PURPOSE: To evaluate the feasibility and accuracy of automated analysis software for use with multidetector computed tomographic angiography (CTA) in the exact grading of internal carotid artery stenosis. METHODS: A retrospective pilot trial was performed using CTA datasets from 87 stenotic carotid arteries in 46 consecutive patients (34 men; median age 73.5 years) with known cerebrovascular disease. Internal carotid artery (ICA) stenosis was graded according to NASCET criteria by 2 experienced vascular radiologists in consensus using axial source images as well as curved planar reformations and digital subtraction angiography (DSA). These results were then compared to those obtained from the automated CTA analysis software and the results of manually adapted automated CTA analysis. RESULTS: Measurements from automated CTA analysis as well as manually adapted automated CTA analysis correlated significantly to those of axial/reformatted CTA and DSA (r=0.53 and r=0.82, r=0.58 and 0.70, respectively, all p<0.05). Compared to axial/reformatted CTA measurements, automated CTA analysis had a median difference of -16%, while manually adapted automated CTA had a difference of -10%. Corresponding differences in a comparison with DSA were +4% and -2%, respectively. Circumferential calcification or kinking of the ICA origin did not significantly interfere with these differences (all p>0.05). Sensitivities for the detection of ICA stenosis >70% by manually adapted automated CTA analysis and automated measurement were 44.2% and 34.9%, respectively, versus axial/reformatted CTA. Compared with DSA as the gold standard, the sensitivities were 54.2% and 62.5%, respectively. Specificities for both methods and gold standards all exceeded 90%. CONCLUSION: Commercially available automated CTA analysis is a feasible tool, but sensitivities are still not sufficient for clinical application.
PURPOSE: To evaluate the feasibility and accuracy of automated analysis software for use with multidetector computed tomographic angiography (CTA) in the exact grading of internal carotid artery stenosis. METHODS: A retrospective pilot trial was performed using CTA datasets from 87 stenotic carotid arteries in 46 consecutive patients (34 men; median age 73.5 years) with known cerebrovascular disease. Internal carotid artery (ICA) stenosis was graded according to NASCET criteria by 2 experienced vascular radiologists in consensus using axial source images as well as curved planar reformations and digital subtraction angiography (DSA). These results were then compared to those obtained from the automated CTA analysis software and the results of manually adapted automated CTA analysis. RESULTS: Measurements from automated CTA analysis as well as manually adapted automated CTA analysis correlated significantly to those of axial/reformatted CTA and DSA (r=0.53 and r=0.82, r=0.58 and 0.70, respectively, all p<0.05). Compared to axial/reformatted CTA measurements, automated CTA analysis had a median difference of -16%, while manually adapted automated CTA had a difference of -10%. Corresponding differences in a comparison with DSA were +4% and -2%, respectively. Circumferential calcification or kinking of the ICA origin did not significantly interfere with these differences (all p>0.05). Sensitivities for the detection of ICA stenosis >70% by manually adapted automated CTA analysis and automated measurement were 44.2% and 34.9%, respectively, versus axial/reformatted CTA. Compared with DSA as the gold standard, the sensitivities were 54.2% and 62.5%, respectively. Specificities for both methods and gold standards all exceeded 90%. CONCLUSION: Commercially available automated CTA analysis is a feasible tool, but sensitivities are still not sufficient for clinical application.
Authors: Chengcheng Zhu; Andrew J Patterson; Owen M Thomas; Umar Sadat; Martin J Graves; Jonathan H Gillard Journal: Int J Cardiovasc Imaging Date: 2012-11-08 Impact factor: 2.357
Authors: H A Marquering; P J Nederkoorn; L Smagge; H A Gratama van Andel; R van den Berg; C B Majoie Journal: AJNR Am J Neuroradiol Date: 2011-12-22 Impact factor: 3.825
Authors: J Borst; H A Marquering; M Kappelhof; T Zadi; A C van Dijk; P J Nederkoorn; R van den Berg; A van der Lugt; C B L M Majoie Journal: AJNR Am J Neuroradiol Date: 2015-08-06 Impact factor: 3.825
Authors: Timo Siepmann; Kristian Barlinn; Thomas Floegel; Jessica Barlinn; Lars-Peder Pallesen; Volker Puetz; Hagen H Kitzler Journal: Front Cardiovasc Med Date: 2021-12-09