Optimizing Synthetic Hematocrit for Cardiac MRI: A Multivariable Model Calibrated with Standardized Venous Sampling 

Optimizing Synthetic Hematocrit for Cardiac MRI: A Multivariable Model Calibrated with Standardized Venous Sampling | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Optimizing Synthetic Hematocrit for Cardiac MRI: A Multivariable Model Calibrated with Standardized Venous Sampling Çağrı Özcan¹, Hasan Yiğit, Mehmet Serkan Çetin, İrem Özcan, Oğuzhan Tokur This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6821726/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 22 Aug, 2025 Read the published version in The International Journal of Cardiovascular Imaging → Version 1 posted 8 You are reading this latest preprint version Abstract Background and Purpose Cardiac magnetic resonance (CMR) is a key modality for non-invasive extracellular volume (ECV) quantification, typically requiring venous hematocrit (Hct). However, fluctuations in Hct can affect accuracy. This study aimed to develop a synthetic Hct model integrating age, sex, and pre- and post-contrast blood pool T1 values, calibrated against an optimized venous Hct reference. Methods Cardiac MRI data from 253 patients, including those with various pathologies and normal findings, were retrospectively analyzed for ECV calculations. To minimize postural fluctuations, venous Hct samples were taken after the patients remained in a stable supine position for at least 15 minutes, just before contrast injection. Hct syn values were derived from pre-contrast and post-contrast T1 mapping sequences. A comprehensive Hct syn model was developed by integrating age, sex, and both pre- and post-contrast blood pool T1 values. Unlike previous studies, these variables were evaluated collectively, and the model was calibrated using an optimized venous Hct reference. The performance of the derived model at high ECV values was further evaluated in the hypertrophic cardiomyopathy (HCM) subcohort. Results The Hct syn model showed strong agreement with measured ECV (bias: 0.31%, RMSE: 1.25, CCC: 0.949). In the HCM subcohort, it maintained high correlation (r = 0.98) with a clinically acceptable bias of − 1.16%. Compared to sex-specific formulas by Chen et al., the model demonstrated improved performance (R² = 0.56) with lower variability. Conclusions This standardized synthetic Hct model enables accurate, non-invasive ECV estimation without blood sampling and demonstrates superior performance, especially in patients with elevated ECV values. Cardiac magnetic resonance imaging (CMR imaging) diffuse interstitial myocardial fibrosis extracellular volüme fraction (ECV fraction) Myocardial Mapping Hematocrit Figures Figure 1 Figure 2 Figure 3 Figure 4 INTRODUCTION Cardiac MRI has become an indispensable diagnostic tool for the non-invasive evaluation of cardiovascular pathologies, offering highly precise assessments, particularly in tissue characterization and volumetric analysis. Among these advanced capabilities, myocardial ECV estimation plays a pivotal role in detecting and monitoring myocardial diseases associated with fibrosis and infiltrative cardiomyopathies. ECV quantifies the myocardial extracellular space, providing insight into processes such as fibrosis, edema, and extracellular matrix expansion. These pathological features are critically involved in the pathophysiology of a wide range of cardiac conditions, including hypertrophic and dilated cardiomyopathy, myocarditis, myocardial infarction, amyloidosis, and hypertensive heart disease.[ 1 ] Traditionally, myocardial ECV is calculated using the R1 (1/T1) values obtained from pre- and post-contrast T1 mapping sequences, in combination with venous Hct, which is required to correct for the contrast agent’s volume distribution in blood. This estimation was originally proposed by Arheden et al.[ 2 , 3 ] using the following equation: \(\:ECV=\left(1-hct\right)\frac{\varDelta\:{R1}_{myo}}{\varDelta\:{R1}_{blood}}=(1-hct)\frac{\left(\frac{1}{{T1}_{myo\:post}}-\frac{1}{{T1}_{myo\:pre}}\right)}{\left(\frac{1}{{T1}_{blood\:post}}-\frac{1}{{T1}_{blood\:pre}}\right)}\) Venous Hct plays a critical role in determining the partition coefficient, which enables the separation of extracellular and intracellular compartments within the myocardial tissue. The Society for Cardiovascular Magnetic Resonance (SCMR) recommends that venous blood sampling for hematocrit (Hct) should be performed within 24 hours prior to the CMR scan to ensure accurate ECV estimation [ 1 ]. However, this approach is not always practical in routine clinical settings, as it requires additional effort, cost, and is prone to human error. One of the major challenges with venous Hct use lies in its susceptibility to diurnal and postural variations. The diurnal fluctuations of Hct throughout the day have been well documented, and Hct levels may vary depending on the time at which blood samples are collected [ 4 ]. Several studies have examined the impact of these temporal fluctuations on Hct measurements, with some reporting that while the effect may not always be statistically significant, it can introduce a minor bias in myocardial ECV estimation [ 5 – 7 ]. In a meta-analysis by Thirup et al., diurnal variability between two consecutive Hct measurements was reported to be as high as 12% [ 4 ]. Based on this, Su et al. highlighted that a theoretical 12% variation in Hct could result in an 8% difference in ECV values [ 5 ]. Additionally, postural changes, such as transitioning from an upright to a supine position, can lead to shifts in plasma volume, resulting in significant fluctuations in Hct levels. Lundvall et al. demonstrated that posture-induced changes in hemoglobin concentration can reach up to 11%, underscoring the magnitude of these effects [ 8 ]. This phenomenon, known as postural pseudoanemia, results in a decrease in Hct due to cranial fluid shifts and consequently leads to an overestimation of non-invasively calculated myocardial ECV. The time required to achieve near-maximal plasma volume shift varies across studies, typically ranging from 15 to 20 minutes [ 8 ]. In a study by Jakob et al., most changes in plasma volume occurred rapidly within the first 10 to 15 minutes after assuming a new posture [ 9 ]. When the effects of postural and diurnal fluctuations on Hct are considered together, significant deviations in myocardial ECV estimations may occur. This is particularly critical in conditions such as cardiac amyloidosis, where even small variations in ECV can have important clinical consequences. In these patients, ECV plays a pivotal role not only in diagnosis but also in monitoring disease progression and assessing therapeutic response [ 10 – 12 ]. Therefore, it is essential that ECV measurements be both accurate and consistent. However, minimizing such errors by standardizing the timing and conditions of venous Hct sampling is often challenging in routine clinical practice. To overcome these limitations, synthetic hematocrit models have been developed in recent years, based on the known linear relationship between the longitudinal relaxation rate of blood (R1 = 1/T1) and hematocrit [ 13 ]. These models eliminate the need for venous blood sampling by estimating Hct from blood pool T1 (T1 Blood ) values obtained through CMR T1 mapping sequences. Using T1 Blood values derived from both pre-contrast and post-contrast images, various regression models—including linear, reciprocal, and Deming regressions—have been formulated to estimate synthetic Hct, yielding promising results when compared to conventional ECV calculations based on venous Hct [ 6 , 7 , 13 – 23 ]. To improve accuracy, Opatril et al. developed models using linear and reciprocal regression formulas that incorporated both native T1Blood (T1-BN) and post-contrast T1Blood (T1-BP) values. Including post-contrast parameters significantly improved the statistical precision of the formula [ 16 ]. While it is well established that hematocrit values differ between sexes, emerging evidence suggests that both sex and age significantly influence T1 mapping parameters, thereby affecting synthetic Hct and ECV estimations [ 14 , 15 , 22 ]. Multiple studies—including our previous research—have demonstrated significant differences in myocardial T1 times between males and females, and have further shown that aging in females is associated with a marked reduction in T1 values [ 24 – 26 ]. These findings highlight the potential importance of incorporating sex and age as variables into synthetic Hct models to enable more accurate and personalized ECV estimations. In our study, to minimize potential variations in venous Hct measurements caused by timing and postural factors, blood samples were collected after patients had remained in a stable supine position for at least 15 minutes prior to contrast administration. This approach ensured the acquisition of reliable Hct values and aimed to enhance the accuracy of the resulting synthetic Hct models. Previous studies have extensively discussed the relationship between synthetic Hct and ECV, as well as the impact of blood pool T1 values and sex on these models. Rather than re-evaluating these variables individually, the present study aimed to develop a comprehensive model that integrates age, sex, and both pre- and post-contrast blood pool T1 values, calibrated against an optimized venous Hct reference. This approach was designed to offer a highly accurate, non-invasive, and clinically applicable method for myocardial ECV estimation. METHODS Study Design and Patient Population This single-center retrospective study was conducted between January 2024 and January 2025 at the Radiology Department of the Healthcare Research and Application Center of the University of Health Sciences in Ankara. Under the Declaration of Helsinki, the ethics committee approval was obtained by the Ankara Training and Research Hospital Clinical Research Ethics Committee on 08 January 2025 (Ethic number: E-24-360). Due to the retrospective nature of the study, obtaining additional consent from the patients was not required. Data from 629 patients who underwent CMR examinations between September 2015 and September 2023 were evaluated for eligibility in this study. Additionally, sex, age, body mass index (BMI), and cardiac diagnoses were collected as supplementary information. A total of 253 patients aged 18 and older, including both healthy individuals and those with various cardiovascular pathologies, were included for myocardial ECV calculation using CMR. Inclusion criteria were: (1) having venous Hct measurements obtained immediately before contrast agent injection; (2) having sufficient LGE-free myocardium in the mid-ventricular septal wall of the left ventricle for quantitative analysis. Exclusion criteria were: (1) patients with a history of cardiac surgery, (2) patients under 18 years of age, (3) patients with technical limitations affecting image quality ( Fig. 1 ) . Patients were randomly assigned to derivation (n = 177) and validation (n = 76) cohorts in a 70–30% ratio. The splitting process was performed using random allocation to simulate real-world generalizability and minimize selection bias. The derivation group was used to train the model and determine variable importance, while the validation group served to assess its performance independently. Baseline clinical and imaging variables were compared between the two groups to ensure balanced distribution. Of these included cases, 44 subjects had normal cardiac MRI findings, with no observed abnormalities in functional, structural, or quantitative analyses (Supplementary Materials) . To evaluate the accuracy of the model in the setting of elevated ECV values, a hypertrophic cardiomyopathy (HCM) subcohort (n = 70) was selected from the full cohort based on clinical diagnosis and cardiac MRI findings. The diagnosis of HCM was confirmed based on cardiac MRI imaging criteria, in accordance with the 2023 European Society of Cardiology guidelines for cardiomyopathies [ 27 ]. The performance of the synthetic Hct model was then specifically assessed within this subcohort. Imaging Protocol All patients' CMR examinations were performed at 1.5T (Magnetom Aera, Siemens Healthineers, Erlangen, Germany) using an 18-channel body coil. Myocardial T1 mapping was performed using an electrocardiogram - triggered modified Look-Locker inversion recovery (MOLLI) pulse sequence, both before and after the intravenous administration of a gadolinium-based contrast agent (0.15 mmol/kg, Gadovist, Bayer AG, Leverkusen, Germany). None of the patients were given any intravenous fluids, except for the small volume administered following the contrast injection. Post-contrast imaging was performed 15 minutes after the injection to ensure distribution equilibrium of the contrast agent. The MOLLI protocol used a 5(3)3 sampling scheme for native T1 mapping and a 4(1)3(1)2 sampling scheme for postcontrast T1 mapping. Scan parameters were as follows: TE/TR 1.12/2.8 ms; flip angle 35°; bandwidth 1085 Hz/Pixel; TI start 100 ms; TI increment 80 ms; slice thickness 5 mm; iPAT factor (GRAPPA) 2. In both pre- and post-contrast sequences, at least three short-axis slices, covering the basal, mid, and apical segments of the left ventricle, were sequentially acquired from the same anatomical locations. Imaging Analysis All cardiac MRI images were reviewed by a radiologist with three years of experience in cardiac MRI and eight years of experience in radiological imaging, using a diagnostic workstation (Leonardo Syngo MR E11, Siemens Healthineers). Regions of interest (ROIs) were manually delineated in the myocardium of the interventricular septum, carefully excluding both the endocardium and epicardium. Additionally, ROIs were drawn within the blood pool, avoiding papillary muscles, on the same midventricular short-axis slice for both pre- and post-contrast T1 mapping. To ensure accurate quantitative analysis, all included patients had a sufficient amount of LGE-free myocardium in the midventricular septum. Patients with less than 20 pixels of LGE-free myocardium available for ROI measurements were excluded from the cohort.[ 28 ] ( Fig. 2 ) After measuring the T1 times on the native and post-contrast maps, the \(\:\varDelta\:\) R1myo and \(\:\varDelta\:\) R1blood values from the ECV formula by Arheden et al.[ 3 ], as mentioned in the introduction section, were calculated. Pre- and post-contrast T1 values from the blood pool were then utilized to estimate synthetic Hct (Hct syn ) using locally derived models, which were subsequently used to quantify ECV. In HCM patients, manual ROI-based measurements were performed in the left ventricular wall, where pathological enhancement is most prominent and reliably quantifiable. These measurements were conducted in accordance with the predefined criteria described above and were evaluated using the same statistical methods applied in the validation subgroup. Hematocrit and Extracellular Volume Measurements Cardiac MRI examinations were generally conducted between 11:30 and 14:00 at our center. Venous Hct samples were collected just before contrast injection and after the patients had remained in a stable supine position for at least 15 minutes, in order to minimize variations caused by postural fluctuations in myocardial ECV calculations. While this standard protocol was applied to most patients, for a limited number of cases, blood samples were collected after a longer supine period due to repeated sequences caused by artifacts in the images obtained before contrast injection. Venous Hct (Hct measured ) values were measured using standard laboratory methods at the central laboratory of our hospital. These values were used in myocardial ECV (ECV measured ) calculations, defined as the reference ECV. This standardization aimed to minimize the effects of time- and position-related variations on ECV calculations. In the literature, numerous synthetic Hct formulas have been derived based on the linear relationship between Hct and the R1 value of blood. Opatril et al. incorporated post-contrast T1 blood values into synthetic Hct calculations, achieving results closer to venous Hct values [ 16 ]. Additionally, Chen et al. developed sex-specific (female/male), device-specific (1.5/3T), and blood pool-specific (LV/RV) models, reporting that “the specific model (Deming Regression)” provided more accurate results for synthetic Hct calculations [ 15 ]. Several studies have demonstrated significant myocardial and blood T1 differences between males and females, as well as a marked decrease in T1 times in women as they age [ 24 – 26 ]. In our latest study, we found significantly lower myocardial native T1 times in women over 50, considered postmenopausal, compared to younger women [ 24 ]. These findings suggest that considering sex and age in synthetic Hct calculations could be critical for achieving more personalized and accurate ECV estimates. This study aimed to develop a synthetic hematocrit model based on an optimized venous reference, integrating age, sex, and both pre- and post-contrast T1 blood values, to achieve more accurate estimations of synthetic Hct (Hct syn ) and ECV (ECV syn ). This model, along with the specific model developed by Chen et al.