Automated patient couch removal algorithm on CT images.

Area of Science:

• Medical Imaging
• Image Processing
• Radiology

Background:

• Patient couches in Computed Tomography (CT) scans are often considered artifacts.
• These artifacts can hinder the interpretation of medical images, particularly with 3D rendering techniques.
• Automated removal of these artifacts is desirable to improve image quality and diagnostic accuracy.

Purpose of the Study:

• To propose and validate a novel automated method for patient couch removal from CT images.
• To address the challenge of patient couch artifacts in medical imaging.
• To enhance the utility of CT scans for visualization and analysis.

Main Methods:

• The method employs similarity measurements between selected axial slices.
• It assumes the patient couch remains consistent across different slices.
• It identifies and accounts for couch bending via sagittal movement analysis in consecutive slices after initial segmentation.

Main Results:

• The developed method successfully removes patient couches from CT images.
• Validation on real medical data demonstrates the effectiveness of the automated approach.
• The technique operates without requiring user interaction.

Conclusions:

• The proposed automated patient couch removal method is effective for CT images.
• This technique offers a valuable tool for artifact reduction in medical imaging.
• The method has the potential to improve image quality for clinical applications and research.