Abstract 

This study investigated the feasibility and prognostic relevance of threshold-based quantification of myocardial delayed enhancement (MDE) on CT in patients with nonischemic dilated cardiomyopathy (NIDCM).

Materials and Methods

Forty-three patients with NIDCM (59.3 ± 17.1 years; 21 male) were included in the study and underwent cardiac CT and MRI. MDE was quantified manually and with a threshold-based quantification method using cutoffs of 2, 3, and 4 standard deviations (SDs) on three sets of CT images (100 kVp, 120 kVp, and 70 keV). Interobserver agreement in MDE quantification was assessed using the intraclass correlation coefficient (ICC). Agreement between CT and MRI was evaluated using the Bland-Altman method and the concordance correlation coefficient (CCC). Patients were followed up for the subsequent occurrence of the primary composite outcome, including cardiac death, heart transplantation, heart failure hospitalization, or appropriate use of an implantable cardioverter-defibrillator. The Kaplan–Meier method was used to estimate event-free survival according to MDE levels.

Results

Late gadolinium enhancement (LGE) was observed in 29 patients (67%, 29/43), and the mean LGE found with the 5-SD threshold was 4.1% ± 3.6%. The 4-SD threshold on 70-keV CT showed excellent interobserver agreement (ICC = 0.810) and the highest concordance with MRI (CCC = 0.803). This method also yielded the smallest bias with the narrowest range of 95% limits of agreement compared to MRI (bias, -0.119%; 95% limits of agreement, -4.216% to 3.978%). During a median follow-up of 1625 days (interquartile range, 712–1430 days), 10 patients (23%, 10/43) experienced the primary composite outcome. Event-free survival significantly differed between risk subgroups divided by the optimal MDE cutoff of 4.3% (log-rank P = 0.005).

Conclusion

The 4-SD threshold on 70-keV monochromatic CT yielded results comparable to those of MRI for quantifying MDE as a marker of myocardial fibrosis, which showed prognostic value in patients with NIDCM.

Keywords

Myocardial fibrosis; Myocardial delayed enhancement; Semi-automated quantification; Nonischemic dilated cardiomyopathy

Journal: Korean Journal of Radiology 

Author

Na Young Kim, Dong Jin Im, Yoo Jin Hong, Byoung Wook Choi, Seok-Min Kang, Jong-Chan Youn, and Hye-Jeong Lee

DOI: 10.3348/kjr.2023.1271

Abstract

Cardiac diffusion weighted-magnetic resonance imaging (DWI) has slowly developed due to its technical difficulties. However, this limitation could be overcome by advanced techniques, including a stimulated echo technique and a gradient moment nulling technique. This study aimed to develop and validate a high-order DWI sequence, using echo-planar imaging (EPI) and second-order motion-compensated (M012) diffusion gradient applied to cardiac imaging in small-sized animals with fast heart and respiratory rates, and to investigate the feasibility of cardiac DWI, diagnosing acute myocardial injury in isoproterenol-induced myocardial injury rat models. The M012 diffusion gradient sequence was designed for diffusion tensor imaging of the rat myocardium and validated in the polyvinylpyrrolidone phantom. Following sequence optimization, 23 rats with isoproterenol-induced acute myocardial injury and five healthy control rats underwent cardiac MRI, including cine imaging, T1 mapping, and DWI. Diffusion gradient was applied using a 9.4-T MRI scanner (Bruker, BioSpec 94/20, gradient amplitude = 440 mT/m, maximum slew rate = 3440 T/m/s) with double gating (electrocardiogram and respiratory gating). Troponin I was used as a serum biomarker for myocardial injury. Histopathologic examination of the heart was subsequently performed. The developed DWI sequence using EPI and M012 provided the interpretable images of rat hearts. The apparent diffusion coefficient (ADC) values were significantly higher in rats with acute myocardial injury than in the control group (1.847 ± 0.326 * 10–3 mm2/s vs. 1.578 ± 0.144 * 10–3 mm2/s, P < 0.001). Troponin I levels were increased in the blood samples of rats with acute myocardial injury (P < 0.001). Histopathologic examinations detected myocardial damage and subendocardial fibrosis in rats with acute myocardial injury. The newly developed DWI technique has the ability to detect myocardial injury in small animal models, representing high ADC values on the myocardium with isoproterenol-induced injury.

