An Interactive App with Multi-parametric MRI - Whole-Mount Histology Correlation for Enhanced Prostate MRI Training...

Area of Science:

• Radiology
• Medical Education
• Oncology

Background:

• Prostate cancer detection relies heavily on accurate interpretation of multi-parametric Magnetic Resonance Imaging (mpMRI).
• Radiologists require specialized training to effectively identify and characterize prostate cancer lesions on MRI, correlating findings with histology.

Purpose of the Study:

• To evaluate the educational effectiveness of a novel interactive web-based learning application, LearnRadiology, in improving prostate cancer detection skills among radiologists.
• To assess the impact of the learning application on diagnostic performance, including sensitivity and positive predictive value, in a simulated observer study.

Main Methods:

• An interactive learning application was developed featuring 20 curated prostate MRI cases with corresponding whole-mount histology.
• Three radiologists (one radiologist, two residents) performed an observer study, marking suspected cancer areas and providing confidence scores before and after using the learning app.
• Diagnostic performance was assessed by comparing radiologist interpretations with whole-mount pathology, with a minimum one-month washout period between assessments.

Main Results:

• All participating radiologists demonstrated improved sensitivity and positive predictive value for detecting prostate cancer after using the LearnRadiology app.
• Significant improvements were observed in the confidence scores for identifying true positive cancer lesions across all radiologists.
• Specific improvements in sensitivity were noted for residents (R2: 44% to 59%, P=0.03; R3: 62% to 72%, P=0.04), and positive predictive value for residents (R2: 52% to 79%, P=0.01; R3: 48% to 65%, P=0.04).

Conclusions:

• The web-based LearnRadiology application is an effective educational tool for enhancing prostate cancer detection in radiologists.
• The app demonstrably improves diagnostic performance and confidence, supporting postgraduate radiology training.
• Interactive learning resources like LearnRadiology can significantly contribute to better prostate cancer diagnosis through improved radiologist training.