Search

Purpose: Radiological reporting is transitioning to quantitative analysis, requiring large-scale multi-center validation of biomarkers. A major prerequisite and bottleneck for this task is the voxelwise annotation of image data, which is time-consuming for large cohorts. In this study, we propose an iterative training workflow to support and facilitate such segmentation tasks, specifically for high-resolution thoracic CT data., Methods: Our study included 132 thoracic CT scans from clinical practice, annotated by 13 radiologists. In three iterative training experiments, we aimed to improve and accelerate segmentation of the heart and mediastinum. Each experiment started with manual segmentation of 5-25 CT scans, which served as training data for a nnU-Net. Further iterations incorporated AI pre-segmentation and human correction to improve accuracy, accelerate the annotation process, and reduce human involvement over time., Results: Results showed consistent improvement in AI model quality with each iteration. Resampled datasets improved the Dice similarity coefficients for both the heart (DCS 0.91 [0.88; 0.92]) and the mediastinum (DCS 0.95 [0.94; 0.95]). Our AI models reduced human interaction time by 50 % for heart and 70 % for mediastinum segmentation in the most potent iteration. A model trained on only five datasets achieved satisfactory results (DCS > 0.90)., Conclusions: The iterative training workflow provides an efficient method for training AI-based segmentation models in multi-center studies, improving accuracy over time and simultaneously reducing human intervention. Future work will explore the use of fewer initial datasets and additional pre-processing methods to enhance model quality., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 The Authors. Published by Elsevier B.V. All rights reserved.)

Objectives: Robustness of radiomic features in physiological tissue is an important prerequisite for quantitative analysis of tumor biology and response assessment. In contrast to previous studies which focused on different tumors with mostly short scan-re-scan intervals, this study aimed to evaluate the robustness of radiomic features in cancer-free patients and over a clinically encountered inter-scan interval., Materials and Methods: Patients without visible tumor burden who underwent at least two portal-venous phase dual energy CT examinations of the abdomen between May 2016 and January 2020 were included, while macroscopic tumor burden was excluded based upon follow-up imaging for all patients (≥3 months). Further, patients were excluded if no follow-up imaging was available, or if the CT protocol showed deviations between repeated examinations. Circular regions of interest were placed and proofread by two board-certified radiologists (4 years and 5 years experience) within the liver (segments 3 and 6), the psoas muscle (left and right), the pancreatic head, and the spleen to obtain radiomic features from normal-appearing organ parenchyma using PyRadiomics. Radiomic feature robustness was tested using the concordance correlation coefficient with a threshold of 0.75 considered indicative for deeming a feature robust., Results: In total, 160 patients with 480 repeated abdominal CT examinations (range: 2-4 per patient) were retrospectively included in this single-center, IRB-approved study. Considering all organs and feature categories, only 4.58 % (25/546) of all features were robust with the highest rate being found in the first order feature category (20.37 %, 22/108). Other feature categories (grey level co-occurrence matrix, grey level dependence matrix, grey level run length matrix, grey level size zone matrix, and neighborhood gray-tone difference matrix) yielded an overall low percentage of robust features (range: 0.00 %-1.19 %). A subgroup analysis revealed the reconstructed field of view and the X-ray tube current as determinants of feature robustness (significant differences in subgroups for all organs, p < 0.001) as well as the size of the region of interest (no significant difference for the pancreatic head with p = 0.135, significant difference with p < 0.001 for all other organs)., Conclusion: Radiomic feature robustness obtained from cancer-free subjects with repeated examinations using a consistent protocol and CT scanner was limited, with first order features yielding the highest proportion of robust features., Competing Interests: Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Nils Große Hokamp: Research support, talk honoraria (Philips), consultancy (Bristol Myers Squibb); David Zopfs: Research support, talk honoraria (Philips); Kristina Sonnabend: Employee (Philips); Clemens Hentschke: Employee (Mint Medical GmbH); Bettina Baeßler: Speaker (Bayer Vital GmbH), founder and CEO of Lernrad GmbH; Liliana Lourenco Caldeira: Research support (Philips)., (Copyright © 2024 The Authors. Published by Elsevier B.V. All rights reserved.)

