Artificial intelligence in multiparametric prostate cancer imaging with focus on deep-learning methods.

Area of Science:

• Radiology
• Medical Imaging
• Artificial Intelligence

Background:

• Prostate cancer diagnosis relies on multiparametric imaging, which can be burdensome for radiologists.
• Combining multiple imaging techniques is complex and time-consuming for accurate diagnosis.

Purpose of the Study:

• To review advances in computer-aided diagnostic (CAD) tools for prostate cancer imaging.
• To highlight the impact of deep learning on CAD for prostate cancer.
• To compare methods for delivering CAD output to clinicians.

Main Methods:

• Review of recent literature on CAD for prostate cancer.
• Focus on deep learning techniques applied in the last few years.
• Analysis of methods for integrating CAD results into clinical workflow.

Main Results:

• Significant progress in CAD tools for prostate cancer detection and characterization.
• Deep learning has shown promise in enhancing the performance of multiparametric imaging analysis.
• Various methods exist for presenting CAD output to aid radiologists' decision-making.

Conclusions:

• CAD tools, particularly those using deep learning, are crucial for improving prostate cancer imaging efficiency and accuracy.
• Further research is needed to optimize the integration of CAD into clinical practice.
• Advanced CAD systems can alleviate radiologist workload and enhance diagnostic performance.