Related Experiment Video
Updated: Apr 3, 2026
06:08
A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound
Published on: March 21, 2025
2.0K
A fully automatic computer aided diagnosis system for peripheral zone prostate cancer detection using
Valentina Giannini1, Simone Mazzetti1, Anna Vignati1
1Department of Radiology at Candiolo Cancer Institute-FPO, IRCCS, Strada Provinciale 142 km 3.95, 10060 Candiolo, Italy.
Summary
A new computer-aided diagnosis (CAD) system enhances multiparametric MRI for prostate cancer (PCa) detection. This automated tool improves accuracy and reduces radiologist workload in diagnosing and staging PCa.
Area of Science:
- Radiology
- Medical Imaging
- Oncology
Background:
- Multiparametric MRI (mp-MRI) is crucial for prostate cancer (PCa) diagnosis and staging.
- Interpreting mp-MRI is time-consuming and requires specialized radiologists, leading to potential variability.
Purpose of the Study:
- To develop and evaluate a fully automatic computer-aided diagnosis (CAD) system for prostate cancer detection using mp-MRI.
- To improve the accuracy, reproducibility, and efficiency of PCa diagnosis and staging.
Main Methods:
- A two-stage automatic CAD system was developed: 1) voxel-wise malignancy probability mapping, and 2) candidate segmentation of suspected regions.
- The system was trained and tested using whole-mount histological sections as the reference standard on 56 patients (65 lesions).
Main Results:
- The voxel-wise step achieved an area under the ROC curve of 0.91.
- The second step demonstrated a per-patient sensitivity of 97% with a median of 3 false positive regions per prostate.
Conclusions:
- The proposed CAD system shows potential as a first or second reader for PCa management, reducing reporting time and inter-reader variability.
- This system could assist radiologists in identifying biopsy targets for MRI-guided biopsies.
