Yuan Lihua, Lu Jaming, Shu Xin, Liang Kun, Wang Cheng, Chen Jiu, Wang Zhishun
Department of Interventional Radiography, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing 210008, China.
Department of Radiography, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing 210008, China.
Diagnostics (Basel). 2025 May 1;15(9):1157. doi: 10.3390/diagnostics15091157.
: This study evaluates the diagnostic efficacy of the apparent diffusion coefficient (ADC) and T1-weighted contrast-enhanced (T1W + C) and T2-weighted (T2W) imaging modalities in differentiating vestibular schwannomas (VSs) and cerebellopontine angle meningiomas (CPAMs), aiming to optimize clinical imaging protocols for these tumors. : A retrospective analysis was conducted on 97 surgically and pathologically confirmed cases (65 VS, 32 CPAM) from Nanjing Drum Tower Hospital. Imaging features from ADC, T1W + C, and T2W sequences were extracted using medical imaging software. A support vector machine (SVM) model was trained to classify tumors based on these features, focusing on first-, second-, and third-order radiomic characteristics. : The ADC images demonstrated the highest classification efficiency, particularly with third-order features (AUC = 0.9817). The T2W images achieved the best accuracy (87.63%) using second-order features. Multimodal analysis revealed that ADC alone outperformed combinations with T1W + C or T2W sequences, suggesting limited added value from multi-sequence integration. : Diffusion-weighted imaging (DWI) sequences, particularly ADC maps, exhibit superior diagnostic utility compared to T1W + C and T2W sequences in distinguishing VS and CPAM. The findings advocate prioritizing DWI in clinical imaging workflows to enhance diagnostic accuracy and streamline protocols.
本研究评估表观扩散系数(ADC)以及T1加权增强(T1W + C)和T2加权(T2W)成像模式在鉴别前庭神经鞘瘤(VS)和桥小脑角脑膜瘤(CPAM)中的诊断效能,旨在优化这些肿瘤的临床成像方案。对南京鼓楼医院97例经手术和病理证实的病例(65例VS,32例CPAM)进行回顾性分析。使用医学成像软件提取ADC、T1W + C和T2W序列的成像特征。训练支持向量机(SVM)模型,根据这些特征对肿瘤进行分类,重点关注一阶、二阶和三阶影像组学特征。ADC图像显示出最高的分类效率,尤其是三阶特征(AUC = 0.9817)。使用二阶特征时,T2W图像的准确率最高(87.63%)。多模态分析显示,单独使用ADC优于与T1W + C或T2W序列的组合,表明多序列整合的附加值有限。在区分VS和CPAM方面,扩散加权成像(DWI)序列,尤其是ADC图,与T1W + C和T2W序列相比具有更高的诊断效用。研究结果提倡在临床成像工作流程中优先使用DWI,以提高诊断准确性并简化方案。