ABSTRACT
Abdominal aortic aneurysm (AAA) is a condition where the weakening of the aortic wall leads to its widening and the generation of a thrombus. To prevent a possible rupture of the aortic wall, AAA can be treated non-invasively by means of the endovascular aneurysm repair technique (EVAR), consisting of placing a stent-graft inside the aorta by a cateter to exclude the aneurysm sac from the blood circulation. A major complication is the presence of liquid blood turbulences, called endoleaks, in the thrombus formed in the space between the aortic wall and the stent-graft. In this paper we propose an automatic method for the detection of type II endoleaks in computer tomography angiography (CTA) images. The lumen and thrombus in the aneurysm area are first segmented using a radial model approach. Then, these regions are split into Thrombus Connected Components (TCCs) using a watershed-based segmentation and geometric and image content-based characteristics are obtained for each TCC. Finally, TCCs are classified into endoleaks and non-endoleaks using a multilayer Perceptron (MLP) trained on manual labeled sample TCCs provided by experts.
Subject(s)
Angiography/methods , Aortic Aneurysm, Abdominal/surgery , Endoleak/diagnosis , Image Processing, Computer-Assisted/methods , Tomography, X-Ray Computed/methods , Algorithms , Blood Vessel Prosthesis , Endoleak/diagnostic imaging , Humans , Neural Networks, Computer , StentsABSTRACT
OBJECTIVE: To evaluate the quantification of liver iron concentration using 1-Tesla magnetic resonance imaging (MRI) and its ability to diagnose or rule out hemochromatosis. To evaluate the role of 1.5-Tesla MRI in inconclusive cases. MATERIAL AND METHOD: Between 2002 and 2006, we used 1-Tesla MRI (Gandon method) and liver biopsy to quantify the liver iron concentration in 31 patients. Moreover, we used 1.5-Tesla MRI (according to Alústiza's model) and liver biopsy to determine the liver iron concentration in 10 additional patients and to check the results of 10 patients in whom 1-Tesla MRI detected iron overload. RESULTS: In the first group of 31 patients, liver biopsy classified the liver iron concentration as normal (<36 micromol.Fe/g) in 11 patients, as hemosiderosis (36-80 micromol.Fe/g) in 15, and as hemochromatosis (>80 micromol.Fe/g) in 5. The correlation with the values calculated at MRI was 100% in the 5 cases with hemochromatosis; in the 15 patients with hemosiderosis, 5 were correctly classified and the liver iron concentration was overestimated in 10; of the 11 patients with normal liver iron concentration, 6 were correctly classified and 5 were overestimated. Quantification >80 at MRI has a sensitivity and negative predictive value of 100% and specificity of 50% for the diagnosis of hemochromatosis. Quantification <36 at MRI has a positive predictive value and specificity of 100% to identify the absence of iron overload. In the 10 patients with liver biopsy that underwent 1.5-Tesla MRI, there was a high correlation between the two techniques. CONCLUSION: The reliability of the evaluation of liver iron concentration using 1-Tesla MRI is useful for ruling out hemochromatosis and identifying patients without iron overload. We observed a tendency to overestimate liver iron concentration in both patients with overload and in those without, and this limits the reliability of the technique. 1.5-Tesla MRI is a good alternative for quantifying liver iron concentration more precisely.
