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

Shi, Rong; Wu, Rui; An, Dong‐Aolei; Chen, B.-H.; Wu, Chong-Wen; Du, Liang; Jiang, Meng; Xu, Jianrong; Wu, Lian‐Ming

doi:10.1016/j.crad.2020.11.001

Cited by 17 publications

(31 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Amano et al 18 , using radiomic features from T2-weighted STIR images, have found that a GLRLM-GrayLevelNonUniformity value of 64.7 was the best discriminator between HCM patients and controls with an area under the receiver operating characteristic curve of 0.93. Notably, only a few pilot HCM studies [19][20][21]36 have applied radiomic analysis to T1 maps. Two studies 19,21 have focused on the diagnostic capability of radiomic analysis to discriminate between HCM and HHD patients.…”

Section: Discussionmentioning

confidence: 99%

“…Increased T2 values, probably related to low grade myocardial in ammation and dilated lymphatic channels 13,14 , have been demonstrated in HCM patients and have been proven to be related to an increased arrhythmic risk 15,16 , as well as to more severe myocardial injury 17 . However, the usefulness of these mapping techniques is limited by a partial overlap of T1 and T2 values between HCM patients and healthy controls, as well as between HCM patients and patients with other myocardial diseases [18][19][20][21] .…”

Section: Introductionmentioning

confidence: 99%

“…While radiomic analysis has been mainly applied in the eld of oncology, there is a growing interest in improving the diagnostic accuracy and prognostic value of CMR imaging by exploiting radiomic techniques [25][26][27][28][29][30] . Recently, radiomics has aroused as a useful tool for unveiling myocardial tissue characteristics in HCM patients [18][19][20][21][31][32][33][34][35][36] . In this regard, Baeßler et al 31 have reported signi cant differences in texture features obtained from T1weighted images of HCM patients and healthy subjects, identifying a single feature (i.e., gray level non uniformity from the gray level run length matrix) as the best discriminator.…”

Section: Introductionmentioning

confidence: 99%

“…Given their inherent quantitative nature, T1 and T2 maps might be particularly suitable for radiomic analysis. So far, few studies have applied radiomic analysis to T1 maps in patients with HCM [19][20][21]36 . In two studies of the same group, radiomic analysis of native T1 maps has allowed discriminating between hypertensive heart disease (HHD) and HCM 19 , as well as between LGE + and LGE-patients, identifying a subgroup of HCM patients without scar in whom gadolinium administration could be avoided 36 .…”

Section: Introductionmentioning

confidence: 99%

“…Moreover, Wang et al 20 have found that radiomic analysis of native T1 maps can discriminate HCM patients with different mutations of sarcomere-related genes. Shi et al 21 have shown that texture analysis of native T1 and ECV maps could distinguish between patients (HCM and HHD) and healthy subjects, as well as between HCM and HHD, with diagnostic accuracies comparable with strain parameters.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Image Preprocessing and Filtering Effect on the Estimate of Myocardial Radiomic Features From T1 and T2 Mapping in Hypertrophic Cardiomyopathy

Marfisi

Tessa

Marzi

et al. 2021

Preprint

View full text Add to dashboard Cite

Radiomics is emerging as a promising and useful tool in cardiac magnetic resonance (CMR) imaging applications. Accordingly, the purpose of this study was to investigate, for the first time, the effect of image preprocessing and filtering on radiomic features estimation from quantitative CMR T1 and T2 mapping. Specifically, T1 and T2 maps of 26 patients with hypertrophic cardiomyopathy (HCM) were used to estimate 98 radiomic features for 7 different resampling voxel sizes (at fixed bin width), 9 different bin widths (at fixed resampling voxel size), and 7 different spatial filters (at fixed resampling voxel size/bin width). While we found a remarkable dependence of myocardial radiomic features from T1 and T2 mapping on image filters, many radiomic features showed a limited sensitivity to resampling voxel size/bin width, in terms of intraclass correlation coefficient (>0.75) and coefficient of variation (<30%). The estimate of several textural radiomic features showed a linear significant (p<0.05) correlation with resampling voxel size/bin width. Overall, radiomic features from T2 maps have proven to be less sensitive to image preprocessing than those from T1 maps, especially when varying bin width. Our results might corroborate the potential of radiomics from T1/T2 mapping in HCM and hopefully in other myocardial diseases.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Image Preprocessing and Filtering Effect on the Estimate of Myocardial Radiomic Features From T1 and T2 Mapping in Hypertrophic Cardiomyopathy

Marfisi

Tessa

Marzi

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

An Integrated Algorithm for Differentiating Hypertrophic Cardiomyopathy From Hypertensive Heart Disease

