A new study conducted by researchers at the First Affiliated Hospital of Jinan University demonstrates that deep learning-based CT-FFR (DL-CTFFR) can reduce the need for invasive coronary angiography (ICA) and improve outcomes in patients with stable coronary artery disease (CAD). The study results were published in the paper, “A 2-year Investigation of the Impact of the Computed Tomography-derived Fractional Flow Reserve Calculated Using a Deep Learning Algorithm on Routine Decision-making for Coronary Artery Disease Management” available in the peer-reviewed medical journal, European Radiology*.
In the study, researchers investigated the use of DL-CTFFR as an alternative to ICA in the selection of patients for coronary intervention. Coronary computed tomography angiogram (CCTA) scans from patients were processed using Keya Medical’s deep learning-based CT-FFR software, DeepVessel FFR.
Deep Learning-based CT-FFR as an Alternative to Invasive Coronary Angiography
In this single-center retrospective study, researchers at the First Affiliated Hospital of Jinan University investigated the value of a new DL-CTFFR algorithm using ICA-guided revascularization as a reference. The DL-CTFFR algorithm provided by Keya Medical utilizes deep learning to learn patterns of stenotic coronary arteries and FFR, improving the efficiency of FFR prediction when compared to existing methods derived from CCTA.
A total of 243 patients with symptomatic coronary artery disease (CAD) who had a documented degree of stenosis greater than 50% on CCTA were included in the study. After admission, patients received ICA-guided intervention, and DL-CTFFR measurements were performed retrospectively by an independent core laboratory at Keya Medical. During the two-year follow-up process, the impact of DL-CTFFR on clinical decision-making was compared to that of ICA-guided care for symptomatic CAD patients.
DL-CTFFR Demonstrates Effectiveness in Guiding Revascularization
The study results verified the safety and effectiveness of DL-CTFFR as an alternative to ICA. According to the study findings, up to 72% of invasive diagnostic procedures could have been avoided by using a DL-CTFFR value greater than 0.8 as a cut-off for intervention without increasing major adverse cardiovascular events (MACE).
Additionally, revascularization for symptomatic CAD patients guided by the combination of CCTA stenosis greater than 50% and DL-CTFFR less than 0.8 was associated with a MACE rate similar to that when guided by ICA stenosis in a 2-year follow-up investigation. The results demonstrate that DL-CTFFR could potentially improve the efficiency of CCTA-guided decision-making for invasive inspection.
It is also important to note that there was not a significant difference between the decision-making for intervention in tandem lesions based on the positive criteria for both ICA and DL-CTFFR, demonstrating that DL-CTFFR in coronary intervention for tandem lesions is safe.
Collaborate with Us
*European Radiology is a peer-reviewed medical journal published by Springer Science+Business Media that features original articles and state-of-the-art reviews written by leading radiologists.