Affiliations: [a]
Department of Ultrasound, Zhongshan Hospital, Fudan University, Shanghai, China
| [b]
Present address: Department of Diagnostic Ultrasound, Second University Hospital of Sichuan University, Cheng Du, Sichuan, China
| [c]
Shanghai Institute of Medical Imaging, Shanghai, China
| [d]
Medical Imaging College, Shanghai University of Medicine and Health Sciences, Shanghai, China
Correspondence:
[*]
Corresponding author: Hong Ding, MD, Department of Ultrasound, Zhongshan Hospital, Fudan University, Shanghai 200032, China. Tel.: +86 136 5188 6013; Fax: +86 21 6422 0319; E-mail: ding.hong@zs-hospital.sh.cn and Jinzhu Zhou, Professor, Medical Imaging College, Shanghai University of Medicine and Health Sciences, Shanghai 201318, China. Tel.: +86 137 0171 1490; Fax: +86 21 6588 7512; E-mail: zhoujz@sumhs.edu.cn.
Abstract: BACKGROUND:Savitzky-Golay filter is a digital filter used in data smoothing, we introduced a contrast-enhanced ultrasound (CEUS) quantification software based on the filter (SGCQ). OBJECTIVE:To explore the methodology of analyzing hepatic tumors hemodynamics applying SGCQ software and the correlation between SGCQ parameters and hepatocellular carcinoma (HCC) angiogenesis. METHODS:Nighty-seven right-lobe located hepatic mass cases (15 females and 82 males, mean age 58±10y, mean lesion size: 39.9±11.6 mm) underwent CEUS scan preoperatively and had a final diagnosis of HCC (n = 52) or colorectal cancer metastatic liver tumors (MLT, n = 45) were included. CEUS was carried out using a 1–5 MHz convex probe. The CEUS clips were recorded and analyzed off-line to obtain the parameters. The parameters were analyzed by 3 observers separately to investigate inter-observer variability. Parameters were compared between tumor and adjacent liver and between different tumors. Immunohistochemistry was used to evaluate the microvessel density (MVD) of HCC, and the correlation between the parameters and MVD was analyzed. RESULTS:Intraclass correlation coefficient (ICC) of all parameters were greater than 0.75 except wash-in slope (a3) and time to peak (TTP) of adjacent liver. The parameters of a3, wash-out slope (a2), perfusion time (PT) and area under the curve (AUC) were significantly different between HCC and liver. The a2, a3, AUC, PT and enhanced intensity (EI) were significantly different between MLT and liver. AUC, a2, a3, da1 and dPI were significantly different between HCC and MLT. AUC, a2 and EI correlated with MVD. CONCLUSION:SGCQ software quantification have good consistency among three observers, the parameters of SGCQ can display the hemodynamics of HCC and MLT and the difference between them. AUC and EI can serve as useful biomarkers in tumor angiogenesis evaluation.
