医学
肝细胞癌
肝病学
放射科
病理
对比度(视觉)
内科学
计算机科学
人工智能
作者
Kazushi Numata,Hiroyuki Fukuda,Hiromi Nihonmatsu,Masaaki Kondo,Akito Nozaki,Makoto Chuma,Manabu Morimoto,Takashi Oshima,Masahiro Okada,Takamichi Murakami,Shigeo Takebayashi,Shin Maeda,Yoshiaki Inayama,Masayuki Nakano,Katsuaki Tanaka
标识
DOI:10.1007/s00261-015-0489-6
摘要
We evaluated the use of tumor vessel patterns observed during arterial-phase contrast-enhanced ultrasonography (US) to differentiate regenerative nodules (RN) from early hepatocellular carcinoma (HCC) or high-grade dysplastic nodules (HGDN) in patients with chronic liver disease. Pathologically confirmed lesions (83 early HCC, 6 HGDN, and 13 RN with mean maximal diameters of 15.4, 15.3, and 16.2 mm, respectively) were enrolled in this retrospective study. We performed contrast-enhanced US using a perflubutane-based contrast agent. We then classified the tumor vessels observed during the arterial phase of contrast-enhanced US into two patterns: peripheral vessels (centripetal pattern) and central vessels (centrifugal pattern). Eighty-one (97.6%) of the 83 early HCC exhibited various enhancement patterns (hypovascular, 44.6%; isovascular, 25.3%; and hypervascular, 27.7%) and a peripheral vessel pattern, while the remaining 2 lesions (2.4%) exhibited hypovascular enhancement and a central vessel pattern. All 6 HGDN lesions were hypovascular with a peripheral vessel pattern. Twelve (92.3%) of the 13 RN were hypovascular with a central vessel pattern, and the remaining one (7.7%) was hypervascular with a central vessel pattern. When lesions exhibiting a central vessel pattern during arterial-phase contrast-enhanced US were diagnosed as RN, the sensitivity, specificity, and accuracy of these diagnoses were 100%, 97.8%, and 98.0%, respectively. The tumor vessel patterns observed during arterial-phase contrast-enhanced US may be useful for differentiating RN from early HCC or HGDN in patients with chronic liver disease.
科研通智能强力驱动
Strongly Powered by AbleSci AI