《计算机应用》唯一官方网站 ›› 2026, Vol. 46 ›› Issue (2): 640-651.DOI: 10.11772/j.issn.1001-9081.2025020220
• 前沿与综合应用 • 上一篇
收稿日期:2025-03-06
修回日期:2025-05-18
接受日期:2025-05-27
发布日期:2025-08-08
出版日期:2026-02-10
通讯作者:
邱兆文
作者简介:崔克俭(2001—),男,辽宁盘锦人,硕士研究生,CCF会员,主要研究方向:计算机视觉、医疗影像处理、深度学习基金资助:
Kejian CUI, Zhiming WANG, Zhaowen QIU(
)
Received:2025-03-06
Revised:2025-05-18
Accepted:2025-05-27
Online:2025-08-08
Published:2026-02-10
Contact:
Zhaowen QIU
About author:CUI Kejian, born in 2001, M. S. candidate. His research interests include computer vision, medical image processing, deep learning.Supported by:摘要:
针对现有的视网膜分割模型在三维特征提取时易丢失血管拓扑信息导致分支断裂、二维分割结果结构连续性不足以及血管分析与疾病预测跨模态关联缺失的三大问题,提出协同框架MA_DNet(Multi-scale topology-Aware Disease Network)。该框架由增强分割模型MA_Net+与疾病预测模块组成。MA_Net+是在所提模型MA_Net(Multi-scale topology-Aware Network)的基础上,通过一个中间特征再训练模块细化并连接断裂的血管分支,从而优化拓扑完整性。首先,通过三维模块GMSF(Gated Multi-Scale Fusion)模块提取多尺度空间卷积和融合复杂分支特征,并结合ResMamba模块建模血管的长程拓扑依赖关系以增强三维特征表征,进而抑制分割结果中拓扑断裂的发生;其次,利用二维模块MA_Net+的卷积层进一步优化局部血管结构的连续性;最后,设计级联预测模块,并将形态学参数与临床指标相结合,从而构建影像特征与冠心病(CAD)风险的跨模态关联。实验结果表明,MA_Net+框架在OCTA-500公开数据集的一个数据子集上的Dice系数和Jaccard指数分别达到了93.02%和87.04%,较IPN-V2+(Image Projection Network V2+)模型分别提升了0.28和0.37个百分点;在OCTA-500的另一子集上MA_Net+框架的两个指标分别为89.76%和81.52%,较IPN-V2+模型分别提升了0.35和0.57个百分点;MA_Net+框架的疾病预测模块在某个私有数据集上的AUC(Area Under Curve)达到了86.23%。可见,MA_DNet框架通过三维拓扑建模与多尺度融合机制,能够有效提升血管分割的连续性;同时,该框架还探索了视网膜影像与CAD风险之间的跨模态关联预测,为无创心血管诊断提供了新方案。
中图分类号:
崔克俭, 王志明, 邱兆文. 基于光学相干断层扫描血管成像的视网膜血管分割与冠心病预测方法[J]. 计算机应用, 2026, 46(2): 640-651.
Kejian CUI, Zhiming WANG, Zhaowen QIU. Method for retinal vessel segmentation and coronary artery disease prediction using optical coherence tomography angiography[J]. Journal of Computer Applications, 2026, 46(2): 640-651.
