Related Experiment Video
Updated: May 8, 2026
07:43
In Vivo Quantification of Hip Arthrokinematics during Dynamic Weight-bearing Activities using Dual Fluoroscopy
Published on: July 2, 2021
Joint surface reconstruction and 4D deformation estimation from sparse data and prior knowledge for marker-less
Benjamin Berkels1, Sebastian Bauer, Svenja Ettl
1Institute for Numerical Simulation, Rheinische Friedrich-Wilhelms-Universität Bonn, 53115 Bonn, Germany. benjamin.berkels@ins.unibonn.de
Medical Physics
|September 7, 2013
Summary
This study introduces a novel sparse-to-dense registration method for real-time 3D body surface reconstruction and 4D motion field estimation. The approach accurately tracks respiratory motion for improved image-guided interventions.
Area of Science:
- Medical Imaging
- Biomedical Engineering
- Computational Anatomy
Background:
- Intraprocedural tracking of respiratory motion is crucial for enhancing image-guided diagnosis and interventions.
- Accurate real-time monitoring of patient movement during medical procedures remains a significant challenge.
Purpose of the Study:
- To develop and validate a sparse-to-dense registration approach for recovering 3D body surface and estimating 4D surface motion fields.
- To enable precise tracking of respiratory motion using sparse sampling data and patient-specific prior shape knowledge.
Main Methods:
- Utilized a marker-less, laser-based active triangulation (AT) sensor for sparse, real-time 3D measurements.
- Registered sparse measurements with dense reference surfaces from planning data to recover a dense, spatio-temporal 4D deformation field.
- Validated the method on a 4D CT respiration phantom and evaluated on real and synthetic data.
Main Results:
- Achieved a mean surface reconstruction accuracy of ±0.23 mm, significantly reducing initial surface mismatch.
- Estimated surface motion fields with a 95th percentile residual mesh-to-mesh distance not exceeding 1.17 mm.
- Demonstrated a mean runtime of 2.3 seconds per frame, outperforming related work.
Conclusions:
- The developed approach shows potential for patient monitoring during external beam radiation therapy using reconstructed surfaces.
- Enables motion-compensated dose delivery by integrating the 4D surface motion field with external-internal correlation models.
