论文部分内容阅读
目的:探讨基于表观弥散系数(ADC)值的直方图分析方法在鉴别中耳胆脂瘤和慢性化脓性中耳炎中的最佳指标与诊断效能。方法:回顾性分析。纳入2017年1月—2019年12月首都医科大学附属北京同仁医院放射科经手术病理证实的64例中耳胆脂瘤(胆脂瘤组)及56例慢性化脓性中耳炎(中耳炎组)患者MRI资料。采用ImageJ软件在ADC图病变显示的最大层面勾画感兴趣区(ROI)，通过SPSS统计软件对直方图分析获得ADC值的均数、标准差、偏度、峰度、第10百分位数(n P10)和第90百分位数(n P90)等指标进行统计学分析，比较两组直方图各参数指标的差异，并绘制有差异的ADC值指标鉴别两种病变的受试者操作特征曲线(ROC)，获得各指标的ROC下面积(AUC)值。采用logistic回归分析方法探讨鉴别两种病变的最佳独立诊断指标。n 结果:胆脂瘤组和中耳炎组间ADC值的均数、标准差、偏度、n P10和n P90差异均有统计学意义(n P值均<0.01)，且鉴别两种病变的AUC值分别为0.912、0.988、0.809、0.793和0.957，其中ADC值标准差的AUC值最大，灵敏度和特异度分别是95.31%和96.43%。Logistic回归分析结果显示，ADC值的标准差为鉴别中耳胆脂瘤和慢性化脓性中耳炎的最佳独立诊断指标(n OR＝1.034，95%可信区间1.020～1.048，n P<0.01)。n 结论:基于ADC值的直方图分析方法有助于提高鉴别中耳胆脂瘤和慢性化脓性中耳炎的能力，ADC值标准差的鉴别效能最佳。“,”Objective:This study aimed to investigate the best index and diagnostic efficiency in the differential diagnosis of middle ear cholesteatoma and chronic suppurative otitis media using histogram analysis based on the apparent diffusion coefficient (ADC) value.Methods:The MRI images of 64 patients with middle ear cholesteatoma (cholesteatoma group) and 56 patients with chronic suppurative otitis media (otitis media group) confirmed by pathology at the Department of Radiology, Beijing Tongren Hospital, Capital Medical University from January 2017 to December 2019 were retrospectively analyzed. The region of interest was drawn covering the lesion at the section showing the maximum area of the lesion on the ADC map using ImageJ software. The mean, standard deviation, skewness, kurtosis, 10th percentile (n P10), and 90th percentile (n P90) of ADC value were obtained by SPSS statistical software for statistical analysis. The differences of histogram indexes were compared between the two groups. The ROC of each index of ADC value was obtained, and the area under the curve (AUC) of each index was obtained. Logistic regression analysis was used to investigate the best independent diagnostic index of the two groups.n Results:Significant differences in the mean, standard deviation, skewness, n P10, and n P90 of ADC value were observed between the cholesteatoma group and the otitis media group (n P<0.01). The AUCs of the mean, standard deviation, deviation,n P10, and n P90 in differentiating the two groups were 0.912, 0.988, 0.809, 0.793, and 0.957, respectively, in which the standard deviation of ADC value had the largest AUC. The sensitivity and specificity of the standard deviation of ADC were 95.31% and 96.43%, respectively. The results of logistic regression analysis showed that the standard deviation of ADC value (n OR=1.034, 95% confidence interval from 1.020 to 1.048, n P<0.01) was the best independent diagnostic indicator in distinguishing the cholesteatoma group from the otitis media group.n Conclusions:The histogram analysis based on ADC value can improve the ability of discriminating middle ear cholesteatoma from chronic suppurative otitis media, in which the standard deviation of ADC value is the best indicator.