| 子集 | MA_Net | MA_Net+ | |||||||
|---|---|---|---|---|---|---|---|---|---|
| 输入体积 | 分割块大小 | 批量大小 | 迭代次数 | 模型保存频率 | 输入图片大小 | 批量大小 | 迭代次数 | 模型保存频率 | |
| OCTA_3M | 304×304×640 | 76×76×160 | 3 | 25 000 | 300∶1 | 304×304 | 2 | 3 000 | 50∶1 |
| OCTA_6M | 400×600×640 | 100×100×160 | 3 | 25 000 | 300∶1 | 400×400 | 2 | 3 000 | 50∶1 |
表 1 MA_Net和MA_Net+在两个子数据集上的网络配置
Tab. 1 Network configuration of MA_Net and MA_Net+ on two subsets
| 子集 | MA_Net | MA_Net+ | |||||||
|---|---|---|---|---|---|---|---|---|---|
| 输入体积 | 分割块大小 | 批量大小 | 迭代次数 | 模型保存频率 | 输入图片大小 | 批量大小 | 迭代次数 | 模型保存频率 | |
| OCTA_3M | 304×304×640 | 76×76×160 | 3 | 25 000 | 300∶1 | 304×304 | 2 | 3 000 | 50∶1 |
| OCTA_6M | 400×600×640 | 100×100×160 | 3 | 25 000 | 300∶1 | 400×400 | 2 | 3 000 | 50∶1 |
| 基线 | GMSF | ResMamba | Dice | Jac | Bacc | Pre | Rec |
|---|---|---|---|---|---|---|---|
| ✓ | 91.71 | 84.80 | 95.36 | 92.29 | 89.43 | ||
| ✓ | ✓ | 90.55 | 82.88 | 94.72 | 91.27 | 90.07 | |
| ✓ | ✓ | 92.05 | 85.38 | 95.07 | 93.70 | 90.58 | |
| ✓ | ✓ | ✓ | 92.77 | 86.60 | 95.67 | 93.89 | 91.78 |
表2 OCTA_3M子集上的消融实验结果 (%)
Tab. 2 Ablation experiment results on OCTA_3M subset
| 基线 | GMSF | ResMamba | Dice | Jac | Bacc | Pre | Rec |
|---|---|---|---|---|---|---|---|
| ✓ | 91.71 | 84.80 | 95.36 | 92.29 | 89.43 | ||
| ✓ | ✓ | 90.55 | 82.88 | 94.72 | 91.27 | 90.07 | |
| ✓ | ✓ | 92.05 | 85.38 | 95.07 | 93.70 | 90.58 | |
| ✓ | ✓ | ✓ | 92.77 | 86.60 | 95.67 | 93.89 | 91.78 |
| 基线 | GMSF | ResMamba | Dice | Jac | Bacc | Pre | Rec |
|---|---|---|---|---|---|---|---|
| ✓ | 88.19 | 78.98 | 93.22 | 88.99 | 87.57 | ||
| ✓ | ✓ | 88.82 | 80.02 | 92.09 | 93.46 | 84.80 | |
| ✓ | ✓ | 89.05 | 80.38 | 92.58 | 92.67 | 85.88 | |
| ✓ | ✓ | ✓ | 89.40 | 80.94 | 93.81 | 90.40 | 88.62 |
表3 OCTA_6M子集上的消融实验结果 (%)
Tab. 3 Ablation experiment results on OCTA_6M subset
| 基线 | GMSF | ResMamba | Dice | Jac | Bacc | Pre | Rec |
|---|---|---|---|---|---|---|---|
| ✓ | 88.19 | 78.98 | 93.22 | 88.99 | 87.57 | ||
| ✓ | ✓ | 88.82 | 80.02 | 92.09 | 93.46 | 84.80 | |
| ✓ | ✓ | 89.05 | 80.38 | 92.58 | 92.67 | 85.88 | |
| ✓ | ✓ | ✓ | 89.40 | 80.94 | 93.81 | 90.40 | 88.62 |
| 模型 | Dice | Jac | Bacc |
|---|---|---|---|
| U-Net[ | 88.55±3.43 | 79.62±5.08 | 93.31±1.68 |
| U-Net++[ | 88.61±3.07 | 79.67±4.59 | 91.93±1.74 |
| Attention U-Net[ | 88.71±3.19 | 79.84±4.74 | 93.14±1.49 |
| Joint-seg[ | 91.13±2.09 | 83.78±3.40 | — |
| IPN[ | 90.62±5.96 | 83.25±7.78 | 93.87±4.19 |
| IPN-V2[ | 92.46±3.93 | 86.19±5.83 | 95.34±3.16 |
| IPN-V2+[ | 92.74±3.95 | 86.67±5.88 | 95.22±3.12 |
| FARGO[ | 91.68±2.05 | 84.70±3.34 | — |
| MA_Net | 92.77±2.43 | 86.60±3.90 | 95.67±2.07 |
| MA_Net+ | 93.02±2.40 | 87.04±3.89 | 95.80±2.00 |
表4 所提模型与其他OCTA模型在OCTA_3M子集上的对比结果 (%)
Tab. 4 Comparison results of proposed and other OCTA models on OCTA_3M subset
| 模型 | Dice | Jac | Bacc |
|---|---|---|---|
| U-Net[ | 88.55±3.43 | 79.62±5.08 | 93.31±1.68 |
| U-Net++[ | 88.61±3.07 | 79.67±4.59 | 91.93±1.74 |