[ 15 ] (Hct reported ; ECV reported ), were tested in the validation cohort, and ECV measured was compared with ECV derived from synthetic Hcts and ECV reported . In accordance with the method described by Chen et al. [ 15 ], sex-specific formulas were applied separately to male and female participants to calculate ECV values. The Hct reported was calculated using the following formulas: for females, Hct = 1048 x (1/T1blood) − 0.2630 , and for male, Hct = 1319 x (1/T1blood) − 0.4318 . These sex-adjusted values were collectively analyzed under the heading of ‘ECV reported ’. Statistical Analysis Continuous variables were expressed as mean ± standard deviation or median with interquartile range, depending on data distribution. Categorical variables were presented as frequencies and percentages. Between-group comparisons were made using the Student’s t-test or Mann-Whitney U test for continuous variables, and the chi-square test or Fisher’s exact test for categorical variables, as appropriate. Variable selection and model development were carried out in the derivation cohort. Initially, univariate linear regression analyses were performed for each candidate predictor, including age, sex, native and post-contrast T1 values of blood, body mass index (BMI), and myocardial T1 values. To reduce overfitting and identify the most relevant predictors, least absolute shrinkage and selection operator (LASSO) regression was applied. Only variables retained by LASSO were entered into the multivariable linear regression model. The final model’s coefficients were determined using standard ordinary least squares estimation within the derivation group. Internal validation was performed by applying the model to the validation cohort and evaluating its predictive accuracy using the coefficient of determination (R²), root mean square error (RMSE), mean bias, and Lin’s concordance correlation coefficient (CCC). Agreement between measured and synthetic ECV values was first evaluated using Pearson correlation coefficients and scatter plots, followed by Bland–Altman analysis to assess bias and limits of agreement. A priori sample size estimation was performed based on Green’s rule for multiple linear regression (n ≥ 50 + 8k), which recommends a minimum sample size of 90 for five predictors (age, sex, pre- and post-contrast T1 values of blood, and BMI) [ 29 ]. Appropriate attention was paid to sample size considerations within the derivation cohort, aiming to exceed the recommended minimum and ensure adequate statistical power and stability of coefficient estimates. All descriptive and comparative statistical analyses were conducted using SPSS version 25 (IBM Corp., Armonk, NY). RESULTS Baseline Characteristics The overall cohort represented a diverse clinical population, including 142 patients with non-ischemic cardiomyopathy, 44 with normal cardiac findings, 18 with myocarditis, 16 with ischemic cardiomyopathy, and 11 with isolated heart failure diagnosed by CMR ( Table 1 ) . Table 1 Distribution of Cardiac Diagnoses in the Study Cohort Diagnosis Category Patient Count (n, %) Non-Ischemic CMP 142, 56.1% Normal 44, 17.4% Myocarditis 18, 7.1% Ischemic CMP 16, 6.3% Isolated Heart Failure (Identified by CMR) 11, 4.3% Myocarditis and Pericarditis 11, 4.3% Fabry Disease-Related Cardiomyopathy 3, 1.2% ARVC 2, 0.8% Cardiac Tumors 1, 0.4% Congenital Heart Diseases 1, 0.4% Infective Endocarditis 1, 0.4% Pericarditis 1, 0.4% Sarcoidosis 1, 0.4% Valvular Diseases 1, 0.4% Abbrevations : ARVC, Arrhythmogenic right ventricular cardiomyopathy; CMP, cardiomyopathy; CMR, cardiac magnetic resonance Baseline characteristics of the derivation cohort and validation cohort are summarized in Table 2 . The two cohorts were similar in age, sex, and baseline cardiac MRI parameters, with no statistically significant differences in Hct or ECV values. In addition, the baseline characteristics of the HCM subcohort are presented in Table 3 to provide detailed demographic and imaging information for this pathologic population. In our center, the reference range for native myocardial T1 values is 955.73–1056.14 ms for males and 984.52–1086.94 ms for females. In the normal subgroup of our study population, the mean native T1 values observed for both sexes remained within these reference limits. Additionally, the mean ECVmeasured was 24.42 ± 2.88% in males and 25.70 ± 2.62% in females, consistent with previously reported normal ranges. These findings suggest that both myocardial tissue characteristics and ECV values in the normal cohort were within expected physiological limits (Supplementary Materials). In the overall cohort, native myocardial T1 values were significantly higher in females than in males (1037.76 ± 37.16 ms vs. 1007.22 ± 43.20 ms, respectively; p = 8.38 × 10⁻⁹). Moreover, venous hematocrit levels were also significantly higher in males compared to females (p = 2.56 × 10⁻²⁴), and a modest inverse correlation was observed between age and Hct (r = − 0.26, p = 2.25 × 10⁻⁵). Derivation of the Synthetic Hematocrit (Hct) Model In the derivation cohort, a multivariable linear regression model was developed to estimate synthetic hematocrit (Hct syn ) by integrating age, sex, native T1 blood ( T 1 − BN ), and post-contrast T1 blood ( T 1 − BP ) values. Candidate predictors were first screened using least absolute shrinkage and selection operator (LASSO) regression to avoid overfitting and to identify the most predictive variables. Among the initial variables, age, sex, T 1 − BN , and T 1 − BP were retained by LASSO and subsequently entered into the final model. (Table 4 ) Table 2. Baseline Characteristics of the Derivation and Validation Cohorts Derivation Cohort N=177 (70%) Validation Cohort N=76 (30%) Variable Mean±SD Mean±SD P value Female Sex, N (%) 66 (37.3) 33 (43.4) 0.360 Age, years 42.5±17.2 42.9±16.5 0.881 BMI, kg/m 2 26.3±4.7 27.8±5.8 0.055 Heart Rate, bmp 74.3±12.3 73.8±11.3 0.794 LVEF, % 60.5±12.4 62.8±10.0 0.170 EDVI, ml/m 2 73.5±22.6 69.5±16.0 0.165 ESVI, ml/m 2 30.9±21.7 26.7±12.7 0.119 SVI, ml/m 2 42.6±8.9 42.8±8.5 0.869 CI, ml/min/m 2 3.1±0.7 3.1±0.6 0.790 T 1-BN , msn 1567.8±102.7 1567.9±97.3 0.996 T 1-BP , msn 286.4±59.2 281.7±47.5 0.539 T 1-MN , msn 1018.2±44.7 1021.6±40.8 0.569 T 1-MP , msn 440.2±56.6 433.8±46.1 0.390 Hct measured , % 42.2±4.7 42.2±4.4 0.929 Hct reported , % 41.7±4.2 41.5±3.9 0.739 ECV measured , % 25.8±3.6 26.2±4.1 0.408 ECV reported , % 26.0±3.7 26.5±3.8 0.338 Abbreviations: BMI, body-mass index; Bmp, beats per minute; CI, cardiac index; ECV, extracellular volume; EDVI, end-diastolic volume index; ESVI, end-systolic volume index; Hct, hematocrit; LVEF, left ventricular ejection fraction; SVI, stroke volume index; T 1-BN , native T1 blood; T 1-BP , post-contrast T1 blood; T 1-MN , native T1 myocard; T 1-MP , post-contrast T1 myocard Table 3. Demographic and Imaging Parameters in the HCM Subcohort (n=70) Variable Mean±SD Female Sex, N (%) 24 (34.29%) Age, years 52±15.2 BMI, kg/m 2 27.5 ± 4.3 LVEF, % 64.2±14.5 EDVI, ml/m 2 71.9±25.6 ESVI, ml/m 2 28.0±22.5 SVI, ml/m 2 43.9±11.4 T 1-BN , msn 1573.07 ± 109.83 T 1-BP , msn 297.6 ± 51.57 T 1-MN , msn 1124.31 ± 72.18 T 1-MP , msn 356.9 ± 77.25 Hct measured , % 42.5±5.9 Hct reported , % 41.2±5.3 Hct syn , % 44.4±3.7 ECV measured , % 41.8±13.0 ECV reported , % 43.0±13.9 ECV syn , % 40.6±12.8 Abbreviations: BMI, body-mass index; CI, cardiac index; ECV, extracellular volume; EDVI, end-diastolic volume index; ESVI, end-systolic volume index; Hct, hematocrit; LVEF, left ventricular ejection fraction; SVI, stroke volume index; T 1-BN , native T1 blood; T 1-BP , post-contrast T1 blood; T 1-MN , native T1 myocard; T 1-MP , post-contrast T1 myocard Table 4. Derivation of the Synthetic Hematocrit (Hct syn ) Model Using Age, Sex, and Blood T1 Values Variables LASSO Regression Univariate Analysis Multivariate Analysis ß Selected ß (95% CI) P value ß (95% CI) P value Standardized ß Age -0.413 Yes -0.083 (-0.122--0.044) <0.001 -0.027 (-0.056-0.001) 0.061 -0.099 Female ( Sex ) -1.452 Yes -5.604 (-6.789--4.419) <0.001 -3.097 (-4.188--2.006) <0.001 -0.318 T 1-BN -2.174 Yes -0.030 (-0.036--0.025) <0.001 -0.022 (-0.027--0.016) <0.001 -0.471 T 1-BP 0.644 Yes 0.020 (0.008-0.031) 0.001 0.012 (0.004-0.021) 0.004 0.149 BMI 0.0 No Abbreviations: BMI, body-mass index; T 1-BN , native T1 blood; T 1-BP , post-contrast T1 blood Candidate variables were first evaluated using Least Absolute Shrinkage and Selection Operator (LASSO) regression for selection. Only variables retained in LASSO were included in multivariable modeling. Univariate analyses are shown for descriptive purposes only. Despite a borderline p-value (0.061), age was retained in the multivariable model due to its selection by LASSO and its known clinical relevance in hematocrit variation. Adjusted R 2 of the model is 0.560. Hct syn = 77.906 - 0.022 x T 1-BN - 3.097 (if female) + 0.012 x T 1-BP - 0.027 x Age In univariate analyses, all four selected variables demonstrated significant associations with Hct syn (p < 0.001). In the multivariate model, T 1 − BN emerged as the strongest predictor (standardized ß = − 0.471), followed by female sex (ß = − 3.097, p < 0.001), T 1 − BP (ß = 0.012, p = 0.004), and age (ß = − 0.027, p = 0.061). Despite the borderline statistical significance of age, it was retained in the final model due to its known clinical relevance and consistent selection in the LASSO procedure. The final regression equation was: Hct syn = 77.906 − 0.022 x T 1 − BN − 3.097 (if female) + 0.012 x T 1 − BP − 0.027 x Age The model demonstrated good predictive performance, with an adjusted R² of 0.560 in the derivation set. This indicates that the model explained approximately 56% of the variability in venous Hct measurements. The inclusion of both pre- and post-contrast blood T1 values, as well as demographic variables, contributed to the robustness and individualized accuracy of the Hct syn prediction. Validation Performance In the validation cohort, ECV syn showed excellent correlation with ECV measured , with a Pearson correlation coefficient of r = 0.96, p < 0.001. ECV reported also showed a strong correlation (r = 0.95, p < 0.001) ( Fig. 3 ) . Performance metrics are summarized in Table 5 . ECV syn demonstrated an R² of 0.91, mean bias of + 0.31%, RMSE of 1.25%, and Lin's CCC of 0.949. In contrast, ECV reported showed an R² of 0.88, mean bias of + 0.40%, RMSE of 1.39%, and CCC of 0.942. Table 5. Performance Metrics of Synthetic and Reported ECV Compared to Measured ECV in the Validation Cohort Independent Variable R-squared Mean Bias (%) RMSE (%) Lin's CCC ECV reported 0.88 +0.40 1.39 0.942 ECV synthetic 0.91 +0.31 1.25 0.949 Abbreviations: ECV, extracellular volume; Lin's CCC, Lin's concordance correlation coefficient; RMSE, root mean square error Table 6. Performance Metrics of Synthetic and Reported ECV Compared to Measured ECV in the HCM Subcohort Independent Variable R-squared Mean Bias (%) RMSE (%) Lin's CCC ECV synthetic 0.953 -1.16 2.91 0.989 ECV reported 0.955 +1.21 2.81 0.992 Abbreviations: ECV, extracellular volume; Lin's CCC, Lin's concordance correlation coefficient; RMSE, root mean square error. Bland–Altman analysis revealed a mean bias of 0.31% and limits of agreement from − 2.08% to + 2.70%. For ECV reported , the mean bias was 0.40% with wider limits of agreement (− 2.23% to + 3.02%) ( Fig. 3 ) . Performance of Synthetic and Reported ECV Models in the HCM Subcohort In the HCM subcohort consisting of 70 individuals, ECV values calculated using the proposed synthetic hematocrit model (ECV syn ) were compared with reference values obtained using venous hematocrit (ECV measured ) and with literature-based values calculated using the sex-specific Chen’s model (ECV reported ). The average ECV measured in this subcohort was 41.82 ± 13.03%, reflecting a high burden of myocardial fibrosis. Based on standard ECV thresholds, 44.3% of patients exhibited severely elevated ECV (> 40%), while 47.1% fell within the 30–40% range, indicating significant fibrosis. Compared to the measured reference, ECV syn demonstrated a mean bias of − 1.16%, with RMSE of 2.91%, Lin’s CCC of 0.989, and a Pearson correlation coefficient (r) of 0.979 (p < 0.001) ( Table 6 ) . A strong linear correlation was observed with tight clustering along the regression line and narrow 95% confidence intervals. Bland–Altman analysis revealed limits of agreement ranging from − 6.43% to + 4.10% ( Fig. 4 ) . The mean ECV syn value was 40.65 ± 12.84%. In comparison, ECV reported showed a mean bias of + 1.21%, RMSE of 2.81%, CCC of 0.992, and r of 0.985. Limits of agreement for ECV reported ranged from − 3.80% to + 6.22%, with a mean value of 43.03 ± 13.99%. ( Fig. 4 )( Table 6 ) While both models showed strong correlation with ECV measured , the multivariable model produced more balanced estimates with slightly narrower dispersion, despite a modest underestimation trend. A freely accessible online calculator for synthetic ECV estimation based on the proposed model is available at https://syntheticecvcalculator.shinyapps.io/syntheticecvapp , with a corresponding QR code provided in the supplementary material. DISCUSSION This study introduces a novel synthetic hematocrit model that integrates age, sex, and both pre- and post-contrast blood pool T1 values, calibrated using an optimized venous Hct reference. The model was developed using 1.5T MOLLI-based cardiac MRI data and demonstrated clinical applicability in the HCM subcohort characterized by high ECV values. The key findings of this study are as follows: A new synthetic Hct model was developed that integrates age, sex, and both pre- and post-contrast blood T1 values, calibrated using an optimized venous Hct reference. Venous Hct measurements were obtained after patients remained supine for at least 15 minutes, minimizing posture-related hemodynamic variability and providing a physiologically reliable reference. The ECVsyn values derived from this model demonstrated lower bias (0.31%), lower RMSE (1.25), and stronger agreement (CCC = 0.949) compared to the ECVreported values calculated using the model of Chen et al.[ 15 ]. In the HCM subcohort, characterized by elevated ECV, the model showed a strong correlation (r = 0.98). Although a slight negative bias (− 1.16%) was observed, it remained within clinically acceptable limits. Calibration Approach in Venous Blood Sampling One of the major strengths of our study is the methodological rigor in venous Hct sampling. Unlike previous studies, all samples were obtained from patients after remaining in the supine position for at least 15 minutes, thereby minimizing posture-related hemodilution. This approach improved the physiological reliability of the model by addressing a factor often overlooked in previous studies such as those by Chen, Lim, and Yin.[ 14 , 15 , 18 ] Opatril et al. [ 16 ] showed that including post-contrast T1 values improved prediction accuracy; however, their model did not include important demographic variables like age and sex, and did not standardize venous Hct collection. Similarly, Lim et al. [ 18 ] developed a more accurate sex-specific model, but it did not incorporate post-contrast T1 values. Our study is the first to integrate all these parameters into a single model, enhancing its clinical adaptability. Model Performance and Comparison with the Literature In terms of explanatory power, our synthetic Hct model outperformed previously published models. Among all models developed using 1.5T and 3T MRI systems, ours achieved the highest coefficient of determination (R² = 0.56). This value was higher than those reported by Opatril (R² = 0.49), Su (R² = 0.51), Kammerlander (R² = 0.284), and Treibel (R² = 0.50) [ 5 , 13 , 16 , 19 , 21 , 30 ]. In the sex-specific models developed by Chen et al., R² was 0.43 for males and 0.27 for females [ 15 ]. These findings underscore the value of incorporating demographic and imaging-based variables, which appears to significantly enhance both the accuracy and generalizability of the model. Our model achieved high accuracy for ECV estimation, with r = 0.974, CCC = 0.949, and narrow Bland–Altman limits of agreement (− 2.08% to + 2.70%). This accuracy was maintained even in the HCM subcohort (r = 0.98), demonstrating the model’s robustness against pathological variation. In healthy individuals, myocardial ECV values are typically reported as 25.3 ± 3.5% and 25.4 ± 2.5%, and an error range of ± 2.5–3.5% is generally considered acceptable [ 15 , 31 ]. In our study, the prediction error of the proposed model remained within this accepted range even in the fibrotic HCM subgroup. Our findings are consistent with the growing literature on synthetic ECV and contribute additional advancements. In a large 1.5T and 3T study by Chen et al., synthetic ECVs were reported to correlate well with histologically measured collagen volume fractions, similarly to conventional measured ECVs (R² = 0.87–0.88). The average difference between synthetic and conventional ECV was less than 0.2%, with agreement limits around ± 4%, comparable to our findings [ 15 ]. Compared with ECV reported values derived using their sex-specific model, our model yielded lower bias, stronger agreement, and slightly narrower limits of agreement ( ~ ± 2.5%), which we attribute to our rigorous imaging protocol and the inclusion of variables beyond pre-contrast blood T1. Robinson et al. [ 6 ] also reported a strong correlation between synthetic and measured ECV (r = 0.92), with a mean bias of 0.23% and agreement limits between − 2.82% and + 3.27%. Although this model also performed well, its total LoA width of 6.09% suggests more variability. In contrast, our model provided slightly higher correlation (r = 0.974), a modestly higher bias (0.31%), but a narrower LoA (− 2.08% to + 2.70%). These results