Measurement of Cardiothoracic Ratio on Chest X-rays Using Artificial Intelligence-A Systematic Review and...

Area of Science:

• Medical Imaging and Diagnostics
• Artificial Intelligence in Healthcare
• Cardiology and Radiology

Background:

• Chest X-rays (CXRs) are essential for diagnosing conditions like cardiomegaly using the cardiothoracic ratio (CTR).
• Automating CTR determination with artificial intelligence (AI) offers potential for improved patient care and diagnostic efficiency.

Purpose of the Study:

• To systematically review and meta-analyze the performance of AI models in automated CTR assessment.
• To compare AI-driven CTR determination against human assessments and identify optimal models for clinical use.

Main Methods:

• A comprehensive database search was conducted in June 2023, adhering to the PICO framework.
• Included studies focused on AI-assisted CTR assessment from standing-position CXRs published in the last decade.
• Fourteen studies (70,472 CXRs) were analyzed using PRISMA 2020 guidelines and Cochrane risk of bias assessment.

Main Results:

• AI models demonstrated a pooled AUC of 0.959 (95% CI 0.944-0.975) for cardiomegaly detection.
• The pooled standardized mean difference for CTR measurement was 0.0353 (95% CI 0.147-0.0760).
• Significant heterogeneity was observed (I² 89.97%), with no detected publication bias.

Conclusions:

• AI shows significant promise for accurate CTR measurement and cardiomegaly detection in CXRs.
• Standardizing AI methodologies and reporting is critical for reliable clinical integration and advancing medical imaging diagnostics.