| Attention U-Net[ | 88.71±3.19 | 79.84±4.74 | 93.14±1.49 |
| Joint-seg[ | 91.13±2.09 | 83.78±3.40 | — |
| IPN[ | 90.62±5.96 | 83.25±7.78 | 93.87±4.19 |
| IPN-V2[ | 92.46±3.93 | 86.19±5.83 | 95.34±3.16 |
| IPN-V2+[ | 92.74±3.95 | 86.67±5.88 | 95.22±3.12 |
| FARGO[ | 91.68±2.05 | 84.70±3.34 | — |
| MA_Net | 92.77±2.43 | 86.60±3.90 | 95.67±2.07 |
| MA_Net+ | 93.02±2.40 | 87.04±3.89 | 95.80±2.00 |
| 模型 | Dice | Jac | Bacc |
|---|---|---|---|
| U-Net[ | 88.55±3.43 | 79.62±5.08 | 93.31±1.68 |
| U-Net++[ | 88.61±3.07 | 79.67±4.59 | 91.93±1.74 |
| Attention U-Net[ | 88.71±3.19 | 79.84±4.74 | 93.14±1.49 |
| Joint-seg[ | 89.72±3.21 | 81.17±3.11 | — |
| IPN[ | 88.64±3.21 | 79.73±4.92 | 93.07±2.42 |
| IPN-V2[ | 89.08±2.73 | 80.41±4.29 | 93.52±2.13 |
| IPN-V2+[ | 89.41±2.74 | 80.95±4.32 | 93.46±2.12 |
| PAENet[ | 89.36±2.70 | 80.43±4.15 | 94.05±1.95 |
| PAENet+[ | 89.69±2.77 | 81.42±4.39 | 93.68±2.08 |
| FARGO[ | 89.15±2.39 | 80.50±3.75 | — |
| MA_Net | 89.40±2.73 | 80.94±4.27 | 93.81±2.09 |
| MA_Net+ | 89.76±2.64 | 81.52±4.16 | 94.09±1.77 |
表5 所提模型与其他OCTA模型在OCTA_6M子集上的对比结果 (%)
Tab. 5 Comparison results of proposed and other OCTA models on OCTA_6M subset
| 模型 | Dice | Jac | Bacc |
|---|---|---|---|
| U-Net[ | 88.55±3.43 | 79.62±5.08 | 93.31±1.68 |
| U-Net++[ | 88.61±3.07 | 79.67±4.59 | 91.93±1.74 |
| Attention U-Net[ | 88.71±3.19 | 79.84±4.74 | 93.14±1.49 |
| Joint-seg[ | 89.72±3.21 | 81.17±3.11 | — |
| IPN[ | 88.64±3.21 | 79.73±4.92 | 93.07±2.42 |
| IPN-V2[ | 89.08±2.73 | 80.41±4.29 | 93.52±2.13 |
| IPN-V2+[ | 89.41±2.74 | 80.95±4.32 | 93.46±2.12 |
| PAENet[ | 89.36±2.70 | 80.43±4.15 | 94.05±1.95 |
| PAENet+[ | 89.69±2.77 | 81.42±4.39 | 93.68±2.08 |
| FARGO[ | 89.15±2.39 | 80.50±3.75 | — |
| MA_Net | 89.40±2.73 | 80.94±4.27 | 93.81±2.09 |
| MA_Net+ | 89.76±2.64 | 81.52±4.16 | 94.09±1.77 |
| 模型 | 总参数量/106 | 模型 | 总参数量/106 |
|---|---|---|---|
| IPN-V2[ | 2.25 | IPN-V2+ResMamba | 2.33 |
| IPN-V2+[ | 3.81 | MA_Net | 3.36 |
| IPN-V2+GMSF | 2.34 | MA_Net+ | 4.86 |
表6 OCTA-500数据集上现有网络的模型复杂度比较
Tab. 6 Model complexity comparison of existing networks on OCTA-500 dataset
| 模型 | 总参数量/106 | 模型 | 总参数量/106 |
|---|---|---|---|
| IPN-V2[ | 2.25 | IPN-V2+ResMamba | 2.33 |
| IPN-V2+[ | 3.81 | MA_Net | 3.36 |
| IPN-V2+GMSF | 2.34 | MA_Net+ | 4.86 |
| 指标 | 数值 | 指标 | 数值 |
|---|---|---|---|
| 准确率 | 71.93 | Bacc | 71.71 |
| AUC | 86.23 | 召回率 | 69.23 |
表7 冠心病的预测结果 (%)
Tab. 7 Prediction results of CAD
| 指标 | 数值 | 指标 | 数值 |
|---|---|---|---|
| 准确率 | 71.93 | Bacc | 71.71 |
| AUC | 86.23 | 召回率 | 69.23 |
| [1] | WINTHER S, SCHMIDT S E, MAYRHOFER T, et al. Incorporating coronary calcification into pre-test assessment of the likelihood of coronary artery disease[J]. Journal of the American College of Cardiology, 2020, 76(21): 2421-2432. |
| [2] | MASSIN P, ERGINAY A, HAOUCHINE B, et al. Retinal thickness in healthy and diabetic subjects measured using optical coherence tomography mapping software[J]. European Journal of Ophthalmology, 2002, 12(2): 102-108. |