suggest that while both models perform well, our model offers advantages in minimizing both systematic error and variability. In the study by Kammerlander et al., bias was negligible (+ 0.007%), but agreement limits were wider (− 4.32% to + 4.33%). Additionally, our model achieved a stronger correlation (r = 0.96, R² = 0.92) compared to their reported r = 0.943 and R² = 0.889 [ 19 ]. Although Yin et al. [ 14 ] reported a very low bias (+ 0.05%), their LoA also ranged widely (− 4.22% to + 4.31%). These results collectively suggest that while all synthetic models fall within clinically acceptable limits, our approach may offer more reliable and consistent results at the individual patient level. This advantage likely stems from the integration of age, sex, and both pre- and post-contrast T1 blood values, along with a carefully optimized Hct sampling protocol. The ECV syn model demonstrated excellent agreement with the reference ECV measured values in the HCM subcohort (bias: −1.16%, CCC: 0.989), with clinically acceptable agreement limits (− 6.43% to + 4.10%). Although the ECV reported model also showed strong concordance (CCC: 0.992), it exhibited a slight positive bias (+ 1.21%) and wider LoA [ 15 ]. The high mean ECV in the HCM subcohort (41.82%) confirms that the model was tested in a fibrotic myocardial population. The slight underestimation may be attributable to the model being trained predominantly on normal-to-intermediate ECV ranges. Nevertheless, the ECV syn model yielded highly correlated and balanced predictions, even in patients with high ECV values. These findings were further supported by high coefficients of determination (R² = 0.953 for ECV syn and R² = 0.955 for ECV reported ), confirming the strong linear relationship of both models with ECV measured . Collectively, these results support the feasibility of accurate synthetic ECV estimation in fibrotic patient populations without the need for invasive blood sampling. All synthetic models, including ours, assume a fixed timing for post-contrast T1 mapping. Although we maintained a consistent post-contrast interval during image acquisition, variation in this timing in clinical practice may affect blood contrast equilibrium and alter the T1–Hct relationship. This could reduce the accuracy of models incorporating post-contrast values. Therefore, protocol standardization is essential for reliable application. Although a prototype Shiny web interface for ECV syn calculation was developed, ideal implementation would involve integration into standard CMR post-processing platforms or directly into scanner software. For example, vendors could provide automated “synthetic ECV” generation immediately following T1 mapping, eliminating the need for venous sampling and enabling seamless integration into clinical workflow. Clinical Implications T1 mapping–based CMR is considered the gold standard for noninvasive evaluation of diffuse myocardial fibrosis. However, the need for venous Hct and its biological variability can limit clinical use. Our synthetic model eliminates the need for blood sampling during CMR, significantly streamlining the ECV quantification process. In addition to saving time and resources, this also enhances patient comfort and safety by avoiding an invasive step. Real-time synthetic ECV computation allows the reporting physician to access tissue characterization metrics immediately upon image acquisition. In cases with borderline or unexpected values, additional sequences or validation can be performed while the patient is still on the table. Furthermore, a generalizable synthetic Hct model enables ECV quantification in settings without immediate access to laboratory services or Hct devices, such as in after-hours emergency scans. It may also be useful in retrospective studies where blood samples are missing but T1 maps are available. Inclusion of age- and sex-specific variables improves precision across diverse physiological conditions. Ultimately, routine adoption of synthetic ECV could expand the role of quantitative CMR tissue characterization and increase its availability in both clinical and research settings. Given that ECV is a prognostically relevant marker of diffuse myocardial fibrosis, this model could facilitate broader integration of ECV into patient care pathways. The high agreement and low bias observed in our study indicate that synthetic ECV may be a reliable substitute for conventional ECV in clinical practice. Limitations Despite the promising performance of our model, certain limitations should be acknowledged. This was a single-center study using a single 1.5T scanner and MOLLI T1 mapping sequence. Although internal validation was robust, external performance remains untested. However, inclusion of a diagnostically heterogeneous cohort, comprising non-ischemic and ischemic cardiomyopathies, myocarditis, isolated heart failure, and structurally normal hearts, provides preliminary evidence of model generalizability across myocardial conditions. Differences in T1 mapping techniques or patient demographics may require model recalibration. As previously noted in the literature, synthetic Hct models may be affected by hardware configuration, magnetic field strength, and sequence parameters, potentially necessitating site-specific calibration [ 13 , 21 ]. Some studies suggest reduced accuracy of synthetic ECV at hematocrit extremes [ 5 , 15 ]. Although our cohort did not exclude patients based on Hct, severely anemic or polycythemic individuals were not present, and thus model performance in these settings remains uncertain. Similarly, infiltrative cardiomyopathies such as cardiac amyloidosis were underrepresented, limiting our assessment in this group. Future multicenter studies are warranted to confirm generalizability and evaluate model performance across diverse Hct values and pathologies. Finally, our analysis included only adult patients. Previous studies, including that by Raucci et al. (2017), have highlighted the risk of clinically significant error when using synthetic Hct models in pediatric populations [ 22 ]. Thus, separate validation is necessary for pediatric use. CONCLUSION This study presents a synthetic hematocrit model based on age, sex, and both pre- and post-contrast T1 values, calibrated using an optimized venous reference. The resulting ECV syn estimates demonstrated superior accuracy and agreement compared to previous models. This approach offers a noninvasive and clinically viable alternative for ECV quantification and brings synthetic ECV one step closer to routine clinical use. Abbreviations BMI: Body mass index CCC: Concordance correlation coefficient CMR: Cardiac magnetic resonance ECV: Extracellular volume HCM: Hypertrophic cardiomyopathy Hct: Hematocrit LASSO: Least absolute shrinkage and selection operator MOLLI: Modified look-locker inversion recovery RMSE: Root mean square error SCMR: The Society for Cardiovascular Magnetic Resonance Declarations Acknowledgements Not applicable. Author contributions Conception and study design: Çağrı Özcan (ÇÖ), Hasan Yiğit (HY). Conduction of experiments: ÇÖ, HY, İrem Özcan. Interpretation of data: ÇÖ, HY, Mehmet Serkan Çetin, Oğuzhan Tokur. Draft of manuscript: ÇÖ. All authors read and approved the final manuscript. Funding Not applicable. Ethics approval and consent to participate Under the Declaration of Helsinki, the ethics committee approval was obtained by the Ankara Training and Research Hospital Clinical Research Ethics Committee on 08 January 2025. The ethics committee application number was E-24-360. Consent for publication Due to the retrospective nature of the study, obtaining additional consent from the patients was not required. Compenting interests The authors declare that they have no competing interests. Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this work, the author(s) used OpenAI’s ChatGPT in order to assist with English language editing, clarity improvements, and academic phrasing during the drafting of the manuscript’s Introduction, Methods, Discussion and Conclusion sections. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication. References Messroghli DR, Moon JC, Ferreira VM, et al. Clinical recommendations for cardiovascular magnetic resonance mapping of T1, T2, T2* and extracellular volume: a consensus statement by the Society for Cardiovascular Magnetic Resonance (SCMR) endorsed by the European Association for Cardiovascular Imaging (EACVI). Journal of Cardiovascular Magnetic Resonance 2017; 19:1-24 Arheden Hk, Saeed M, Higgins CB, et al. Reperfused rat myocardium subjected to various durations of ischemia: estimation of the distribution volume of contrast material with echo-planar MR imaging. Radiology 2000; 215:520-528 Arheden Hk, Saeed M, Higgins CB, et al. Measurement of the distribution volume of gadopentetate dimeglumine at echo-planar MR imaging to quantify myocardial infarction: comparison with 99mTc-DTPA autoradiography in rats. Radiology 1999; 211:698-708 Thirup P. Haematocrit: within-subject and seasonal variation. Sports Medicine 2003; 33:231-243 Su M-Y, Huang Y-S, Niisato E, et al. Is a timely assessment of the hematocrit necessary for cardiovascular magnetic resonance–derived extracellular volume measurements? Journal of Cardiovascular Magnetic Resonance 2020; 22:77 Robison S, Karur GR, Wald RM, Thavendiranathan P, Crean AM, Hanneman K. Noninvasive hematocrit assessment for cardiovascular magnetic resonance extracellular volume quantification using a point-of-care device and synthetic derivation. Journal of Cardiovascular Magnetic Resonance 2018; 20:19 Engblom H, Kanski M, Kopic S, et al. Importance of standardizing timing of hematocrit measurement when using cardiovascular magnetic resonance to calculate myocardial extracellular volume (ECV) based on pre-and post-contrast T1 mapping. Journal of Cardiovascular Magnetic Resonance 2018; 20:46 Lundvall J, Bjerkhoel P, Quittenbaum S, Lindgren P. Rapid plasma volume decline upon quiet standing reflects large filtration capacity in dependent limbs. Acta physiologica scandinavica 1996; 158:161-167 Jacob G, Ertl AC, Shannon JR, Furlan R, Robertson RM, Robertson D. Effect of standing on neurohumoral responses and plasma volume in healthy subjects. Journal of Applied Physiology 1998; 84:914-921 Fontana M, Banypersad SM, Treibel TA, et al. Differential myocyte responses in patients with cardiac transthyretin amyloidosis and light-chain amyloidosis: a cardiac MR imaging study. Radiology 2015; 277:388-397 Banypersad SM, Sado DM, Flett AS, et al. Quantification of myocardial extracellular volume fraction in systemic AL amyloidosis: an equilibrium contrast cardiovascular magnetic resonance study. Circulation: Cardiovascular Imaging 2013; 6:34-39 Syed IS, Glockner JF, Feng D, et al. Role of cardiac magnetic resonance imaging in the detection of cardiac amyloidosis. JACC: cardiovascular imaging 2010; 3:155-164 Treibel TA, Nasis A, Fontana M, et al. An instantaneous ECV with no blood sampling: using native blood T1 for hematocrit is as good as standard ECV. Journal of Cardiovascular Magnetic Resonance 2015; 17:1-2 Yin J, Qin J, Liu W, et al. A comparative study of synthetic and venous hematocrit for calculating cardiovascular magnetic resonance-derived extracellular volume. The International Journal of Cardiovascular Imaging 2024:1-10 Chen W, Doeblin P, Al-Tabatabaee S, et al. Synthetic extracellular volume in cardiac magnetic resonance without blood sampling: a reliable tool to replace conventional extracellular volume. Circulation: Cardiovascular Imaging 2022; 15:e013745 Opatril L, Panovsky R, Machal J, et al. Extracellular volume quantification using synthetic haematocrit assessed from native and post-contrast longitudinal relaxation T1 times of a blood pool. BMC Cardiovascular Disorders 2021; 21:363 Shang Y, Zhang X, Zhou X, Wang J. Extracellular volume fraction measurements derived from the longitudinal relaxation of blood-based synthetic hematocrit may lead to clinical errors in 3 T cardiovascular magnetic resonance. Journal of Cardiovascular Magnetic Resonance 2018; 20:56 Lim E-H, Le T-T, Bryant J, et al. Importance of sex-specific regression models to estimate synthetic hematocrit and extracellular volume fraction. JACC: Cardiovascular Imaging 2018; 11:1366-1367 Kammerlander AA, Duca F, Binder C, et al. Extracellular volume quantification by cardiac magnetic resonance imaging without hematocrit sampling: ready for prime time? Wiener klinische Wochenschrift 2018; 130:190-196 Fent GJ, Garg P, Foley JR, et al. Synthetic myocardial extracellular volume fraction. JACC: Cardiovascular Imaging 2017; 10:1402-1404 Treibel TA, Fontana M, Maestrini V, et al. Automatic measurement of the myocardial interstitium: synthetic extracellular volume quantification without hematocrit sampling. JACC: Cardiovascular Imaging 2016; 9:54-63 Raucci Jr FJ, Parra DA, Christensen JT, et al. Synthetic hematocrit derived from the longitudinal relaxation of blood can lead to clinically significant errors in measurement of extracellular volume fraction in pediatric and young adult patients. Journal of Cardiovascular Magnetic Resonance 2016; 19:58 Bluemke DA, Kawel-Boehm N. Can a MR imaging scanner accurately measure hematocrit to determine ECV fraction? In: American College of Cardiology Foundation Washington, DC, 2016:64-66 Özcan Ç, Yiğit H, Çetin MS, Özcan İ. Analysis of myocardial T1, T2, and T2* values by age, sex, and cardiac segments in normal population: a prospective study. The International Journal of Cardiovascular Imaging 2024; Gottbrecht M, Kramer CM, Salerno M. Native T1 and extracellular volume measurements by cardiac MRI in healthy adults: a meta-analysis. Radiology 2019; 290:317 Piechnik SK, Ferreira VM, Lewandowski AJ, et al. Normal variation of magnetic resonance T1 relaxation times in the human population at 1.5 T using ShMOLLI. Journal of Cardiovascular Magnetic Resonance 2013; 15:13 Arbelo E, Protonotarios A, Gimeno JR, et al. 2023 ESC Guidelines for the management of cardiomyopathies. Eur Heart J 2023; 44:3503-3626 Schulz-Menger J, Bluemke DA, Bremerich J, et al. Standardized image interpretation and post-processing in cardiovascular magnetic resonance - 2020 update: Society for Cardiovascular Magnetic Resonance (SCMR): Board of Trustees Task Force on Standardized Post-Processing. Journal of Cardiovascular Magnetic Resonance 2020; 22:19 Green SB. How many subjects does it take to do a regression analysis. Multivariate behavioral research 1991; 26:499-510 Biso S, Weber J, Philip S, Omar K. Excellent Reproducibility of Synthetic ECV Without Blood Extraction Across Different Cardiomyopathies Using Published Regression. Journal of Cardiovascular Magnetic Resonance 2024; 26 Kellman P, Wilson JR, Xue H, et al. Extracellular volume fraction mapping in the myocardium, part 2: initial clinical experience. J Cardiovasc Magn Reson 2012; 14:64 Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterials.docx Cite Share Download PDF Status: Published Journal Publication published 22 Aug, 2025 Read the published version in The International Journal of Cardiovascular Imaging → Version 1 posted Editorial decision: Revision requested 09 Aug, 2025 Reviews received at journal 07 Aug, 2025 Reviewers agreed at journal 21 Jul, 2025 Reviewers agreed at journal 16 Jun, 2025 Reviewers invited by journal 10 Jun, 2025 Editor assigned by journal 04 Jun, 2025 Submission checks completed at journal 04 Jun, 2025 First submitted to journal 04 Jun, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6821726","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":469587899,"identity":"ce0adb52-06b8-440a-880c-942fd63d9a8c","order_by":0,"name":"Çağrı Özcan¹","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA70lEQVRIiWNgGAWjYFCCBDDJA+FUADEzcwMpWs6AtDASpwUCGNvAJH4tuu0JjB9+/Lonw89/+OHnwnm10fztQC0/Krbh1GJ25gGzZG9fMY9kwzFj6ZnbjufOOMzYwNhz5jZuLTcS2Bh4exJ4DA72MEjzbjuW2wDUwszYhl8L41+QlsM8zL955xzLnU+MFmaeH0Atx3jYpHkbanI3ENRy5mGztGxDAo9kD5uZNc+xA7kbgVoO4vXL8eSDH9/8SbAHhtjj2zw1dbnzzh8++OBHBW4t4FiARAcYHAaTB/Coh4I/cFYdYcWjYBSMglEw4gAAsW9ZH6TohtwAAAAASUVORK5CYII=","orcid":"","institution":"Ankara Etimesgut Şehit Sait Ertürk State Hospital","correspondingAuthor":true,"prefix":"","firstName":"Çağrı","middleName":"","lastName":"Özcan¹","suffix":""},{"id":469587900,"identity":"a052d196-4997-4929-9117-1597a6e0358b","order_by":1,"name":"Hasan Yiğit","email":"","orcid":"","institution":"Ankara Training and Research Hospital","correspondingAuthor":false,"prefix":"","firstName":"Hasan","middleName":"","lastName":"Yiğit","suffix":""},{"id":469587901,"identity":"d601a7da-7393-4ee7-8ab6-b33a5b134faa","order_by":2,"name":"Mehmet Serkan Çetin","email":"","orcid":"","institution":"Ankara Bilkent City Hospital","correspondingAuthor":false,"prefix":"","firstName":"Mehmet","middleName":"Serkan","lastName":"Çetin","suffix":""},{"id":469587902,"identity":"1f73728d-8d10-4cd5-93a8-306bf8415902","order_by":3,"name":"İrem Özcan","email":"","orcid":"","institution":"Ankara Gülhane Training and Research Hospital","correspondingAuthor":false,"prefix":"","firstName":"İrem","middleName":"","lastName":"Özcan","suffix":""},{"id":469587903,"identity":"d8228aec-f65b-4f49-98d9-65d0c167d6ce","order_by":4,"name":"Oğuzhan Tokur","email":"","orcid":"","institution":"Kütahya Health Sciences University","correspondingAuthor":false,"prefix":"","firstName":"Oğuzhan","middleName":"","lastName":"Tokur","suffix":""}],"badges":[],"createdAt":"2025-06-04 14:53:35","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6821726/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6821726/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10554-025-03504-9","type":"published","date":"2025-08-22T16:28:55+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":84670233,"identity":"67ba4955-b145-49ea-8dd5-cc65d5e59bdd","added_by":"auto","created_at":"2025-06-16 06:36:11","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":170762,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eWorkflow. \u003c/strong\u003eOut of the 376 excluded patients, 211 were excluded due to unavailable Hct values for ECV calculation. Nine patients had a history of cardiac surgery. Twenty-five were excluded due to poor image quality, defined as artifacts in the mid-ventricular septum affecting pre- or post-contrast T1 mapping measurements. Thirty-two patients lacked sufficient LGE-free myocardium in the mid-ventricular septal wall for quantitative analysis. Additionally, 99 patients were excluded for being under 18 years of age. The remaining patients were randomly assigned to derivation (n=177) and validation (n=76) cohorts in a 70% to 30% ratio. ECV, extracellular volume; Hct, hematocrit.