This is a retrospective, single-center study to develop and validate an updated algorithm for computing fractional flow reserve from intravascular ultrasound images (IvusFFR v2.0). The study will use existing intravascular ultrasound (IVUS) images and corresponding invasive fractional flow reserve (FFR) measurements obtained from approximately 100 patients who underwent clinically indicated coronary angiography, IVUS, and FFR assessment at Fuwai Hospital. All data will be anonymized prior to analysis. The primary objective is to assess the diagnostic accuracy of IvusFFR v2.0 in identifying hemodynamically significant coronary stenosis, using wire-based FFR ≤0.80 as the reference standard. This study involves no prospective intervention or additional patient contact.

Inclusion Criteria

• Age ≥18 years.
• Underwent clinically indicated coronary angiography, intravascular ultrasound (IVUS) imaging, and invasive fractional flow reserve (FFR) measurement at Fuwai Hospital.
• IVUS pullback imaging of sufficient quality for analysis, covering the entire lesion segment of interest.
• Valid FFR measurement recorded.
• Presence of at least one native coronary artery stenosis with visual diameter stenosis between 30% and 90%.

Exclusion Criteria

• Chronic total occlusion (CTO) of the target vessel.
• Prior coronary artery bypass grafting (CABG) involving the target vessel.
• IVUS pullback did not cover the entire lesion.
• Severe myocardial bridge or vessel spasm in the interrogated vessel.
• Substantial thrombosis or dissection identified by IVUS.
• Image quality deemed inadequate for analysis by two independent analysts.

This retrospective study includes approximately 100 adult patients with suspected or known coronary artery disease who underwent clinically indicated invasive coronary angiography, intravascular ultrasound (IVUS) imaging, and fractional flow reserve (FFR) measurement at Fuwai Hospital. The cohort consists of patients with intermediate coronary lesions (30-90% diameter stenosis by visual estimation) for whom both anatomical assessment by IVUS and physiological assessment by FFR were deemed clinically necessary. Data were collected from existing medical records and imaging archives, and all personal identifiers have been removed prior to analysis.