Related Experiment Video
Updated: Dec 30, 2025
09:24
Multiparametric Optical Mapping of the Langendorff-perfused Rabbit Heart
Published on: September 13, 2011
23.2K
Dynamic coronary roadmapping via catheter tip tracking in X-ray fluoroscopy with deep learning based Bayesian
Hua Ma1, Ihor Smal2, Joost Daemen3
1Biomedical Imaging Group Rotterdam, Erasmus MC, University Medical Center Rotterdam, Rotterdam, the Netherlands.
Medical Image Analysis
|January 25, 2020
Summary
This study introduces a new dynamic coronary roadmapping technique for percutaneous coronary intervention (PCI). It improves visualization and reduces contrast agent use by compensating for heart and breathing motion.
Area of Science:
- Medical Imaging
- Interventional Cardiology
- Deep Learning
Background:
- Percutaneous coronary intervention (PCI) relies on X-ray angiography with contrast agents.
- High contrast agent use increases kidney failure risk, necessitating non-contrast methods.
- Current non-contrast methods require mental reconstruction of coronary anatomy.
Purpose of the Study:
- To develop a novel dynamic coronary roadmapping approach for PCI.
- To enhance visual feedback and minimize contrast agent usage.
- To improve interventional cardiologist guidance during procedures.
Main Methods:
- Developed a dynamic coronary roadmapping approach compensating for cardiac and respiratory motion.
- Utilized ECG alignment for cardiac motion and catheter tip tracking for respiratory motion.
- Proposed a deep learning-based Bayesian filtering method for accurate catheter tip tracking integrating CNNs and particle filtering.
Main Results:
- Validated the approach on clinical X-ray images, demonstrating accurate catheter tip tracking.
- Achieved accurate performance in dynamic coronary roadmapping experiments.
- The system operates in real-time on a single GPU.
Conclusions:
- The novel dynamic coronary roadmapping approach offers improved visual guidance during PCI.
- It has the potential to significantly reduce contrast agent use in clinical workflows.
- This technology can enhance patient safety by mitigating risks associated with contrast agents.
