Share：
Share this content in WeChat
X
[Chinese][PDF]72481
• Clinical Article •
The feasibility study of MRI texture analysis in predicting delayed enhancement status in cardiac amyloidosis
WU Xi  TANG Lu  DENG Qiao  WU Tao  LIU Nian  ZHANG Tianjing  CHEN Yucheng  HUANG Xiaohua  SUN Jiayu 

Cite this article as: Wu X, Tang L, Deng Q, et al. The feasibility study of MRI texture analysis in predicting delayed enhancement status in cardiac amyloidosis[J]. Chin J Magn Reson Imaging, 2021, 12(12): 6-11. DOI:10.12015/issn.1674-8034.2021.12.002.


[Abstract] Objective To explore the value of magnetic resonance imaging (MRI) texture analysis in predicting delayed enhancement status in patients with cardiac amyloidosis (CA). Materials and Methods: One hundred and thirty-two patients with CA confirmed by pathology were retrospectively analyzed, including presence (87 cases) and absence (45 cases) of late gadolinium enhancement (LGE) groups, and Sixty-six health volunteers were recruited. Regions of interest (ROIs) on native T1 mapping were drawn by two radiologists using open source software ITK-SNAP. FeAture Explorer (FAE) software was used to extract and select features. Finally, eight features were selected to build the model of support vector machine (SVM) A, which was used to differentiate the myocardium between CA patients without LGE and health volunteers. By using the same method of feature dimension reduction, nine features were selected to construct the model of SVM B to further predict the presence or absence of LGE in patients with CA. Finally, sensitivity, specificity, positive predictive value, negative predictive value, accuracy and area under the receiver operating characteristic curve (AUC) were calculated to evaluate the differential diagnostic efficacy of the two models.Results The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of training set of the model of SVM A were 0.839, 0.957, 0.929, 0.898 and 0.909, respectively, and those of test set were 0.786, 0.950, 0.917, 0.864 and 0.882, respectively. The AUC of the training set and the test set were 0.948 (95% CI: 0.890—0.991) and 0.918 (95% CI: 0.780—1.000), respectively. The sensitivity, specificity, positive predictive value, negative predictive value and accuracy of training set of the model of SVM B were 0.853, 0.710, 0.853, 0.710 and 0.804, respectively, and those of test set were 0.846, 0.714, 0.846, 0.714 and 0.800, respectively. The AUC of the training set and the test set were 0.762 (95% CI: 0.639—0.875) and 0.758 (95% CI: 0.565—0.937), respectively.Conclusions The radiomics models based on MRI texture analysis without contrast agent have a reasonable diagnostic performance in differentiating the myocardium between CA patients without LGE and health volunteers and predicting whether the myocardium of CA patients has delayed enhancement.
[Keywords] cardiac amyloidosis；radiomics；texture analysis；cardiovascular magnetic resonance

WU Xi1   TANG Lu2   DENG Qiao1   WU Tao2   LIU Nian1   ZHANG Tianjing3   CHEN Yucheng4   HUANG Xiaohua1   SUN Jiayu2*  

1 Department of Radiology, Affiliated Hospital of North Sichuan Medical College, Nanchong 637000, China

2 Department of Radiology, West China Hospital, Sichuan University, Chengdu 610041, China

3 Philips (China) Investment Co., Ltd, Guangzhou 510180, China

4 Department of Cardiology, West China Hospital, Sichuan University, Chengdu 610041, China

Sun JY, E-mail: sunjiayu@wchscu.cn

Conflicts of interest   None.

ACKNOWLEDGMENTS Key Research and Development Projects in Sichuan Province (No. 2020YFS0123).
Received  2021-06-23
Accepted  2021-10-12
DOI: 10.12015/issn.1674-8034.2021.12.002
Cite this article as: Wu X, Tang L, Deng Q, et al. The feasibility study of MRI texture analysis in predicting delayed enhancement status in cardiac amyloidosis[J]. Chin J Magn Reson Imaging, 2021, 12(12): 6-11. DOI:10.12015/issn.1674-8034.2021.12.002.

