期刊文献+

基于胼胝体MRI影像组学在精神分裂症诊断中的应用研究

Research on the Application of Radiomics Based on Corpus Callosum MRI Imaging in the Diagnosis of Schizophrenia
下载PDF
导出
摘要 目的利用胼胝体MRI影像组学特征构建影像组学模型用于诊断精神分裂症。方法回顾性分析山西白求恩医院高分辨率T1WI矢状位图像,纳入精神分裂症患者72例,健康对照组70例。T1WI图像手动勾画整个胼胝体为ROI,提取组学特征后采用mRMR筛选10个特征,使用不同特征数运用XGBoost算法建立10个影像组学模型并进行交叉验证选出最优模型。采用ROC曲线对模型性能评估,绘制校正曲线并计算Brier score评估模型的一致性,采用决策曲线分析(DCA)进一步探讨模型的临床应用价值。结果当选取6个特征时,交叉验证集AUC值最高为0.794。影像组学模型在训练集和测试集上的AUC分别为0.867(95%CI 0.754-0.945)和0.833(95%CI 0.721-0.912)。校准曲线显示出良好的一致性。决策曲线分析也具有较高的净收益。结论基于胼胝体MRI影像组学特征构建影像组学模型具有诊断精神分裂症价值,有助于实现个体化医疗和精准医疗的目标。 Objective Constructing a radiomics model using corpus callosum MRI radiomic features for the diagnosis of schizophrenia.Methods A retrospective analysis of high-resolution T1-weighted sagittal images from Shanxi Bethune Hospital included 72 schizophrenia patients and 70 healthy controls.The entire corpus callosum was manually delineated as the ROI on T1WI images,and radiomic features were extracted.Ten features were selected using the mRMR method,and radiomics models were constructed using different numbers of features with the XGBoost algorithm and cross-validated to select the optimal model.The performance of the model was assessed using ROC curves,calibration curves were plotted,and the Brier score was calculated to evaluate the consistency of the model.Decision curve analysis(DCA)was further used to explore the clinical application value of the model.Results When six features were selected,the highest AUC value in the cross-validation set was 0.794.The model exhibited AUCs of 0.867(95%CI 0.754-0.945)and 0.833(95%CI 0.721-0.912)in the training and test sets,respectively.The calibration curve demonstrated good consistency.Decision curve analysis also indicated a high net benefit.Conclusion Constructing a radiomics model based on corpus callosum MRI radiomic features has value in diagnosing schizophrenia,contributing to the goals of personalized and precision medicine.
作者 裴栩瑶 黄忠江 陈艳军 张妍 PEI Xu-yao;HUANG Zhong-jiang;CHEN Yan-jun;ZHANG Yan(Department of Psychiatry,Shanxi Bethune Hospital,Shanxi Academy of Medical Sciences,Tongji Shanxi Hospital,Third Hospital of Shanxi Medical University,Taiyuan 030032,Shanxi Province,China;Department of Imaging,Shanxi Traditional Chinese Medical Hospital,Taiyuan 030001,Shanxi Province,China)
出处 《中国CT和MRI杂志》 2024年第11期17-19,共3页 Chinese Journal of CT and MRI
关键词 精神分裂症 MRI 影像组学 胼胝体 Schizophrenia MRI Radiomics Corpus Callosum
  • 相关文献

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部