Purpose: To assess the performance of semi-automated volumetry of solid pulmonary nodules on single-energy tin-filtered ultralow dose (ULD) chest CT scans at a radiation dose equivalent to chest X-ray relative to standard dose (SD) chest CT scans and assess the impact of kernel and iterative reconstruction selection., Methods: Ninety-four consecutive patients from a prospective single-center study were included and underwent clinically indicated SD chest CT (1.9 ± 0.8 mSv) and additional ULD chest CT (0.13 ± 0.01 mSv) in the same session. All scans were reconstructed with a soft tissue (Br40) and lung (Bl64) kernel as well as with Filtered Back Projection (FBP) and Iterative Reconstruction (ADMIRE-3 and ADMIRE-5). One hundred and forty-eight solid pulmonary nodules were identified and analysed by semi-automated volumetry on all reconstructions. Nodule volumes were compared amongst all reconstructions thereby focusing on the agreement between SD and ULD scans., Results: Nodule volumes ranged from 58.5 (28.8-126) mm 3 for ADMIRE-5 Br40 ULD reconstructions to 72.5 (39-134) mm 3 for FBP Bl64 SD reconstructions with significant differences between reconstructions (p < 0.001). Interscan agreement of volumes between two given reconstructions ranged from ICC = 0.605 to ICC = 0.999. Between SD and ULD scans, agreement of nodule volumes was highest for FBP Br40 (ICC = 0.995), FBP Bl64 (ICC = 0.939) and ADMIRE-5 Bl64 (ICC = 0.994) reconstructions. ADMIRE-3 reconstructions exhibited reduced interscan agreement of nodule volumes (ICCs from 0.788 - 0.882)., Conclusions: The interscan agreement of node volumes between SD and ULD is high depending on the choice of kernel and reconstruction algorithm. However, caution should be exercised when comparing two image series that were not identically reconstructed., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2022 The Author(s). Published by Elsevier B.V. All rights reserved.)

Background: 4D flow imaging can be used to evaluate vortex formation in the pulmonary artery seen in patients with pulmonary hypertension. We evaluated if a k-t accelerated multi-VENC (velocity encoding) 4D flow acquisition improves image quality, inter-reader agreement and correlation with hemodynamic parameters., Methods: A total of 14 patients with pulmonary hypertension (5 females, 9 males; mean age 61 ± 16 years) underwent 4D flow MRI (magnetic resonance imaging) and right heart catheterization. In addition to that, 13 healthy volunteers (2 females, 11 males, mean age 33 ± 12 years) also underwent 4D flow MRI. Multi- and single-VENC datasets were reconstructed and evaluated for vortex formation and vortex duration by two blinded readers and image quality was rated on a 5-point scale., Results: Both readers rated image quality as significantly higher on multi-VENC datasets (3.96 ± 0.71 vs. 2.56 ± 0.93, p < 0.001; 4.70 ± 0.61 vs. 4.07 ± 0.92, p = 0.003). Inter-reader correlation for vortex duration quantification was higher on multi-VENC datasets compared to single-VENC datasets (r = 0.63 vs. r = 0.44). No significant correlation was found between vortex duration and mean pulmonary artery pressure in patients with PH., Conclusion: Multi-VENC 4D flow MRI significantly improves image quality and inter-reader agreement for the evaluation of vortex formation in the pulmonary artery., (Copyright © 2021 Elsevier B.V. All rights reserved.)

Objectives: To evaluate whether a magnetic resonance imaging (MRI) radiomics-based machine learning classifier can predict postoperative pancreatic fistula (POPF) after pancreaticoduodenectomy (PD) and to compare its performance to T1 signal intensity ratio (T1 SIratio)., Methods: Sixty-two patients who underwent 3 T MRI before PD between 2008 and 2018 were retrospectively analyzed. POPF was graded and split into clinically relevant POPF (CR-POPF) vs. biochemical leak or no POPF. On T1- and T2-weighted images, 2 regions of interest were placed in the pancreatic corpus and cauda. 173 radiomics features were extracted using pyRadiomics. Additionally, the pancreas-to-muscle T1 SIratio was measured. The dataset was augmented and split into training (70 %) and test sets (30 %). A Boruta algorithm was used for feature reduction. For prediction of CR-POPF models were built using a gradient-boosted tree (GBT) and logistic regression from the radiomics features, T1 SIratio and a combination of the two. Diagnostic accuracy of the models was compared using areas under the receiver operating characteristics curve (AUCs)., Results: Five most important radiomics features were identified for prediction of CR-POPF. A GBT using these features achieved an AUC of 0.82 (95 % Confidence Interval [CI]: 0.74 - 0.89) when applied on the original (non-augmented) dataset. Using T1 SIratio, a GBT model resulted in an AUC of 0.75 (CI: 0.63 - 0.84) and a logistic regression model delivered an AUC of 0.75 (CI: 0.63 - 0.84). A GBT model combining radiomics features and T1 SIratio resulted in an AUC of 0.90 (CI 0.84 - 0.95)., Conclusion: MRI-radiomics with routine sequences provides promising prediction of CR-POPF., (Copyright © 2021 Elsevier B.V. All rights reserved.)