Journal: Scientific Report

Author

Chul Hwan Park, Pan Ki Kim, Yoonjung Kim, Tae Hoon Kim, Yoo Jin Hong, Eunkyung Ahn, Yoon Jin Cha & Byoung Wook Choi 

DOI: doi.org/10.1038/s41598-024-52746-5

Abstract 

A short-time cardiac imaging protocol added to breast MRI achieved excellent results for interscan and intercoil reproducibility and variability in ventricular functional or tissue characterization parameters in both healthy controls and participants with breast cancer who underwent chemotherapy.

Purpose

To investigate the feasibility and interscan variability of short-time cardiac MRI protocol after chemotherapy in individuals with breast cancer.

Materials and Methods

A total of 13 healthy female controls (mean age, 52.4 years ± 13.2 [SD]) and 85 female participants with breast cancer (mean age, 51.8 years ± 9.9) undergoing chemotherapy prospectively underwent routine breast MRI and short-time cardiac MRI using a 3-T scanner with peripheral pulse gating in the prone position. Interscan, intercoil, and interobserver reproducibility and variability of native T1 and extracellular volume (ECV), as well as ventricular functional parameters, were measured using the intraclass correlation coefficient (ICC), standard error of measurement (SEM), or coefficient of variation (CoV).

Results

Left ventricular functional parameters had excellent interscan reproducibility (ICC ≥ 0.80). Left ventricular ejection fraction showed low interscan variability in control and chemotherapy participants (SEM, 2.0 and 1.2; CoV, 3.1 and 1.9, respectively). Native T1 showed excellent interscan (ICC, 0.75) and intercoil (ICC, 0.81) reproducibility in the control group and good interscan reproducibility (ICC, 0.72 and 0.73, respectively) in the participants undergoing immediate and remote chemotherapy. Interscan reproducibility for ECV was excellent in the control group and in the remote chemotherapy group (ICC, 0.93 and 0.88, respectively) and fair in the immediate chemotherapy group (ICC, 0.52). In the regional analysis, interscan repeatability and variability of native T1 and ECV were superior in the anteroseptum or inferoseptum than in other segments in the immediate chemotherapy group. Native T1 and ECV had good to excellent interobserver agreement across all groups.

Conclusion

Short-time cardiac MRI showed excellent results for interscan, intercoil, and interobserver reproducibility and variability for ventricular functional or tissue characterization parameters, suggesting that this modality is feasible for routine surveillance of cardiotoxicity evaluation in individuals with breast cancer.

Keywords: Cardiac MRI, Heart, Cardiomyopathy

Journal: Radiology: Cardiothoracic Imaging 

Author

Yoo Jin Hong, Kyunghwa Han, Hye-Jeong Lee, Jin Hur, Young Jin Kim, Min Jung Kim, and Byoung Wook Choi 

DOI: 10.1148/ryct.220229

Abstract

Background: The reliability and diagnostic performance of deep learning (DL)-based automated T2 measurements on T2 map of 3.0-T cardiac magnetic resonance imaging (MRI) using multi-institutional datasets have not been investigated. We aimed to evaluate the performance of a DL-based software for measuring automated T2 values from 3.0-T cardiac MRI obtained at two centers.

Methods: Eighty-three subjects were retrospectively enrolled from two centers (42 healthy subjects and 41 patients with myocarditis) to validate a commercial DL-based software that was trained to segment the left ventricular myocardium and measure T2 values on T2 mapping sequences. Manual reference T2 values by two experienced radiologists and those calculated by the DL-based software were obtained. The segmentation performance of the DL-based software and the non-inferiority of automated T2 values were assessed compared with the manual reference standard per segment level. The software’s performance in detecting elevated T2 values was assessed by calculating the sensitivity, specificity, and accuracy per segment.

Results: The average Dice similarity coefficient for segmentation of myocardium on T2 maps was 0.844. The automated T2 values were non-inferior to the manual reference T2 values on a per-segment analysis (45.35 vs. 44.32 ms). The DL-based software exhibited good performance (sensitivity: 83.6–92.8%; specificity: 82.5–92.0%; accuracy: 82.7–92.2%) in detecting elevated T2 values.

Conclusions: The DL-based software for automated T2 map analysis yields non-inferior measurements at the per-segment level and good performance for detecting myocardial segments with elevated T2 values compared with manual analysis.