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6821726/v1/ac59f4727d2db9c32d10a173.png"},{"id":84671353,"identity":"32feb120-9f46-45cd-97b0-7428a28842bb","added_by":"auto","created_at":"2025-06-16 06:44:11","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":670985,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRepresentative CMR images demonstrating ROI placement for ECV and synthetic Hct measurements.\u003c/strong\u003eIn the same mid-ventricular short-axis slice, ROIs were manually placed in the interventricular septum on both (A) pre-contrast and (C) post-contrast T1 maps, carefully excluding the endocardium and epicardium. Blood pool ROIs were also drawn in the same slice, avoiding papillary muscles. (B) PSIR images were used to identify and exclude areas with late gadolinium enhancement (LGE). Only slices with artifact-free images and at least 20 pixels of LGE-free myocardium were included for T1 measurements. The obtained values were used to calculate ∆R1myo and ∆R1blood, which were then applied in both measured and synthetic ECV calculations.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6821726/v1/88a4c805641497b72a3472f2.png"},{"id":84670234,"identity":"f34dfd73-d13a-4030-b8e7-289b8f238775","added_by":"auto","created_at":"2025-06-16 06:36:11","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":227898,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCorrelation analyses between synthetic/reported and measured ECV values in the validation cohort.\u003c/strong\u003e \u003cstrong\u003e(A)\u003c/strong\u003e Scatter plot illustrating the strong linear relationship between ECV\u003csub\u003esyn\u003c/sub\u003e and ECV\u003csub\u003emeasured\u003c/sub\u003e values in the validation cohort (n = 76). The red dashed line represents the linear regression fit with 95% confidence interval (shaded area). The Pearson correlation coefficient was \u003cem\u003er\u003c/em\u003e = 0.96 (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001), indicating excellent agreement.\u003cstrong\u003e (B)\u003c/strong\u003e Scatter plot showing the correlation between ECV\u003csub\u003ereported\u003c/sub\u003e and ECV\u003csub\u003emeasured\u003c/sub\u003e (\u003cem\u003er = 0.95, p \u0026lt; 0.001\u003c/em\u003e), indicating a comparably strong association. \u003cstrong\u003e(C)\u003c/strong\u003e Bland–Altman plot comparing ECV\u003csub\u003esyn\u003c/sub\u003e and ECV\u003csub\u003emeasured\u003c/sub\u003e in the validation cohort. The mean bias was 0.31% with limits of agreement ranging from −2.08% to 2.70%. Dashed lines indicate the mean bias (blue) and ±1.96 standard deviations (red).\u003cstrong\u003e (D)\u003c/strong\u003e Bland–Altman plot for ECV\u003csub\u003ereported\u003c/sub\u003e vs. ECV\u003csub\u003emeasured\u003c/sub\u003e, showing a mean bias of 0.40% (limits of agreement: −2.23% to +3.02%).\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6821726/v1/ca0613eb1b21480a9ac9be01.png"},{"id":84670225,"identity":"4643490e-e2ee-46f4-bd75-2155e94a8d6c","added_by":"auto","created_at":"2025-06-16 06:36:10","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":341268,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCorrelation analyses between synthetic/reported and measured ECV values in the HCM subcohort.\u003c/strong\u003e \u003cstrong\u003e(A)\u003c/strong\u003eScatter plot demonstrating the strong linear relationship between ECV\u003csub\u003esyn\u003c/sub\u003e and ECV\u003csub\u003emeasured\u003c/sub\u003e (n = 70). The red dashed line represents the linear regression fit with the shaded area indicating the 95% confidence interval. Pearson correlation was r = 0.98 (p\u0026lt;0.001), suggesting excellent agreement. \u003cstrong\u003e(B)\u003c/strong\u003e Scatter plot showing the correlation between ECV\u003csub\u003ereported\u003c/sub\u003e and ECV\u003csub\u003emeasured \u003c/sub\u003e(r = 0.98, p\u0026lt;0.001), similarly indicating a strong linear association. \u003cstrong\u003e(C)\u003c/strong\u003e Bland–Altman plot comparing ECV\u003csub\u003esyn\u003c/sub\u003e with ECV\u003csub\u003emeasured\u003c/sub\u003e. The mean bias was −1.16% with limits of agreement ranging from −6.43% to +4.10%.\u003cbr\u003e\n\u003cstrong\u003e(D)\u003c/strong\u003e Bland–Altman plot for ECV\u003csub\u003ereported\u003c/sub\u003e vs. ECV\u003csub\u003emeasured\u003c/sub\u003e, showing a mean bias of +1.21% and limits of agreement between −3.80% and +6.22%. Dashed lines indicate the mean bias (blue) and ±1.96 SD range (red).\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6821726/v1/d092d5ff7365ad023f593a46.png"},{"id":89847159,"identity":"5d15a21d-be18-4ad6-b861-a66780a70467","added_by":"auto","created_at":"2025-08-25 16:41:27","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3118038,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6821726/v1/0ca73210-ab1f-4904-9c20-dbc4699ee45e.pdf"},{"id":84670221,"identity":"0ed947dc-90ac-43aa-b26d-236e6b7588c4","added_by":"auto","created_at":"2025-06-16 06:36:10","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1766737,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterials.docx","url":"https://assets-eu.researchsquare.com/files/rs-6821726/v1/6bf5b932730ee05e146a90df.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Optimizing Synthetic Hematocrit for Cardiac MRI: A Multivariable Model Calibrated with Standardized Venous Sampling ","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eCardiac MRI has become an indispensable diagnostic tool for the non-invasive evaluation of cardiovascular pathologies, offering highly precise assessments, particularly in tissue characterization and volumetric analysis. Among these advanced capabilities, myocardial ECV estimation plays a pivotal role in detecting and monitoring myocardial diseases associated with fibrosis and infiltrative cardiomyopathies. ECV quantifies the myocardial extracellular space, providing insight into processes such as fibrosis, edema, and extracellular matrix expansion. These pathological features are critically involved in the pathophysiology of a wide range of cardiac conditions, including hypertrophic and dilated cardiomyopathy, myocarditis, myocardial infarction, amyloidosis, and hypertensive heart disease.[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eTraditionally, myocardial ECV is calculated using the R1 (1/T1) values obtained from pre- and post-contrast T1 mapping sequences, in combination with venous Hct, which is required to correct for the contrast agent\u0026rsquo;s volume distribution in blood. This estimation was originally proposed by Arheden et al.[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e] using the following equation:\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:ECV=\\left(1-hct\\right)\\frac{\\varDelta\\:{R1}_{myo}}{\\varDelta\\:{R1}_{blood}}=(1-hct)\\frac{\\left(\\frac{1}{{T1}_{myo\\:post}}-\\frac{1}{{T1}_{myo\\:pre}}\\right)}{\\left(\\frac{1}{{T1}_{blood\\:post}}-\\frac{1}{{T1}_{blood\\:pre}}\\right)}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e \u003cp\u003eVenous Hct plays a critical role in determining the partition coefficient, which enables the separation of extracellular and intracellular compartments within the myocardial tissue.\u003c/p\u003e \u003cp\u003eThe Society for Cardiovascular Magnetic Resonance (SCMR) recommends that venous blood sampling for hematocrit (Hct) should be performed within 24 hours prior to the CMR scan to ensure accurate ECV estimation [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. However, this approach is not always practical in routine clinical settings, as it requires additional effort, cost, and is prone to human error. One of the major challenges with venous Hct use lies in its susceptibility to diurnal and postural variations. The diurnal fluctuations of Hct throughout the day have been well documented, and Hct levels may vary depending on the time at which blood samples are collected [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Several studies have examined the impact of these temporal fluctuations on Hct measurements, with some reporting that while the effect may not always be statistically significant, it can introduce a minor bias in myocardial ECV estimation [\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. In a meta-analysis by Thirup et al., diurnal variability between two consecutive Hct measurements was reported to be as high as 12% [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Based on this, Su et al. highlighted that a theoretical 12% variation in Hct could result in an 8% difference in ECV values [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Additionally, postural changes, such as transitioning from an upright to a supine position, can lead to shifts in plasma volume, resulting in significant fluctuations in Hct levels. Lundvall et al. demonstrated that posture-induced changes in hemoglobin concentration can reach up to 11%, underscoring the magnitude of these effects [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. This phenomenon, known as postural pseudoanemia, results in a decrease in Hct due to cranial fluid shifts and consequently leads to an overestimation of non-invasively calculated myocardial ECV. The time required to achieve near-maximal plasma volume shift varies across studies, typically ranging from 15 to 20 minutes [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. In a study by Jakob et al., most changes in plasma volume occurred rapidly within the first 10 to 15 minutes after assuming a new posture [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWhen the effects of postural and diurnal fluctuations on Hct are considered together, significant deviations in myocardial ECV estimations may occur. This is particularly critical in conditions such as cardiac amyloidosis, where even small variations in ECV can have important clinical consequences. In these patients, ECV plays a pivotal role not only in diagnosis but also in monitoring disease progression and assessing therapeutic response [\u003cspan additionalcitationids=\"CR11\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Therefore, it is essential that ECV measurements be both accurate and consistent. However, minimizing such errors by standardizing the timing and conditions of venous Hct sampling is often challenging in routine clinical practice.\u003c/p\u003e \u003cp\u003eTo overcome these limitations, synthetic hematocrit models have been developed in recent years, based on the known linear relationship between the longitudinal relaxation rate of blood (R1\u0026thinsp;=\u0026thinsp;1/T1) and hematocrit [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. These models eliminate the need for venous blood sampling by estimating Hct from blood pool T1 (T1\u003csub\u003eBlood\u003c/sub\u003e) values obtained through CMR T1 mapping sequences. Using T1\u003csub\u003eBlood\u003c/sub\u003e values derived from both pre-contrast and post-contrast images, various regression models\u0026mdash;including linear, reciprocal, and Deming regressions\u0026mdash;have been formulated to estimate synthetic Hct, yielding promising results when compared to conventional ECV calculations based on venous Hct [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan additionalcitationids=\"CR14 CR15 CR16 CR17 CR18 CR19 CR20 CR21 CR22\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. To improve accuracy, Opatril et al. developed models using linear and reciprocal regression formulas that incorporated both native T1Blood (T1-BN) and post-contrast T1Blood (T1-BP) values. Including post-contrast parameters significantly improved the statistical precision of the formula [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWhile it is well established that hematocrit values differ between sexes, emerging evidence suggests that both sex and age significantly influence T1 mapping parameters, thereby affecting synthetic Hct and ECV estimations [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Multiple studies\u0026mdash;including our previous research\u0026mdash;have demonstrated significant differences in myocardial T1 times between males and females, and have further shown that aging in females is associated with a marked reduction in T1 values [\u003cspan additionalcitationids=\"CR25\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. These findings highlight the potential importance of incorporating sex and age as variables into synthetic Hct models to enable more accurate and personalized ECV estimations.\u003c/p\u003e \u003cp\u003eIn our study, to minimize potential variations in venous Hct measurements caused by timing and postural factors, blood samples were collected after patients had remained in a stable supine position for at least 15 minutes prior to contrast administration. This approach ensured the acquisition of reliable Hct values and aimed to enhance the accuracy of the resulting synthetic Hct models.\u003c/p\u003e \u003cp\u003ePrevious studies have extensively discussed the relationship between synthetic Hct and ECV, as well as the impact of blood pool T1 values and sex on these models. Rather than re-evaluating these variables individually, the present study aimed to develop a comprehensive model that integrates age, sex, and both pre- and post-contrast blood pool T1 values, calibrated against an optimized venous Hct reference. This approach was designed to offer a highly accurate, non-invasive, and clinically applicable method for myocardial ECV estimation.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy Design and Patient Population\u003c/h2\u003e \u003cp\u003eThis single-center retrospective study was conducted between January 2024 and January 2025 at the Radiology Department of the Healthcare Research and Application Center of the University of Health Sciences in Ankara. Under the Declaration of Helsinki, the ethics committee approval was obtained by the Ankara Training and Research Hospital Clinical Research Ethics Committee on 08 January 2025 (Ethic number: E-24-360). Due to the retrospective nature of the study, obtaining additional consent from the patients was not required.\u003c/p\u003e \u003cp\u003eData from 629 patients who underwent CMR examinations between September 2015 and September 2023 were evaluated for eligibility in this study. Additionally, sex, age, body mass index (BMI), and cardiac diagnoses were collected as supplementary information. A total of 253 patients aged 18 and older, including both healthy individuals and those with various cardiovascular pathologies, were included for myocardial ECV calculation using CMR. Inclusion criteria were: (1) having venous Hct measurements obtained immediately before contrast agent injection; (2) having sufficient LGE-free myocardium in the mid-ventricular septal wall of the left ventricle for quantitative analysis. Exclusion criteria were: (1) patients with a history of cardiac surgery, (2) patients under 18 years of age, (3) patients with technical limitations affecting image quality \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003ePatients were randomly assigned to derivation (n\u0026thinsp;=\u0026thinsp;177) and validation (n\u0026thinsp;=\u0026thinsp;76) cohorts in a 70\u0026ndash;30% ratio. The splitting process was performed using random allocation to simulate real-world generalizability and minimize selection bias. The derivation group was used to train the model and determine variable importance, while the validation group served to assess its performance independently. Baseline clinical and imaging variables were compared between the two groups to ensure balanced distribution. Of these included cases, 44 subjects had normal cardiac MRI findings, with no observed abnormalities in functional, structural, or quantitative analyses \u003cb\u003e(Supplementary Materials)\u003c/b\u003e. To evaluate the accuracy of the model in the setting of elevated ECV values, a hypertrophic cardiomyopathy (HCM) subcohort (n\u0026thinsp;=\u0026thinsp;70) was selected from the full cohort based on clinical diagnosis and cardiac MRI findings. The diagnosis of HCM was confirmed based on cardiac MRI imaging criteria, in accordance with the 2023 European Society of Cardiology guidelines for cardiomyopathies [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. The performance of the synthetic Hct model was then specifically assessed within this subcohort.