| [3] | LAVIA C, BONNIN S, MAULE M, et al. Vessel density of superficial, intermediate, and deep capillary plexuses using optical coherence tomography angiography[J]. Retina, 2019, 39(2): 247-258. |
| [4] | VILLAPLANA-VELASCO A, PIGEYRE M, ENGELMANN J, et al. Fine-mapping of retinal vascular complexity loci identifies notch regulation as a shared mechanism with myocardial infarction outcomes[J]. Communications Biology, 2023, 6: No.523. |
| [5] | WONG T Y, ISLAM F M A, KLEIN R, et al. Retinal vascular caliber, cardiovascular risk factors, and inflammation: the Multi-Ethnic Study of Atherosclerosis (MESA)[J]. Investigative Ophthalmology and Visual Science, 2006, 47(6): 2341-2350. |
| [6] | MacGILLIVRAY T J, TRUCCO E, CAMERON J R, et al. Retinal imaging as a source of biomarkers for diagnosis, characterization and prognosis of chronic illness or long-term conditions[J]. The British Journal of Radiology, 2014, 87(1040): No.20130832. |
| [7] | WANG S B, MITCHELL P, LIEW G, et al. A spectrum of retinal vasculature measures and coronary artery disease[J]. Atherosclerosis, 2018, 268: 215-224. |
| [8] | ISTVÁN L, CZAKÓ C, ÉLŐ Á, et al. Imaging retinal microvascular manifestations of carotid artery disease in older adults: from diagnosis of ocular complications to understanding microvascular contributions to cognitive impairment[J]. GeroScience, 2021, 43(4): 1703-1723. |
| [9] | ARNOULD L, GUENANCIA C, BINQUET C, et al. Caractéristiques vasculaires rétiniennes: modifications lors du vieillissement et en pathologie vasculaire systémique (cardiaque et cérébrale)[J]. Journal Français d’Ophtalmologie, 2022, 45(1): 104-118. |
| [10] | IORGA R E, COSTIN D, MUNTEANU-DĂNULESCU R S, et al. Non-invasive retinal vessel analysis as a predictor for cardiovascular disease[J]. Journal of Personalized Medicine, 2024, 14(5): No.501. |
| [11] | LEE S J V, GOH Y Q, ROJAS-CARABALI W, et al. Association between retinal vessels caliber and systemic health: a comprehensive review[J]. Survey of Ophthalmology, 2025, 70(2): 184-199. |
| [12] | ARNOULD L, MERIAUDEAU F, GUENANCIA C, et al. Using artificial intelligence to analyse the retinal vascular network: the future of cardiovascular risk assessment based on oculomics? a narrative review[J]. Ophthalmology and Therapy, 2023, 12(2): 657-674. |
| [13] | SHIROMANI S, AlBADRI A, LINDEKE-MYERS A, et al. Reduced retinal microvascular density in women with coronary microvascular dysfunction: a pilot study[J]. American Heart Journal Plus: Cardiology Research and Practice, 2025, 51: No.100502. |
| [14] | ZEPPENFELD K, TFELT-HANSEN J, DE RIVA M, et al. 2022 ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: developed by the task force for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death of the European Society of Cardiology (ESC) Endorsed by the Association for European Paediatric and Congenital Cardiology (AEPC)[J]. European Heart Journal, 2022, 43(40): 3997-4126. |
