RESUMEN
OBJECTIVE: Differences of attenuation and enhancement patterns in focal nodular hyperplasia and hepatocellular adenoma were evaluated and quantified using triphasic single-slice helical CT. MATERIALS AND METHODS: Forty-five histologically proven focal nodular hyperplasias in 27 patients and 18 hepatocellular adenomas in six patients were examined with helical CT. Quantitative evaluation included the following: attenuation of lesions, scar, and liver parenchyma during unenhanced, arterial (20 sec after injection), and portal venous phases (70 sec after injection); relative enhancement of lesions and liver (the ratio between attenuation in arterial phase and portal venous phase, respectively, and attenuation in unenhanced phase); and the prevalence of scar and its central vessel in focal nodular hyperplasia. RESULTS: The study showed no significant difference between mean attenuation values of focal nodular hyperplasia (mean +/- SD, 51.2 +/- 5.9 H) and hepatocellular adenoma (mean +/- SD, 56.3 +/- 7.8 H) in the unenhanced phase. In the arterial phase attenuation values were significantly higher in focal nodular hyperplasia (mean +/- SD, 117.9 +/- 15.1 H) than in hepatocellular adenoma (mean +/- SD, 80.1 +/- 10.5 H). In the portal venous phase no significant differences in attenuation values were detected between focal nodular hyperplasia (mean +/- SD, 112.1 +/- 20.4 H) and hepatocellular adenoma (mean +/- SD, 110.2 +/- 12.9 H). For enhancement parameter thresholds separating focal nodular hyperplasia from hepatocellular adenoma, the following were found: the relative enhancement was higher in 100% of the focal nodular hyperplasias and lower than or equal to 1.6 (accuracy, 96%) in 87% of the hepatocellular adenomas. CONCLUSION: Triphasic helical CT combined with quantitative evaluation of liver lesions offers the possibility of detecting differences in liver lesions that are visually similar on CT. The attenuation and relative enhancement in the arterial phase show significant differences that make accurate differentiation between focal nodular hyperplasia and hepatocellular adenoma possible.
Asunto(s)
Adenoma de Células Hepáticas/diagnóstico por imagen , Hiperplasia Nodular Focal/diagnóstico por imagen , Neoplasias Hepáticas/diagnóstico por imagen , Tomografía Computarizada por Rayos X/métodos , Adulto , Medios de Contraste , Diagnóstico Diferencial , Femenino , Humanos , Yohexol/análogos & derivadosRESUMEN
This paper presents a hybrid-unsupervised and supervised-classifier for land use classification of remote sensing images. The entire satellite image is quantized by an unsupervised Neural Gas process and the resulting codebook is labeled by a supervised majority voting process using the ground truth. The performance of the classifier is similar to that of Maximum Likelihood and is only a little worse than Multilayer Perceptions while training and classifying requires no expert knowledge after collecting the ground truth. The hybrid classifier is much better suited to classifications with complex non-normally distributed classes than Maximum Likelihood. The main advantage of the Neural Gas classifier, however, is that it requires much less user interaction than other classifiers, especially Maximum Likelihood.