Deep-Learning-Based Disease Classification in Patients Undergoing Cine Cardiac MRI.

BACKGROUND: Automated approaches may allow for fast, reproducible clinical assessment of cardiovascular diseases from MRI. PURPOSE: To develop an MRI-based deep learning (DL) disease classification algorithm to distinguish among normal subjects (NORM), patients with dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy (HCM), and ischemic heart disease (IHD). STUDY TYPE: Retrospective. POPULATION: A total of 1337 subjects (55% female), comprising normal subjects (N = 568), and patients with DCM (N = 151), HCM (N = 177), and IHD (N = 441). FIELD STRENGTH/SEQUENCE: Balanced steady-state free precession cine sequence at 1.5/3.0 T. ASSESSMENT: Bi-ventricular morphological and functional features and global and segmental left ventricular strain features were automatically extracted from short- and long-axis cine images. Variational autoencoder models were trained on the extracted features and compared against consensus disease label provided by two expert readers (13 and 14 years of experience). Adding unlabeled, normal data to the training was explored to increase specificity of NORM class. STATISTICAL TESTS: Tenfold cross-validation for model development; mean, standard deviation (SD) for measurements; classification metrics: area under the curve (AUC), confusion matrix, accuracy, specificity, precision, recall; 95% confidence intervals; Mann-Whitney U test for significance. RESULTS: AUCs of 0.952 for NORM, 0.881 for DCM, 0.908 for HCM, and 0.856 for IHD and overall accuracy of 0.778 were obtained, with specificity of 0.908 for the NORM class using both SAX and LAX features. Longitudinal strain features slightly improved classification metrics by 0.001 to 0.03 points, except for HCM-AUC. Differences in accuracy, metrics for NORM class and HCM-AUC were statistically significant. Cotraining using unlabeled data increased the specificity for the NORM class to 0.961. DATA CONCLUSION: Cardiac function features automatically extracted from cine MRI have potential to be used for disease classification, especially for normal-abnormal classification. Feature analyses showed that strain features were important for disease labeling. Cotraining using unlabeled data may help to increase specificity for normal-abnormal classification. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY: Stage 1.

Artificial intelligence-based fully automated stress left ventricular ejection fraction as a prognostic marker in patients undergoing stress cardiovascular magnetic resonance.

AIMS: This study aimed to determine in patients undergoing stress cardiovascular magnetic resonance (CMR) whether fully automated stress artificial intelligence (AI)-based left ventricular ejection fraction (LVEFAI) can provide incremental prognostic value to predict death above traditional prognosticators. METHODS AND RESULTS: Between 2016 and 2018, we conducted a longitudinal study that included all consecutive patients referred for vasodilator stress CMR. LVEFAI was assessed using AI algorithm combines multiple deep learning networks for LV segmentation. The primary outcome was all-cause death assessed using the French National Registry of Death. Cox regression was used to evaluate the association of stress LVEFAI with death after adjustment for traditional risk factors and CMR findings. In 9712 patients (66 ± 15 years, 67% men), there was an excellent correlation between stress LVEFAI and LVEF measured by expert (LVEFexpert) (r = 0.94, P < 0.001). Stress LVEFAI was associated with death [median (interquartile range) follow-up 4.5 (3.7-5.2) years] before and after adjustment for risk factors [adjusted hazard ratio, 0.84 (95% confidence interval, 0.82-0.87) per 5% increment, P < 0.001]. Stress LVEFAI had similar significant association with death occurrence compared with LVEFexpert. After adjustment, stress LVEFAI value showed the greatest improvement in model discrimination and reclassification over and above traditional risk factors and stress CMR findings (C-statistic improvement: 0.11; net reclassification improvement = 0.250; integrative discrimination index = 0.049, all P < 0.001; likelihood-ratio test P < 0.001), with an incremental prognostic value over LVEFAI determined at rest. CONCLUSION: AI-based fully automated LVEF measured at stress is independently associated with the occurrence of death in patients undergoing stress CMR, with an additional prognostic value above traditional risk factors, inducible ischaemia and late gadolinium enhancement.

