文献速递消化道肿瘤放疗放疗物理与技术

基于小波图像特征预测MRI引导自适应放疗中每日在线自适应再计划的必要性

Predicting necessity of daily online adaptive replanning based on wavelet image features for MRI guided adaptive radiation therapy.
PubMed 文献 · Radiother Oncol · 2022-10-07

该研究针对MRI引导自适应放疗(MRgART)中在线自适应再计划(OLAR)耗时费力的问题,利用24例胰腺癌患者共119次每日MRI的小波多尺度纹理特征构建机器学习分类器预测OLAR必要性。最佳三特征组合模型预测OLAR必要性的交叉验证AUC为0.98。

要点

  • 基于每日MRI小波特征构建的贝叶斯分类器可预测胰腺癌MRgART中OLAR的必要性。
  • 最佳模型为3特征组合,预测OLAR必要性的CV-AUC达0.98。
  • 该方法旨在自动客观判断是否需要OLAR,避免不必要的操作。

核心结论

  • 基于胰腺癌每日MRI提取的小波特征构建的机器学习分类器模型,可自动客观地判断某个治疗分次是否需要OLAR,从而避免MRgART中不必要的操作。

研究设计与样本

  • 研究开发了一种机器学习分类器,基于每日MRI提取的小波多尺度纹理特征预测OLAR必要性。
  • 模型训练与测试使用24例胰腺癌患者在1.5 T MR-Linac上接受MRgART期间采集的119个每日MRI数据集。
  • 采用Spearman相关性、组内相关系数(ICC)、变异系数(COV)、t检验(p < 0.05)、自组织映射(SOM)和最大稳定极值区域(MSER)算法筛选候选特征,并用贝叶斯分类器构建预测模型。
  • 模型性能以ROC曲线的AUC判定。

主要发现

  • Spearman相关性识别出123个非冗余特征(r < 0.9)。
  • 其中82个特征在重定位方面显示高ICC > 0.6,67个特征对OLAR的COV大于9%。
  • 38个特征通过t检验,其中25个通过SOM,12个通过MSER,最终12个特征用于构建分类器模型。
  • 每次使用2-3个特征组合构建分类器模型,最佳性能模型为3特征组合,预测OLAR必要性的CV-AUC为0.98。

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)。

原文

PURPOSE: Online adaptive replanning (OLAR) is generally labor-intensive and time-consuming during MRI-guided adaptive radiation therapy (MRgART). This work aims to develop a method to determine OLAR necessity during MRgART. METHODS: A machine learning classifier was developed to predict OLAR necessity based on wavelet multiscale texture features extracted from daily MRIs and was trained and tested with data from 119 daily MRI datasets acquired during MRgART for 24 pancreatic cancer patients treated on a 1.5 T MR-Linac. Spearman correlations, interclass correlation (ICC), coefficient of variance (COV), t-test (p < 0.05), self-organized map (SOM) and maximum stable extremal region (MSER) algorithm were used to determine candidate features, which were used to build the prediction models using Bayesian classifiers. The model performance was judged using the AUC of the ROC curve. RESULTS: Spearman correlation identified 123 features that were not redundant (r < 0.9). Of them 82 showed high ICC for repositioning > 0.6, 67 had a COV greater than 9% for OLAR. Among the 38 features passed the t-test, 25 passed the SOM and 12 passed the MSER. These final 12 features were used to build the classifier model. The combination of 2-3 features at a time was used to build the classifier models. The best performing model was a 3-feature combination, which can predict OLAR necessity with a CV-AUC of 0.98. CONCLUSIONS: A machine learning classifier model based on the wavelet features extracted from daily MRI for pancreatic cancer was developed to automatically and objectively determine if OLAR is necessary for a treatment fraction avoiding unnecessary effort during MRgART.

来源:Radiother Oncol,查看原文。版权归原作者所有。

引用

  1. Nasief HG, Parchur AK, Omari E, et al. Predicting necessity of daily online adaptive replanning based on wavelet image features for MRI guided adaptive radiation therapy. Radiother Oncol. 2022. PMID: 36216299. doi:10.1016/j.radonc.2022.10.001

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。