Related Experiment Video
Updated: Feb 28, 2026
A Cognitive Fusion-guided Prostate Biopsy Using Multiparametric Magnetic Resonance Imaging and Transrectal Ultrasound
Published on: March 21, 2025
Classification of Clinical Significance of MRI Prostate Findings Using 3D Convolutional Neural Networks
Alireza Mehrtash1,2, Alireza Sedghi3, Mohsen Ghafoorian1,4
1Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States.
Abstract:
Prostate cancer (PCa) remains a leading cause of cancer mortality among American men. Multi-parametric magnetic resonance imaging (mpMRI) is widely used to assist with detection of PCa and characterization of its aggressiveness. Computer-aided diagnosis (CADx) of PCa in MRI can be used as clinical decision support system to aid radiologists in interpretation and reporting of mpMRI. We report on the development of a convolution neural network (CNN) model to support CADx in PCa based on the appearance of prostate tissue in mpMRI, conducted as part of the SPIE-AAPM-NCI PROSTATEx challenge. The performance of different combinations of mpMRI inputs to CNN was assessed and the best result was achieved using DWI and DCE-MRI modalities together with the zonal information of the finding. On the test set, the model achieved an area under the receiver operating characteristic curve of 0.80.
Insights
A new convolution neural network (CNN) model aids prostate cancer (PCa) detection using multi-parametric magnetic resonance imaging (mpMRI). Combining diffusion-weighted imaging (DWI) and dynamic contrast-enhanced MRI (DCE-MRI) with zonal data improved diagnostic accuracy.
Area of Science:
- Medical imaging
- Artificial intelligence in oncology
- Radiology
Background:
- Prostate cancer (PCa) is a significant cause of cancer mortality in men.
- Multi-parametric magnetic resonance imaging (mpMRI) is crucial for PCa detection and aggressiveness assessment.
- Computer-aided diagnosis (CADx) systems can enhance radiologists' interpretation of mpMRI.
Purpose of the Study:
- To develop a convolution neural network (CNN) model for PCa computer-aided diagnosis (CADx).
- To evaluate the performance of different mpMRI input combinations for PCa detection.
- To support radiologists in interpreting mpMRI for PCa.
Main Methods:
- Development of a CNN model for PCa CADx using mpMRI data.
- Utilized data from the SPIE-AAPM-NCI PROSTATEx challenge.
- Assessed performance using diffusion-weighted imaging (DWI), dynamic contrast-enhanced MRI (DCE-MRI), and zonal information.
Main Results:
- The best CNN model performance was achieved by combining DWI and DCE-MRI modalities with zonal information.
- The model demonstrated an area under the receiver operating characteristic curve (AUC) of 0.80 on the test set.
- This indicates a promising level of diagnostic accuracy for the developed CADx system.
Conclusions:
- A CNN model integrating DWI, DCE-MRI, and zonal data shows strong potential for PCa CADx.
- This approach can serve as a valuable clinical decision support tool for radiologists.
- Further development and validation could enhance PCa diagnosis and patient outcomes.