AI-based, automated chamber volumetry from gated, non-contrast CT.

BACKGROUND: Accurate chamber volumetry from gated, non-contrast cardiac CT (NCCT) scans can be useful for potential screening of heart failure. OBJECTIVES: To validate a new, fully automated, AI-based method for cardiac volume and myocardial mass quantification from NCCT scans compared to contrasted CT Angiography (CCTA). METHODS: Of a retrospectively collected cohort of 1051 consecutive patients, 420 patients had both NCCT and CCTA scans at mid-diastolic phase, excluding patients with cardiac devices. Ground truth values were obtained from the CCTA scans. RESULTS: The NCCT volume computation shows good agreement with ground truth values. Volume differences [95% CI ] and correlation coefficients were: -9.6 [-45; 26] mL, r â€‹= â€‹0.98 for LV Total, -5.4 [-24; 13] mL, r â€‹= â€‹0.95 for LA, -8.7 [-45; 28] mL, r â€‹= â€‹0.94 for RV, -5.2 [-27; 17] mL, r â€‹= â€‹0.92 for RA, -3.2 [-42; 36] mL, r â€‹= â€‹0.91 for LV blood pool, and -6.7 [-39; 26] g, r â€‹= â€‹0.94 for LV wall mass, respectively. Mean relative volume errors of less than 7% were obtained for all chambers. CONCLUSIONS: Fully automated assessment of chamber volumes from NCCT scans is feasible and correlates well with volumes obtained from contrast study.

Deep learning-based left ventricular segmentation demonstrates improved performance on respiratory motion-resolved whole-heart reconstructions.

Introduction: Deep learning (DL)-based segmentation has gained popularity for routine cardiac magnetic resonance (CMR) image analysis and in particular, delineation of left ventricular (LV) borders for LV volume determination. Free-breathing, self-navigated, whole-heart CMR exams provide high-resolution, isotropic coverage of the heart for assessment of cardiac anatomy including LV volume. The combination of whole-heart free-breathing CMR and DL-based LV segmentation has the potential to streamline the acquisition and analysis of clinical CMR exams. The purpose of this study was to compare the performance of a DL-based automatic LV segmentation network trained primarily on computed tomography (CT) images in two whole-heart CMR reconstruction methods: (1) an in-line respiratory motion-corrected (Mcorr) reconstruction and (2) an off-line, compressed sensing-based, multi-volume respiratory motion-resolved (Mres) reconstruction. Given that Mres images were shown to have greater image quality in previous studies than Mcorr images, we hypothesized that the LV volumes segmented from Mres images are closer to the manual expert-traced left ventricular endocardial border than the Mcorr images. Method: This retrospective study used 15 patients who underwent clinically indicated 1.5 T CMR exams with a prototype ECG-gated 3D radial phyllotaxis balanced steady state free precession (bSSFP) sequence. For each reconstruction method, the absolute volume difference (AVD) of the automatically and manually segmented LV volumes was used as the primary quantity to investigate whether 3D DL-based LV segmentation generalized better on Mcorr or Mres 3D whole-heart images. Additionally, we assessed the 3D Dice similarity coefficient between the manual and automatic LV masks of each reconstructed 3D whole-heart image and the sharpness of the LV myocardium-blood pool interface. A two-tail paired Student's t-test (alpha = 0.05) was used to test the significance in this study. Results & Discussion: The AVD in the respiratory Mres reconstruction was lower than the AVD in the respiratory Mcorr reconstruction: 7.73 ± 6.54 ml vs. 20.0 ± 22.4 ml, respectively (n = 15, p-value = 0.03). The 3D Dice coefficient between the DL-segmented masks and the manually segmented masks was higher for Mres images than for Mcorr images: 0.90 ± 0.02 vs. 0.87 ± 0.03 respectively, with a p-value = 0.02. Sharpness on Mres images was higher than on Mcorr images: 0.15 ± 0.05 vs. 0.12 ± 0.04, respectively, with a p-value of 0.014 (n = 15). Conclusion: We conclude that the DL-based 3D automatic LV segmentation network trained on CT images and fine-tuned on MR images generalized better on Mres images than on Mcorr images for quantifying LV volumes.