Keywords: Magnetic resonance imaging (MRI); heart; deep learning (DL); T2 map; myocarditis

Journal: Quantitative Imaging in Medicine and Surgery

Author

Hwan Kim, Young Joong Yang, Kyunghwa Han, Pan Ki Kim, Byoung Wook Choi, Jin Young Kim, Young Joo Suh

DOI: 10.21037/qims-23-375

Abstract

Objective

To quantitatively analyze the cardiac magnetic resonance imaging (CMR) characteristics of chemotherapy-related cardiac dysfunction (CTRCD) and explore their prognostic value for major adverse cardiovascular events (MACE).

Materials and Methods

A total of 145 patients (male:female = 76:69, mean age = 63.0 years) with cancer and heart failure who underwent CMR between January 2015 and January 2021 were included. CMR was performed using a 3T scanner (Siemens). Biventricular functions, native T1 T2, extracellular volume fraction (ECV) values, and late gadolinium enhancement (LGE) of the left ventricle (LV) were compared between those with and without CTRCD. These were compared between patients with mild-to-moderate CTRCD and those with severe CTRCD. Cox proportional hazard regression analysis was used to evaluate the association between the CMR parameters and MACE occurrence during follow-up in the CTRCD patients.

Results

Among 145 patients, 61 had CTRCD and 84 did not have CTRCD. Native T1, ECV, and T2 were significantly higher in the CTRCD group (1336.9 ms, 32.5%, and 44.7 ms, respectively) than those in the non-CTRCD group (1303.4 ms, 30.5%, and 42.0 ms, respectively; P = 0.013, 0.010, and < 0.001, respectively). They were not significantly different between patients with mild-to-moderate and severe CTRCD. Indexed LV mass was significantly smaller in the CTRCD group (65.0 g/m2 vs. 78.9 g/m2; P < 0.001). According to the multivariable Cox regression analysis, T2 (hazard ratio [HR]: 1.14, 95% confidence interval [CI]: 1.01–1.27; P = 0.028) and quantified LGE (HR: 1.07, 95% CI: 1.01–1.13; P = 0.021) were independently associated with MACE in the CTRCD patients.

Conclusion

Quantitative parameters from CMR have the potential to evaluate myocardial changes in CTRCD. Increased T2 with reduced LV mass was demonstrated in CTRCD patients even before the development of severe cardiac dysfunction. T2 and quantified LGE may be independent prognostic factors for MACE in patients with CTRCD.

Keywords

Cardiotoxicity; Magnetic resonance imaging; T1 mapping; T2 mapping; Late gadolinium enhancement

Journal: Korean Journal of Radiology

Author

Jinhee Kim, Yoo Jin Hong, Kyunghwa Han, Jin Young Kim, Hye-Jeong Lee, Jin Hur, Young Jin Kim, and Byoung Wook Choi

DOI: 10.3348/kjr.2023.0095

Abstract

Importance  Dual-energy chest radiography exhibits better sensitivity than single-energy chest radiography, partly due to its ability to remove overlying anatomical structures.

Objectives  To develop and validate a deep learning–based synthetic bone-suppressed (DLBS) nodule-detection algorithm for pulmonary nodule detection on chest radiographs.

Design, Setting, and Participants  This decision analytical modeling study used data from 3 centers between November 2015 and July 2019 from 1449 patients. The DLBS nodule-detection algorithm was trained using single-center data (institute 1) of 998 chest radiographs. The DLBS algorithm was validated using 2 external data sets (institute 2, 246 patients; and institute 3, 205 patients). Statistical analysis was performed from March to December 2021.

Exposures  DLBS nodule-detection algorithm.

Main Outcomes and Measures  The nodule-detection performance of DLBS model was compared with the convolution neural network nodule-detection algorithm (original model). Reader performance testing was conducted by 3 thoracic radiologists assisted by the DLBS algorithm or not. Sensitivity and false-positive markings per image (FPPI) were compared.

Results  Training data consisted of 998 patients (539 men [54.0%]; mean [SD] age, 54.2 [9.82] years), and 2 external validation data sets consisted of 246 patients (133 men [54.1%]; mean [SD] age, 55.3 [8.7] years) and 205 patients (105 men [51.2%]; mean [SD] age, 51.8 [9.1] years). Using the external validation data set of institute 2, the bone-suppressed model showed higher sensitivity compared with that of the original model for nodule detection (91.5% [109 of 119] vs 79.8% [95 of 119]; P < .001). The overall mean of FPPI with the bone-suppressed model was reduced compared with the original model (0.07 [17 of 246] vs 0.09 [23 of 246]; P < .001). For the observer performance testing with the data of institute 3, the mean sensitivity of 3 radiologists was 77.5% (95% [CI], 69.9%-85.2%), whereas that of radiologists assisted by DLBS modeling was 92.1% (95% CI, 86.3%-97.3%; P < .001). The 3 radiologists had a reduced number of FPPI when assisted by the DLBS model (0.071 [95% CI, 0.041-0.111] vs 0.151 [95% CI, 0.111-0.210]; P < .001).

