| 张长江,陈峻帆,黄豆豆,蒋雯丽,罗银灯.体素内不相干运动直方图参数预测直肠癌神经周围浸润[J].中国医学影像技术,2025,41(1):99~103 |
| 体素内不相干运动直方图参数预测直肠癌神经周围浸润 |
| Intravoxel incoherent motion histogram parameters for predicting perineural invasion of rectal cancer |
| 投稿时间：2024-06-06 修订日期：2024-09-25 |
| DOI：10.13929/j.issn.1003-3289.2025.01.021 |
| 中文关键词: 直肠肿瘤 磁共振成像 体素内不相干运动 |
| 英文关键词:rectal neoplasms magnetic resonance imaging intravoxel incoherent motion |
| 基金项目: |
|
| 摘要点击次数: 357 |
| 全文下载次数: 116 |
| 中文摘要: |
| 目的 观察体素内不相干运动(IVIM)直方图参数预测直肠癌神经周围浸润(PNI)的价值。方法 回顾性分析55例直肠癌患者,根据有无PNI将其分为阳性组(n=23)与阴性组(n=32);基于MR IVIM图像分别获取表观扩散系数(ADC)、灌注分数(f)、真实扩散系数(D)及伪扩散系数(D*)直方图参数,包括均值(mean)、偏度(skewness)、峰度(kurtosis)、熵(entropy),以及第25、50、75百分位数(25th、50th、75th),并进行组间比较。将组间差异有统计学意义的参数纳入二元逐步logistic回归分析,筛选预测直肠癌PNI状态的独立因素,构建联合参数模型;绘制受试者工作特征曲线,计算曲线下面积(AUC),评估各IVIM直方图参数及联合参数模型预测直肠癌PNI的效能。结果 阳性组ADCmean、ADC25th、ADC50th、ADC75th、fmean、f25th、f50th、f75th及Dentropy*均低于阴性组(P均＜0.05);以上述参数预测直肠癌PNI阳性的AUC分别为0.693、0.665、0.701、0.675、0.831、0.847、0.835、0.722及0.785。基于f25th及Dentropy*构建的联合参数模型预测直肠癌PNI阳性的敏感度、特异度及AUC分别为95.65%、65.63%及0.897。结论 利用IVIM直方图参数可有效预测直肠癌PNI。 |
| 英文摘要: |
| Objective To observe the value of intravoxel incoherent motion (IVIM) histogram parameters for predicting perineural invasion (PNI) of rectal cancer. Methods Fifty-five patients with rectal cancer were retrospectively enrolled and divided into positive group (n=23) and negative group (n=32) according to PNI or not. Histogram parameters of apparent diffusion coefficient (ADC), perfusion fraction (f), true diffusion coefficient (D) and pseudo diffusion coefficient (D*) were obtained based on MR IVIM images, including the mean, skewness, kurtosis, entropy, as well as the 25th, 50th and 75th percentiles (25th, 50th, 75th). The above IVIM histogram parameters were compared between groups, and those with significant difference were included in binary stepwise logistic regression analysis to screen independent factors for predicting PNI status of rectal cancer. Then a combined parameter model was established. Receiver operating characteristic curve was drawn, and the area under the curve (AUC) was calculated to evaluate the efficacy of IVIM histogram parameters and combined parameter model for predicting PNI of rectal cancer. Results ADCmean, ADC25th, ADC50th, ADC75th, fmean, f25th, f50th, f75th and D*entropy in positive group were all lower than those in negative group (all P＜0.05). AUC of the above parameters for predicting PNI of rectal cancer was 0.693, 0.665, 0.701, 0.675, 0.831, 0.847, 0.835, 0.722 and 0.785, respectively. The sensitivity, specificity and AUC of combined parameter model based on f25th and D*entropy for predicting PNI of rectal cancer was 95.65%, 65.63% and 0.897, respectively. Conclusion IVIM histogram parameters could be used to predict PNI of rectal cancer effectively. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
