
浏览全部资源
扫码关注微信
1.中国科学技术大学生物医学工程学院(苏州)/生命科学与医学部,合肥 230026
2.中国科学院苏州生物医学工程技术研究所/生医工人工智能研究室,苏州 215163
3.中山大学肿瘤防治中心影像科,广州 510060
沈菁菁,Email:jjshen@mail.ustc.edu.cn,研究方向:肺癌智能影像计算。
高 欣,Email:xingaosam@163.com。
收稿日期:2025-04-16,
网络出版日期:2025-09-03,
移动端阅览
沈菁菁,张睿,李琼等.基于多序列光谱CT影像组学与深度学习的肺癌EGFR突变预测[J].重庆医科大学学报,
Shen Jingjing,Zhang Rui,Li Qiong,et al.Prediction of EGFR mutations in lung cancer based on multi-sequence spectral CT using radiomics and deep learning[J].Journal of Chongqing Medical University,
沈菁菁,张睿,李琼等.基于多序列光谱CT影像组学与深度学习的肺癌EGFR突变预测[J].重庆医科大学学报, DOI:10.13406/j.cnki.cyxb..
Shen Jingjing,Zhang Rui,Li Qiong,et al.Prediction of EGFR mutations in lung cancer based on multi-sequence spectral CT using radiomics and deep learning[J].Journal of Chongqing Medical University, DOI:10.13406/j.cnki.cyxb..
目的
2
探究多序列光谱计算机断层扫描(computed tomography,CT)对于术前无创预测肺癌患者表皮生长因子受体(epidermal growth factor receptor,EGFR)突变状态的价值。
方法
2
回顾性收集2021年4月至2022年5月通过光谱CT增强扫描的236例肺癌患者的数据,按照8∶2的比例,随机划分训练集(188例)和验证集(48例)。基于常规CT图像及光谱CT的6种序列图像提取并筛选肿瘤的影像组学和深度学习特征,采用最小绝对收缩和选择算子以及决策融合策略建立了多个预测模型,并通过融合临床特征构建了临床-影像模型。采用受试者工作特征曲线评价模型预测效能,DeLong检验比较曲线下面积(area under the curve,AUC)。
结果
2
光谱CT模型(AUC=0.801)相比常规CT模型(AUC=0.705)的预测效能有显著提升(
P
=0.034)。在所有构建的模型中,临床-影像模型(AUC=0.804)预测效能最高。
结论
2
相比常规CT,多序列的光谱CT在EGFR突变预测方面更具优势,其联合常规CT和临床特征构建的模型,有望为无创预测EGFR突变提供新思路,提升诊断效率。
Objective
2
To explore the value of multi-sequence spectral computed tomography(CT) in non-invasive preoperative prediction of epidermal growth factor receptor(EGFR) mutation status in patients with lung cancer.
Methods
2
The data of 236 patients with lung cancer who underwent enhanced spectral CT scan from April 2021 to May 2022 were retrospectively collected,and the patients were randomly divided into a training set(
n
=188) and a validation set(
n
=48) at a ratio of 8∶2. Radiomic and deep learning features of tumors were extracted and selected based on conventional CT images and six sequences of spectral CT images. Multiple prediction models were established using the least absolute shrinkage and selection operator and decision fusion strategy,and the clinical-imaging model was constructed by combining clinical features. Receiver operating characteristic curve was used to evaluate the prediction performance of each model. DeLong test was used to compare the differences in area under the curve(AUC).
Results
2
The prediction performance of the spectral CT model(AUC=0.801) was significantly improved compared with the conventional CT model(AUC=0.705)(
P
=0.034). Among all the constructed models,the clinical-imaging model(AUC=0.804) had the highest prediction performance.
Conclusion
2
Compared with conventional CT,multi-sequence spectral CT has an advantage in EGFR mutation prediction. The model constructed by combining multi-sequence spectral CT with conventional CT and clinical features is expected to provide a new idea for non-invasive prediction of EGFR mutations and improve diagnostic efficiency.
Han BF , Zheng RS , Zeng HM , et al . Cancer incidence and mortality in China,2022 [J ] . J Natl Cancer Cent , 2024 , 4 ( 1 ): 47 - 53 .
Vaid AK , Gupta A , Momi G . Overall survival in stage IV EGFR mutation-positive NSCLC:Comparing first-,second-and third-generation EGFR-TKIs(Review) [J ] . Int J Oncol , 2021 , 58 ( 2 ): 171 - 184 .
Bodalal Z , Trebeschi S , Nguyen-Kim TDL , et al . Radiogenomics:bridging imaging and genomics [J ] . Abdom Radiol (NY) , 2019 , 44 ( 6 ): 1960 - 1984 .
Jia TY , Xiong JF , Li XY , et al . Identifying EGFR mutations in lung adenocarcinoma by noninvasive imaging using radiomics features and random forest modeling [J ] . Eur Radiol , 2019 , 29 ( 9 ): 4742 - 4750 .
Morgado J , Pereira T , Silva F , et al . Machine learning and feature selection methods for EGFR mutation status prediction in lung cancer [J ] . Appl Sci , 2021 , 11 ( 7 ): 3273 .
Shiri I , Maleki H , Hajianfar G , et al . Next-generation radiogenomics sequencing for prediction of EGFR and KRAS mutation status in NSCLC patients using multimodal imaging and machine learning algorithms [J ] . Mol Imaging Biol , 2020 , 22 ( 4 ): 1132 - 1148 .