| [15] | HU D, PAN L, CHEN X, et al. A novel vessel segmentation algorithm for pathological en-face images based on matched filter[J]. Physics in Medicine and Biology, 2023, 68(5): No.055014. |
| [16] | GHENCIU L A, DIMA M, STOICESCU E R, et al. Retinal imaging-based oculomics: artificial intelligence as a tool in the diagnosis of cardiovascular and metabolic diseases[J]. Biomedicines, 2024, 12(9): No.2150. |
| [17] | LISA GRACIA M, VIEITEZ SANTIAGO M, SALMÓN GONZALEZ Z, et al. La hipertensión arterial y el score de Framingham de riesgo vascular en la obstrucción venosa retiniana[J]. Hipertensión y Riesgo Vascular, 2019, 36(4): 193-198. |
| [18] | VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook: Curran Associates Inc., 2017: 6000-6010. |
| [19] | DEVLIN J, CHANG M W, LEE K, et al. BERT: pre-training of deep bidirectional Transformers for language understanding[C]// Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). Stroudsburg: ACL, 2019: 4171-4186. |
| [20] | XING Z, YU L, WAN L, et al. NestedFormer: nested modality-aware Transformer for brain tumor segmentation[EB/OL]. [2024-11-05].. |
| [21] | GU A, DAO T. Mamba: linear-time sequence modeling with selective state spaces[EB/OL]. [2024-10-12].. |
| [22] | LIU Y, TIAN Y, ZHAO Y, et al. VMamba: visual state space model[EB/OL]. [2024-09-30].. |
| [23] | BASAR T. A new approach to linear filtering and prediction problems[M]. [S.l.]:Wiley-IEEE Press, 2001: 167-179. |
| [24] | MA J, LI F, WANG B. U-Mamba: enhancing long-range dependency for biomedical image segmentation[EB/OL]. [2024-12-15].. |
| [25] | 孙颖,丁卫平,黄嘉爽,等. RCAR-UNet:基于粗糙通道注意力机制的视网膜血管分割网络[J]. 计算机研究与发展, 2023, 60(4): 947-961. |
| SUN Y, DING W P, HUANG J S, et al. RCAR-UNet: retinal vessels segmentation network based on rough channel attention mechanism[J]. Journal of Computer Research and Development, 2023, 60(4): 947-961. | |
| [26] | 梁礼明,黄朝林,石霏,等. 融合形状先验的水平集眼底图像血管分割[J]. 计算机学报, 2018, 41(7): 1678-1692. |
| LIANG L M, HUANG C L, SHI F, et al. Retinal vessel segmentation in fundus images using level set combined with shape priori[J]. Chinese Journal of Computers, 2018, 41(7): 1678-1692. | |
| [27] | 解立志,周明全,田沄,等. 基于区域增长与局部自适应C-V模型的脑血管分割[J]. 软件学报, 2013, 24(8): 1927-1936. |
| XIE L Z, ZHOU M Q, TIAN Y, et al. Cerebrovascular segmentation based on region growing and local adaptive C-V model[J]. Journal of Software, 2013, 24(8): 1927-1936. | |
| [28] | 梁礼明,刘博文,杨海龙,等. 基于多特征融合的有监督视网膜血管提取[J]. 计算机学报, 2018, 41(11): 2566-2580. |
| LIANG L M, LIU B W, YANG H L, et al. Supervised blood vessel extraction in retinal images based on multi-feature fusion[J]. Chinese Journal of Computers, 2018, 41(11): 2566-2580. | |
| [29] | MAIER H, FAGHIHROOHI S, NAVAB N. A line to align: deep dynamic time warping for retinal OCT segmentation[C]// Proceedings of the 2021 International Conference on Medical Image Computing and Computer-Assisted Intervention, LNCS 12901. Cham: Springer, 2021: 709-719. |
