Related Experiment Video
Updated: Apr 11, 2026
06:08
A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound
Published on: March 21, 2025
2.0K
Multimodal image-guided prostate fusion biopsy based on automatic deformable registration
Oliver Zettinig1, Amit Shah2, Christoph Hennersperger2
1Computer-Aided Medical Procedures, Technische Universität München, Boltzmannstr. 3, 85748, Garching, Germany. oliver.zettinig@tum.de.
Summary
This study introduces a new PET-MRI and ultrasound fusion system for prostate cancer biopsies. The framework accurately maps suspicious regions, improving diagnostic accuracy for prostate cancer.
Area of Science:
- Medical Imaging
- Biomedical Engineering
- Radiology
Background:
- Transrectal ultrasound (TRUS)-guided prostate biopsy is the standard for prostate cancer diagnosis but has low sensitivity.
- Positron Emission Tomography (PET) imaging with prostate-specific membrane antigen (PSMA) radiotracers and Magnetic Resonance Imaging (MRI) enhances the detection of suspicious prostate cancer areas.
Purpose of the Study:
- To develop and evaluate a multimodal image-guided biopsy framework combining PET-MRI and TRUS.
- To enable accurate real-time guidance for prostate biopsies by fusing advanced imaging modalities.
Main Methods:
- Automatic segmentation of prostate TRUS images using a Hough transform-based random forest.
- Elastic surface alignment and deformation field propagation using the Coherent Point Drift algorithm for image registration.
- Integration of PET-MRI data with TRUS for multimodal image fusion.
Main Results:
- The proposed framework demonstrates minimal clinical workflow impact and temporal overhead.
- Evaluations using phantoms and clinical data from 13 patients confirm the approach's validity.
- The system successfully maps suspicious regions identified in PET/MRI to the TRUS image space.
Conclusions:
- The multimodal fusion system effectively bridges information from PET-MRI to TRUS-guided interventions.
- This approach enhances the precision of targeted prostate biopsies by leveraging complementary imaging data.
