Computer-aided simple triage (CAST) for coronary CT angiography (CCTA).

Area of Science:

• Cardiology
• Medical Imaging
• Artificial Intelligence

Background:

• Computer-aided detection/diagnosis (CAD) systems are evolving.
• Computer-aided simple triage (CAST) is a new subclass of CAD.
• Coronary CT angiography (CCTA) is crucial for diagnosing coronary artery disease.

Purpose of the Study:

• To present a CAST software system for fully automatic initial interpretation of CCTA.
• To demonstrate the system's CAST compliance and suitability for emergency room (ER) chest pain patient triage.

Main Methods:

• The system segments the coronary artery tree and labels major arteries.
• It detects significant stenotic lesions (>50% stenosis) and assesses image quality.
• For multiphase studies, the system automatically selects the optimal phase.

Main Results:

• The system achieved >90% per patient sensitivity and 40-70% per patient specificity.
• In the chest pain ER population, specificity was 60-70% with >98% NPV.
• The system demonstrated high diagnostic performance without human interaction.

Conclusions:

• The CCTA CAD system meets CAST requirements for efficient 24/7 ER triage.
• This is the first fully operational, clinically validated, CAST-compliant CAD system for CCTA analysis.
• The system enables automatic detection of significant stenosis, improving patient care.