Share：
Share this content in WeChat
X
[Chinese] [PDF] 44 1
• Reviews •
Research advances in artificial intelligence-assisted cardiac magnetic resonance imaging analysis
ZHANG Tianyue  FU Bing  YANG Zhi 

Cite this article as ZHANG T Y, FU B, YANG Z. Research advances in artificial intelligence-assisted cardiac magnetic resonance imaging analysis[J]. Chin J Magn Reson Imaging, 2026, 17(8): 177-183. DOI:10.12015/issn.1674-8034.2026.08.020.


[Abstract] Cardiac magnetic resonance (CMR) enables comprehensive assessment of cardiac structure, function, perfusion, and myocardial tissue characteristics, and plays an important role in the diagnosis, risk stratification, and therapeutic evaluation of cardiovascular diseases. However, CMR image post-processing is relatively complex, and conventional manual analysis is time-consuming, subjective, and limited in reproducibility, which restricts its further application in large-scale studies and efficient clinical practice. In recent years, artificial intelligence (AI) has been increasingly applied to CMR image analysis and has shown promising potential in automatic segmentation, functional quantification, tissue characterization, disease identification, risk prediction, dynamic spatiotemporal modeling, image reconstruction, and data augmentation. This review is organized around major CMR image analysis tasks and focuses on recent advances in AI-based automatic segmentation and functional quantification, tissue characterization and disease differentiation, risk prediction, dynamic spatiotemporal modeling, image reconstruction, and data augmentation. It also summarizes the applicable scenarios, representative findings, and current limitations of different techniques. Current AI-CMR research remains challenged by data heterogeneity, inconsistent annotation standards, insufficient external validation, limited cross-center generalizability, and inadequate interpretability of model outputs. This review aims to provide a reference for the design of AI-CMR-related studies, model optimization, and subsequent applications, with the goal of advancing CMR image analysis toward greater standardization, automation, and precision, while offering new perspectives for precision diagnosis, risk prediction, and individualized management of cardiovascular diseases.
[Keywords] artificial intelligence；cardiac magnetic resonance；radiomics；deep learning；image segmentation；risk prediction

ZHANG Tianyue   FU Bing   YANG Zhi*  

Department of Radiology, Chengdu Fifth People's Hospital, Chengdu 611130, China

Corresponding author: YANG Z, E-mail: yangzhicdwy@qq.com

Conflicts of interest   None.

Received  2026-05-12
Accepted  2026-07-29
DOI: 10.12015/issn.1674-8034.2026.08.020
Cite this article as ZHANG T Y, FU B, YANG Z. Research advances in artificial intelligence-assisted cardiac magnetic resonance imaging analysis[J]. Chin J Magn Reson Imaging, 2026, 17(8): 177-183. DOI:10.12015/issn.1674-8034.2026.08.020.