Moreno S , Bonfante M , Zurek E , et al . A radiogenomics ensemble to predict EGFR and KRAS mutations in NSCLC [J ] . Tomography , 2021 , 7 ( 2 ): 154 - 168 .
黄栎有 , 徐 璐 , 温林春 , 等 . CT影像组学模型及深度学习技术预测肺腺癌EGFR突变 [J ] . 放射学实践 , 2022 , 37 ( 8 ): 971 - 976 .
Huang LY , Xu L , Wen LC , et al . Value of CT radiomics models and deep learning technique in predicting EGFR mutation of lung adenocarcinoma [J ] . Radiol Pract , 2022 , 37 ( 8 ): 971 - 976 .
中华放射学杂志双层探测器光谱CT临床应用协作组 . 双层探测器光谱CT临床应用中国专家共识(第一版) [J ] . 中华放射学杂志 , 2020 , 54 ( 7 ): 635 - 643 .
Chinese Journal of Radiology Cooperative Group of Clinical Application of Dual-layer Spectral Detector CT . China expert consensus on clinical application of dual-layer spectral detector CT [J ] . Chin J Radiol , 2020 , 54 ( 7 ): 635 - 643 .
任海燕 , 甄艳华 , 郑加贺 . 双层光谱探测器CT临床应用与研究进展 [J ] . 中国医学影像技术 , 2020 , 36 ( 10 ): 1555 - 1558 .
Ren HY , Zhen YH , Zheng JH . Clinical applications and progresses of dual-layer spectral detector CT [J ] . Chin J Med Imag Technol , 2020 , 36 ( 10 ): 1555 - 1558 .
余 烨 , 林了一 , 华小兰 , 等 . 肺浸润性腺癌能谱CT定量参数与EGFR基因突变相关性研究 [J ] . 中国医学计算机成像杂志 , 2020 , 26 ( 3 ): 242 - 246 .
Yu Y , Lin LY , Hua XL , et al . The correlation study of spectral CT and epidermal growth factor receptor mutation in invasive pulmonary adenocarcinoma [J ] . Chin Comput Med Imag , 2020 , 26 ( 3 ): 242 - 246 .
Zhang GJ , Cao YT , Zhang J , et al . Epidermal growth factor receptor mutations in lung adenocarcinoma:associations between dual-energy spectral CT measurements and histologic results [J ] . J Cancer Res Clin Oncol , 2021 , 147 ( 4 ): 1169 - 1178 .
Li FL , Qi LL , Cheng SN , et al . Predicting epidermal growth factor receptor mutations in non-small cell lung cancer through dual-layer spectral CT:a prospective study [J ] . Insights Imaging , 2024 , 15 ( 1 ): 109 .
Huang XM , Sun YL , Tan MY , et al . Three-dimensional convolutional neural network-based prediction of epidermal growth factor receptor expression status in patients with non-small cell lung cancer [J ] . Front Oncol , 2022 , 12 : 772770 .
Isensee F , Jaeger PF , Kohl SAA , et al . nnU-Net:a self-configuring method for deep learning-based biomedical image segmentation [J ] . Nat Methods , 2021 , 18 ( 2 ): 203 - 211 .
朱 标 , 敖红艳 , 洪 琪 , 等 . 能谱CT定量和多参数磁共振成像对肺腺癌EGFR突变相关性的诊断价值 [J ] . 中国现代医生 , 2024 , 62 ( 15 ): 74 - 78 .
Zhu B , Ao HY , Hong Q , et al . Diagnostic value of quantitative energy spectrum CT and multi-parameter magnetic resonance imaging on EGFR mutation in lung adenocarcinoma [J ] . China Mod Dr , 2024 , 62 ( 15 ): 74 - 78 .
Lin LY , Cheng JJ , Tang DQ , et al . The associations among quantitative spectral CT parameters,Ki-67 expression levels and EGFR mutation status in NSCLC [J ] . Sci Rep , 2020 , 10 ( 1 ): 3436 .
周建忠 , 付雪林 , 邹红烨 , 等 . 基于静脉期双能CT的影像组学模型预测肺腺癌EGFR基因突变状态 [J ] . 临床放射学杂志 , 2021 , 40 ( 8 ): 1516 - 1520 .
Zhou JZ , Fu XL , Zou HY , et al . Radiomics analysis of venous phase dual-energy CT imaging for predicting EGFR mutation status in lung adenocarcinoma [J ] . J Clin Radiol , 2021 , 40 ( 8 ): 1516 - 1520 .
Ma JW , Jiang X , Wang YM , et al . Dual-energy CT-based radiomics in predicting EGFR mutation status non-invasively in lung adenocarcinoma [J ] . Heliyon , 2024 , 10 ( 2 ): e24372 .
Neri S , Menju T , Sowa T , et al . Prognostic impact of microscopic vessel invasion and visceral pleural invasion and their correlations with epithelial-mesenchymal transition,cancer stemness,and treatment failure in lung adenocarcinoma [J ] . Lung Cancer , 2019 , 128 : 13 - 19 .
Cheng ZH , Shan F , Yang YS , et al . CT characteristics of non-small cell lung cancer with epidermal growth factor receptor mutation:a systematic review and meta-analysis [J ] . BMC Med Imaging , 2017 , 17 ( 1 ): 5 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621