# Markerless Suture Needle 6D Pose Tracking with Robust Uncertainty Estimation for Autonomous Minimally Invasive Robotic Surgery

**Authors:** Zih-Yun Chiu, Albert Z Liao, Florian Richter, Bjorn Johnson, Michael C. Yip  
**arXiv:** [2109.12722](https://arxiv.org/abs/2109.12722)  
**Date:** 2021-09-26  
**Summary page:** https://www.summarizepaper.com/en/paper/2109.12722v1/

## Key points

- Accurate localization of suture needles is crucial for successful operations in autonomous suturing.
- Existing approaches often rely on markers or complex feature extraction operations, limiting their applicability in real-world environments.
- The authors propose a novel markerless suture needle pose tracking method using Bayesian filters.
- The proposed approach utilizes a data-efficient feature point detector trained to extract feature points on the needle without markers.
- These detected feature points are used to construct an observation model that measures the overlap between detections and the expected projection of the needle.
- The observation model can be calculated efficiently, enabling real-time tracking.
- An approximation for the covariance of observation noise is derived to enhance robustness to uncertainty in detections.
- Experimental results show low tracking errors in position and orientation, both in simulation and real-world environments.
- This research eliminates markers and simplifies feature extraction operations, presenting a promising advancement in markerless suture needle pose tracking for autonomous minimally invasive robotic surgery.

## Summary

In the field of autonomous suturing, accurate localization of suture needles is crucial for successful operations. However, existing approaches often rely on markers or complex feature extraction operations, limiting their applicability in real-world environments. To address this challenge, the authors propose a novel markerless suture needle pose tracking method using Bayesian filters. The proposed approach utilizes a data-efficient feature point detector that is trained to extract feature points on the needle without the need for markers. These detected feature points are then used to construct a novel observation model that measures the overlap between the detections and the expected projection of the needle. Importantly, this observation model can be calculated efficiently, enabling real-time tracking. To enhance robustness to uncertainty in detections, the authors derive an approximation for the covariance of observation noise in their proposed method. This allows for more accurate tracking even in challenging scenarios where detection uncertainties may arise. Experimental results conducted in simulation demonstrate that the proposed observation model achieves low tracking errors of approximately 1.5mm in position and 1 degree in orientation. Furthermore, qualitative results obtained from real-world environments show high consistency between the projected tracked pose and the actual pose of the needle. Overall, this research presents a promising advancement in markerless suture needle pose tracking for autonomous minimally invasive robotic surgery by eliminating markers and simplifying feature extraction operations. This approach has potential applications in real-world surgical settings where precise and reliable needle localization is essential for successful outcomes.