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eImaging Protocol\u003c/h3\u003e\n\u003cp\u003eAll patients' CMR examinations were performed at 1.5T (Magnetom Aera, Siemens Healthineers, Erlangen, Germany) using an 18-channel body coil. Myocardial T1 mapping was performed using an electrocardiogram - triggered modified Look-Locker inversion recovery (MOLLI) pulse sequence, both before and after the intravenous administration of a gadolinium-based contrast agent (0.15 mmol/kg, Gadovist, Bayer AG, Leverkusen, Germany). None of the patients were given any intravenous fluids, except for the small volume administered following the contrast injection. Post-contrast imaging was performed 15 minutes after the injection to ensure distribution equilibrium of the contrast agent. The MOLLI protocol used a 5(3)3 sampling scheme for native T1 mapping and a 4(1)3(1)2 sampling scheme for postcontrast T1 mapping. Scan parameters were as follows: TE/TR 1.12/2.8 ms; flip angle 35\u0026deg;; bandwidth 1085 Hz/Pixel; TI start 100 ms; TI increment 80 ms; slice thickness 5 mm; iPAT factor (GRAPPA) 2. In both pre- and post-contrast sequences, at least three short-axis slices, covering the basal, mid, and apical segments of the left ventricle, were sequentially acquired from the same anatomical locations.\u003c/p\u003e\n\u003ch3\u003eImaging Analysis\u003c/h3\u003e\n\u003cp\u003eAll cardiac MRI images were reviewed by a radiologist with three years of experience in cardiac MRI and eight years of experience in radiological imaging, using a diagnostic workstation (Leonardo Syngo MR E11, Siemens Healthineers). Regions of interest (ROIs) were manually delineated in the myocardium of the interventricular septum, carefully excluding both the endocardium and epicardium. Additionally, ROIs were drawn within the blood pool, avoiding papillary muscles, on the same midventricular short-axis slice for both pre- and post-contrast T1 mapping. To ensure accurate quantitative analysis, all included patients had a sufficient amount of LGE-free myocardium in the midventricular septum. Patients with less than 20 pixels of LGE-free myocardium available for ROI measurements were excluded from the cohort.[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] \u003cb\u003e(\u003c/b\u003eFig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e\u003cb\u003e)\u003c/b\u003e\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eAfter measuring the T1 times on the native and post-contrast maps, the \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\varDelta\\:\\)\u003c/span\u003e\u003c/span\u003eR1myo and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\varDelta\\:\\)\u003c/span\u003e\u003c/span\u003eR1blood values from the ECV formula by Arheden et al.[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e], as mentioned in the introduction section, were calculated. Pre- and post-contrast T1 values from the blood pool were then utilized to estimate synthetic Hct (Hct\u003csub\u003esyn\u003c/sub\u003e) using locally derived models, which were subsequently used to quantify ECV.\u003c/p\u003e \u003cp\u003eIn HCM patients, manual ROI-based measurements were performed in the left ventricular wall, where pathological enhancement is most prominent and reliably quantifiable. These measurements were conducted in accordance with the predefined criteria described above and were evaluated using the same statistical methods applied in the validation subgroup.\u003c/p\u003e\n\u003ch3\u003eHematocrit and Extracellular Volume Measurements\u003c/h3\u003e\n\u003cp\u003eCardiac MRI examinations were generally conducted between 11:30 and 14:00 at our center. Venous Hct samples were collected just before contrast injection and after the patients had remained in a stable supine position for at least 15 minutes, in order to minimize variations caused by postural fluctuations in myocardial ECV calculations. While this standard protocol was applied to most patients, for a limited number of cases, blood samples were collected after a longer supine period due to repeated sequences caused by artifacts in the images obtained before contrast injection. Venous Hct (Hct\u003csub\u003emeasured\u003c/sub\u003e) values were measured using standard laboratory methods at the central laboratory of our hospital. These values were used in myocardial ECV (ECV\u003csub\u003emeasured\u003c/sub\u003e) calculations, defined as the reference ECV. This standardization aimed to minimize the effects of time- and position-related variations on ECV calculations.\u003c/p\u003e \u003cp\u003eIn the literature, numerous synthetic Hct formulas have been derived based on the linear relationship between Hct and the R1 value of blood. Opatril et al. incorporated post-contrast T1 blood values into synthetic Hct calculations, achieving results closer to venous Hct values [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Additionally, Chen et al. developed sex-specific (female/male), device-specific (1.5/3T), and blood pool-specific (LV/RV) models, reporting that \u0026ldquo;the specific model (Deming Regression)\u0026rdquo; provided more accurate results for synthetic Hct calculations [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Several studies have demonstrated significant myocardial and blood T1 differences between males and females, as well as a marked decrease in T1 times in women as they age [\u003cspan additionalcitationids=\"CR25\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. In our latest study, we found significantly lower myocardial native T1 times in women over 50, considered postmenopausal, compared to younger women [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. These findings suggest that considering sex and age in synthetic Hct calculations could be critical for achieving more personalized and accurate ECV estimates.\u003c/p\u003e \u003cp\u003eThis study aimed to develop a synthetic hematocrit model based on an optimized venous reference, integrating age, sex, and both pre- and post-contrast T1 blood values, to achieve more accurate estimations of synthetic Hct (Hct\u003csub\u003esyn\u003c/sub\u003e) and ECV (ECV\u003csub\u003esyn\u003c/sub\u003e). This model, along with the specific model developed by Chen et al.[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] (Hct\u003csub\u003ereported\u003c/sub\u003e; ECV\u003csub\u003ereported\u003c/sub\u003e), were tested in the validation cohort, and ECV\u003csub\u003emeasured\u003c/sub\u003e was compared with ECV derived from synthetic Hcts and ECV\u003csub\u003ereported\u003c/sub\u003e. In accordance with the method described by Chen et al. [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], sex-specific formulas were applied separately to male and female participants to calculate ECV values. The Hct\u003csub\u003ereported\u003c/sub\u003e was calculated using the following formulas: for females, \u003cem\u003eHct\u0026thinsp;=\u0026thinsp;1048 x (1/T1blood) \u0026minus;\u0026thinsp;0.2630\u003c/em\u003e, and for male, \u003cem\u003eHct\u0026thinsp;=\u0026thinsp;1319 x (1/T1blood) \u0026minus;\u0026thinsp;0.4318\u003c/em\u003e. These sex-adjusted values were collectively analyzed under the heading of \u0026lsquo;ECV\u003csub\u003ereported\u003c/sub\u003e\u0026rsquo;.\u003c/p\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eContinuous variables were expressed as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation or median with interquartile range, depending on data distribution. Categorical variables were presented as frequencies and percentages. Between-group comparisons were made using the Student\u0026rsquo;s t-test or Mann-Whitney U test for continuous variables, and the chi-square test or Fisher\u0026rsquo;s exact test for categorical variables, as appropriate.\u003c/p\u003e \u003cp\u003eVariable selection and model development were carried out in the derivation cohort. Initially, univariate linear regression analyses were performed for each candidate predictor, including age, sex, native and post-contrast T1 values of blood, body mass index (BMI), and myocardial T1 values. To reduce overfitting and identify the most relevant predictors, least absolute shrinkage and selection operator (LASSO) regression was applied. Only variables retained by LASSO were entered into the multivariable linear regression model. The final model\u0026rsquo;s coefficients were determined using standard ordinary least squares estimation within the derivation group. Internal validation was performed by applying the model to the validation cohort and evaluating its predictive accuracy using the coefficient of determination (R\u0026sup2;), root mean square error (RMSE), mean bias, and Lin\u0026rsquo;s concordance correlation coefficient (CCC). Agreement between measured and synthetic ECV values was first evaluated using Pearson correlation coefficients and scatter plots, followed by Bland\u0026ndash;Altman analysis to assess bias and limits of agreement.\u003c/p\u003e \u003cp\u003eA priori sample size estimation was performed based on Green\u0026rsquo;s rule for multiple linear regression (n\u0026thinsp;\u0026ge;\u0026thinsp;50\u0026thinsp;+\u0026thinsp;8k), which recommends a minimum sample size of 90 for five predictors (age, sex, pre- and post-contrast T1 values of blood, and BMI) [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Appropriate attention was paid to sample size considerations within the derivation cohort, aiming to exceed the recommended minimum and ensure adequate statistical power and stability of coefficient estimates.\u003c/p\u003e \u003cp\u003eAll descriptive and comparative statistical analyses were conducted using SPSS version 25 (IBM Corp., Armonk, NY).\u003c/p\u003e \u003c/div\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\n \u003ch2\u003eBaseline Characteristics\u003c/h2\u003e\n \u003cp\u003eThe overall cohort represented a diverse clinical population, including 142 patients with non-ischemic cardiomyopathy, 44 with normal cardiac findings, 18 with myocarditis, 16 with ischemic cardiomyopathy, and 11 with isolated heart failure diagnosed by CMR \u003cstrong\u003e(\u003c/strong\u003eTable \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e\u003cstrong\u003e)\u003c/strong\u003e.\u003c/p\u003e\n \u003cdiv\u003e\u0026nbsp;\u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eDistribution of Cardiac Diagnoses in the Study Cohort\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eDiagnosis Category\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePatient Count (n, %)\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNon-Ischemic CMP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e142, 56.1%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNormal\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e44, 17.4%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMyocarditis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e18, 7.1%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIschemic CMP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e16, 6.3%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eIsolated Heart Failure (Identified by CMR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11, 4.3%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMyocarditis and Pericarditis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11, 4.3%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eFabry Disease-Related Cardiomyopathy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3, 1.2%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eARVC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2, 0.8%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCardiac Tumors\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1, 0.4%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCongenital Heart Diseases\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1, 0.4%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInfective Endocarditis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1, 0.4%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePericarditis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1, 0.4%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSarcoidosis\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1, 0.4%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eValvular Diseases\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1, 0.4%\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbbrevations\u003c/strong\u003e: ARVC, Arrhythmogenic right ventricular cardiomyopathy; CMP, cardiomyopathy; CMR, cardiac magnetic resonance\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003eBaseline characteristics of the derivation cohort and validation cohort are summarized in Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e. The two cohorts were similar in age, sex, and baseline cardiac MRI parameters, with no statistically significant differences in Hct or ECV values. In addition, the baseline characteristics of the HCM subcohort are presented in Table \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e to provide detailed demographic and imaging information for this pathologic population.\u003c/p\u003e\n \u003cp\u003eIn our center, the reference range for native myocardial T1 values is 955.73\u0026ndash;1056.14 ms for males and 984.52\u0026ndash;1086.94 ms for females. In the normal subgroup of our study population, the mean native T1 values observed for both sexes remained within these reference limits. Additionally, the mean ECVmeasured was 24.42\u0026thinsp;\u0026plusmn;\u0026thinsp;2.88% in males and 25.70\u0026thinsp;\u0026plusmn;\u0026thinsp;2.62% in females, consistent with previously reported normal ranges. These findings suggest that both myocardial tissue characteristics and ECV values in the normal cohort were within expected physiological limits \u003cstrong\u003e(Supplementary Materials).\u003c/strong\u003e\u003c/p\u003e\n \u003ctable id=\"Tab2\" border=\"1\"\u003e\u003c/table\u003e\n \u003cp\u003eIn the overall cohort, native myocardial T1 values were significantly higher in females than in males (1037.76\u0026thinsp;\u0026plusmn;\u0026thinsp;37.16 ms vs. 1007.22\u0026thinsp;\u0026plusmn;\u0026thinsp;43.20 ms, respectively; p\u0026thinsp;=\u0026thinsp;8.38 \u0026times; 10⁻⁹). Moreover, venous hematocrit levels were also significantly higher in males compared to females (p\u0026thinsp;=\u0026thinsp;2.56 \u0026times; 10⁻\u0026sup2;⁴), and a modest inverse correlation was observed between age and Hct (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.26, p\u0026thinsp;=\u0026thinsp;2.25 \u0026times; 10⁻⁵).\u003c/p\u003e\n \u003ch3\u003eDerivation of the Synthetic Hematocrit (Hct) Model\u003c/h3\u003e\n \u003cp\u003eIn the derivation cohort, a multivariable linear regression model was developed to estimate synthetic hematocrit (Hct\u003csub\u003esyn\u003c/sub\u003e) by integrating age, sex, native T1 blood (\u003cem\u003eT\u003c/em\u003e\u003csub\u003e\u003cem\u003e1\u0026thinsp;\u0026minus;\u0026thinsp;BN\u003c/em\u003e\u003c/sub\u003e), and post-contrast T1 blood (\u003cem\u003eT\u003c/em\u003e\u003csub\u003e\u003cem\u003e1\u0026thinsp;\u0026minus;\u0026thinsp;BP\u003c/em\u003e\u003c/sub\u003e) values. Candidate predictors were first screened using least absolute shrinkage and selection operator (LASSO) regression to avoid overfitting and to identify the most predictive variables. Among the initial variables, age, sex, \u003cem\u003eT\u003c/em\u003e\u003csub\u003e\u003cem\u003e1\u0026thinsp;\u0026minus;\u0026thinsp;BN\u003c/em\u003e\u003c/sub\u003e, and \u003cem\u003eT\u003c/em\u003e\u003csub\u003e\u003cem\u003e1\u0026thinsp;\u0026minus;\u0026thinsp;BP\u003c/em\u003e\u003c/sub\u003e were retained by LASSO and subsequently entered into the final model. (Table \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e)\u003c/p\u003e\n \u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"565\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 2.\u003c/strong\u003e\u0026nbsp; Baseline Characteristics of the Derivation and Validation Cohorts\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDerivation Cohort\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eN=177 (70%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eValidation Cohort\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eN=76 (30%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMean\u0026plusmn;SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMean\u0026plusmn;SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFemale Sex, N (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e66 (37.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e33 (43.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.360\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge, years\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e42.5\u0026plusmn;17.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e42.9\u0026plusmn;16.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.881\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBMI, kg/m\u003csup\u003e2\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e26.3\u0026plusmn;4.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e27.8\u0026plusmn;5.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.055\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHeart Rate, bmp\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e74.3\u0026plusmn;12.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e73.8\u0026plusmn;11.