Related Experiment Video
Updated: Sep 15, 2025
09:10
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
1.9K
Landmark-free automatic digital twin registration in robot-assisted partial nephrectomy using a generic end-to-end
Kilian Chandelon1,2, Alice Pitout3, Mathieu Souchaud4
1EnCoV, Institut Pascal, UMR6602 CNRS, UCA, Clermont-Ferrand University Hospital, Clermont-Ferrand, France. kilian.chandelon@gmail.com.
Summary
This study introduces a novel, landmark-free registration method for kidney surgery using augmented reality. The approach accurately aligns preoperative models with surgical images, enabling enhanced precision in partial nephrectomy procedures.
Area of Science:
- Medical Imaging
- Surgical Technology
- Computer-Aided Surgery
Background:
- Augmented Reality (AR) has advanced minimally invasive surgery for organs like the liver and uterus.
- Accurate registration between preoperative digital twins and surgical camera images is crucial for AR in surgery.
- Kidney surgery presents unique challenges for AR registration due to a lack of visible anatomical landmarks.
Purpose of the Study:
- To develop and evaluate a landmark-free registration method for Augmented Reality in kidney surgery.
- To address the unresolved challenges in aligning preoperative kidney models with intraoperative surgical images.
Main Methods:
- A novel, landmark-free registration approach utilizing a generic kidney model.
- An end-to-end neural network trained on a custom dataset to directly regress registration from surgical RGB images.
Main Results:
- The method demonstrated strong concordance with expert-labelled registration across four clinical cases.
- Achieved an average tumor contour alignment error of 7.3 ± 4.1 mm within 9.4 ± 0.2 ms.
- Showed robustness despite anatomical variations and intraoperative motion.
Conclusions:
- The proposed landmark-free registration method meets clinical requirements for accuracy, speed, and resource efficiency.
- This approach is a promising tool for Augmented Reality-Assisted Partial Nephrectomy.
- Enables precise alignment for complex kidney procedures where traditional landmarks are absent.
