文献速递消化道肿瘤放疗放疗物理与技术

DeepTOP实现精准肿瘤分割与治疗结局预测

Accurate tumor segmentation and treatment outcome prediction with DeepTOP.
PubMed 文献 · Radiother Oncol · 2023-02-21

研究基于III期多中心随机临床试验(NCT01211210)中99例直肠癌新辅助治疗患者的1889张MRI切片,开发深度学习工具DeepTOP,实现从肿瘤分割到结局预测的自动流程,预测病理完全缓解的准确率0.789、特异性0.725、敏感性0.812。

要点

  • DeepTOP整合U-Net分割模型与三层卷积神经网络预测模型,可自动完成肿瘤分割与结局预测。
  • 在直肠癌新辅助放化疗MRI数据中,肿瘤分割Dice系数0.79、IoU 0.75、切片特异性敏感度0.98。
  • 预测病理完全缓解的准确率0.789、特异性0.725、敏感性0.812,性能优于其他竞争算法。

核心结论

  • DeepTOP为临床环境中开发其他分割与预测工具提供了可操作的开放框架。
  • 基于DeepTOP的肿瘤评估可为临床决策提供参考,并促进影像标志物驱动的试验设计。

研究设计与样本

  • 使用深度学习开发DeepTOP工具,构建从肿瘤分割到结局预测的自动流程。
  • 分割模型采用U-Net编码解码结构,预测模型为三层卷积神经网络,并应用权重分布算法优化性能。
  • 训练与验证数据为III期多中心随机临床试验(NCT01211210)中99例直肠癌新辅助治疗患者的1889张MRI切片。

主要发现

  • DeepTOP在肿瘤分割中Dice系数为0.79、IoU为0.75、切片特异性敏感度为0.98。
  • DeepTOP预测放化疗后病理完全缓解的准确率为0.789、特异性为0.725、敏感性为0.812。
  • 在临床试验中通过多种设计的流程系统优化和验证DeepTOP,其性能优于其他竞争算法。

关键数值与推荐

  • 肿瘤分割性能:Dice系数0.79、IoU 0.75、切片特异性敏感度0.98。
  • 病理完全缓解预测性能:准确率0.789、特异性0.725、敏感性0.812。

临床怎么用

  • DeepTOP可避免手动标注和特征提取,利用原始MRI图像实现自动肿瘤分割与治疗结局预测。
  • 基于DeepTOP的肿瘤评估可为临床决策提供参考,并促进影像标志物驱动的试验设计。

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

BACKGROUND: Accurate outcome prediction prior to treatment can facilitate trial design and clinical decision making to achieve better treatment outcome. METHOD: We developed the DeepTOP tool with deep learning approach for region-of-interest segmentation and clinical outcome prediction using magnetic resonance imaging (MRI). DeepTOP was constructed with an automatic pipeline from tumor segmentation to outcome prediction. In DeepTOP, the segmentation model used U-Net with a codec structure, and the prediction model was built with a three-layer convolutional neural network. In addition, the weight distribution algorithm was developed and applied in the prediction model to optimize the performance of DeepTOP. RESULTS: A total of 1889 MRI slices from 99 patients in the phase III multicenter randomized clinical trial (NCT01211210) on neoadjuvant treatment for rectal cancer was used to train and validate DeepTOP. We systematically optimized and validated DeepTOP with multiple devised pipelines in the clinical trial, demonstrating a better performance than other competitive algorithms in accurate tumor segmentation (Dice coefficient: 0.79; IoU: 0.75; slice-specific sensitivity: 0.98) and predicting pathological complete response to chemo/radiotherapy (accuracy: 0.789; specificity: 0.725; and sensitivity: 0.812). DeepTOP is a deep learning tool that could avoid manual labeling and feature extraction and realize automatic tumor segmentation and treatment outcome prediction by using the original MRI images. CONCLUSION: DeepTOP is open to provide a tractable framework for the development of other segmentation and predicting tools in clinical settings. DeepTOP-based tumor assessment can provide a reference for clinical decision making and facilitate imaging marker-driven trial design.

来源:Radiother Oncol,查看原文。版权归原作者所有。

引用

  1. Li L, Xu B, Zhuang Z, et al. Accurate tumor segmentation and treatment outcome prediction with DeepTOP. Radiother Oncol. 2023. PMID: 36813177. doi:10.1016/j.radonc.2023.109550

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。