Background: To evaluate the feasibility of a k-t accelerated multi-VENC 4D phase contrast flow MRI acquisition of the main heart-surrounding vessels, its benefits over a traditional single-VENC acquisition and to present reference flow and velocity values in a large cohort of volunteers., Methods: 44 healthy volunteers were examined on a 3 T MRI scanner (Ingenia, Philips, Best, The Netherlands). 4D flow measurements were obtained with a FOV including the aorta and the pulmonary arteries. VENC values were set to 40, 100 and 200 cm/s and unfolded based on an MRI signal model. Unfolded multi-VENC data was compared to the single-VENC with VENC 200 cm/s. Flow and velocity quantification was performed in several regions of interest (ROI) placed in the ascending aorta and in the main pulmonary artery. Conservation of mass analysis was performed for single- and multi-VENC datasets. Values for mean and maximal flow velocity and stroke volume were calculated and compared to the literature., Results: Mean scan time was 13.8 ± 4 min. Differences between stroke volumes between the ascending aorta and the main pulmonary artery were significantly lower in multi-VENC datasets compared to single-VENC datasets (9.6 ± 7.8 mL vs. 25.4 ± 26.4 mL, p < 0.001). This was also true for differences in stroke volume between up- and downstream ROIs in the ascending aorta and pulmonary artery. Values for mean and maximal velocities and stroke volume were in-line with previous studies. To highlight potential clinical applications two exemplary 4D flow measurements in patients with different pathologies are shown and compared to single-VENC datasets., Conclusions: k-t accelerated multi-VENC 4D phase contrast flow MRI acquisition of the great vessels is feasible in a clinically acceptable scan duration. It offers improvements over traditional single-VENC 4D flow, expectedly being valuable when vessels with different flow velocities or complex flow phenomena are evaluated., (Copyright © 2021 Elsevier B.V. All rights reserved.)

Purpose: Severe aortic stenosis (AS) is known to be associated with substantial myocardial remodelling, leading to diffuse myocardial fibrosis (DMF). Native myocardial T1 is emerging as a novel imaging biomarker for the non-invasive assessment of DMF. In contrast, no studies exist elucidating changes of myocardial T2 reflecting myocardial oedema in the presence of AS. The purpose of the present study was to combine native T1 and T2 mapping in order to characterize myocardial tissue changes in the setting of severe AS., Methods: After obtaining ethical approval and informed consent, a total of 26 prospectively selected patients with severe AS (13 women, mean age 81 ± 7 years) and 17 healthy controls (12 women, mean age 63 ± 6 years) underwent cardiac magnetic resonance (CMR) imaging on a clinical 3 T scanner. The CMR protocol included a native Modified Look-Locker (MOLLI) T1 mapping and a Gradient Spin Echo (GraSE) T2-mapping sequence in three short-axis slices and one long-axis view. After segmentation, myocardial T1 and T2 values were averaged over the entire myocardium. Statistical analysis was performed using Wilcoxon sum-rank test, Welch's independent t-test and Pearson's correlation coefficient., Results: Global native myocardial T1 was significantly increased in AS patients when compared to controls (1305 ± 39 vs. 1272 ± 21 ms, p = .005). Similarly, mean myocardial T2 was significantly elevated in AS patients (51 ± 4 vs. 46 ± 2 ms, p < .001) and showed a strong correlation with native T1 (r = .60, p < .001). An overlap was observed between T1 of both groups, whereas T2 discriminated nearly perfectly between the two groups (area under the curve in ROC analyses: 0.76 for T1, 0.87 for T2)., Conclusions: Patients with severe AS exhibit significantly elevated native myocardial T1, which has previously been shown to correlate with the amount of myocardial collagen. Adding to this evidence, the present study is the first to show that native T1 and T2 are both significantly elevated and correlated in AS patients, pointing towards a potential role of oedematous/inflammatory processes in the pathophysiology of myocardial remodelling associated with AS., (Copyright © 2018 Elsevier B.V. All rights reserved.)

