Early Prediction of Left Ventricular Reverse Remodeling in First-Diagnosed Idiopathic Dilated Cardiomyopathy: A Comparison of Linear Model, Random Forest, and Extreme Gradient Boosting

The Prognostic Significance of the DBIL/HDLC Ratio in Patients With Dilated Cardiomyopathy

Background: In cardiovascular pathology, both direct bilirubin (DBIL) and high-density lipoprotein cholesterol (HDLC) have been associated with adverse clinical outcomes. However, the prognostic significance of these biomarkers in the context of dilated cardiomyopathy (DCM) remains unclear. To address this gap, this study conducted a retrospective analysis to evaluate the prognostic value of the DBIL/HDLC ratio in patients diagnosed with DCM. Methods and Results: A total of 986 consecutive DCM patients were retrospectively enrolled from January 2010 to December 2019 and divided into two groups based on the DBIL/HDLC ratio cut-off value: ≤ 4.45 (n = 483) and > 4.45 (n = 503). Patients with lower DBIL/HDLC (≤ 4.45) experienced lower in-hospital mortality, long-term mortality, and major adverse clinical events (MACEs) (0.8%, 32.9%, and 12.2%, respectively) compared to those with higher DBIL/HDLC (> 4.45) (6.4%, 59.1%, and 16.7%, respectively). Multivariate analysis identified DBIL/HDLC as an independent risk factor for long-term mortality (odds ratio: 1.026; 95% confidence interval (CI): 1.005-1.048; p = 0.016) and all-cause mortality over a median follow-up of 67 ± 1.8 months (hazard ratio: 1.011; 95% CI: 1.005-1.018; p < 0.001). The receiver operating characteristic curve showed good discrimination for long-term mortality (area under the curve (AUC): 0.675; 95% CI: 0.692-0.708; p < 0.001). The Kaplan-Meier survival analysis demonstrated a better prognosis for patients with DBIL/HDLC ≤ 4.45 (log-rank χ 2 = 40.356, p < 0.001). Furthermore, the impact of additional variables on the results was investigated by a subgroup analysis. Conclusion: The DBIL/HDLC ratio could serve as a simple and cost-effective tool for evaluating prognosis in DCM.

Transient Left Ventricular Dysfunction from Cardiomyopathies to Myocardial Viability: When and Why Cardiac Function Recovers

Transient left ventricular dysfunction (TLVD), a temporary condition marked by reversible impairment of ventricular function, remains an underdiagnosed yet significant contributor to morbidity and mortality in clinical practice. Unlike the well-explored atherosclerotic disease of the epicardial coronary arteries, the diverse etiologies of TLVD require greater attention for proper diagnosis and management. The spectrum of disorders associated with TLVD includes stress-induced cardiomyopathy, central nervous system injuries, histaminergic syndromes, various inflammatory diseases, pregnancy-related conditions, and genetically determined syndromes. Furthermore, myocardial infarction with non-obstructive coronary arteries (MINOCA) origins such as coronary artery spasm, coronary thromboembolism, and spontaneous coronary artery dissection (SCAD) may also manifest as TLVD, eventually showing recovery. This review highlights the range of ischemic and non-ischemic clinical situations that lead to TLVD, gathering conditions like Tako-Tsubo Syndrome (TTS), Kounis syndrome (KS), Myocarditis, Peripartum Cardiomyopathy (PPCM), and Tachycardia-induced cardiomyopathy (TIC). Differentiation amongst these causes is crucial, as they involve distinct clinical, instrumental, and genetic predictors that bode different outcomes and recovery potential for left ventricular function. The purpose of this review is to improve everyday clinical approaches to treating these diseases by providing an extensive survey of conditions linked with TLVD and the elements impacting prognosis and outcomes.

Design and Analysis of a Polymeric Left Ventricular Simulator via Computational Modelling

Preclinical testing of medical devices is an essential step in the product life cycle, whereas testing of cardiovascular implants requires specialised testbeds or numerical simulations using computer software Ansys 2016. Existing test setups used to evaluate physiological scenarios and test cardiac implants such as mock circulatory systems or isolated beating heart platforms are driven by sophisticated hardware which comes at a high cost or raises ethical concerns. On the other hand, computational methods used to simulate blood flow in the cardiovascular system may be simplified or computationally expensive. Therefore, there is a need for low-cost, relatively simple and efficient test beds that can provide realistic conditions to simulate physiological scenarios and evaluate cardiovascular devices. In this study, the concept design of a novel left ventricular simulator made of latex rubber and actuated by pneumatic artificial muscles is presented. The designed left ventricular simulator is geometrically similar to a native left ventricle, whereas the basal diameter and long axis length are within an anatomical range. Finite element simulations evaluating left ventricular twisting and shortening predicted that the designed left ventricular simulator rotates approximately 17 degrees at the apex and the long axis shortens around 11 mm. Experimental results showed that the twist angle is 18 degrees and the left ventricular simulator shortens 5 mm. Twist angles and long axis shortening as in a native left ventricle show it is capable of functioning like a native left ventricle and simulating a variety of scenarios, and therefore has the potential to be used as a test platform.