Prognostic impact of artificial intelligence-based fully automated global circumferential strain in patients undergoing stress CMR.

AIMS: To determine whether fully automated artificial intelligence-based global circumferential strain (GCS) assessed during vasodilator stress cardiovascular (CV) magnetic resonance (CMR) can provide incremental prognostic value. METHODS AND RESULTS: Between 2016 and 2018, a longitudinal study included all consecutive patients with abnormal stress CMR defined by the presence of inducible ischaemia and/or late gadolinium enhancement. Control subjects with normal stress CMR were selected using a propensity score-matching. Stress-GCS was assessed using a fully automatic machine-learning algorithm based on featured-tracking imaging from short-axis cine images. The primary outcome was the occurrence of major adverse clinical events (MACE) defined as CV mortality or nonfatal myocardial infarction. Cox regressions evaluated the association between stress-GCS and the primary outcome after adjustment for traditional prognosticators. In 2152 patients [66 ± 12 years, 77% men, 1:1 matched patients (1076 with normal and 1076 with abnormal CMR)], stress-GCS was associated with MACE [median follow-up 5.2 (4.8-5.5) years] after adjustment for risk factors in the propensity-matched population [adjusted hazard ratio (HR), 1.12 (95% CI, 1.06-1.18)], and patients with normal CMR [adjusted HR, 1.35 (95% CI, 1.19-1.53), both P < 0.001], but not in patients with abnormal CMR (P = 0.058). In patients with normal CMR, an increased stress-GCS showed the best improvement in model discrimination and reclassification above traditional and stress CMR findings (C-statistic improvement: 0.14; NRI = 0.430; IDI = 0.089, all P < 0.001; LR-test P < 0.001). CONCLUSION: Stress-GCS is not a predictor of MACE in patients with ischaemia, but has an incremental prognostic value in those with a normal CMR although the absolute event rate remains low.

AI-Based Fully Automated Left Atrioventricular Coupling Index as a Prognostic Marker in Patients Undergoing Stress CMR.

BACKGROUND: The left atrioventricular coupling index (LACI) is a strong and independent predictor of heart failure (HF) in individuals without clinical cardiovascular disease. Its prognostic value is not established in patients with cardiovascular disease. OBJECTIVES: This study sought to determine in patients undergoing stress cardiac magnetic resonance (CMR) whether fully automated artificial intelligence-based LACI can provide incremental prognostic value to predict HF. METHODS: Between 2016 and 2018, the authors conducted a longitudinal study including all consecutive patients with abnormal (inducible ischemia or late gadolinium enhancement) vasodilator stress CMR. Control subjects with normal stress CMR were selected using propensity score matching. LACI was defined as the ratio of left atrial to left ventricular end-diastolic volumes. The primary outcome included hospitalization for acute HF or cardiovascular death. Cox regression was used to evaluate the association of LACI with the primary outcome after adjustment for traditional risk factors. RESULTS: In 2,134 patients (65 ± 12 years, 77% men, 1:1 matched patients [1,067 with normal and 1,067 with abnormal CMR]), LACI was positively associated with the primary outcome (median follow-up: 5.2 years [IQR: 4.8-5.5 years]) before and after adjustment for risk factors in the overall propensity-matched population (adjusted HR: 1.18 [95% CI: 1.13-1.24]), in patients with abnormal CMR (adjusted HR per 0.1% increment: 1.22 [95% CI: 1.14-1.30]), and in patients with normal CMR (adjusted HR per 0.1% increment: 1.12 [95% CI: 1.05-1.20]) (all P < 0.001). After adjustment, a higher LACI of ≥25% showed the greatest improvement in model discrimination and reclassification over and above traditional risk factors and stress CMR findings (C-index improvement: 0.16; net reclassification improvement = 0.388; integrative discrimination index = 0.153, all P < 0.001; likelihood ratio test P < 0.001). CONCLUSIONS: LACI is independently associated with hospitalization for HF and cardiovascular death in patients undergoing stress CMR, with an incremental prognostic value over traditional risk factors including inducible ischemia and late gadolinium enhancement.