Purpose/introduction: This study aims to evaluate the incremental diagnostic value of cardiac magnetic resonance (CMR) feature tracking (FT) derived atrial and ventricular strain-analysis in patients with acute myocarditis (myocarditis) as an additional tool to established Lake-Louise criteria (LLC)., Material and Methods: A total of 86 patients with clinically proven myocarditis and 30 healthy controls underwent a comprehensive CMR protocol. In addition to established LLC, FT derived strain parameters from the left (LA) and right atrium (RA) as well as the left (LV) and right ventricle (RV) were assessed. Receiver operating characteristics analysis was performed to compare diagnostic performance., Results: Patients with myocarditis showed significantly reduced LA passive strain (LA ε e : 26.3 ± 14.5 vs. 33.5 ± 10.1%, p = .007), LA peak early negative strain rate (LA SRe: -1.94 ± 0.59 1/s vs. -1.46 ± 0.62 1/s, p < .001), LV global longitudinal strain (LV GLS: -17.2 ± 4.9% vs. -13.3 ± 6.2%, p < .001), LV midventricular circumferential strain (LV mid CS: -25.9 ± 4.7% vs. -22.0 ± 6.5%, p < .001), and an increased RV basal circumferential SR (RV basal CSR: -0.70 ± 028 vs. -0.58 ± 0.34 1/s, p = .096) compared to healthy controls. In a subgroup analysis of patients with myocarditis and preserved LV function, RV basal CSR was also significantly increased compared to healthy controls (-0.74 ± 0.27 vs. -0.57 ± 0.26 1/s; p = .035) whereas LA SRe (-1.49 ± 0.59 vs. -1.32 ± 0.74%; p = .005) was significantly reduced. In multinominal logistic regression analysis, LA SRe and RV basal CSR proved to be the best independent predictors of myocarditis with preserved LV function. Combined with LLC, strain parameters enhanced the diagnostic performance in such patients (Areas under the curve (AUC): LLC: 0.78, LLC + LV GLS + LA SRe: 0.86), whereas LA SRe was the best performing single parameter (AUC: 0.72)., Conclusion: Combining quantitative CMR derived atrial and ventricular strain parameters with established LLC parameters can improve the diagnostic performance in patients with suspected myocarditis, including those with preserved LV function. Further investigations should focus on LA function, which appears to be more sensitive to early functional changes than LV function., (Copyright © 2018 Elsevier B.V. All rights reserved.)

Purpose: To test in a first proof-of-concept study whether texture analysis (TA) allows for the detection of myocardial tissue alterations in hypertrophic cardiomyopathy (HCM) on non-contrast T1-weighted cardiac magnetic resonance (CMR) images using machine learning based approaches., Methods: This retrospective, IRB-approved study included 32 patients with known HCM. Thirty patients with normal CMR served as controls. Regions-of-interest for TA encompassing the left ventricle were drawn on short-axis non-contrast T1-weighted images using a freely available software package. Step-wise dimension reduction and texture feature selection was performed for selecting features enabling the detection of myocardial tissue alterations in HCM patients on non-contrast T1-weighted CMR images., Results: Comparing HCM patients and controls, four texture features were identified showing significant differences between groups (Grey-level Non-uniformity [GLevNonU]: 74 ± 17 vs. 38 ± 9, p < .001; Energy of wavelet coefficients in low-frequency sub-bands [WavEnLL]: 58 ± 5 vs. 48 ± 10, p < .001; Fraction: 0.70 ± 0.07 vs. 0.78 ± 0.05, p < .001; Sum Average: 16.6 ± 0.4 vs. 17.0 ± 0.5, p = .007). A model containing the single parameter GLevNonU proved to be the best for differentiating between HCM patients and controls with a sensitivity/specificity of 91%/93%. A cut-off of GLevNonU ≥46 allowed for distinguishing HCM patients from controls with a sensitivity/specificity of 94%/90%. Even in patients without late gadolinium enhancement (LGE), the defined cut-off led to a differentiation of LGE- patients from healthy controls with 100% sensitivity and 90% specificity., Conclusions: TA on non-contrast T1-weighted images allows for the detection of myocardial tissue alterations in the setting of HCM with excellent accuracy, delivering potential novel parameters for a non-contrast assessment of myocardial texture alterations., (Copyright © 2018 Elsevier B.V. All rights reserved.)