[1]
VON KNOBELSDORFF-BRENKENHOFF F, SCHULZ-MENGER J. Role of cardiovascular magnetic resonance in the guidelines of the European Society of Cardiology[J/OL]. J Cardiovasc Magn Reson, 2016, 18: 6 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/26800662/. DOI: 10.1186/s12968-016-0225-6.
[2]
MORALES M A, MANNING W J, NEZAFAT R. Present and future innovations in AI and cardiac MRI[J/OL]. Radiology, 2024, 310(1): e231269 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/38193835/. DOI: 10.1148/radiol.231269.
[3]
JAFARI M, SHOEIBI A, KHODATARS M, et al. Automated diagnosis of cardiovascular diseases from cardiac magnetic resonance imaging using deep learning models: a review[J/OL]. Comput Biol Med, 2023, 160: 106998 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/37182422/. DOI: 10.1016/j.compbiomed.2023.106998.
[4]
CASTELLACCIO A, ALMEIDA AROSTEGUI N, PALOMO JIMÉNEZ M, et al. Artificial intelligence in cardiovascular magnetic resonance imaging[J]. Radiologia (Engl Ed), 2025, 67(2): 239-247. DOI: 10.1016/j.rxeng.2025.03.001.
[5]
ALADWANI F, PERELLI A, MORDI I, et al. Deep learning for cardiac MRI: performance evidence and barriers to clinical integration. A Systematic Review and Meta-Analysis[J/OL]. Eur Heart J Imaging Meth Pract, 2026, 4(1): qyag045 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/41877690/. DOI: 10.1093/ehjimp/qyag045.
[6]
ZHANG Q, FOTAKI A, GHADIMI S, et al. Improving the efficiency and accuracy of cardiovascular magnetic resonance with artificial intelligence: review of evidence and proposition of a roadmap to clinical translation[J/OL]. J Cardiovasc Magn Reson, 2024, 26(2): 101051 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/38909656/. DOI: 10.1016/j.jocmr.2024.101051.
[7]
AROMIWURA A A, CAVALCANTE J L, KWONG R Y, et al. The role of artificial intelligence in cardiovascular magnetic resonance imaging[J/OL]. Prog Cardiovasc Dis, 2024, 86: 13-25 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/38925255/. DOI: 10.1016/j.pcad.2024.06.004.
[8]
BAEßLER B, ENGELHARDT S, HEKALO A, et al. Perfect match: radiomics and artificial intelligence in cardiac imaging[J/OL]. Circ Cardiovasc Imaging, 2024, 17(6): e015490 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/38889216/. DOI: 10.1161/CIRCIMAGING.123.015490.
[9]
POLIDORI T, DE SANTIS D, RUCCI C, et al. Radiomics applications in cardiac imaging: a comprehensive review[J]. Radiol Med, 2023, 128(8): 922-933. DOI: 10.1007/s11547-023-01658-x.
[10]
DENG J, ZHOU L T, LIAO B H, et al. Challenges in clinical translation of cardiac magnetic resonance imaging radiomics in non-ischemic cardiomyopathy: a narrative review[J]. Cardiovasc Diagn Ther, 2024, 14(6): 1210-1227. DOI: 10.21037/cdt-24-138.
[11]
MASTRODICASA D, VAN ASSEN M, HUISMAN M, et al. Use of AI in cardiac CT and MRI: a scientific statement from the ESCR, EuSoMII, NASCI, SCCT, SCMR, SIIM, and RSNA[J/OL]. Radiology, 2025, 314(1): e240516 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/39873607/. DOI: 10.1148/radiol.240516.
[12]
BAI W J, SINCLAIR M, TARRONI G, et al. Automated cardiovascular magnetic resonance image analysis with fully convolutional networks[J/OL]. J Cardiovasc Magn Reson, 2018, 20(1): 65 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/30217194/. DOI: 10.1186/s12968-018-0471-x.
[13]
BERNARD O, LALANDE A, ZOTTI C, et al. Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved [J]. IEEE Trans Med Imaging, 2018, 37(11): 2514-2525. DOI: 10.1109/TMI.2018.2837502.
[14]
CAMPELLO V M, GKONTRA P, IZQUIERDO C, et al. Multi-centre, multi-vendor and multi-disease cardiac segmentation: the M&Ms challenge[J]. IEEE Trans Med Imaging, 2021, 40(12): 3543-3554. DOI: 10.1109/TMI.2021.3090082.
[15]
