[1]陈炜,辛梓睿,陈硒,等.三维深度网络结合迁移学习的房颤患者冠脉CTA术前分类[J].中国医学物理学杂志,2025,42(9):1245-1254.[doi:DOI:10.3969/j.issn.1005-202X.2025.09.017]
　CHEN Wei,,et al.Three-dimensional deep neural network integrating transfer learning for preoperative coronary CTA classification in atrial fibrillation patients[J].Chinese Journal of Medical Physics,2025,42(9):1245-1254.[doi:DOI:10.3969/j.issn.1005-202X.2025.09.017]
点击复制

三维深度网络结合迁移学习的房颤患者冠脉CTA术前分类()
分享到：

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
42
期数:
2025年第9期
页码:
1245-1254
栏目:
医学人工智能
出版日期:
2025-09-30

文章信息/Info

Title:
Three-dimensional deep neural network integrating transfer learning for preoperative coronary CTA classification in atrial fibrillation patients
文章编号:
1005-202X（2025）09-1245-10
作者:
陈炜1; 2; 3; 4; 辛梓睿1; 3; 4; 陈硒1; 4; 刘振江1; 4; 罗爱静1; 2; 3; 4
1.中南大学湘雅二医院， 湖南 长沙 410011； 2.中南大学生命科学学院， 湖南 长沙 410013； 3.医学信息研究湖南省普通高等学校重点实验室（中南大学）， 湖南 长沙 410013； 4.湖南省心血管智能医疗临床医学研究中心， 湖南 长沙 410011
Author(s):
CHEN Wei1;  2;  3;  4;  XIN Zirui1;  3;  4;  CHEN Xi1;  4;  LIU Zhenjiang1;  4;  LUO Aijing1;  2;  3;  4
1. The Second Xiangya Hospital of Central South University, Changsha 410011, China 2. School of Life Sciences, Central South University, Changsha 410013, China 3. Key Laboratory of Medical Information Research in Hunan Province (Central South University), Changsha 410013, China 4. Clinical Medical Research Center for Cardiovascular Intelligent Medicine in Hunan Province, Changsha 410011, China
关键词:
心房颤动; 冠脉CTA; 三维深度网络; 迁移学习; 导管消融; 手术策略
Keywords:
Keywords: atrial fibrillation coronary computed tomography angiography three-dimensional deep neural network transfer learning catheter ablation surgical strategy
分类号:
R318；R541.75
DOI:
DOI:10.3969/j.issn.1005-202X.2025.09.017
文献标志码:
A
摘要:
目的：基于三维深度网络构建房颤患者冠脉CTA术前分类模型，探讨迁移学习技术对医学影像分类模型性能的影响，为导管消融提供术前决策支持，推动房颤治疗向精确化、个性化转变。方法：分别以3D ConvNet和3D-ResNet作为骨干网络，提取冠脉CTA序列的三维分类特征；利用公开预训练权重进行迁移学习，并通过混淆矩阵、准确率、AUC等指标评估模型分类效果，对比分析迁移学习模型与初始化训练模型之间的性能差异。结果：迁移学习显著改善了模型性能，与初始化训练模型相比，各个迁移学习模型的分类AUC提升9.1%~16.7%，准确率提高6.2%~23.5%，其中结合MedicalNet预训练权重的3D-ResNet18模型表现最佳，AUC达0.77，准确率为0.71。结论：结合迁移学习的三维深度网络能通过术前冠脉CTA有效识别需在肺静脉隔离基础上附加额外消融的房颤患者，这将有助于引导临床医生优化手术策略并改善治疗效果，从而降低术后远期复发率。
Abstract:
Abstract: Objective To develop a three-dimensional (3D) deep neural network based preoperative classification model for coronary computed tomography angiography (CTA) in atrial fibrillation patients, and to explore the effects of transfer learning on the performance of medical image classification models, thereby providing preoperative decision support for catheter ablation to advance atrial fibrillation treatment toward precision and personalization. Methods Utilizing 3D ConvNet and 3D ResNet as backbone network, the three-dimensional classification features were extracted from coronary CTA sequences. The publicly available pre-trained weights were used for transfer learning. The model performance was evaluated through metrics such as confusion matrix, classification accuracy, and area under the curve (AUC). A comparative analysis was also conducted to evaluate the performance differences between the transfer learning model and the initialized training model. Results Transfer learning yielded significant performance improvements over the initialized training models, attaining AUC improvement of 9.1%-16.7% and accuracy enhancement of 6.2%-23.5%. Among all models, 3D-ResNet18 model with MedicalNet pre-training weights performed the best, achieving an AUC of 0.77 and an accuracy of 0.71. Conclusion The proposed three-dimensional deep network enhanced by transfer learning can effectively identify atrial fibrillation patients requiring additional ablation besides pulmonary vein isolation through preoperative coronary CTA, which will assist clinicians in optimizing surgical strategies and improving treatment outcomes, thereby reducing long-term postoperative recurrence rates.