| [30] | LI M, CHEN Y, JI Z, et al. Image projection network: 3D to 2D image segmentation in OCTA images[J]. IEEE Transactions on Medical Imaging, 2020, 39(11): 3343-3354. |
| [31] | LI M, HUANG K, XU Q, et al. OCTA-500: a retinal dataset for optical coherence tomography angiography study[J]. Medical Image Analysis, 2024, 93: No.103092. |
| [32] | QUAN X, HOU G, YIN W, et al. A multi-modal and multi-stage fusion enhancement network for segmentation based on OCT and OCTA images[J]. Information Fusion, 2025, 113: No.102594. |
| [33] | YANG C, LI B, XIAO Q, et al. LA-Net: layer attention network for 3D-to-2D retinal vessel segmentation in OCTA images[J]. Physics in Medicine and Biology, 2024, 69(4): No.045019. |
| [34] | TUN Y Z, AIMMANEE P. Machine learning-based retinal neovascularization localization in en face optical coherence tomography angiography images using vessel features[J]. IEEE Access, 2025, 13: 2045-2057. |
| [35] | FRAZ M M, REMAGNINO P, HOPPE A, et al. Blood vessel segmentation methodologies in retinal images: a survey[J]. Computer Methods and Programs in Biomedicine, 2012, 108(1): 407-433. |
| [36] | BRUMMER A B, HUNT D, SAVAGE V. Improving blood vessel tortuosity measurements via highly sampled numerical integration of the Frenet-Serret equations[J]. IEEE Transactions on Medical Imaging, 2021, 40(1): 297-309. |
| [37] | CHE AZEMIN M Z, HAMID F AB, WANG J J, et al. Box-counting fractal dimension algorithm variations on retina images[C]// Advanced Computer and Communication Engineering Technology: Proceedings of ICOCOE 2015, LNEE 362. Cham: Springer, 2016: 337-343. |
| [38] | DASH J, BHOI N. Retinal blood vessel segmentation using Otsu thresholding with principal component analysis[C]// Proceedings of the 2nd International Conference on Inventive Systems and Control. Piscataway: IEEE, 2018: 933-937. |
| [39] | NAGASATO D, TABUCHI H, MASUMOTO H, et al. Automated detection of a nonperfusion area caused by retinal vein occlusion in optical coherence tomography angiography images using deep learning[J]. PLoS ONE, 2019, 14(11): No.e0223965. |
| [40] | LUISI J, LIU W, ZHANG W, et al. En-face optical coherence tomography angiography for longitudinal monitoring of retinal injury[J]. Applied Sciences, 2019, 9(13): No.2617. |
| [41] | HUANG H, LIN L, TONG R, et al. UNet 3+: a full-scale connected UNet for medical image segmentation[C]// Proceedings of the 2020 IEEE International Conference on Acoustics, Speech and Signal Processing. Piscataway: IEEE, 2020: 1055-1059. |
| [42] | RONNEBERGER O, FISCHER P, BROX T. U-Net: convolutional networks for biomedical image segmentation[C]// Proceedings of the 2015 International Conference on Medical Image Computing and Computer-Assisted Intervention, LNCS 9351. Cham: Springer, 2015: 234-241. |