YALCINKAYA D M, YOUSSEF K, HEYDARI B, et al. Improved robustness for deep learning-based segmentation of multi-center myocardial perfusion cardiovascular MRI datasets using data-adaptive uncertainty-guided space-time analysis[J/OL]. J Cardiovasc Magn Reson, 2024, 26(2): 101082 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/39142567/. DOI: 10.1016/j.jocmr.2024.101082.
[16]
LEITE J, BOLLACHE E, NGUYEN V, et al. Deep learning-based segmentation and strain analysis of left heart chambers from long-axis CMR images[J/OL]. Eur Heart J Imaging Methods Pract, 2024, 2(4): qyaf070 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/40657429/. DOI: 10.1093/ehjimp/qyaf070.
[17]
PEZESHKI P S, GHORASHI S M, HOUSHMAND G, et al. Feature tracking cardiac magnetic resonance imaging to assess cardiac manifestations of systemic diseases[J]. Heart Fail Rev, 2023, 28(5): 1189-1199. DOI: 10.1007/s10741-023-10321-6.
[18]
SCHILLING M, UNTERBERG-BUCHWALD C, LOTZ J, et al. Assessment of deep learning segmentation for real-time free-breathing cardiac magnetic resonance imaging at rest and under exercise stress[J/OL]. Sci Rep, 2024, 14: 3754 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/38355969/. DOI: 10.1038/s41598-024-54164-z.
[19]
CHADALAVADA S, RAUSEO E, et al. Quality control of cardiac magnetic resonance imaging segmentation, feature tracking, aortic flow, and native T1 analysis using automated batch processing in the UK Biobank study[J/OL]. Eur Heart J Imaging Meth Pract, 2024, 2(3): qyae094 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/39385845/. DOI: 10.1093/ehjimp/qyae094.
[20]
FAHMY A S, ROWIN E J, JAAFAR N, et al. Radiomics of late gadolinium enhancement reveals prognostic value of myocardial scar heterogeneity in hypertrophic cardiomyopathy[J]. JACC Cardiovasc Imaging, 2024, 17(1): 16-27. DOI: 10.1016/j.jcmg.2023.05.003.
[21]
PU C L, HU X, LV S Y, et al. Identification of fibrosis in hypertrophic cardiomyopathy: a radiomic study on cardiac magnetic resonance cine imaging[J]. Eur Radiol, 2023, 33(4): 2301-2311. DOI: 10.1007/s00330-022-09217-0.
[22]
WANG Z C, FAN Z Z, LIU X Y, et al. Deep learning for discrimination of hypertrophic cardiomyopathy and hypertensive heart disease on MRI native T1 maps[J]. J Magn Reson Imaging, 2024, 59(3): 837-848. DOI: 10.1002/jmri.28904.
[23]
RAISI-ESTABRAGH Z, MARTIN-ISLA C, NISSEN L, et al. Radiomics analysis enhances the diagnostic performance of CMR stress perfusion: a proof-of-concept study using the Dan-NICAD dataset[J/OL]. Front Cardiovasc Med, 2023, 10: 1141026 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/37781298/. DOI: 10.3389/fcvm.2023.1141026.
[24]
CICEK V, BAGCI U. AI-powered contrast-free cardiovascular magnetic resonance imaging for myocardial infarction[J/OL]. Front Cardiovasc Med, 2024, 11: 1457498 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/39639975/. DOI: 10.3389/fcvm.2024.1457498.
[25]
GERMAIN P, VARDAZARYAN A, PADOY N, et al. Classification of cardiomyopathies from MR cine images using convolutional neural network with transfer learning[J/OL]. Diagnostics, 2021, 11(9): 1554 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/34573896/. DOI: 10.3390/diagnostics11091554.
[26]
DIAO K Y, LIANG H Q, YIN H K, et al. Multi-channel deep learning model-based myocardial spatial-temporal morphology feature on cardiac MRI cine images diagnoses the cause of LVH[J/OL]. Insights Imaging, 2023, 14(1): 70 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/37093501/. DOI: 10.1186/s13244-023-01401-0.
[27]
WANG Y J, YANG K, WEN Y, et al. Screening and diagnosis of cardiovascular disease using artificial intelligence-enabled cardiac magnetic resonance imaging[J]. Nat Med, 2024, 30(5): 1471-1480. DOI: 10.1038/s41591-024-02971-2.
[28]
