| 魏守奕,李欣颖,张维,等.基于深度学习算法的腹盆部CT辐射剂量自动评估的可行性研究[J].中华放射医学与防护杂志,2024,44(8):699-703.Wei Shouyi,Li Xinying,Zhang Wei,et al.Feasibility study of automatic assessment of abdominal and pelvic CT radiation dose based on deep learning algorithm[J].Chin J Radiol Med Prot,2024,44(8):699-703 |
| 基于深度学习算法的腹盆部CT辐射剂量自动评估的可行性研究 |
| Feasibility study of automatic assessment of abdominal and pelvic CT radiation dose based on deep learning algorithm |
| 投稿时间：2023-10-24 |
| DOI：10.3760/cma.j.cn112271-20231024-00136 |
| 中文关键词: 辐射剂量 容积CT剂量指数 分割 回归 深度学习 |
| 英文关键词:Radiation dose CTDIvol Segmentation Regression Deep learning |
| 基金项目: |
|
| 摘要点击次数: 2603 |
| 全文下载次数: 1041 |
| 中文摘要: |
| 目的 探讨深度学习模型为基础的腹盆部CT辐射剂量指数自动评估的可行性。方法 回顾性分析2021年2月至2022年2月连续采集的临床腹盆部CT数据,共有1 084例患者图像,成像设备为西门子SOMATOM Defination Flash CT、飞利浦iCT、通用电气 lightspeed VCT。容积CT剂量指数(CTDIvol)预测模型由器官分割和剂量预测两个功能模块组成。以腹盆部位实际扫描区域分割结果为基础,通过剂量回归预测模块对CTDIvol进行自动评估。将纳入研究的1 084例患者图像分为训练集784例、验证集196例和测试集104例。以Dice系数为混合模型腹盆部位分割性能的评价指标,以准确个数占比和均方根对数误差(RMSLE)为CTDIvol估算模型性能的评价指标。结果 在测试集中,深度学习模型在CT腹部图像分割任务的Dice系数高达0.998,同时CTDIvol回归模型在估算辐射剂量时的RMSLE为9.41%,且估算正确占比达到92%。散点图分析显示部分CTDIvol估算值存在较大误差,提示模型在这些情况下可能需要进一步优化。结论 深度学习模型可准确自动分割CT腹部图像并估算辐射剂量,可用于临床辐射剂量的监测与管理。 |
| 英文摘要: |
| Objective To explore the feasibility of automatic assessment of abdominal and pelvic CT radiation dose index (CTDIvol) based on deep learning models. Methods A retrospective analysis was conducted on clinical abdominal and pelvic CT data collected continuously from February 2021 to February 2022. A total of 1 084 sets of patient images were obtained using equipment of Siemens SOMATOM Definition Flash CT, Philips iCT, and GE lightspeed VCT. The volume CT dose index (CTDIvol) prediction model consisted of two functional modules: organ segmentation and dose prediction. Based on the result of actual scanning area segmentation in the abdominal and pelvic area, CTDIvol was evaluated automatically by dose regression prediction module. The images of 1 084 patients included in the study were randomly divided into a training set of 784, a validation set of 196 and a test set of 104. Dice coefficient was used to evaluate the abdominal and pelvic segmentation performance of the hybrid model, and accurate number proportion and root-mean-square logarithm error (RMSLE) were used as the evaluation index of the CTDIvol estimation model performance. Results In the test set, the Dice coefficient of the deep learning model in the task of CT abdominal image segmentation was as high as 0.998, and the RMSLE of the CTDIvol regression model in estimation of radiation dose was 9.41%, with an accuracy rate of 92%. Scatter plot analysis showed that some CTDIvol estimates had significant errors, indicating that the model might need to be further optimized in these situations. Conclusions The deep learning models can accurately and automatically segment CT abdominal images and estimate radiation dose, which can be used for clinical radiation dose monitoring and management. |
| HTML 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|