ADR--Anatomy-Driven Reformation.

Area of Science:

• Medical Imaging and Visualization
• Computational Anatomy
• Computer-Aided Diagnosis

Background:

• Advanced visualization techniques are crucial for modern computer-aided medical applications.
• Multiplanar Reformation (MPR) and Curved Planar Reformation (CPR) are established tools aiding medical diagnosis and therapy.
• Current methods may lack the ability to adapt to complex individual patient anatomy.

Purpose of the Study:

• To present a novel reformation approach generalizing MPR to curved surfaces for patient-specific anatomical geometry.
• To develop a framework for generating accurate, geometry-driven reformatted medical volumes.
• To improve diagnostic, navigation, and annotation tasks through enhanced visualization.

Main Methods:

• A non-linear as-rigid-as-possible volumetric deformation scheme using triangular surface meshes as input.
• Introduction of importance maps to control distortion and enhance visual quality in critical areas.
• Integration with existing slice-based medical dataset inspection workflows.

Main Results:

• Successful generation of reformatted medical volumes tailored to individual patient anatomy.
• Demonstrated ability to create flat views of curved anatomical structures with controlled distortions.
• Improved visual quality in areas of clinical interest through importance maps.

Conclusions:

• The proposed reformation framework offers a generalized approach for curved surfaces, enhancing medical visualization.
• The method seamlessly integrates with current medical imaging workflows, potentially increasing efficiency.
• The developed system and use cases demonstrate significant benefits for diagnosis, navigation, and annotation.