Kong

Wang

et al. 2023

Magnetic Resonance Imaging

View full text Add to dashboard Cite

BackgroundDifferentiating hypertrophic cardiomyopathy (HCM) from hypertensive heart disease (HHD) is challenging.PurposeTo identify differences between HCM and HHD on a patient basis using MRI.Study TypeRetrospective.PopulationA total of 219 subjects, 148 in phase I (baseline data and algorithm development: 75 HCM, 33 HHD, and 40 controls) and 71 in phase II (algorithm validation: 56 HCM and 15 HHD).Field Strength/SequenceContrast‐enhanced inversion‐prepared gradient echo and cine‐balanced steady‐state free precession sequences at 3.0 T.AssessmentMRI parameters assessed included left ventricular (LV) ejection fraction (LVEF), LV end systolic and end diastolic volumes (LVESV and LVEDV), mean maximum LV wall thickness (MLVWT), LV global longitudinal and circumferential strain (GRS, GLS, and GCS), and native T1. Parameters, which were significantly different between HCM and HHD in univariable analysis, were entered into a principal component analysis (PCA). The selected components were then introduced into a multivariable regression analysis to model an integrated algorithm (IntA) for screening the two disorders. IntA performance was assessed for patients with and without LGE in phase I (development) and phase II (validation).Statistical TestsUnivariable regression, PCA, receiver operating curve (ROC) analysis. A P value <0.05 was considered statistically significant.ResultsDerived IntA formulation included LVEF, LVESV, LVEDV, MLVWT, and GCS. In LGE‐positive subjects in phase l, the cutoff point of IntA ≥81 indicated HCM (83% sensitivity and 91% specificity), with the area under the ROC curve (AUC) of 0.900. In LGE‐negative subjects, a higher possibility of HCM was indicated by a cutoff point of IntA ≥84 (100% sensitivity and 82% specificity), with an AUC of 0.947. Validation of IntA in phase II resulted in an AUC of 0.846 in LGE‐negative subjects and 0.857 in LGE‐positive subjects.Data ConclusionA per‐patient‐based IntA algorithm for differentiating HCM and HHD was generated from MRI data and incorporated FT, LGE and morphologic parameters.Evidence Level3.Technical EfficacyStage 2.

show abstract

Deep Learning for Discrimination of Hypertrophic Cardiomyopathy and Hypertensive Heart Disease on MRI Native T1 Maps

Wang

Fan

Liu

et al. 2023

Magnetic Resonance Imaging

View full text Add to dashboard Cite

BackgroundNative T1 and radiomics were used for hypertrophic cardiomyopathy (HCM) and hypertensive heart disease (HHD) differentiation previously. The current problem is that global native T1 remains modest discrimination performance and radiomics requires feature extraction beforehand. Deep learning (DL) is a promising technique in differential diagnosis. However, its feasibility for discriminating HCM and HHD has not been investigated.PurposeTo examine the feasibility of DL in differentiating HCM and HHD based on T1 images and compare its diagnostic performance with other methods.Study TypeRetrospective.Population128 HCM patients (men, 75; age, 50 years ± 16) and 59 HHD patients (men, 40; age, 45 years ± 17).Field Strength/Sequence3.0T; Balanced steady‐state free precession, phase‐sensitive inversion recovery (PSIR) and multislice native T1 mapping.AssessmentCompare HCM and HHD patients baseline data. Myocardial T1 values were extracted from native T1 images. Radiomics was implemented through feature extraction and Extra Trees Classifier. The DL network is ResNet32. Different input including myocardial ring (DL‐myo), myocardial ring bounding box (DL‐box) and the surrounding tissue without myocardial ring (DL‐nomyo) were tested. We evaluate diagnostic performance through AUC of ROC curve.Statistical TestsAccuracy, sensitivity, specificity, ROC, and AUC were calculated. Independent t test, Mann–Whitney U‐test and Chi‐square test were adopted for HCM and HHD comparison. P < 0.05 was considered statistically significant.ResultsDL‐myo, DL‐box, and DL‐nomyo models showed an AUC (95% confidential interval) of 0.830 (0.702–0.959), 0.766 (0.617–0.915), 0.795 (0.654–0.936) in the testing set. AUC of native T1 and radiomics were 0.545 (0.352–0.738) and 0.800 (0.655–0.944) in the testing set.Data ConclusionThe DL method based on T1 mapping seems capable of discriminating HCM and HHD. Considering diagnostic performance, the DL network outperformed the native T1 method. Compared with radiomics, DL won an advantage for its high specificity and automated working mode.Level of Evidence4Technical Efficacy Stage2

show abstract

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

Cited by 17 publications

References 24 publications

Image Preprocessing and Filtering Effect on the Estimate of Myocardial Radiomic Features From T1 and T2 Mapping in Hypertrophic Cardiomyopathy

Image Preprocessing and Filtering Effect on the Estimate of Myocardial Radiomic Features From T1 and T2 Mapping in Hypertrophic Cardiomyopathy

An Integrated Algorithm for Differentiating Hypertrophic Cardiomyopathy From Hypertensive Heart Disease

Deep Learning for Discrimination of Hypertrophic Cardiomyopathy and Hypertensive Heart Disease on MRI Native T1 Maps

Contact Info

Product

Resources

About