AMYAR A, AL-DEIRI D, SROUBEK J, et al. Radiomic cardiac MRI signatures for predicting ventricular arrhythmias in patients with nonischemic dilated cardiomyopathy[J/OL]. JACC Adv, 2025, 4(4): 101684 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/40127609/. DOI: 10.1016/j.jacadv.2025.101684.
[29]
YUE X Z, CUI J N, HUANG S C, et al. An interpretable radiomics-based machine learning model for predicting reverse left ventricular remodeling in STEMI patients using late gadolinium enhancement of myocardial scar[J]. Eur Radiol, 2025, 35(10): 6302-6312. DOI: 10.1007/s00330-025-11419-1.
[30]
ZHAO K K, ZHU Y J, CHEN X Y, et al. Machine learning in hypertrophic cardiomyopathy: nonlinear model from clinical and CMR features predicting cardiovascular events[J]. JACC Cardiovasc Imaging, 2024, 17(8): 880-893. DOI: 10.1016/j.jcmg.2024.04.013.
[31]
LAI C X, YIN M L, KHOLMOVSKI E G, et al. Multimodal AI to forecast arrhythmic death in hypertrophic cardiomyopathy[J]. Nat Cardiovasc Res, 2025, 4(7): 891-903. DOI: 10.1038/s44161-025-00679-1.
[32]
BELLO G A, DAWES T J W, DUAN J M, et al. Deep-learning cardiac motion analysis for human survival prediction[J]. Nat Mach Intell, 2019, 1(2): 95-104. DOI: 10.1038/s42256-019-0019-2.
[33]
CHEN Y T, XIE W, ZHANG J W, et al. Myocardial segmentation of cardiac MRI sequences with temporal consistency for coronary artery disease diagnosis[J/OL]. Front Cardiovasc Med, 2022, 9: 804442 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/35282363/. DOI: 10.3389/fcvm.2022.804442.
[34]
FERDIAN E, SUINESIAPUTRA A, FUNG K, et al. Fully automated myocardial strain estimation from cardiovascular MRI-tagged images using a deep learning framework in the UK biobank[J/OL]. Radiol Cardiothorac Imaging, 2020, 2(1): e190032 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/32715298/. DOI: 10.1148/ryct.2020190032.
[35]
LYU Q, SHAN H M, XIE Y B, et al. Cine cardiac MRI motion artifact reduction using a recurrent neural network[J]. IEEE Trans Med Imaging, 2021, 40(8): 2170-2181. DOI: 10.1109/TMI.2021.3073381.
[36]
KLEMENZ A C, REICHARDT L, GORODEZKY M, et al. Accelerated cardiac MRI with deep learning-based image reconstruction for cine imaging[J/OL]. Radiol Cardiothorac Imaging, 2024, 6(6): e230419 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/39540821/. DOI: 10.1148/ryct.230419.
[37]
CHEN J N, MEI J R, LI X H, et al. TransUNet: Rethinking the U-Net architecture design for medical image segmentation through the lens of transformers[J/OL]. Med Image Anal, 2024, 97: 103280 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/39096845/. DOI: 10.1016/j.media.2024.103280.
[38]
LI B J, YANG T J, ZHAO X. NVTrans-UNet: Neighborhood vision transformer based U-Net for multi-modal cardiac MR image segmentation[J/OL]. J Appl Clin Med Phys, 2023, 24(3): e13908 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/36651634/. DOI: 10.1002/acm2.13908.
[39]
LYU J, LI G Y, WANG C Y, et al. Region-focused multi-view transformer-based generative adversarial network for cardiac cine MRI reconstruction[J/OL]. Med Image Anal, 2023, 85: 102760 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/36720188/. DOI: 10.1016/j.media.2023.102760.
[40]
JACOB A J, BORGOHAIN I, CHITIBOI T, et al. Towards a cardiovascular magnetic resonance foundation model for multi-task cardiac image analysis[J/OL]. J Cardiovasc Magn Reson, 2025, 27(2): 101967 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/41046013/. DOI: 10.1016/j.jocmr.2025.101967.
[41]
MACHADO I, PUYOL-ANTÓN E, HAMMERNIK K, et al. A deep learning-based integrated framework for quality-aware undersampled cine cardiac MRI reconstruction and analysis[J]. IEEE Trans Biomed Eng, 2024, 71(3): 855-865. DOI: 10.1109/TBME.2023.3321431.
[42]
