Deep learning, optimal subtraction technique for coronary stent evaluation by CTA

August 15, 2022 – According to the ARRS American Journal of Roentgenology (AJR), the combination of deep learning reconstruction (DLR) and a subtraction technique yielded optimal diagnostic performance for the detection of restenosis intra-stent by coronary CTA.

Noting that these findings could guide patient selection for invasive coronary stent assessment, combining DLR with a two-breath subtraction technique “may help overcome challenges with stent-related bloom artifact,” added corresponding author Yi-Ning Wang of the State Key Laboratory of Severe and Rare Complex Diseases at Peking Union Medical College Hospital in China.

Between March 2020 and August 2021, Wang and his team studied 30 patients (22 male, 8 female; mean age, 63.6 years) with a total of 59 coronary stents who underwent coronary angioplasty using the two-breath technique. (i.e. no-contrast and contrast-enhanced acquisitions). Conventional and subtraction images were reconstructed for hybrid iterative reconstruction (HIR) and DLR, while the maximum visible in-stent lumen diameter was measured. Two readers independently assessed the images for in-stent restenosis (≥50% stenosis).

Ultimately, coronary CTA using DLR and subtraction technique – with combined interpretation (conventional and subtraction images) – yielded PPV, VPN and accuracy for in-stent restenosis for reader 1 of 73.3%, 93.2%, and 88.1%, and for Reader 2 75.0%, 84.3%, and 83.1%, respectively.

Recognizing that the two-breath subtraction technique requires additional acquisition without contrast (and therefore a higher radiation dose), “DLR allows for a reduction in radiation exposure, while improving image quality”, underlined the authors of this article of the AJR.

For more information:

Comments are closed.