Shallow and deep learning classifiers in medical image analysis

An increasingly strong connection between artificial intelligence and medicine has enabled the development of predictive models capable of supporting physicians' decision-making. Artificial intelligence encompasses much more than machine learning, which nevertheless is its most cited and used sub-branch in the last decade. Since most clinical problems can be modeled through machine learning classifiers, it is essential to discuss their main elements. This review aims to give primary educational insights on the most accessible and widely employed classifiers in radiology field, distinguishing between "shallow" learning (i.e., traditional machine learning) algorithms, including support vector machines, random forest and XGBoost, and "deep" learning architectures including convolutional neural networks and vision transformers. In addition, the paper outlines the key steps for classifiers training and highlights the differences between the most common algorithms and architectures. Although the choice of an algorithm depends on the task and dataset dealing with, general guidelines for classifier selection are proposed in relation to task analysis, dataset size, explainability requirements, and available computing resources. Considering the enormous interest in these innovative models and architectures, the problem of machine learning algorithms interpretability is finally discussed, providing a future perspective on trustworthy artificial intelligence.Relevance statement The growing synergy between artificial intelligence and medicine fosters predictive models aiding physicians. Machine learning classifiers, from shallow learning to deep learning, are offering crucial insights for the development of clinical decision support systems in healthcare. Explainability is a key feature of models that leads systems toward integration into clinical practice. Key points • Training a shallow classifier requires extracting disease-related features from region of interests (e.g., radiomics).• Deep classifiers implement automatic feature extraction and classification.• The classifier selection is based on data and computational resources availability, task, and explanation needs.

The potential predictive value of cardiac mechanics for left ventricular reverse remodelling in dilated cardiomyopathy

AIMS

Left ventricular reverse remodelling (LVRR) is an important objective of optimal medical management for dilated cardiomyopathy (DCM) patients, as it is associated with favourable long-term outcomes. Cardiac magnetic resonance (CMR) can comprehensively assess cardiac structure and function. We aimed to assess the CMR parameters at baseline and investigate independent variables to predict LVRR in DCM patients.

METHODS AND RESULTS

Nighty-eight initially diagnosed DCM patients who underwent CMR and echocardiography examinations at baseline were included. CMR parameters and feature tracking (FT) based left ventricular (LV) global strain (nStrain) and nStrain indexed to LV cardiac mass index (rStrain) were measured. The predictors of LVRR were determined by multivariate logistic regression analyses. Receiver operating characteristic (ROC) curves were used to evaluate the diagnostic performance of CMR parameters and were compared by the DeLong test. At a median follow-up time of 9 [interquartile range, 7-12] months, 35 DCM patients (36%) achieved LVRR. The patients with LVRR had lower LV volume, mass, LGE extent and stroke volume index (LVSVi) and higher left ventricular remodelling index (LVRI), nStrains, rStrains, and peak systolic strain rate (PSSR) in the longitudinal direction and rStrains in the circumferential direction at baseline (all P < 0.05). In the multivariate logistic regression analyses, LVRI [per SD, odds ratio (OR) 1.79; 95% confidence interval (CI) 1.08-2.98; P = 0.024] and the ratio of global longitudinal peak strain (rGLPS) (per SD, OR 1.88; 95% CI 1.18-3.01; P = 0.008) were independent predictors of LVRR. The combination of LVSVi, LVRI, and rGLPS had a greater area under the curve (AUC) than the combination of LVSVi and LVRI (0.75 vs. 0.68), but not significantly (P = 0.09).

CONCLUSIONS

Patients with LVRR had a lower LV volume index, lower LVSV index, lower LGE extent, higher LVRI, and preserved myocardial deformation in the longitudinal direction at baseline. LVRI and rGLPS at baseline were independent determinants of LVRR.

The Predictive Value of Myocardial Native T1 Mapping Radiomics in Dilated Cardiomyopathy: A Study in a Chinese Population

BACKGROUND

Investigation of the factors influencing dilated cardiomyopathy (DCM) prognosis is important as it could facilitate risk stratification and guide clinical decision-making.

PURPOSE

To assess the prognostic value of magnetic resonance imaging (MRI) radiomics analysis of native T1 mapping in DCM.

STUDY TYPE

Prospective.

SUBJECTS

Three hundred and thirty consecutive patients with non-ischemic DCM (mean age 48.42 ± 14.20 years, 247 males).

