HASTE 扩散加权 MR 成像对头颈部病变表观扩散系数的各种定量方法的比较。
Comparison of various methods for quantification of apparent diffusion coefficient of head and neck lesions with HASTE diffusion-weighted MR imaging.
机构信息
Department of Oral and Maxillofacial Radiology, Tokyo Dental College, Chiba, Japan.
出版信息
Oral Surg Oral Med Oral Pathol Oral Radiol. 2012 Aug;114(2):266-76. doi: 10.1016/j.oooo.2012.03.015.
OBJECTIVES
The purposes of this retrospective study were to compare various methods of apparent diffusion coefficient (ADC) measurement for head and neck lesions in half-Fourier single-shot turbo spin-echo (HASTE) diffusion-weighted imaging (DWI) and determine the threshold ADC value for predicting malignancy.
STUDY DESIGN
HASTE DW images of 46 lesions (10 cysts, 14 benign tumors, and 22 malignant tumors) were studied retrospectively. The ADC values were compared among the 0-1000 method, 500-1000 method, and weighted linear regression (WLR) fit.
RESULTS
The highest overall accuracies of 83.3%, 86.1%, and 88.9% were obtained when ADC values of 1.24 × 10(-3) mm(2)/s (0-1000 method), 0.98 × 10(-3) mm(2)/s (500-1000 method), and 1.23 × 10(-3) mm(2)/s (WLR fit), respectively, were used for the threshold.
CONCLUSIONS
The present results indicate that ADC measurement with HASTE DWI is useful in predicting malignancy of head and neck lesions.
目的
本回顾性研究旨在比较半傅里叶单激发涡轮自旋回波（HASTE）弥散加权成像（DWI）中头颈部病变表观弥散系数（ADC）的各种测量方法，并确定预测恶性肿瘤的 ADC 值阈值。
研究设计
对 46 个病变（10 个囊肿、14 个良性肿瘤和 22 个恶性肿瘤）的 HASTE DWI 图像进行回顾性研究。比较了 0-1000 法、500-1000 法和加权线性回归（WLR）拟合的 ADC 值。
结果
当使用阈值为 1.24×10(-3)mm(2)/s（0-1000 法）、0.98×10(-3)mm(2)/s（500-1000 法）和 1.23×10(-3)mm(2)/s（WLR 拟合）时，ADC 值的总体准确率分别为 83.3%、86.1%和 88.9%，最高。
结论
本研究结果表明，HASTE DWI 中的 ADC 测量有助于预测头颈部病变的恶性程度。
Zotero 插件