利用扩散MRI水限制模式的影像组学分析鉴别良性肝硬化结节与肝细胞癌
本研究基于2中心扩散MRI影像组学,纳入328例患者共517个LI-RADS 2-5结节(242个良性、275个HCC),提取1434个特征并筛选16个关键特征构建分类器。支持向量机在测试集鉴别良性结节与HCC的准确率0.92、敏感度0.94、特异度0.86;作者建议在多中心数据集进一步评估以解决 harmonization 挑战。
要点
- 扩散MRI影像组学支持向量机模型鉴别良性结节与HCC,测试集准确率0.92、敏感度0.94、特异度0.86。
- 良性结节依据LR-2、3的随访影像判定,HCC依据LR-4、5的病理结果判定。
- 作者建议在多中心数据集进一步评估以解决 harmonization 挑战。
核心结论
- 基于病灶扩散特性的支持向量机模型在鉴别良性结节与HCC中表现良好,作者认为具有临床潜力,并建议在多中心数据集进一步评估以解决 harmonization 挑战。
研究设计与样本
- 回顾性研究,纳入328例患者共517个LI-RADS 2-5结节(242个良性、275个HCC),MRI来自3 T和1.5 T设备,按80%训练集和20%测试集随机划分,预处理包括重采样和归一化,采用ComBat harmonization校正扫描仪变异,经高相关滤波、PCA滤波和LASSO筛选特征,10折交叉验证选择最佳分类器并在测试集评估准确率。
诊断标准与关键数值
- 从扩散图像病灶感兴趣体积提取1434个特征,基于筛选出的16个最重要特征构建分类器;支持向量机在测试集鉴别良性结节与HCC的准确率0.92、敏感度0.94、特异度0.86。
局限性
- 作者建议在多中心数据集进一步评估以解决 harmonization 挑战。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
OBJECTIVES: Current study aimed to investigate radiomics features derived from 2-centre diffusion-MRI to differentiate benign and hepatocellular carcinoma (HCC) liver nodules.
METHODS: A total of 328 patients with 517 Liver Imaging Reporting and Data System (LI-RADS) 2-5 nodules were included. MR images were retrospectively collected from 3 T and 1.5 T MRI vendors. Lesions were categorized into 242 benign and 275 HCC based on follow-up imaging for LR-2,3 and pathology results for LR-4,5 nodules and randomly divided into training (80%) and test (20%) sets. Preprocessing included resampling and normalization. Radiomics features were extracted from lesion volume-of-interest (VOI) on diffusion images. Scanner variability was corrected using ComBat harmonization method followed by high-correlation filter, PCA filter, and LASSO to select important features. The best classifier model was selected by 10-fold cross-validation, and accuracy was assessed on the test dataset.
RESULTS: In total, 1434 features were extracted, and subsequent classifiers were constructed based on the 16 most important selected features. Notably, support-vector machine (SVM) demonstrated better performance in the test dataset in distinguishing between benign and HCC nodules, achieving an accuracy of 0.92, sensitivity of 0.94, and specificity of 0.86.
CONCLUSIONS: Utilizing diffusion-MRI radiomics, our study highlights the performance of SVM, trained on lesions' diffusivity characteristics, in distinguishing benign and HCC nodules, ensuring clinical potential. It is suggested that further evaluations be conducted on multicentre datasets to address harmonization challenges.
ADVANCES IN KNOWLEDGE: Integration of diffusion radiomics for monitoring water restriction patterns as tumour histopathological index, with machine learning models demonstrates potential for achieving a reliable noninvasive method to improve the current diagnostic criteria.
引用
- Arian A, Fotouhi M, Samadi Khoshe Mehr F, et al. Differentiating benign and hepatocellular carcinoma cirrhotic nodules: radiomics analysis of water restriction patterns with diffusion MRI. Br J Radiol. 2025. PMID: 40419462. doi:10.1093/bjr/tqaf106
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。