Conclusions and Relevance  This decision analytical modeling study found that the DLBS model was more sensitive to detecting pulmonary nodules on chest radiographs compared with the original model. These findings suggest that the DLBS model could be beneficial to radiologists in the detection of lung nodules in chest radiographs without need of the specialized equipment or increase of radiation dose.

Journal: JAMA Network

Author

Hwiyoung Kim, Kye Ho Lee, Kyunghwa Han, Ji Won Lee, Jin Young Kim, Dong Jin Im, Yoo Jin Hong, Byoung Wook Choi, Jin Hur

DOI: 10.1001/jamanetworkopen.2022.53820

Abstract

With the increased availability of magnetic resonance imaging (MRI) and a progressive rise in the frequency of cardiac device implantation, there is an increased chance that patients with implanted cardiac devices require MRI examination during their lifetime. Though MRI is generally contraindicated in patients who have undergone pacemaker implantation with electronic circuits, the recent introduction of MR Conditional pacemaker allows physicians to take advantage of MRI to assess these patients during diagnosis and treatment. When MRI examinations of patients with pacemaker are requested, physicians must confirm whether the device is a conventional pacemaker or an MR Conditional pacemaker by reviewing chest radiographs or the electronic medical records (EMRs). The purpose of this study was to evaluate the utility of a deep convolutional neural network (DCNN) trained to detect pacemakers on chest radiographs and to determine the device’s subclassification. The DCNN perfectly detected pacemakers on chest radiographs and the accuracy of the subclassification of pacemakers using the internal and external test datasets were 100.0% (n = 106/106) and 90.1% (n = 279/308). The DCNN can be applied to the radiologic workflow for double-checking purposes, thereby improving patient safety during MRI and preventing busy physicians from making errors.

Journal: Journal of Medical Systems

Author

Yoonah Do, Soo Ho Ahn, Sungjun Kim, Jin Kyem Kim, Byoung Wook Choi, Hwiyoung Kim & Young Han Lee

DOI: doi.org/10.1007/s10916-023-01981-w

Abstract

Objective

This study aimed to develop and validate models using radiomics features on a native T1 map from cardiac magnetic resonance (CMR) to predict left ventricular reverse remodeling (LVRR) in patients with nonischemic dilated cardiomyopathy (NIDCM).

Materials and Methods

Data from 274 patients with NIDCM who underwent CMR imaging with T1 mapping at Severance Hospital between April 2012 and December 2018 were retrospectively reviewed. Radiomic features were extracted from the native T1 maps. LVRR was determined using echocardiography performed ≥ 180 days after the CMR. The radiomics score was generated using the least absolute shrinkage and selection operator logistic regression models. Clinical, clinical + late gadolinium enhancement (LGE), clinical + radiomics, and clinical + LGE + radiomics models were built using a logistic regression method to predict LVRR. For internal validation of the result, bootstrap validation with 1000 resampling iterations was performed, and the optimism-corrected area under the receiver operating characteristic curve (AUC) with 95% confidence interval (CI) was computed. Model performance was compared using AUC with the DeLong test and bootstrap.

Results

Among 274 patients, 123 (44.9%) were classified as LVRR-positive and 151 (55.1%) as LVRR-negative. The optimism-corrected AUC of the radiomics model in internal validation with bootstrapping was 0.753 (95% CI, 0.698–0.813). The clinical + radiomics model revealed a higher optimism-corrected AUC than that of the clinical + LGE model (0.794 vs. 0.716; difference, 0.078 [99% CI, 0.003–0.151]). The clinical + LGE + radiomics model significantly improved the prediction of LVRR compared with the clinical + LGE model (optimism-corrected AUC of 0.811 vs. 0.716; difference, 0.095 [99% CI, 0.022–0.139]).

Conclusion

The radiomic characteristics extracted from a non-enhanced T1 map may improve the prediction of LVRR and offer added value over traditional LGE in patients with NIDCM. Additional external validation research is required.