FIELD STRENGTH/SEQUENCE

Balanced steady-state free precession and modified Look-Locker inversion recovery T1 mapping sequences at 3 T.

ASSESSMENT

Clinical characteristics, conventional MRI parameters (ventricular volumes, function, and mass), native myocardial T1, and radiomics features extracted from native T1 mapping were obtained. The study endpoint was defined as all-cause mortality or heart transplantation. Models were developed based on 1) clinical data; 2) radiomics data based on T1 mapping; 3) clinical and conventional MRI data; 4) clinical, conventional MRI, and native T1 data; and 5) clinical, conventional MRI, and radiomics T1 mapping data. Each prediction model was trained according to follow-up results with AdaBoost, random forest, and logistic regression classifiers.

STATISTICAL TESTS

The predictive performance was evaluated using the area under the receiver operating characteristic curve (AUC) and F1 score by 5-fold cross-validation.

RESULTS

During a median follow-up of 53.5 months (interquartile range, 41.6-69.5 months), 77 patients with DCM experienced all-cause mortality or heart transplantation. The random forest model based on radiomics combined with clinical and conventional MRI parameters achieved the best performance, with AUC and F1 score of 0.95 and 0.89, respectively.

DATA CONCLUSION

A machine-learning framework based on radiomics analysis of T1 mapping prognosis prediction in DCM.

LEVEL OF EVIDENCE

1 TECHNICAL EFFICACY: Stage 2.

Regression of cardiac hypertrophy in health and disease: mechanisms and therapeutic potential

Left ventricular hypertrophy is a leading risk factor for cardiovascular morbidity and mortality. Although reverse ventricular remodelling was long thought to be irreversible, evidence from the past three decades indicates that this process is possible with many existing heart disease therapies. The regression of pathological hypertrophy is associated with improved cardiac function, quality of life and long-term health outcomes. However, less than 50% of patients respond favourably to most therapies, and the reversibility of remodelling is influenced by many factors, including age, sex, BMI and disease aetiology. Cardiac hypertrophy also occurs in physiological settings, including pregnancy and exercise, although in these cases, hypertrophy is associated with normal or improved ventricular function and is completely reversible postpartum or with cessation of training. Studies over the past decade have identified the molecular features of hypertrophy regression in health and disease settings, which include modulation of protein synthesis, microRNAs, metabolism and protein degradation pathways. In this Review, we summarize the evidence for hypertrophy regression in patients with current first-line pharmacological and surgical interventions. We further discuss the molecular features of reverse remodelling identified in cell and animal models, highlighting remaining knowledge gaps and the essential questions for future investigation towards the goal of designing specific therapies to promote regression of pathological hypertrophy.

Understanding post-surgical decline in left ventricular function in primary mitral regurgitation using regression and machine learning models

Background

Class I echocardiographic guidelines in primary mitral regurgitation (PMR) risks left ventricular ejection fraction (LVEF) < 50% after mitral valve surgery even with pre-surgical LVEF > 60%. There are no models predicting LVEF < 50% after surgery in the complex interplay of increased preload and facilitated ejection in PMR using cardiac magnetic resonance (CMR).

Objective

Use regression and machine learning models to identify a combination of CMR LV remodeling and function parameters that predict LVEF < 50% after mitral valve surgery.

Methods

CMR with tissue tagging was performed in 51 pre-surgery PMR patients (median CMR LVEF 64%), 49 asymptomatic (median CMR LVEF 63%), and age-matched controls (median CMR LVEF 64%). To predict post-surgery LVEF < 50%, least absolute shrinkage and selection operator (LASSO), random forest (RF), extreme gradient boosting (XGBoost), and support vector machine (SVM) were developed and validated in pre-surgery PMR patients. Recursive feature elimination and LASSO reduced the number of features and model complexity. Data was split and tested 100 times and models were evaluated via stratified cross validation to avoid overfitting. The final RF model was tested in asymptomatic PMR patients to predict post-surgical LVEF < 50% if they had gone to mitral valve surgery.

Results

Thirteen pre-surgery PMR had LVEF < 50% after mitral valve surgery. In addition to LVEF (P = 0.005) and LVESD (P = 0.13), LV sphericity index (P = 0.047) and LV mid systolic circumferential strain rate (P = 0.024) were predictors of post-surgery LVEF < 50%. Using these four parameters, logistic regression achieved 77.92% classification accuracy while RF improved the accuracy to 86.17%. This final RF model was applied to asymptomatic PMR and predicted 14 (28.57%) out of 49 would have post-surgery LVEF < 50% if they had mitral valve surgery.

Conclusions

These preliminary findings call for a longitudinal study to determine whether LV sphericity index and circumferential strain rate, or other combination of parameters, accurately predict post-surgical LVEF in PMR.