| [43] | ZHOU Z, RAHMAN SIDDIQUEE M M, TAJBAKHSH N, et al. UNet++: a nested U-Net architecture for medical image segmentation[C]// Proceedings of the 2018 International Workshop on Deep Learning in Medical Image Analysis, LNCS 11045. Cham: Springer, 2018: 3-11. |
| [44] | OKTAY O, SCHLEMPER J, FOLGOC L L, et al. Attention U-Net: learning where to look for the pancreas[EB/OL]. [2024-11-28].. |
| [45] | HU K, JIANG S, ZHANG Y, et al. Joint-Seg: treat foveal avascular zone and retinal vessel segmentation in OCTA images as a joint task[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: No.4007113. |
| [46] | WU Z, WANG Z, ZOU W, et al. PAENet: a progressive attention-enhanced network for 3D to 2D retinal vessel segmentation[C]// Proceedings of the 2021 IEEE International Conference on Bioinformatics and Biomedicine. Piscataway: IEEE, 2021: 1579-1584. |
| [47] | PENG L, LIN L, CHENG P, et al. FARGO: a joint framework for FAZ and RV segmentation from OCTA images[C]// Proceedings of the 2021 International Workshop on Ophthalmic Medical Image Analysis. Cham: Springer, 2021: 42-51. |
| [48] | LIU M, LOVERN C, LYCETT K, et al. The association between markers of inflammation and retinal microvascular parameters: a systematic review and meta-analysis[J]. Atherosclerosis, 2021, 336: 12-22. |
| [1] | 韩锋, 卜永丰, 梁浩翔, 黄舒雯, 张朝阳, 孙士杰. 基于多层次时空交互依赖的车辆轨迹异常检测[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 604-612. |
| [2] | 姜皓骞, 张东, 李冠宇, 陈恒. 基于结构增强的层次化任务导向提示策略的对话推荐系统SetaCRS[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 368-377. |
| [3] | 陈晓雷, 郑芷薇, 黄雪, 曲振彬. 结合球面对齐与自适应几何校正的全景视频超分辨率网络[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 528-535. |
| [4] | 刘权捷, 顾兆一, 王春源. 复杂光照条件下的不安全驾驶行为检测[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 613-619. |
| [5] | 林金娇, 张灿舜, 陈淑娅, 王天鑫, 连剑, 徐庸辉. 基于改进图注意力网络的车险欺诈检测方法[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 437-444. |
| [6] | 李名, 王孟齐, 张爱丽, 任花, 窦育强. 基于条件生成对抗网络和混合注意力机制的图像隐写方法[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 475-484. |
| [7] | 曹杰, 谢凌锋, 王丙金, 张昌河, 余紫东, 邓超. 考虑类不平衡和背景多样性问题的青少年脊柱侧弯筛查方法[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 630-639. |
| [8] | 郭泽一, 李凤莲, 徐利春. 基于双重决策机制的深度符号回归算法[J]. 《计算机应用》唯一官方网站, 2026, 46(2): 406-415. |
| [9] | 边小勇, 袁培洋, 胡其仁. 双编码空频混合的红外小目标检测方法[J]. 《计算机应用》唯一官方网站, 2026, 46(1): 252-259. |
| [10] | 昝志辉, 王雅静, 李珂, 杨智翔, 杨光宇. 基于SAA-CNN-BiLSTM网络的多特征融合语音情感识别方法[J]. 《计算机应用》唯一官方网站, 2026, 46(1): 69-76. |
| [11] | 景攀峰, 梁宇栋, 李超伟, 郭俊茹, 郭晋育. 基于师生学习的半监督图像去雾算法[J]. 《计算机应用》唯一官方网站, 2025, 45(9): 2975-2983. |
| [12] | 李维刚, 邵佳乐, 田志强. 基于双注意力机制和多尺度融合的点云分类与分割网络[J]. 《计算机应用》唯一官方网站, 2025, 45(9): 3003-3010. |
| [13] | 许志雄, 李波, 边小勇, 胡其仁. 对抗样本嵌入注意力U型网络的3D医学图像分割[J]. 《计算机应用》唯一官方网站, 2025, 45(9): 3011-3016. |
| [14] | 张宏俊, 潘高军, 叶昊, 陆玉彬, 缪宜恒. 结合深度学习和张量分解的多源异构数据分析方法[J]. 《计算机应用》唯一官方网站, 2025, 45(9): 2838-2847. |
| [15] | 李进, 刘立群. 基于残差Swin Transformer的SAR与可见光图像融合[J]. 《计算机应用》唯一官方网站, 2025, 45(9): 2949-2956. |
| 阅读次数 | ||||||
|
全文 |
|
|||||
|
摘要 |
|
|||||