Purpose: The present study aims at evaluating the feasibility and reproducibility of cardiac magnetic resonance (CMR) feature tracking (FT) derived strain and strain rate (SR) parameters of the left and right atrium (LA, RA) in patients with acute myocarditis as well as their potential to detect diastolic dysfunction. In addition, the diagnostic value of LA and RA strain parameters in the setting of acute myocarditis is investigated., Methods: CMR cine data of 30 patients with CMR-positive acute myocarditis were retrospectively analyzed. 25 age- and gender-matched healthy individuals served as a control. Analysis of longitudinal strain and SR of both atria was performed in two long-axis views using a dedicated FT-software. LA and RA deformation was analyzed including reservoir function (total strain [ε s ], peak positive SR [SR s ]), conduit function (passive strain [ε e ], peak early negative SR [SR e ]) and booster pump function (active strain [ε a ], peak late negative SR [SR a ]). Intra- and inter-observer reproducibility was assessed for all strain and SR parameters using Bland-Altman analyses, intra-class correlation coefficients (ICCs) and coefficients of variation (CV)., Results: FT analyses of both atria were feasible in all patients and controls. Reproducibility was good for reservoir and conduit function parameters and moderate for booster pump function parameters. Myocarditis patients demonstrated an impaired LA reservoir and conduit function when compared to healthy controls (LA ε s : 32±17 vs. 46±13, p=0.019; LA SR s : 1.5±0.5 vs. 1.8±0.5, p=0.117; LA SR e : -1.3±0.5 vs. -1.9±0.5, p<0.001), while LA booster pump function was preserved. In logistic regression and ROC-analyses, LA SR e proved to be the best independent predictor of acute myocarditis (AUC 0.80), and using LA SR e with a cut-off of -1.6s -1 resulted in a diagnostic sensitivity of 83% and a specificity of 80%. Changes in RA phasic function parameters showed a tendency to parallel those of the LA and showed no additional effect with respect to the diagnostic potential in acute myocarditis., Conclusions: Myocarditis patients exhibit an impaired atrial reservoir and conduit function, what might be indicative of ventricular diastolic dysfunction. LA SR e was the best predictor for the presence of acute myocarditis in our study, pointing towards the discriminative power of atrial strain analysis in the CMR-based diagnosis of acute myocarditis., (Copyright © 2017 Elsevier B.V. All rights reserved.)

Objectives: To investigate the reproducibility of regional and global strain and strain rate (SR) parameters of both ventricles and to determine sample sizes for all investigated strain and SR parameters in order to generate a practical reference for future studies., Materials and Methods: The study population consisted of 20 healthy individuals and 20 patients with acute myocarditis. Cine sequences in three horizontal long axis views and a stack of short axis views covering the entire left and right ventricle (LV, RV) were retrospectively analysed using a dedicated feature tracking (FT) software algorithm (TOMTEC). For intra-observer analysis, one observer analysed CMR images of all patients and volunteers twice. For inter-observer analysis, three additional blinded observers analysed the same datasets once. Intra- and inter-observer reproducibility were tested in all patients and controls using Bland-Altman analyses, intra-class correlation coefficients (ICCs) and coefficients of variation., Results: Intra-observer reproducibility of global LV strain and SR parameters was excellent (range of ICCs: 0.81-1.00), the only exception being global radial SR with a poor reproducibility (ICC 0.23). On a regional level, basal and midventricular strain and SR parameters were more reproducible when compared to apical parameters. Inter-observer reproducibility of all LV parameters was slightly lower than intra-observer reproducibility, yet still good to excellent for all global and regional longitudinal and circumferential strain and SR parameters (range of ICCs: 0.66-0.93). Similar to the LV, all global RV longitudinal and circumferential strain and SR parameters showed an excellent reproducibility, (range of ICCs: 0.75-0.97). Radial strain and SR parameters were less reproducible in the LV as well as in the RV when compared to circumferential or longitudinal parameters., Conclusion: CMR FT using the TOMTEC algorithm is highly reproducible in health and disease in both, the LV and RV on a global and regional level. The only exceptions are radial strain and SR parameters, which should be used with caution within clinical studies. The sample sizes estimated on the basis of the present study might serve as a reference during the planning of future FT studies., (Copyright © 2017 Elsevier B.V. All rights reserved.)