[1]
Jurcuţ R, Onciul S, Adam R, et al. Multimodality imaging in cardiac amyloidosis: a primer for cardiologists[J]. Eur Heart J-Card Img, 2020, 21(8): 833-844. DOI: 10.1093/ehjci/jeaa063.
[2]
Wan K, Sun JY, Yang D, et al. Left Ventricular Myocardial Deformation on Cine MR Images: Relationship to Severity of Disease and Prognosis in Light-Chain Amyloidosis[J]. Radiology, 2018, 288(1): 73-80. DOI: 10.1148/radiol.2018172435.
[3]
Nicol M, Baudet M, Brun S, et al. Diagnostic score of cardiac involvement in AL amyloidosis[J]. Eur Heart J-Card Img, 2020, 21(5): 542-548. DOI: 10.1093/ehjci/jez180.
[4]
Wu XQ, Feng YL, Li CP, et al. The advances in cardiovascular magnetic resonance imaging for light-chain and transthyretin-related amyloidosis. Chin J Magn Reson Imaging, 2020, 11(8): 707-711. DOI: 10.12015/issn.1674-8034.2020.08.028
[5]
Dorbala S, Ando Y, Bokhari S, et al. ASNC/AHA/ASE/EANM/HFSA/ISA/SCMR/SNMMI Expert Consensus Recommendations for Multimodality Imaging in Cardiac Amyloidosis: Part 1 of 2—Evidence Base and Standardized Methods of Imaging[J]. Journal of Cardiac Failure, 2019, 25(11): e1-e39. DOI: 10.1016/j.cardfail.2019.08.001.
[6]
Raina S, Lensing SY, Nairooz RS, et al. Prognostic Value of Late Gadolinium Enhancement CMR in Systemic Amyloidosis[J]. JACC Cardiovasc Imaging, 2016, 9(11): 1267-1277. DOI: 10.1016/j.jcmg.2016.01.036.
[7]
Hassani C, Saremi F, Varghese BA, et al. Myocardial Radiomics in Cardiac MRI[J]. American journal of roentgenology (1976), 2020, 214(3): 536. DOI: 10.2214/AJR.19.21986
[8]
Baessler B, Luecke C, Lurz J, et al. Cardiac MRI Texture Analysis of T1 and T2 Maps in Patients with Infarctlike Acute Myocarditis[J]. Radiology, 2018, 289(2): 357-365. DOI: 10.1148/radiol.2018180411.
[9]
Shi RY, Wu R, An DL, et al. Texture analysis applied in T1 maps and extracellular volume obtained using cardiac MRI in the diagnosis of hypertrophic cardiomyopathy and hypertensive heart disease compared with normal controls[J]. Clin Radiol, 2021, 76(3): 236-239. DOI: 10.1016/j.crad.2020.11.001.
[10]
Baessler B, Mannil M, Maintz D, et al. Texture analysis and machine learning of non-contrast T1-weighted MR images in patients with hypertrophic cardiomyopathy-Preliminary results[J]. Eur J Radiol, 2018, 102: 61-67. DOI: 10.1016/j.ejrad.2018.03.013.
[11]
Song Y, Zhang J, Zhang YD, et al. FeAture Explorer (FAE): A tool for developing and comparing radiomics models[J]. PLoS One, 2020, 15(8): e237587. DOI: 10.1371/journal.pone.0237587.
[12]
Wu T, Sun HQ, Liu XM, et al. Myocardial Magnetic Resonance Texture Characteristcs of Healthy Volunteers.[J]. Journal of Sichuan University (medical science edition), 2019, 50(4): 494-499. DOI: 10.13464/j.scuxbyxb.2019.04.008
[13]
Karamitsos TD, Piechnik SK, Banypersad SM, et al. Noncontrast T1 Mapping for the Diagnosis of Cardiac Amyloidosis[J]. JACC: Cardiovascular Imaging, 2013, 6(4): 488-497. DOI: 10.1016/j.jcmg.2012.11.013.
[14]
Banypersad SM, Fontana M, Maestrini V, et al. T1 mapping and survival in systemic light-chain amyloidosis[J]. European Heart Journal, 2015, 36(4): 244-251. DOI: 10.1093/eurheartj/ehu444.
[15]
Pinamonti B, Picano E, Ferdeghini EM, et al. Quantitative texture analysis in two-dimensional echocardiography: Application to the diagnosis of myocardial amyloidosis[J]. Journal of the American College of Cardiology, 1989, 14(3): 666-671. DOI: https://doi.org/10.1016/0735-1097(89)90108-3.
[16]
Neisius U, El-Rewaidy H, Nakamori S, et al. Radiomic Analysis of Myocardial Native T1 Imaging Discriminates Between Hypertensive Heart Disease and Hypertrophic Cardiomyopathy[J]. JACC: Cardiovascular Imaging, 2019, 12(10): 1946-1954. DOI: 10.1016/j.jcmg.2018.11.024.
[17]
Wang J, Yang FY, Liu WT, et al. Radiomic Analysis of Native T1 Mapping Images Discriminates Between MYH7 and MYBPC3-Related Hypertrophic Cardiomyopathy[J]. J Magn Reson Imaging, 2020, 52(6): 1714-1721. DOI: 10.1002/jmri.27209.
[18]
Baessler B, Luecke C, Lurz J, et al. Cardiac MRI and Texture Analysis of Myocardial T1 and T2 Maps in Myocarditis with Acute versus Chronic Symptoms of Heart Failure[J]. Radiology, 2019, 292(3): 608-617. DOI: 10.1148/radiol.2019190101.
[19]
Mancio J, Pashakhanloo F, El-Rewaidy H, et al. Machine learning phenotyping of scarred myocardium from cine in hypertrophic cardiomyopathy[J]. European Heart Journal - Cardiovascular Imaging, 2021. DOI: 10.1093/ehjci/jeab056.
[20]
Ma QM, Ma Y, Yu TT, et al. Radiomics of Non-Contrast-Enhanced T1 Mapping: Diagnostic and Predictive Performance for Myocardial Injury in Acute ST-Segment-Elevation Myocardial Infarction[J]. Korean J Radiol, 2021, 22(4): 535-546. DOI: 10.3348/kjr.2019.0969.
[21]
Baessler B, Mannil M, Oebel S, et al. Subacute and Chronic Left Ventricular Myocardial Scar: Accuracy of Texture Analysis on Nonenhanced Cine MR Images[J]. Radiology, 2018, 286(1): 103-112. DOI: 10.1148/radiol.2017170213.
[22]
Larroza A, López-Lereu MP, Monmeneu JV, et al. Texture analysis of cardiac cine magnetic resonance imaging to detect nonviable segments in patients with chronic myocardial infarction[J]. Medical Physics, 2018, 45(4): 1471-1480. DOI: 10.1002/mp.12783.
[23]
Mannil M, Kato K, Manka R, et al. Prognostic value of texture analysis from cardiac magnetic resonance imaging in patients with Takotsubo syndrome: a machine learning based proof-of-principle approach[J]. Sci Rep, 2020, 10(1): 20537. DOI: 10.1038/s41598-020-76432-4.

PREV Initial application of synthetic MRI in evaluating brain maturation of preterm infants
NEXT Diagnostic efficacy of novel imaging markers and its correlation with clinical symptoms for idiopathic normal pressure hydrocephalus
  


LINK


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