Fig. 5: Diagnostic workflow and performance: before versus after AI implementation. | Nature Communications

Fig. 5: Diagnostic workflow and performance: before versus after AI implementation.

From: Automated MRI system for clinically significant prostate cancer detection development validation and real-world implementation

Fig. 5

A, B The prospective design flowchart and (C) result comparison. All statistical tests were two-sided, with P < 0.05 indicative of a statistically significant difference. AUC, AUCs; PPV, positive predictive value; NPV, negative predictive value; AI, ProAI software. Source data are provided as a Source Data file. Note: B AI candidates required reader endorsement to enter the decision; unendorsed candidates were not counted as positives. C Performance metrics: AUC was derived from continuous scores. Sensitivity/specificity utilized a fixed PI-RADS threshold (≥3).

Back to article page