Can Automated Feeder Detection Software Improve Detection of Feeding Arteries?

Feeding artery detection rates during transarterial chemoembolization of HCC lesions were significantly improved by automated feeder detection (AFD) software, according to a new study. AFD also showed better user reproducibility than cone-beam CT alone.

The study included 44 patients undergoing transarterial chemoembolization for the first time. The patients enrolled between May 2012 and July 2013 and had treatment for a total of 85 hepatocellular carcinoma (HCC) tumors.

With the help of digital subtraction angiography (DSA), cone-beam computed tomography (CT), and AFD software with the option of manual adjustment, 147 feeding arteries were detected. To evaluate the performance of the testing methods, three independent interventional radiologists analyzed the cone-beam CT images retrospectively with and without AFD and MA.

Results showed that the use of AFD software led to significant improvement in detection of feeding arteries. With cone-beam CT, sensitivity was 70% and 100 ± 3.5 feeding arteries were detected, with 68.6% agreement among readers. On the other hand, AFD software detected 127±0.6 feeding arteries and had an 86% sensitivity and 99.7% reader agreement. AFD software also reduced the number of false negatives from an average of 47 ± 3.5 to 20 ± 0.6.

Additionally, manual adjustment of the AFD results led to similar feeding artery detection rates (127 ± 5.1, 86% sensitivity). There was lower interreader agreement (91.6%) and slightly fewer false positives.

“When used in conjunction with DSA, cone-beam CT þAFD achieves the highest possible feeding artery detection rates, making this the preferred feeding artery detection method to maximize the efficacy of transarterial chemoembolization procedures targeting HCC lesions,” the study’s authors concluded.

—Lauren LeBano


Chiaradia M, Izamis ML, Radaelli A. Sensitivity and Reproducibility of Automated Feeding Artery Detection Software during Transarterial Chemoembolization of Hepatocellular Carcinoma. J Vasc Interv Radiol. 2018 Feb 2. pii: S1051-0443(17)30964-8. doi: 10.1016/j.jvir.2017.10.025. [Epub ahead of print].