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.794\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLVEF, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e60.5\u0026plusmn;12.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e62.8\u0026plusmn;10.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.170\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEDVI, ml/m\u003csup\u003e2\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e73.5\u0026plusmn;22.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e69.5\u0026plusmn;16.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.165\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eESVI, ml/m\u003csup\u003e2\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e30.9\u0026plusmn;21.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e26.7\u0026plusmn;12.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.119\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSVI, ml/m\u003csup\u003e2\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e42.6\u0026plusmn;8.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e42.8\u0026plusmn;8.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.869\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCI, ml/min/m\u003csup\u003e2\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e3.1\u0026plusmn;0.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e3.1\u0026plusmn;0.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.790\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT\u003csub\u003e1-BN\u003c/sub\u003e, msn\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e1567.8\u0026plusmn;102.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1567.9\u0026plusmn;97.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.996\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT\u003csub\u003e1-BP\u003c/sub\u003e, msn\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e286.4\u0026plusmn;59.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e281.7\u0026plusmn;47.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.539\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT\u003csub\u003e1-MN\u003c/sub\u003e, msn\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e1018.2\u0026plusmn;44.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1021.6\u0026plusmn;40.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.569\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT\u003csub\u003e1-MP\u003c/sub\u003e, msn\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e440.2\u0026plusmn;56.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e433.8\u0026plusmn;46.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.390\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHct\u003csub\u003emeasured\u003c/sub\u003e, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e42.2\u0026plusmn;4.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e42.2\u0026plusmn;4.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.929\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHct\u003csub\u003ereported\u003c/sub\u003e, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e41.7\u0026plusmn;4.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e41.5\u0026plusmn;3.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.739\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eECV\u003csub\u003emeasured\u003c/sub\u003e, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e25.8\u0026plusmn;3.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e26.2\u0026plusmn;4.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.408\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eECV\u003csub\u003ereported\u003c/sub\u003e, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 144px;\"\u003e\n \u003cp\u003e26.0\u0026plusmn;3.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e26.5\u0026plusmn;3.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 152px;\"\u003e\n \u003cp\u003e0.338\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"4\" valign=\"top\" style=\"width: 565px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbbreviations:\u003c/strong\u003e BMI, body-mass index; Bmp, beats per minute; CI, cardiac index; ECV, extracellular volume; EDVI, end-diastolic volume index; ESVI, end-systolic volume index; Hct, hematocrit; LVEF, left ventricular ejection fraction; SVI, stroke volume index;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eT\u003csub\u003e1-BN\u003c/sub\u003e, native T1 blood; T\u003csub\u003e1-BP\u003c/sub\u003e, post-contrast T1 blood; T\u003csub\u003e1-MN\u003c/sub\u003e, native T1 myocard; T\u003csub\u003e1-MP\u003c/sub\u003e, post-contrast T1 myocard\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\u0026nbsp;\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"567\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 567px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 3.\u003c/strong\u003e Demographic and Imaging Parameters in the HCM Subcohort (n=70)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMean\u0026plusmn;SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFemale Sex, N (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e24 (34.29%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge, years\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e52\u0026plusmn;15.2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBMI, kg/m\u003csup\u003e2\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e27.5 \u0026plusmn; 4.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLVEF, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e64.2\u0026plusmn;14.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eEDVI, ml/m\u003csup\u003e2\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e71.9\u0026plusmn;25.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eESVI, ml/m\u003csup\u003e2\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e28.0\u0026plusmn;22.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSVI, ml/m\u003csup\u003e2\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e43.9\u0026plusmn;11.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT\u003csub\u003e1-BN\u003c/sub\u003e, msn\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e1573.07 \u0026plusmn; 109.83\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT\u003csub\u003e1-BP\u003c/sub\u003e, msn\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e297.6 \u0026plusmn; 51.57\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT\u003csub\u003e1-MN\u003c/sub\u003e, msn\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e1124.31 \u0026plusmn; 72.18\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT\u003csub\u003e1-MP\u003c/sub\u003e, msn\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e356.9 \u0026plusmn; 77.25\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHct\u003csub\u003emeasured\u003c/sub\u003e, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e42.5\u0026plusmn;5.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHct\u003csub\u003ereported\u003c/sub\u003e, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e41.2\u0026plusmn;5.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHct\u003csub\u003esyn\u003c/sub\u003e, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e44.4\u0026plusmn;3.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eECV\u003csub\u003emeasured\u003c/sub\u003e, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e41.8\u0026plusmn;13.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eECV\u003csub\u003ereported\u003c/sub\u003e, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e43.0\u0026plusmn;13.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eECV\u003csub\u003esyn\u003c/sub\u003e, %\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 432px;\"\u003e\n \u003cp\u003e40.6\u0026plusmn;12.8\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 567px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbbreviations:\u003c/strong\u003e BMI, body-mass index; CI, cardiac index; ECV, extracellular volume; EDVI, end-diastolic volume index; ESVI, end-systolic volume index; Hct, hematocrit; LVEF, left ventricular ejection fraction; SVI, stroke volume index;\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eT\u003csub\u003e1-BN\u003c/sub\u003e, native T1 blood; T\u003csub\u003e1-BP\u003c/sub\u003e, post-contrast T1 blood; T\u003csub\u003e1-MN\u003c/sub\u003e, native T1 myocard; T\u003csub\u003e1-MP\u003c/sub\u003e, post-contrast T1 myocard\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\u003cbr\u003e\n \u003cdiv\u003e\n \u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"624\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"8\" valign=\"top\" style=\"width: 624px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 4.\u003c/strong\u003e Derivation of the Synthetic Hematocrit (Hct\u003csub\u003esyn\u003c/sub\u003e) Model Using Age, Sex, and Blood T1 Values\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariables\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLASSO Regression\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eUnivariate Analysis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" style=\"width: 265px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMultivariate Analysis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026szlig;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSelected\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026szlig; (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026szlig; (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eP value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eStandardized \u0026szlig;\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e-0.413\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e-0.083 (-0.122--0.044)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e-0.027 (-0.056-0.001)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.061\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e-0.099\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFemale (\u003cem\u003eSex\u003c/em\u003e)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e-1.452\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e-5.604 (-6.789--4.419)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e-3.097 (-4.188--2.006)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e-0.318\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT\u003csub\u003e1-BN\u003c/sub\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e-2.174\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e-0.030 (-0.036--0.025)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e-0.022 (-0.027--0.016)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e-0.471\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT\u003csub\u003e1-BP\u003c/sub\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.644\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.020 (0.008-0.031)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.012 (0.004-0.021)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.004\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.149\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBMI\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e0.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"8\" valign=\"top\" style=\"width: 624px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbbreviations:\u003c/strong\u003e BMI, body-mass index; T\u003csub\u003e1-BN\u003c/sub\u003e, native T1 blood; T\u003csub\u003e1-BP\u003c/sub\u003e, post-contrast T1 blood\u003c/p\u003e\n \u003cp\u003eCandidate variables were first evaluated using Least Absolute Shrinkage and Selection Operator (LASSO) regression for selection. Only variables retained in LASSO were included in multivariable modeling.\u003c/p\u003e\n \u003cp\u003eUnivariate analyses are shown for descriptive purposes only.\u003c/p\u003e\n \u003cp\u003eDespite a borderline p-value (0.061), age was retained in the multivariable model due to its selection by LASSO and its known clinical relevance in hematocrit variation.\u003c/p\u003e\n \u003cp\u003eAdjusted R\u003csup\u003e2\u003c/sup\u003e of the model is 0.560.\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eHct\u003csub\u003esyn\u003c/sub\u003e = 77.906 - 0.022 x T\u003csub\u003e1-BN\u0026nbsp;\u003c/sub\u003e- 3.097 (if female) + 0.012 x T\u003csub\u003e1-BP\u0026nbsp;\u003c/sub\u003e- 0.027 x Age\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003eIn univariate analyses, all four selected variables demonstrated significant associations with Hct\u003csub\u003esyn\u003c/sub\u003e (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). In the multivariate model, \u003cem\u003eT\u003c/em\u003e\u003csub\u003e\u003cem\u003e1\u0026thinsp;\u0026minus;\u0026thinsp;BN\u003c/em\u003e\u003c/sub\u003e emerged as the strongest predictor (standardized \u0026szlig; = \u0026minus;\u0026thinsp;0.471), followed by female sex (\u0026szlig; = \u0026minus;\u0026thinsp;3.097, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), \u003cem\u003eT\u003c/em\u003e\u003csub\u003e\u003cem\u003e1\u0026thinsp;\u0026minus;\u0026thinsp;BP\u003c/em\u003e\u003c/sub\u003e (\u0026szlig; = 0.012, p\u0026thinsp;=\u0026thinsp;0.004), and age (\u0026szlig; = \u0026minus;\u0026thinsp;0.027, p\u0026thinsp;=\u0026thinsp;0.061). Despite the borderline statistical significance of age, it was retained in the final model due to its known clinical relevance and consistent selection in the LASSO procedure.\u003c/p\u003e\n \u003cp\u003eThe final regression equation was:\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eHct\u003c/em\u003e \u003csub\u003e\u0026nbsp;\u003cem\u003esyn\u003c/em\u003e\u0026nbsp;\u003c/sub\u003e \u003cem\u003e= 77.906\u0026thinsp;\u0026minus;\u0026thinsp;0.022 x T\u003c/em\u003e\u003csub\u003e\u003cem\u003e1\u0026thinsp;\u0026minus;\u0026thinsp;BN\u003c/em\u003e\u003c/sub\u003e \u003cem\u003e\u0026minus;\u0026thinsp;3.097 (if female)\u0026thinsp;+\u0026thinsp;0.012 x T\u003c/em\u003e\u003csub\u003e\u003cem\u003e1\u0026thinsp;\u0026minus;\u0026thinsp;BP\u003c/em\u003e\u003c/sub\u003e \u003cem\u003e\u0026minus;\u0026thinsp;0.027 x Age\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003eThe model demonstrated good predictive performance, with an adjusted R\u0026sup2; of 0.560 in the derivation set. This indicates that the model explained approximately 56% of the variability in venous Hct measurements. The inclusion of both pre- and post-contrast blood T1 values, as well as demographic variables, contributed to the robustness and individualized accuracy of the Hct\u003csub\u003esyn\u003c/sub\u003e prediction.\u003c/p\u003e\n \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003eValidation Performance\u003c/h2\u003e\n \u003cp\u003eIn the validation cohort, ECV\u003csub\u003esyn\u003c/sub\u003e showed excellent correlation with ECV\u003csub\u003emeasured\u003c/sub\u003e, with a Pearson correlation coefficient of r\u0026thinsp;=\u0026thinsp;0.96, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001. ECV\u003csub\u003ereported\u003c/sub\u003e also showed a strong correlation (r\u0026thinsp;=\u0026thinsp;0.95, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) \u003cstrong\u003e(\u003c/strong\u003eFig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e\u003cstrong\u003e)\u003c/strong\u003e. Performance metrics are summarized in Table \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e. ECV\u003csub\u003esyn\u003c/sub\u003e demonstrated an R\u0026sup2; of 0.91, mean bias of +\u0026thinsp;0.31%, RMSE of 1.25%, and Lin\u0026apos;s CCC of 0.949. In contrast, ECV\u003csub\u003ereported\u003c/sub\u003e showed an R\u0026sup2; of 0.88, mean bias of +\u0026thinsp;0.40%, RMSE of 1.39%, and CCC of 0.942.\u003c/p\u003e\n \u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 576px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 5.\u003c/strong\u003e Performance Metrics of Synthetic and Reported ECV Compared to Measured ECV in the Validation Cohort\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eIndependent Variable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eR-squared\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMean Bias (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRMSE (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLin\u0026apos;s CCC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eECV\u003csub\u003ereported\u003c/sub\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e+0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e1.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.942\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eECV\u003csub\u003esynthetic\u003c/sub\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e+0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e1.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.949\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 576px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbbreviations:\u003c/strong\u003e ECV, extracellular volume; Lin\u0026apos;s CCC, Lin\u0026apos;s concordance correlation coefficient; RMSE, root mean square error\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\u003cbr\u003e\n \u003cdiv\u003e\n \u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 576px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTable 6.