3D digital breast cancer models with multimodal fusion algorithms

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Radiology

Background:

• Accurate 3D digital breast models integrating radiological and surface data are crucial for clinical applications.
• Current methods lack a ground truth for correlating surface and radiological breast information.
• Patient-specific digital breast models can enhance physician interpretation and visualization.

Purpose of the Study:

• To develop and validate a novel image acquisition protocol for creating patient-specific digital breast models.
• To assess the accuracy of fusing breast Magnetic Resonance Imaging (MRI) and 3D surface scan data.
• To evaluate the impact of breast volume on the fusion algorithm's performance.

Main Methods:

• Designed a new protocol to acquire synchronized breast MRI and 3D surface scan data using surface markers.
• Developed a fusion algorithm to integrate radiological images and 3D surface scans into a patient-specific digital breast model.
• Validated the algorithm in 16 breast cancer patients, quantifying registration errors and tumor location accuracy.

Main Results:

• The developed protocol successfully created patient-specific digital breast models integrating real breast shape and tumor location.
• Fusion of single breasts without a biomechanical model demonstrated acceptable registration errors and accurate tumor localization.
• Breast volume did not significantly affect the performance of the MRI/3D surface scan fusion algorithm.

Conclusions:

• The novel image acquisition protocol and fusion algorithm provide accurate patient-specific digital breast models.
• This technology shows promise for improving breast cancer diagnosis and treatment planning.
• Further research and virtual clinical interfaces can facilitate clinical integration of this fusion technology.