Interactive, multi-modality image registrations for combined MRI/MRSI-planned HDR prostate brachytherapy

Area of Science:

• Medical Physics
• Oncology
• Radiotherapy

Background:

• Focal high dose-rate (HDR) brachytherapy for prostate cancer requires precise imaging for accurate dose delivery.
• Image registration is crucial for aligning pre-treatment imaging with planning data, but can be affected by imaging artifacts.

Purpose of the Study:

• To detail the clinical image registration process for prostate HDR brachytherapy.
• To evaluate the effectiveness of image transformations in correcting distortions and improving treatment planning.

Main Methods:

• Utilized Magnetic Resonance Imaging (MRI), Magnetic Resonance Spectroscopic Imaging (MRSI), and Computed Tomography (CT).
• Applied rigid and nonrigid transformations to correct for endorectal coil deformations and align images.
• Employed mutual information as a morphing metric and an inverse planning optimization algorithm.

Main Results:

• The morphing algorithm successfully corrected probe-induced prostatic distortions.
• Mutual information significantly increased after image morphing (p = 0.0071), validating the technique.
• The entire registration procedure added less than 30 minutes to treatment planning.

Conclusions:

• Image transformations and registrations are valuable tools for prostate HDR brachytherapy.
• This methodology enhances the precision of dose delivery to dominant intraprostatic lesions while sparing organs at risk.