\u003c/strong\u003e Performance Metrics of Synthetic and Reported ECV Compared to Measured ECV in the HCM Subcohort\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eIndependent Variable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eR-squared\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMean Bias (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eRMSE (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLin\u0026apos;s CCC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eECV\u003csub\u003esynthetic\u003c/sub\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.953\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e-1.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e2.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.989\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eECV\u003csub\u003ereported\u003c/sub\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.955\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e+1.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e2.81\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 115px;\"\u003e\n \u003cp\u003e0.992\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 576px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbbreviations:\u003c/strong\u003e ECV, extracellular volume; Lin\u0026apos;s CCC, Lin\u0026apos;s concordance correlation coefficient; RMSE, root mean square error.\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003eBland\u0026ndash;Altman analysis revealed a mean bias of 0.31% and limits of agreement from \u0026minus;\u0026thinsp;2.08% to +\u0026thinsp;2.70%. For ECV\u003csub\u003ereported\u003c/sub\u003e, the mean bias was 0.40% with wider limits of agreement (\u0026minus;\u0026thinsp;2.23% to +\u0026thinsp;3.02%) \u003cstrong\u003e(\u003c/strong\u003eFig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e\u003cstrong\u003e)\u003c/strong\u003e.\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n \u003ch2\u003ePerformance of Synthetic and Reported ECV Models in the HCM Subcohort\u003c/h2\u003e\n \u003cp\u003eIn the HCM subcohort consisting of 70 individuals, ECV values calculated using the proposed synthetic hematocrit model (ECV\u003csub\u003esyn\u003c/sub\u003e) were compared with reference values obtained using venous hematocrit (ECV\u003csub\u003emeasured\u003c/sub\u003e) and with literature-based values calculated using the sex-specific Chen\u0026rsquo;s model (ECV\u003csub\u003ereported\u003c/sub\u003e).\u003c/p\u003e\n \u003cp\u003eThe average ECV\u003csub\u003emeasured\u003c/sub\u003e in this subcohort was 41.82\u0026thinsp;\u0026plusmn;\u0026thinsp;13.03%, reflecting a high burden of myocardial fibrosis. Based on standard ECV thresholds, 44.3% of patients exhibited severely elevated ECV (\u0026gt;\u0026thinsp;40%), while 47.1% fell within the 30\u0026ndash;40% range, indicating significant fibrosis.\u003c/p\u003e\n \u003cp\u003eCompared to the measured reference, ECV\u003csub\u003esyn\u003c/sub\u003e demonstrated a mean bias of \u0026minus;\u0026thinsp;1.16%, with RMSE of 2.91%, Lin\u0026rsquo;s CCC of 0.989, and a Pearson correlation coefficient (r) of 0.979 (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) \u003cstrong\u003e(\u003c/strong\u003eTable \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e\u003cstrong\u003e)\u003c/strong\u003e. A strong linear correlation was observed with tight clustering along the regression line and narrow 95% confidence intervals. Bland\u0026ndash;Altman analysis revealed limits of agreement ranging from \u0026minus;\u0026thinsp;6.43% to +\u0026thinsp;4.10% \u003cstrong\u003e(\u003c/strong\u003eFig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e\u003cstrong\u003e)\u003c/strong\u003e. The mean ECV\u003csub\u003esyn\u003c/sub\u003e value was 40.65\u0026thinsp;\u0026plusmn;\u0026thinsp;12.84%.\u003c/p\u003e\n \u003cp\u003eIn comparison, ECV\u003csub\u003ereported\u003c/sub\u003e showed a mean bias of +\u0026thinsp;1.21%, RMSE of 2.81%, CCC of 0.992, and r of 0.985. Limits of agreement for ECV\u003csub\u003ereported\u003c/sub\u003e ranged from \u0026minus;\u0026thinsp;3.80% to +\u0026thinsp;6.22%, with a mean value of 43.03\u0026thinsp;\u0026plusmn;\u0026thinsp;13.99%. \u003cstrong\u003e(\u003c/strong\u003eFig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e\u003cstrong\u003e)(\u003c/strong\u003eTable \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e)\u003c/p\u003e\n \u003cp\u003eWhile both models showed strong correlation with ECV\u003csub\u003emeasured\u003c/sub\u003e, the multivariable model produced more balanced estimates with slightly narrower dispersion, despite a modest underestimation trend.\u003c/p\u003e\n \u003cp\u003eA freely accessible online calculator for synthetic ECV estimation based on the proposed model is available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://syntheticecvcalculator.shinyapps.io/syntheticecvapp\u003c/span\u003e\u003c/span\u003e, with a corresponding QR code provided in the supplementary material.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eThis study introduces a novel synthetic hematocrit model that integrates age, sex, and both pre- and post-contrast blood pool T1 values, calibrated using an optimized venous Hct reference. The model was developed using 1.5T MOLLI-based cardiac MRI data and demonstrated clinical applicability in the HCM subcohort characterized by high ECV values.\u003c/p\u003e \u003cp\u003eThe key findings of this study are as follows:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eA new synthetic Hct model was developed that integrates age, sex, and both pre- and post-contrast blood T1 values, calibrated using an optimized venous Hct reference.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eVenous Hct measurements were obtained after patients remained supine for at least 15 minutes, minimizing posture-related hemodynamic variability and providing a physiologically reliable reference.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe ECVsyn values derived from this model demonstrated lower bias (0.31%), lower RMSE (1.25), and stronger agreement (CCC\u0026thinsp;=\u0026thinsp;0.949) compared to the ECVreported values calculated using the model of Chen et al.[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eIn the HCM subcohort, characterized by elevated ECV, the model showed a strong correlation (r\u0026thinsp;=\u0026thinsp;0.98). Although a slight negative bias (\u0026minus;\u0026thinsp;1.16%) was observed, it remained within clinically acceptable limits.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eCalibration Approach in Venous Blood Sampling\u003c/h2\u003e \u003cp\u003eOne of the major strengths of our study is the methodological rigor in venous Hct sampling. Unlike previous studies, all samples were obtained from patients after remaining in the supine position for at least 15 minutes, thereby minimizing posture-related hemodilution. This approach improved the physiological reliability of the model by addressing a factor often overlooked in previous studies such as those by Chen, Lim, and Yin.[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eOpatril et al. [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] showed that including post-contrast T1 values improved prediction accuracy; however, their model did not include important demographic variables like age and sex, and did not standardize venous Hct collection. Similarly, Lim et al. [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] developed a more accurate sex-specific model, but it did not incorporate post-contrast T1 values. Our study is the first to integrate all these parameters into a single model, enhancing its clinical adaptability.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eModel Performance and Comparison with the Literature\u003c/h2\u003e \u003cp\u003eIn terms of explanatory power, our synthetic Hct model outperformed previously published models. Among all models developed using 1.5T and 3T MRI systems, ours achieved the highest coefficient of determination (R\u0026sup2; = 0.56). This value was higher than those reported by Opatril (R\u0026sup2; = 0.49), Su (R\u0026sup2; = 0.51), Kammerlander (R\u0026sup2; = 0.284), and Treibel (R\u0026sup2; = 0.50) [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. In the sex-specific models developed by Chen et al., R\u0026sup2; was 0.43 for males and 0.27 for females [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. These findings underscore the value of incorporating demographic and imaging-based variables, which appears to significantly enhance both the accuracy and generalizability of the model.\u003c/p\u003e \u003cp\u003e Our model achieved high accuracy for ECV estimation, with r\u0026thinsp;=\u0026thinsp;0.974, CCC\u0026thinsp;=\u0026thinsp;0.949, and narrow Bland\u0026ndash;Altman limits of agreement (\u0026minus;\u0026thinsp;2.08% to +\u0026thinsp;2.70%). This accuracy was maintained even in the HCM subcohort (r\u0026thinsp;=\u0026thinsp;0.98), demonstrating the model\u0026rsquo;s robustness against pathological variation. In healthy individuals, myocardial ECV values are typically reported as 25.3\u0026thinsp;\u0026plusmn;\u0026thinsp;3.5% and 25.4\u0026thinsp;\u0026plusmn;\u0026thinsp;2.5%, and an error range of \u0026plusmn;\u0026thinsp;2.5\u0026ndash;3.5% is generally considered acceptable [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. In our study, the prediction error of the proposed model remained within this accepted range even in the fibrotic HCM subgroup.\u003c/p\u003e \u003cp\u003eOur findings are consistent with the growing literature on synthetic ECV and contribute additional advancements. In a large 1.5T and 3T study by Chen et al., synthetic ECVs were reported to correlate well with histologically measured collagen volume fractions, similarly to conventional measured ECVs (R\u0026sup2; = 0.87\u0026ndash;0.88). The average difference between synthetic and conventional ECV was less than 0.2%, with agreement limits around \u0026plusmn;\u0026thinsp;4%, comparable to our findings [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Compared with ECV\u003csub\u003ereported\u003c/sub\u003e values derived using their sex-specific model, our model yielded lower bias, stronger agreement, and slightly narrower limits of agreement (\u0026thinsp;~\u0026thinsp;\u0026plusmn;\u0026thinsp;2.5%), which we attribute to our rigorous imaging protocol and the inclusion of variables beyond pre-contrast blood T1.\u003c/p\u003e \u003cp\u003eRobinson et al. [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] also reported a strong correlation between synthetic and measured ECV (r\u0026thinsp;=\u0026thinsp;0.92), with a mean bias of 0.23% and agreement limits between \u0026minus;\u0026thinsp;2.82% and +\u0026thinsp;3.27%. Although this model also performed well, its total LoA width of 6.09% suggests more variability. In contrast, our model provided slightly higher correlation (r\u0026thinsp;=\u0026thinsp;0.974), a modestly higher bias (0.31%), but a narrower LoA (\u0026minus;\u0026thinsp;2.08% to +\u0026thinsp;2.70%). These results suggest that while both models perform well, our model offers advantages in minimizing both systematic error and variability. In the study by Kammerlander et al., bias was negligible (+\u0026thinsp;0.007%), but agreement limits were wider (\u0026minus;\u0026thinsp;4.32% to +\u0026thinsp;4.33%). Additionally, our model achieved a stronger correlation (r\u0026thinsp;=\u0026thinsp;0.96, R\u0026sup2; = 0.92) compared to their reported r\u0026thinsp;=\u0026thinsp;0.943 and R\u0026sup2; = 0.889 [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Although Yin et al. [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] reported a very low bias (+\u0026thinsp;0.05%), their LoA also ranged widely (\u0026minus;\u0026thinsp;4.22% to +\u0026thinsp;4.31%). These results collectively suggest that while all synthetic models fall within clinically acceptable limits, our approach may offer more reliable and consistent results at the individual patient level. This advantage likely stems from the integration of age, sex, and both pre- and post-contrast T1 blood values, along with a carefully optimized Hct sampling protocol.\u003c/p\u003e \u003cp\u003eThe ECV\u003csub\u003esyn\u003c/sub\u003e model demonstrated excellent agreement with the reference ECV\u003csub\u003emeasured\u003c/sub\u003e values in the HCM subcohort (bias: \u0026minus;1.16%, CCC: 0.989), with clinically acceptable agreement limits (\u0026minus;\u0026thinsp;6.43% to +\u0026thinsp;4.10%). Although the ECV\u003csub\u003ereported\u003c/sub\u003e model also showed strong concordance (CCC: 0.992), it exhibited a slight positive bias (+\u0026thinsp;1.21%) and wider LoA [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. The high mean ECV in the HCM subcohort (41.82%) confirms that the model was tested in a fibrotic myocardial population. The slight underestimation may be attributable to the model being trained predominantly on normal-to-intermediate ECV ranges. Nevertheless, the ECV\u003csub\u003esyn\u003c/sub\u003e model yielded highly correlated and balanced predictions, even in patients with high ECV values. These findings were further supported by high coefficients of determination (R\u0026sup2; = 0.953 for ECV\u003csub\u003esyn\u003c/sub\u003e and R\u0026sup2; = 0.955 for ECV\u003csub\u003ereported\u003c/sub\u003e), confirming the strong linear relationship of both models with ECV\u003csub\u003emeasured\u003c/sub\u003e. Collectively, these results support the feasibility of accurate synthetic ECV estimation in fibrotic patient populations without the need for invasive blood sampling.\u003c/p\u003e \u003cp\u003eAll synthetic models, including ours, assume a fixed timing for post-contrast T1 mapping. Although we maintained a consistent post-contrast interval during image acquisition, variation in this timing in clinical practice may affect blood contrast equilibrium and alter the T1\u0026ndash;Hct relationship. This could reduce the accuracy of models incorporating post-contrast values. Therefore, protocol standardization is essential for reliable application.\u003c/p\u003e \u003cp\u003eAlthough a prototype Shiny web interface for ECV\u003csub\u003esyn\u003c/sub\u003e calculation was developed, ideal implementation would involve integration into standard CMR post-processing platforms or directly into scanner software. For example, vendors could provide automated \u0026ldquo;synthetic ECV\u0026rdquo; generation immediately following T1 mapping, eliminating the need for venous sampling and enabling seamless integration into clinical workflow.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eClinical Implications\u003c/h2\u003e \u003cp\u003eT1 mapping\u0026ndash;based CMR is considered the gold standard for noninvasive evaluation of diffuse myocardial fibrosis. However, the need for venous Hct and its biological variability can limit clinical use. Our synthetic model eliminates the need for blood sampling during CMR, significantly streamlining the ECV quantification process. In addition to saving time and resources, this also enhances patient comfort and safety by avoiding an invasive step.\u003c/p\u003e \u003cp\u003eReal-time synthetic ECV computation allows the reporting physician to access tissue characterization metrics immediately upon image acquisition. In cases with borderline or unexpected values, additional sequences or validation can be performed while the patient is still on the table. Furthermore, a generalizable synthetic Hct model enables ECV quantification in settings without immediate access to laboratory services or Hct devices, such as in after-hours emergency scans. It may also be useful in retrospective studies where blood samples are missing but T1 maps are available. Inclusion of age- and sex-specific variables improves precision across diverse physiological conditions.\u003c/p\u003e \u003cp\u003eUltimately, routine adoption of synthetic ECV could expand the role of quantitative CMR tissue characterization and increase its availability in both clinical and research settings. Given that ECV is a prognostically relevant marker of diffuse myocardial fibrosis, this model could facilitate broader integration of ECV into patient care pathways. The high agreement and low bias observed in our study indicate that synthetic ECV may be a reliable substitute for conventional ECV in clinical practice.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eLimitations\u003c/h2\u003e \u003cp\u003eDespite the promising performance of our model, certain limitations should be acknowledged.\u003c/p\u003e \u003cp\u003eThis was a single-center study using a single 1.5T scanner and MOLLI T1 mapping sequence. Although internal validation was robust, external performance remains untested. However, inclusion of a diagnostically heterogeneous cohort, comprising non-ischemic and ischemic cardiomyopathies, myocarditis, isolated heart failure, and structurally normal hearts, provides preliminary evidence of model generalizability across myocardial conditions.\u003c/p\u003e \u003cp\u003eDifferences in T1 mapping techniques or patient demographics may require model recalibration. As previously noted in the literature, synthetic Hct models may be affected by hardware configuration, magnetic field strength, and sequence parameters, potentially necessitating site-specific calibration [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eSome studies suggest reduced accuracy of synthetic ECV at hematocrit extremes [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Although our cohort did not exclude patients based on Hct, severely anemic or polycythemic individuals were not present, and thus model performance in these settings remains uncertain. Similarly, infiltrative cardiomyopathies such as cardiac amyloidosis were underrepresented, limiting our assessment in this group. Future multicenter studies are warranted to confirm generalizability and evaluate model performance across diverse Hct values and pathologies.