervention.

Surgical Robotics Augmented Dexterity Needle Pose Estimation Thread Management Suture Automation Extended Kalman Filter

Problem

Previous robot-assisted suturing systems struggle with inaccurate needle tracking, thread tangling, and uneven suture placement, preventing reliable full wound closure.

Approach

The system integrates an Extended Kalman Filter for precise 6D needle pose estimation, automated 3D suture alignment, and coordinated thread management to guide a surgical robot through consistent suture throws.

Key results

Achieved 74.4% average wound closure with 4.87 sutures per trial without intervention
Reduced execution time by 38% while increasing suture count by 66% compared to STITCH 1.0
Reached 100% wound closure averaging six sutures when allowing two human interventions
Introduced automated 3D suture alignment and EKF-based needle tracking to eliminate manual setup and improve pose accuracy

Why it matters

This advancement brings semi-autonomous surgical robots closer to clinical adoption by reliably automating a tedious, skill-dependent subtask to improve patient outcomes and reduce surgeon fatigue.

Abstract

Surgical suturing is a high-precision task that im- pacts patient healing and scarring. Suturing skill varies widely between surgeons, highlighting the need for robot assistance. Previous robot suturing works, such as STITCH 1.0 [1], struggle to fully close wounds due to inaccurate needle tracking and poor thread management. To address these challenges, we present STITCH 2.0, an elevated augmented dexterity pipeline with seven improvements including: improved EKF needle pose estimation, new thread untangling methods, and an automated 3D suture alignment algorithm. Experimental results over 15 trials find that STITCH 2.0 on average achieves 74.4% wound closure with 4.87 sutures per trial, representing 66% more sutures in 38% less time compared to the previous baseline. When two human interventions are allowed, STITCH 2.0 averages six sutures with 100% wound closure rate.

Index terms

Medical Robots and Systems Surgical Robotics: Laparoscopy Computer Vision for Medical Robotics