Robust electromagnetically guided endoscopic procedure using enhanced particle swarm optimization for multimodal...

Area of Science:

• Medical Imaging
• Robotics
• Computer Vision

Background:

• Electromagnetically guided endoscopy requires precise localization using multimodal sensory information.
• Integrating diverse data sources (CT images, video, sensor data) for stable endoscopic guidance remains a challenge.

Purpose of the Study:

• To propose a novel framework for accurate and continuous endoscope localization.
• To enhance electromagnetic guidance by effectively fusing multimodal sensory information.

Main Methods:

• Utilized an enhanced particle swarm optimization (PSO) algorithm.
• Fused preoperative CT images, endoscopic videos, and electromagnetic sensor data.
• Incorporated current observations and adaptive parameters to improve PSO performance.

Main Results:

• Achieved average guidance accuracy of 3.0 mm and 5.6°, outperforming previous methods (3.9 mm, 7.0°).
• Demonstrated superior position and orientation smoothness (1.0 mm, 1.6° vs. 2.0 mm, 2.6°).
• Improved average visual guidance quality to 0.29.

Conclusions:

• Developed a robust electromagnetically guided endoscopy framework using enhanced PSO.
• Significantly reduced guidance errors, demonstrating the efficacy of the proposed method.