\u003c/p\u003e \u003cp\u003eFinally, our analysis included only adult patients. Previous studies, including that by Raucci et al. (2017), have highlighted the risk of clinically significant error when using synthetic Hct models in pediatric populations [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Thus, separate validation is necessary for pediatric use.\u003c/p\u003e \u003c/div\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eThis study presents a synthetic hematocrit model based on age, sex, and both pre- and post-contrast T1 values, calibrated using an optimized venous reference. The resulting ECV\u003csub\u003esyn\u003c/sub\u003e estimates demonstrated superior accuracy and agreement compared to previous models. This approach offers a noninvasive and clinically viable alternative for ECV quantification and brings synthetic ECV one step closer to routine clinical use.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003e\u003cstrong\u003eBMI:\u003c/strong\u003e Body mass index\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCCC:\u003c/strong\u003e Concordance correlation coefficient\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCMR:\u003c/strong\u003e Cardiac magnetic resonance\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eECV:\u003c/strong\u003e Extracellular volume\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eHCM:\u003c/strong\u003e Hypertrophic cardiomyopathy\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHct:\u003c/strong\u003e Hematocrit\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLASSO:\u003c/strong\u003e Least absolute shrinkage and selection operator\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMOLLI:\u003c/strong\u003e Modified look-locker inversion recovery\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRMSE:\u003c/strong\u003e Root mean square error\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSCMR:\u003c/strong\u003e The Society for Cardiovascular Magnetic Resonance\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConception and study design: \u0026Ccedil;ağrı \u0026Ouml;zcan (\u0026Ccedil;\u0026Ouml;), Hasan Yiğit (HY). Conduction of experiments: \u0026Ccedil;\u0026Ouml;, HY, İrem \u0026Ouml;zcan. Interpretation of data: \u0026Ccedil;\u0026Ouml;, HY, Mehmet Serkan \u0026Ccedil;etin, Oğuzhan Tokur. Draft of manuscript: \u0026Ccedil;\u0026Ouml;. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnder the Declaration of Helsinki, the ethics committee approval was obtained by the Ankara Training and Research Hospital Clinical Research Ethics Committee on 08 January 2025. The ethics committee application number was E-24-360.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDue to the retrospective nature of the study, obtaining additional consent from the patients was not required.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompenting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003egenerative AI and AI-assisted technologies in the writing process\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDuring the preparation of this work, the author(s) used OpenAI\u0026rsquo;s ChatGPT in order to assist with English language editing, clarity improvements, and academic phrasing during the drafting of the manuscript\u0026rsquo;s Introduction, Methods, Discussion and Conclusion sections. After using this tool/service, the author(s) reviewed and edited the content as needed and take(s) full responsibility for the content of the publication.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMessroghli DR, Moon JC, Ferreira VM, et al. Clinical recommendations for cardiovascular magnetic resonance mapping of T1, T2, T2* and extracellular volume: a consensus statement by the Society for Cardiovascular Magnetic Resonance (SCMR) endorsed by the European Association for Cardiovascular Imaging (EACVI). \u003cem\u003eJournal of Cardiovascular Magnetic Resonance\u003c/em\u003e 2017; 19:1-24\u003c/li\u003e\n\u003cli\u003eArheden Hk, Saeed M, Higgins CB, et al. Reperfused rat myocardium subjected to various durations of ischemia: estimation of the distribution volume of contrast material with echo-planar MR imaging. \u003cem\u003eRadiology\u003c/em\u003e 2000; 215:520-528\u003c/li\u003e\n\u003cli\u003eArheden Hk, Saeed M, Higgins CB, et al. Measurement of the distribution volume of gadopentetate dimeglumine at echo-planar MR imaging to quantify myocardial infarction: comparison with 99mTc-DTPA autoradiography in rats. \u003cem\u003eRadiology\u003c/em\u003e 1999; 211:698-708\u003c/li\u003e\n\u003cli\u003eThirup P. Haematocrit: within-subject and seasonal variation. \u003cem\u003eSports Medicine\u003c/em\u003e 2003; 33:231-243\u003c/li\u003e\n\u003cli\u003eSu M-Y, Huang Y-S, Niisato E, et al. Is a timely assessment of the hematocrit necessary for cardiovascular magnetic resonance\u0026ndash;derived extracellular volume measurements? \u003cem\u003eJournal of Cardiovascular Magnetic Resonance\u003c/em\u003e 2020; 22:77\u003c/li\u003e\n\u003cli\u003eRobison S, Karur GR, Wald RM, Thavendiranathan P, Crean AM, Hanneman K. Noninvasive hematocrit assessment for cardiovascular magnetic resonance extracellular volume quantification using a point-of-care device and synthetic derivation. \u003cem\u003eJournal of Cardiovascular Magnetic Resonance\u003c/em\u003e 2018; 20:19\u003c/li\u003e\n\u003cli\u003eEngblom H, Kanski M, Kopic S, et al. Importance of standardizing timing of hematocrit measurement when using cardiovascular magnetic resonance to calculate myocardial extracellular volume (ECV) based on pre-and post-contrast T1 mapping. \u003cem\u003eJournal of Cardiovascular Magnetic Resonance\u003c/em\u003e 2018; 20:46\u003c/li\u003e\n\u003cli\u003eLundvall J, Bjerkhoel P, Quittenbaum S, Lindgren P. Rapid plasma volume decline upon quiet standing reflects large filtration capacity in dependent limbs. \u003cem\u003eActa physiologica scandinavica\u003c/em\u003e 1996; 158:161-167\u003c/li\u003e\n\u003cli\u003eJacob G, Ertl AC, Shannon JR, Furlan R, Robertson RM, Robertson D. Effect of standing on neurohumoral responses and plasma volume in healthy subjects. \u003cem\u003eJournal of Applied Physiology\u003c/em\u003e 1998; 84:914-921\u003c/li\u003e\n\u003cli\u003eFontana M, Banypersad SM, Treibel TA, et al. Differential myocyte responses in patients with cardiac transthyretin amyloidosis and light-chain amyloidosis: a cardiac MR imaging study. \u003cem\u003eRadiology\u003c/em\u003e 2015; 277:388-397\u003c/li\u003e\n\u003cli\u003eBanypersad SM, Sado DM, Flett AS, et al. Quantification of myocardial extracellular volume fraction in systemic AL amyloidosis: an equilibrium contrast cardiovascular magnetic resonance study. \u003cem\u003eCirculation: Cardiovascular Imaging\u003c/em\u003e 2013; 6:34-39\u003c/li\u003e\n\u003cli\u003eSyed IS, Glockner JF, Feng D, et al. Role of cardiac magnetic resonance imaging in the detection of cardiac amyloidosis. \u003cem\u003eJACC: cardiovascular imaging\u003c/em\u003e 2010; 3:155-164\u003c/li\u003e\n\u003cli\u003eTreibel TA, Nasis A, Fontana M, et al. An instantaneous ECV with no blood sampling: using native blood T1 for hematocrit is as good as standard ECV. \u003cem\u003eJournal of Cardiovascular Magnetic Resonance\u003c/em\u003e 2015; 17:1-2\u003c/li\u003e\n\u003cli\u003eYin J, Qin J, Liu W, et al. A comparative study of synthetic and venous hematocrit for calculating cardiovascular magnetic resonance-derived extracellular volume. \u003cem\u003eThe International Journal of Cardiovascular Imaging\u003c/em\u003e 2024:1-10\u003c/li\u003e\n\u003cli\u003eChen W, Doeblin P, Al-Tabatabaee S, et al. Synthetic extracellular volume in cardiac magnetic resonance without blood sampling: a reliable tool to replace conventional extracellular volume. \u003cem\u003eCirculation: Cardiovascular Imaging\u003c/em\u003e 2022; 15:e013745\u003c/li\u003e\n\u003cli\u003eOpatril L, Panovsky R, Machal J, et al. Extracellular volume quantification using synthetic haematocrit assessed from native and post-contrast longitudinal relaxation T1 times of a blood pool. \u003cem\u003eBMC Cardiovascular Disorders\u003c/em\u003e 2021; 21:363\u003c/li\u003e\n\u003cli\u003eShang Y, Zhang X, Zhou X, Wang J. Extracellular volume fraction measurements derived from the longitudinal relaxation of blood-based synthetic hematocrit may lead to clinical errors in 3 T cardiovascular magnetic resonance. \u003cem\u003eJournal of Cardiovascular Magnetic Resonance\u003c/em\u003e 2018; 20:56\u003c/li\u003e\n\u003cli\u003eLim E-H, Le T-T, Bryant J, et al. Importance of sex-specific regression models to estimate synthetic hematocrit and extracellular volume fraction. \u003cem\u003eJACC: Cardiovascular Imaging\u003c/em\u003e 2018; 11:1366-1367\u003c/li\u003e\n\u003cli\u003eKammerlander AA, Duca F, Binder C, et al. Extracellular volume quantification by cardiac magnetic resonance imaging without hematocrit sampling: ready for prime time? \u003cem\u003eWiener klinische Wochenschrift\u003c/em\u003e 2018; 130:190-196\u003c/li\u003e\n\u003cli\u003eFent GJ, Garg P, Foley JR, et al. Synthetic myocardial extracellular volume fraction. \u003cem\u003eJACC: Cardiovascular Imaging\u003c/em\u003e 2017; 10:1402-1404\u003c/li\u003e\n\u003cli\u003eTreibel TA, Fontana M, Maestrini V, et al. Automatic measurement of the myocardial interstitium: synthetic extracellular volume quantification without hematocrit sampling. \u003cem\u003eJACC: Cardiovascular Imaging\u003c/em\u003e 2016; 9:54-63\u003c/li\u003e\n\u003cli\u003eRaucci Jr FJ, Parra DA, Christensen JT, et al. Synthetic hematocrit derived from the longitudinal relaxation of blood can lead to clinically significant errors in measurement of extracellular volume fraction in pediatric and young adult patients. \u003cem\u003eJournal of Cardiovascular Magnetic Resonance\u003c/em\u003e 2016; 19:58\u003c/li\u003e\n\u003cli\u003eBluemke DA, Kawel-Boehm N. Can a MR imaging scanner accurately measure hematocrit to determine ECV fraction? In: American College of Cardiology Foundation Washington, DC, 2016:64-66\u003c/li\u003e\n\u003cli\u003e\u0026Ouml;zcan \u0026Ccedil;, Yiğit H, \u0026Ccedil;etin MS, \u0026Ouml;zcan İ. Analysis of myocardial T1, T2, and T2* values by age, sex, and cardiac segments in normal population: a prospective study. \u003cem\u003eThe International Journal of Cardiovascular Imaging\u003c/em\u003e 2024; \u003c/li\u003e\n\u003cli\u003eGottbrecht M, Kramer CM, Salerno M. Native T1 and extracellular volume measurements by cardiac MRI in healthy adults: a meta-analysis. \u003cem\u003eRadiology\u003c/em\u003e 2019; 290:317\u003c/li\u003e\n\u003cli\u003ePiechnik SK, Ferreira VM, Lewandowski AJ, et al. Normal variation of magnetic resonance T1 relaxation times in the human population at 1.5 T using ShMOLLI. \u003cem\u003eJournal of Cardiovascular Magnetic Resonance\u003c/em\u003e 2013; 15:13\u003c/li\u003e\n\u003cli\u003eArbelo E, Protonotarios A, Gimeno JR, et al. 2023 ESC Guidelines for the management of cardiomyopathies. \u003cem\u003eEur Heart J\u003c/em\u003e 2023; 44:3503-3626\u003c/li\u003e\n\u003cli\u003eSchulz-Menger J, Bluemke DA, Bremerich J, et al. Standardized image interpretation and post-processing in cardiovascular magnetic resonance - 2020 update: Society for Cardiovascular Magnetic Resonance (SCMR): Board of Trustees Task Force on Standardized Post-Processing. \u003cem\u003eJournal of Cardiovascular Magnetic Resonance\u003c/em\u003e 2020; 22:19\u003c/li\u003e\n\u003cli\u003eGreen SB. How many subjects does it take to do a regression analysis. \u003cem\u003eMultivariate behavioral research\u003c/em\u003e 1991; 26:499-510\u003c/li\u003e\n\u003cli\u003eBiso S, Weber J, Philip S, Omar K. Excellent Reproducibility of Synthetic ECV Without Blood Extraction Across Different Cardiomyopathies Using Published Regression. \u003cem\u003eJournal of Cardiovascular Magnetic Resonance\u003c/em\u003e 2024; 26\u003c/li\u003e\n\u003cli\u003eKellman P, Wilson JR, Xue H, et al. Extracellular volume fraction mapping in the myocardium, part 2: initial clinical experience. \u003cem\u003eJ Cardiovasc Magn Reson\u003c/em\u003e 2012; 14:64\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"the-international-journal-of-cardiovascular-imaging","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"caim","sideBox":"Learn more about [The International Journal of Cardiovascular Imaging](https://www.springer.com/journal/10554)","snPcode":"10554","submissionUrl":"https://submission.nature.com/new-submission/10554/3","title":"The International Journal of Cardiovascular Imaging","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Cardiac magnetic resonance imaging (CMR imaging), diffuse interstitial myocardial fibrosis, extracellular volüme fraction (ECV fraction), Myocardial Mapping, Hematocrit","lastPublishedDoi":"10.21203/rs.3.rs-6821726/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6821726/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground and Purpose\u003c/h2\u003e \u003cp\u003eCardiac magnetic resonance (CMR) is a key modality for non-invasive extracellular volume (ECV) quantification, typically requiring venous hematocrit (Hct). However, fluctuations in Hct can affect accuracy. This study aimed to develop a synthetic Hct model integrating age, sex, and pre- and post-contrast blood pool T1 values, calibrated against an optimized venous Hct reference.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eCardiac MRI data from 253 patients, including those with various pathologies and normal findings, were retrospectively analyzed for ECV calculations. To minimize postural fluctuations, venous Hct samples were taken after the patients remained in a stable supine position for at least 15 minutes, just before contrast injection. Hct\u003csub\u003esyn\u003c/sub\u003e values were derived from pre-contrast and post-contrast T1 mapping sequences. A comprehensive Hct\u003csub\u003esyn\u003c/sub\u003e model was developed by integrating age, sex, and both pre- and post-contrast blood pool T1 values. Unlike previous studies, these variables were evaluated collectively, and the model was calibrated using an optimized venous Hct reference. The performance of the derived model at high ECV values was further evaluated in the hypertrophic cardiomyopathy (HCM) subcohort.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThe Hct\u003csub\u003esyn\u003c/sub\u003e model showed strong agreement with measured ECV (bias: 0.31%, RMSE: 1.25, CCC: 0.949). In the HCM subcohort, it maintained high correlation (r\u0026thinsp;=\u0026thinsp;0.98) with a clinically acceptable bias of \u0026minus;\u0026thinsp;1.16%. Compared to sex-specific formulas by Chen et al., the model demonstrated improved performance (R\u0026sup2; = 0.56) with lower variability.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThis standardized synthetic Hct model enables accurate, non-invasive ECV estimation without blood sampling and demonstrates superior performance, especially in patients with elevated ECV values.\u003c/p\u003e","manuscriptTitle":"Optimizing Synthetic Hematocrit for Cardiac MRI: A Multivariable Model Calibrated with Standardized Venous Sampling ","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-16 06:36:04","doi":"10.21203/rs.3.rs-6821726/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-08-09T21:07:43+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-07T20:51:18+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"94482349082190209446462644865435680475","date":"2025-07-21T12:46:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"310136040892895979225208367810087035544","date":"2025-06-16T07:45:01+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-06-11T01:35:49+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-06-05T03:14:32+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-06-05T03:13:28+00:00","index":"","fulltext":""},{"type":"submitted","content":"The International Journal of Cardiovascular Imaging","date":"2025-06-04T14:50:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"the-international-journal-of-cardiovascular-imaging","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"caim","sideBox":"Learn more about [The International Journal of Cardiovascular Imaging](https://www.springer.com/journal/10554)","snPcode":"10554","submissionUrl":"https://submission.nature.com/new-submission/10554/3","title":"The International Journal of Cardiovascular Imaging","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d7354c97-f60d-49f0-a24a-710fb918aef4","owner":[],"postedDate":"June 16th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-08-25T16:31:44+00:00","versionOfRecord":{"articleIdentity":"rs-6821726","link":"https://doi.org/10.1007/s10554-025-03504-9","journal":{"identity":"the-international-journal-of-cardiovascular-imaging","isVorOnly":false,"title":"The International Journal of Cardiovascular Imaging"},"publishedOn":"2025-08-22 16:28:55","publishedOnDateReadable":"August 22nd, 2025"},"versionCreatedAt":"2025-06-16 06:36:04","video":"","vorDoi":"10.1007/s10554-025-03504-9","vorDoiUrl":"https://doi.org/10.1007/s10554-025-03504-9","workflowStages":[]},"version":"v1","identity":"rs-6821726","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6821726","identity":"rs-6821726","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}