Keywords

Cardiomyopathy, Dilated; Magnetic resonance imaging; Prognosis; Radiomics

Journal: Journal of Medical Systems

Author

Suyon Chang, Kyunghwa Han, Yonghan Kwon, Lina Kim, Seunghyun Hwang, Hwiyoung Kim, and Byoung Wook Choi

DOI: 10.3348/kjr.2023.0065

Abstract

Objective

T1 mapping provides valuable information regarding cardiomyopathies. Manual drawing is time consuming and prone to subjective errors. Therefore, this study aimed to test a DL algorithm for the automated measurement of native T1 and extracellular volume (ECV) fractions in cardiac magnetic resonance (CMR) imaging with a temporally separated dataset.

Materials and Methods

CMR images obtained for 95 participants (mean age ± standard deviation, 54.5 ± 15.2 years), including 36 left ventricular hypertrophy (12 hypertrophic cardiomyopathy, 12 Fabry disease, and 12 amyloidosis), 32 dilated cardiomyopathy, and 27 healthy volunteers, were included. A commercial deep learning (DL) algorithm based on 2D U-net (Myomics-T1 software, version 1.0.0) was used for the automated analysis of T1 maps. Four radiologists, as study readers, performed manual analysis. The reference standard was the consensus result of the manual analysis by two additional expert readers. The segmentation performance of the DL algorithm and the correlation and agreement between the automated measurement and the reference standard were assessed. Interobserver agreement among the four radiologists was analyzed.

Results

DL successfully segmented the myocardium in 99.3% of slices in the native T1 map and 89.8% of slices in the post-T1 map with Dice similarity coefficients of 0.86 ± 0.05 and 0.74 ± 0.17, respectively. Native T1 and ECV showed strong correlation and agreement between DL and the reference: for T1, r = 0.967 (95% confidence interval [CI], 0.951–0.978) and bias of 9.5 msec (95% limits of agreement [LOA], -23.6–42.6 msec); for ECV, r = 0.987 (95% CI, 0.980–0.991) and bias of 0.7% (95% LOA, -2.8%–4.2%) on per-subject basis. Agreements between DL and each of the four radiologists were excellent (intraclass correlation coefficient [ICC] of 0.98–0.99 for both native T1 and ECV), comparable to the pairwise agreement between the radiologists (ICC of 0.97–1.00 and 0.99–1.00 for native T1 and ECV, respectively).

Conclusion

The DL algorithm allowed automated T1 and ECV measurements comparable to those of radiologists.

Keywords

Magnetic resonance imaging; Heart; Deep learning; T1 mapping; Extracellular volume fraction

Journal: Korean Journal of Radiology

Author

Suyon Chang, Kyunghwa Han, Suji Lee, Young Joong Yang, Pan Ki Kim, Byoung Wook Choi, and Young Joo Suh

DOI: doi.org/10.1007/s00330-022-08936-8

Abstract

We aimed to determine the effects of deep learning-based reconstruction (DLR) on radiomic features obtained from cardiac computed tomography (CT) by comparing with iterative reconstruction (IR), and filtered back projection (FBP). A total of 284 consecutive patients with 285 cardiac CT scans that were reconstructed with DLR, IR, and FBP, were retrospectively enrolled. Radiomic features were extracted from the left ventricular (LV) myocardium, and from the periprosthetic mass if patients had cardiac valve replacement. Radiomic features of LV myocardium from each reconstruction were compared using a fitting linear mixed model. Radiomics models were developed to diagnose periprosthetic abnormality, and the performance was evaluated using the area under the receiver characteristics curve (AUC). Most radiomic features of LV myocardium (73 of 88) were significantly different in pairwise comparisons between all three reconstruction methods (P < 0.05). The radiomics model on IR exhibited the best diagnostic performance (AUC 0.948, 95% CI 0.880–1), relative to DLR (AUC 0.873, 95% CI 0.735–1) and FBP (AUC 0.875, 95% CI 0.731–1), but these differences did not reach significance (P > 0.05). In conclusion, applying DLR to cardiac CT scans yields radiomic features distinct from those obtained with IR and FBP, implying that feature robustness is not guaranteed when applying DLR.

Journal: Scientific Reports

Author

Sei Hyun Chun, Young Joo Suh, Kyunghwa Han, Yonghan Kwon, Aaron Youngjae Kim & Byoung Wook Choi 

DOI: doi.org/10.1038/s41598-022-19546-1