Semiautomated matching and seed position location for implanted ribbons

Area of Science:

• Medical Physics
• Radiotherapy
• Image Analysis

Background:

• Accurate 3-D seed positioning is crucial for effective brachytherapy.
• Traditional methods using individual seed matching in radiographs are prone to ambiguity and errors.
• Implanting seeds within ribbons offers a potential solution to improve localization accuracy.

Purpose of the Study:

• To develop and evaluate a semi-automated method for accurate 3-D localization of brachytherapy seeds implanted within ribbons.
• To overcome the inherent ambiguity in matching individual seed images from radiographs.
• To improve the precision and reduce errors in 3-D seed position determination.

Main Methods:

• A semi-automated procedure for matching ribbon images from two radiographs with different parallax angles.
• Determination of the 3-D loci of ribbons using ray intersection and spanning tree analysis.
• Interpolation along the 3-D ribbon locus to determine individual seed positions.
• Utilized a double application of the assignment algorithm for ribbon image matching.

Main Results:

• The semi-automated ribbon matching technique significantly reduces ambiguity compared to individual seed matching.
• The method accurately determines the 3-D loci of ribbons and individual seed positions.
• Experiments with a phantom demonstrated a systematic error of 3.0 mm and a random error of less than 2 mm in 3-D seed localization.

Conclusions:

• Implanting brachytherapy seeds within ribbons, coupled with a semi-automated matching procedure, enables accurate and unambiguous 3-D localization.
• This technique offers a substantial improvement over methods relying on individual seed image matching.
• The developed procedure provides a reliable solution for precise 3-D seed positioning in brachytherapy.