Purpose: The present study aims to evaluate the diagnostic value of cardiac magnetic resonance (CMR) feature tracking (FT) derived strain-analysis of both ventricles in patients with acute myocarditis (ACM) in order to improve its currently still challenging non-invasive diagnosis., Methods: CMR cine data of 31 patients with clinically suspected ACM and confirmation of diagnosis by CMR according to the Lake Louise criteria as well as 14 patients with clinically diagnosed ACM but inconspicuous CMR were retrospectively analyzed. 20 healthy volunteers (HV) served as a control. Analysis of global longitudinal, circumferential and radial strain and strain rate of both ventricles was performed in one long-axis and three short-axis slices using a dedicated FT-software (TomTec Imaging Systems)., Results: Patients with ACM showed significantly reduced LV longitudinal strain (-12.7 ± 6.5 vs. -16.8 ± 5.9%, p=0.021) and LV circumferential strain (LVCirStrain; -22.9 ± 5.7 vs. -27.8 ± 4.4 %, p<0.001) compared to HV. Conversely, they showed improved basal RV circumferential strain rate (BasalRVCirSR; -0.70 ± 0.23 vs. -0.47 ± 0.31s(-1), p=0.009). In ACM patients with preserved EF, BasalRVCirSR was significantly increased compared to HV while LV strain was not significantly different between both groups. In multinominal logistic regression analysis, LVCirStrain and BasalRVCirSR proved to be the best independent predictors of ACM with preserved EF. A combined cut-off of -0.53s(-1) for BasalRVCirSR and of -29.0% for LVCirStrain allowed a classification of ACM patients with preserved EF with a sensitivity of 89% and a specificity of 80%. Also patients with clinical ACM but inconspicuous CMR showed a significantly improved BasalRVCirSR and a cut-off of -0.77s(-1) allowed a classification of ACM patients with a sensitivity of 70% and a specificity of 90%, while all other CMR parameters were normal., Conclusions: The defined cut-offs for LVCirStrain and BasalRVCirSR allow a prediction of ACM with high sensitivity and specificity, even in patients with preserved EF and in patients with otherwise completely inconspicuous CMR. Our results point to a discriminative power especially of RV strain analysis in the CMR-based diagnosis of ACM., (Copyright © 2015 Elsevier Ireland Ltd. All rights reserved.)

Background: Previous studies showed that myocardial T2 relaxation times measured by cardiac T2-mapping vary significantly depending on sequence and field strength. Therefore, a systematic comparison of different T2-mapping sequences and the establishment of dedicated T2 reference values is mandatory for diagnostic decision-making., Methods: Phantom experiments using gel probes with a range of different T1 and T2 times were performed on a clinical 1.5T and 3T scanner. In addition, 30 healthy volunteers were examined at 1.5 and 3T in immediate succession. In each examination, three different T2-mapping sequences were performed at three short-axis slices: Multi Echo Spin Echo (MESE), T2-prepared balanced SSFP (T2prep), and Gradient Spin Echo with and without fat saturation (GraSEFS/GraSE). Segmented T2-Maps were generated according to the AHA 16-segment model and statistical analysis was performed., Results: Significant intra-individual differences between mean T2 times were observed for all sequences. In general, T2prep resulted in lowest and GraSE in highest T2 times. A significant variation with field strength was observed for mean T2 in phantom as well as in vivo, with higher T2 values at 1.5T compared to 3T, regardless of the sequence used. Segmental T2 values for each sequence at 1.5 and 3T are presented., Conclusions: Despite a careful selection of sequence parameters and volunteers, significant variations of the measured T2 values were observed between field strengths, MR sequences and myocardial segments. Therefore, we present segmental T2 values for each sequence at 1.5 and 3T with the inherent potential to serve as reference values for future studies., (Copyright © 2015 Elsevier Ireland Ltd. All rights reserved.)