相似文献/References:

[1]虞康惠,江桂华,成官迅.多层螺旋CT肺静脉成像在房颤射频消融术中的应用价值[J].中国医学物理学杂志,2016,33(5):515.[doi:10.3969/j.issn.1005-202X.2016.05.017]
　[J].Chinese Journal of Medical Physics,2016,33(9):515.[doi:10.3969/j.issn.1005-202X.2016.05.017]
[2]张悦,陈真诚,梁永波,等.基于光电容积脉搏波的心房颤动识别方法[J].中国医学物理学杂志,2020,37(11):1416.[doi:DOI:10.3969/j.issn.1005-202X.2020.11.014]
　ZHANG Yue,CHEN Zhencheng,LIANG Yongbo,et al.Atrial fibrillation identification based on photoplethysmography[J].Chinese Journal of Medical Physics,2020,37(9):1416.[doi:DOI:10.3969/j.issn.1005-202X.2020.11.014]
[3]徐亮,陶倩,钟菁,等.基于卷积神经网络的房颤患者导管消融术后复发预测[J].中国医学物理学杂志,2022,39(8):1035.[doi:DOI:10.3969/j.issn.1005-202X.2022.08.021]
　XU Liang,TAO Qian,ZHONG Jing,et al.Prediction of recurrence after catheter ablation in patients with atrial fibrillation based on convolutional neural network[J].Chinese Journal of Medical Physics,2022,39(9):1035.[doi:DOI:10.3969/j.issn.1005-202X.2022.08.021]
[4]方东申,叶琪瑶,石少波,等.基于心电长时RR间期序列的心房颤动检测[J].中国医学物理学杂志,2023,40(8):1009.[doi:DOI:10.3969/j.issn.1005-202X.2023.08.014]
　FANG Dongshen,YE Qiyao,SHI Shaobo,et al.Atrial fibrillation detection based on long-term RR interval sequences of electrocardiogram[J].Chinese Journal of Medical Physics,2023,40(9):1009.[doi:DOI:10.3969/j.issn.1005-202X.2023.08.014]
[5]张瑞芳,梁永波,崔谋,等.结合多模态融合方式的脉搏波房颤识别[J].中国医学物理学杂志,2023,40(10):1260.[doi:DOI:10.3969/j.issn.1005-202X.2023.10.013]
　ZHANG Ruifang,LIANG Yongbo,,et al.Multimodal fusion approach to detect atrial fibrillation using PPG[J].Chinese Journal of Medical Physics,2023,40(9):1260.[doi:DOI:10.3969/j.issn.1005-202X.2023.10.013]

备注/Memo

备注/Memo:
【收稿日期】2025-02-25 【基金项目】湖南省科技厅项目（2024SK4005） 【作者简介】陈炜，硕士研究生，研究方向：医学人工智能，E-mail: cg2024@csu.edu.cn 【通信作者】罗爱静，博士，教授，博士生导师，研究方向：医学信息学，E-mail: luoaj@csu.edu.cn
更新日期/Last Update: 2025-09-30