A localization algorithm and error analysis for stereo x-ray image guidance

Area of Science:

• Medical Imaging
• Radiography
• Biomedical Engineering

Background:

• Stereo x-ray radiography is increasingly used in clinical settings.
• Accurate 3D localization is critical for effective breast biopsy procedures.

Purpose of the Study:

• To analyze the 3D localization error in breast biopsy using stereo x-ray.
• To offer guidelines for enhancing the accuracy of these procedures.

Main Methods:

• Developed a CCD-based digital stereo x-ray imaging system prototype.
• Formulated a mathematical model with two x-ray sources and one detector.
• Derived a closed-form least-squares solution for 3D localization of feature points, specifically the biopsy needle tip.

Main Results:

• Analyzed 3D localization error based on object location, measurement error, source separation, and source-to-detector distance.
• Validated theoretical predictions through numerical simulations and experimental testing.
• Demonstrated agreement between simulated and experimental data.

Conclusions:

• The derived formulas accurately predict 3D localization error in stereo x-ray imaging.
• Results offer valuable insights for optimizing system design and imaging protocols for breast biopsies.