A review of 3D/2D registration methods for image-guided interventions

Area of Science:

• Medical Imaging
• Image-Guided Interventions
• Computer-Aided Surgery

Background:

• Accurate registration of pre- and intra-interventional data is crucial for image-guided radiation therapy, radiosurgery, minimally invasive surgery, endoscopy, and interventional radiology.
• 3D/2D image registration enables the fusion of pre-operative (3D) and intra-operative (2D) imaging data.

Purpose of the Study:

• To survey and review 3D/2D data registration methods.
• To categorize these methods based on key characteristics relevant to image-guided interventions.

Main Methods:

• Focus on registration methods using 3D computed tomography (CT) or magnetic resonance (MR) images as pre-interventional data.
• Utilize 2D X-ray projection images as intra-interventional data.
• Review methods based on image modality, dimensionality, registration basis, geometric transformation, user interaction, and optimization procedures.

Main Results:

• Categorization of 3D/2D registration techniques based on various parameters.
• Identification of common approaches and variations in 3D/2D registration for image-guided procedures.
• Overview of the landscape of 3D/2D registration methods in medical imaging.

Conclusions:

• 3D/2D registration is a vital technology for enhancing precision in image-guided medical interventions.
• Understanding the different registration methods is key to selecting appropriate techniques for specific clinical applications.
• This review provides a comprehensive overview to guide future research and development in image registration.