Automated seed localization from CT datasets of the prostate

Area of Science:

• Medical Physics
• Oncology
• Radiotherapy

Background:

• Permanent brachytherapy implants are increasingly used for prostate cancer treatment.
• Accurate post-treatment dose calculation is vital for assessing patient outcomes and planning future treatments.

Purpose of the Study:

• To develop and evaluate an automatic method for seed localization in permanent brachytherapy implants using CT datasets.
• To compare the accuracy and efficiency of the automated method against the traditional three-film technique.

Main Methods:

• Developed an automatic seed localization method using CT datasets of the prostate.
• Tested the method on a phantom with a patient-planned seed distribution.
• Compared the automated method to the three-film technique using three patient datasets.

Main Results:

• The automatic method demonstrated equal or greater accuracy than the three-film technique across 1 mm, 3 mm, and 5 mm contiguous CT slices.
• The automated method eliminated inter- and intra-observer variability inherent in manual localization.
• Reduced user input time and operator dependency compared to manual methods.

Conclusions:

• The automated seed localization method is a less time-consuming, less operator-dependent, and potentially more accurate alternative to the three-film technique.
• This method enhances brachytherapy seed localization, supporting improved dose calculation and patient management in prostate cancer treatment.