利用治疗前MRI影像组学特征预测接受伽玛刀放射外科治疗的黑色素瘤脑转移患者的BRAF突变
本研究从220个黑色素瘤脑转移瘤的钆对比增强T1加权MRI治疗计划图像中提取1962个影像组学特征,用单层神经网络构建预测BRAF突变的模型。未使用SMOTE时模型准确率77.14%、特异度82.44%、敏感度81.85%、AUC 0.79;应用SMOTE后准确率83.1%、特异度87.07%、敏感度78.82%、AUC 0.82。作者认为该神经网络模型预测BRAF突变具有可行性,SMOTE可提升预测性能。
要点
- 基于钆增强T1加权MRI治疗计划图像提取的影像组学特征,可构建预测黑色素瘤脑转移BRAF突变的神经网络模型。
- 应用SMOTE后模型AUC为0.82、准确率83.1%、特异度87.07%、敏感度78.82%。
- 未应用SMOTE时模型AUC为0.79、准确率77.14%、特异度82.44%、敏感度81.85%。
核心结论
- 本研究显示构建高准确率神经网络模型预测BRAF突变具有可行性,SMOTE可改善预测性能,该模型有助于医师在没有基因检测的情况下获得更准确的治疗结果预期。
研究设计与样本
- 研究纳入220个肿瘤的数据,分为BRAF突变已识别组和BRAF突变未识别组,从钆对比增强T1加权MRI治疗计划图像中提取1962个影像组学特征,采用SMOTE处理数据不平衡问题,使用单层神经网络构建预测模型,评估敏感度、特异度、准确率和AUC。
诊断标准与关键数值
- 未使用SMOTE的最终评估:准确率77.14%、特异度82.44%、敏感度81.85%、AUC 0.79;应用SMOTE后:准确率83.1%、特异度87.07%、敏感度78.82%、AUC 0.82。
临床应用与阅片要点
- 作者认为该模型有助于医师在没有基因检测的情况下获得更准确的治疗结果预期;原文未据此给出具体阅片或报告建议。
局限性
- 原文未报告局限性。
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
AIM: To develop a model using radiomics features extracted from magnetic resonance imaging (MRI) images of Gamma Knife radiosurgery (GKRS) to predict the BRAF mutation in patients with melanoma brain metastases (MBM).
MATERIALS AND METHODS: Data of 220 tumours were classified into two groups. One was a group whose BRAF mutation was identified, and the other group whose BRAF mutation was not identified. We extracted 1,962 radiomics features from gadolinium contrast-enhanced T1-weighted MRI treatment-planning images. Synthetic Minority Over-sampling TEchnique (SMOTE) was performed to address the unbalanced data-related issues. A single-layer neural network (NN) was used to build predictive models with radiomics features. The sensitivity, specificity, accuracy, and the area under the curve (AUC) were evaluated to assess the model performance.
RESULTS: The prediction performance for the final evaluation without the SMOTE had an accuracy of 77.14%, a specificity of 82.44%, a sensitivity of 81.85%, and an AUC of 0.79. The application of SMOTE improved the prediction model to an accuracy of 83.1%, a specificity of 87.07%, a sensitivity of 78.82%, and an AUC of 0.82.
CONCLUSION: The current study showed the feasibility of generating a highly accurate NN model for the BRAF mutation prediction. The prediction performance improved with SMOTE. The model assists physicians to obtain more accurate expectations of the treatment outcome without a genetic test.
引用
- Kawahara D, Jensen A, Yuan J, et al. Predicting the BRAF mutation with pretreatment MRI radiomics features for melanoma brain metastases receiving Gamma Knife radiosurgery. Clin Radiol. 2023. PMID: 37690975. doi:10.1016/j.crad.2023.08.012
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。