Murray Law-Based Quantitative Flow Ratio Proficiency in Clinical Practice: Validating Routine Diagnostic Performance...

Area of Science:

• Cardiology
• Medical Imaging
• Artificial Intelligence

Background:

• Second-generation Murray law-based quantitative flow ratio (QFR) utilizes AI for improved angiography-derived physiological assessment.
• The diagnostic reliability of AI-driven QFR in routine practice is influenced by operator learning curves.

Purpose of the Study:

• To evaluate the impact of operator learning curves on the diagnostic reliability of AI-driven Murray law-based QFR.
• To determine the number of cases required to achieve high diagnostic performance and stability.

Main Methods:

• Consecutive patients with suspected myocardial ischemia underwent AI-driven QFR and fractional flow reserve measurement.
• Diagnostic accuracy was assessed using receiver operating characteristic analysis and learning curves were analyzed by sequential data expansion.
• Stabilization of diagnostic performance was evaluated by calculating first-order AUC differences.

Main Results:

• AI-driven QFR demonstrated excellent diagnostic accuracy with vessel-level AUC of 0.92 and patient-level AUC of 0.91.
• High diagnostic performance (AUC ≥0.90) was achieved after analyzing 170 vessels.
• Peak performance and stability were reached around 441 vessels, with no significant deviation in incremental AUC values.

Conclusions:

• Operator experience significantly influences the accuracy of AI-driven Murray law-based QFR.
• A minimum of 170 analyzed vessels ensures high QFR reliability, with optimal performance around 441 cases.
• Structured training is crucial for the effective clinical integration of AI-enhanced coronary physiology tools.