剂量融合深度集成学习改善脑转移瘤立体定向放射外科结局预测
该研究纳入82例患者的114个脑转移瘤(其中26个经活检证实发生局部失败),开发融合立体定向放射外科空间剂量分布、T1增强MRI、计划靶区分割和临床特征的深度集成学习模型,预测SRS后局部控制。模型AUCROC为0.89±0.02,灵敏度和特异度分别为0.82±0.05和0.84±0.11,优于仅用MRI编码器、仅用MRI深度集成及不含临床特征的三通道集成模型。
要点
- 融合剂量分布、T1-CE MRI、PTV分割与临床特征的深度集成模型预测SRS后BM局部失败,AUCROC 0.89±0.02。
- 该模型灵敏度0.82±0.05、特异度0.84±0.11、准确率0.84±0.08,优于仅MRI输入的VGG-19编码器及仅MRI深度集成。
- 不含临床特征的三通道集成模型AUCROC 0.84±0.03,低于含临床特征的集成模型,提示临床变量融合可能有益。
核心结论
- 借助球面图像投影方法,融合剂量分布与临床变量的深度集成模型在预测脑转移瘤SRS后局部失败方面表现出优异性能。
- 作者认为这一新方法可能改进其他放射治疗结局模型,并需要进一步评估。
研究设计与样本
- 研究纳入82例患者的114个脑转移瘤,其中26个经活检证实发生SRS后局部失败。
- 将每个BM的SRS空间剂量分布配准至计划T1增强MRI,提取与BM中心相交的轴位切片,视野由剂量分布中60%等剂量体积确定。
- 采用球面投影将平面图像内容转换至球面,生成T1-CE、剂量分布和PTV分割的三通道投影变量。
- 使用4个VGG-19深度编码器构建集成模型,每个子模型采用不同球面投影公式输入,并将位置编码后的临床特征与VGG-19深度特征融合生成logit结果。
- 通过逻辑回归综合4个子模型结果,共训练10个随机验证样本分配的模型版本以研究模型稳健性。
主要发现
- 集成模型AUCROC为0.89±0.02,灵敏度0.82±0.05,特异度0.84±0.11,准确率0.84±0.08。
- 集成模型优于仅MRI的VGG-19编码器(灵敏度0.35±0.01,AUCROC 0.64±0.08)。
- 集成模型优于仅MRI的深度集成模型(灵敏度0.60±0.09,AUCROC 0.68±0.06)。
- 集成模型优于不含临床特征融合的三通道集成模型(灵敏度0.78±0.08,AUCROC 0.84±0.03)。
临床怎么用
- 原文未据此给出推荐。
局限性
- 作者在结论中指出,该新方法需要进一步评估。
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: To develop a novel deep ensemble learning model for accurate prediction of brain metastasis (BM) local control outcomes after stereotactic radiosurgery (SRS). METHODS AND MATERIALS: A total of 114 brain metastases (BMs) from 82 patients were evaluated, including 26 BMs that developed biopsy-confirmed local failure post-SRS. The SRS spatial dose distribution (Dmap) of each BM was registered to the planning contrast-enhanced T1 (T1-CE) magnetic resonance imaging (MRI). Axial slices of the Dmap, T1-CE, and planning target volume (PTV) segmentation (PTVseg) intersecting the BM center were extracted within a fixed field of view determined by the 60% isodose volume in Dmap. A spherical projection was implemented to transform planar image content onto a spherical surface using multiple projection centers, and the resultant T1-CE/Dmap/PTVseg projections were stacked as a 3-channel variable. Four Visual Geometry Group (VGG-19) deep encoders were used in an ensemble design, with each submodel using a different spherical projection formula as input for BM outcome prediction. In each submodel, clinical features after positional encoding were fused with VGG-19 deep features to generate logit results. The ensemble's outcome was synthesized from the 4 submodel results via logistic regression. In total, 10 model versions with random validation sample assignments were trained to study model robustness. Performance was compared with (1) a single VGG-19 encoder, (2) an ensemble with a T1-CE MRI as the sole image input after projections, and (3) an ensemble with the same image input design without clinical feature inclusion. RESULTS: The ensemble model achieved an excellent area under the receiver operating characteristic curve (AUCROC: 0.89 ± 0.02) with high sensitivity (0.82 ± 0.05), specificity (0.84 ± 0.11), and accuracy (0.84 ± 0.08) results. This outperformed the MRI-only VGG-19 encoder (sensitivity: 0.35 ± 0.01, AUCROC: 0.64 ± 0.08), the MRI-only deep ensemble (sensitivity: 0.60 ± 0.09, AUCROC: 0.68 ± 0.06), and the 3-channel ensemble without clinical feature fusion (sensitivity: 0.78 ± 0.08, AUCROC: 0.84 ± 0.03). CONCLUSIONS: Facilitated by the spherical image projection method, a deep ensemble model incorporating Dmap and clinical variables demonstrated excellent performance in predicting BM post-SRS local failure. Our novel approach could improve other radiation therapy outcome models and warrants further evaluation.
引用
- Zhao J, Vaios E, Wang Y, et al. Dose-Incorporated Deep Ensemble Learning for Improving Brain Metastasis Stereotactic Radiosurgery Outcome Prediction. Int J Radiat Oncol Biol Phys. 2024. PMID: 38615888. doi:10.1016/j.ijrobp.2024.04.006
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。