WECH T, SCHAD O, SAUER S, et al. Joint image reconstruction and segmentation of real-time cardiovascular magnetic resonance imaging in free-breathing using a model based on disentangled representation learning[J/OL]. J Cardiovasc Magn Reson, 2025, 27(1): 101844 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/39864743/. DOI: 10.1016/j.jocmr.2025.101844.
[43]
RAN L P, HUANG L, YAN X H, et al. Clinical application of free breathing with motion correction artificial intelligence cine sequence of cardiac magnetic resonance in patients with heart failure[J]. Chin J Magn Reson Imaging, 2024, 15(3): 62-67, 73. DOI: 10.12015/issn.1674-8034.2024.03.011.
[44]
MALAGI A V, LI X Q, TAO Q, et al. Advanced cardiac MRI: integrating AI and machine learning for post imaging reconstruction[J/OL]. Curr Treat Options Cardiovasc Med, 2025, 27(1): 67 [2026-05-11]. https://link.springer.com/article/10.1007/s11936-025-01116-z. DOI: 10.1007/s11936-025-01116-z.
[45]
DEMIREL O B, GHANBARI F, HOEGER C W, et al. Late gadolinium enhancement cardiovascular magnetic resonance with generative artificial intelligence[J/OL]. J Cardiovasc Magn Reson, 2025, 27(1): 101127 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/39615654/. DOI: 10.1016/j.jocmr.2024.101127.
[46]
AMIRRAJAB S, KHALIL Y AL, LORENZ C, et al. Label-informed cardiac magnetic resonance image synthesis through conditional generative adversarial networks[J/OL]. Comput Med Imaging Graph, 2022, 101: 102123 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/36174308/. DOI: 10.1016/j.compmedimag.2022.102123.
[47]
KHALIL Y AL, AMIRRAJAB S, LORENZ C, et al. On the usability of synthetic data for improving the robustness of deep learning-based segmentation of cardiac magnetic resonance images[J/OL]. Med Image Anal, 2023, 84: 102688 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/36493702/. DOI: 10.1016/j.media.2022.102688.
[48]
GEARHART A, ANJEWIERDEN S, BUDDHE S, et al. Review of the current state of artificial intelligence in pediatric cardiovascular magnetic resonance imaging[J/OL]. Children (Basel), 2025, 12(4): 416 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/40310065/. DOI: 10.3390/children12040416.
[49]
AKPINAR M H, SENGUR A, SALVI M, et al. Synthetic data generation via generative adversarial networks in healthcare: a systematic review of image- and signal-based studies[J/OL]. IEEE Open J Eng Med Biol, 2025, 6: 183-192 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/39698120/. DOI: 10.1109/OJEMB.2024.3508472.
[50]
LI R, ZHAO S H. Advances in artificial intelligence of cardiovascular imaging[J]. Chin J Magn Reson Imaging, 2019, 10(7): 551-555. DOI: 10.12015/issn.1674-8034.2019.07.014.
[51]
JIMÉNEZ-JARA C, SALAS R, DÍAZ-NAVARRO R, et al. AI applied to cardiac magnetic resonance for precision medicine in coronary artery disease: a systematic review[J/OL]. J Cardiovasc Dev Dis, 2025, 12(9): 345 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/41002624/. DOI: 10.3390/jcdd12090345.
[52]
FANG M Z, NAKASHIMA M, SINGH K, et al. Cardiac magnetic resonance imaging-large language model Meta AI: a finetuned large language model for identifying findings and associated attributes in cardiac magnetic resonance imaging reports[J/OL]. J Cardiovasc Magn Reson, 2025, 27(2): 101968 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/41241006/. DOI: 10.1016/j.jocmr.2025.101968.
[53]
LINARDOS A, KUSHIBAR K, WALSH S, et al. Federated learning for multi-center imaging diagnostics: a simulation study in cardiovascular disease[J/OL]. Sci Rep, 2022, 12: 3551 [2026-05-11]. https://pubmed.ncbi.nlm.nih.gov/35241683/. DOI: 10.1038/s41598-022-07186-4.

PREV Progress in multimodal cardiac magnetic resonance for the evaluation of myocardial injury in obese patients
NEXT Advances in CT and MRI for the assessment of progressive pulmonary fibrosis
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn