Predicting ventricular tachyarrhythmia in patients with systolic heart failure based on texture features of the gray zone from contrast-enhanced magnetic resonance imaging

Entropy of left ventricular late gadolinium enhancement and its prognostic value in hypertrophic cardiomyopathy a new CMR assessment method

PURPOSE

As a novel metric, entropy generated from late gadolinium enhancement (LGE) cardiac magnetic resonance (CMR) can be utilized to assess tissue heterogeneity. However, it is unknown if it can be utilized for risk stratification in hypertrophic cardiomyopathy (HCM). In addition, it is unknown if LGE entropy correlates with LGE mass%, which is commonly utilized for fibrosis assessment. This research was done to investigate these issues.

MATERIALS AND METHODS

Patients with HCM who underwent 3.0-T CMR between January 2015 and January 2020 were prospectively enrolled and classified into low- and high-risk groups according to the AHA/ACC risk stratification guideline for 2020. The LGE entropy was automatically estimated using a generic Python package algorithm. On CMR imaging, the LGE mass% was determined using the CVI 42 software. Endpoint events included sudden cardiac death (SCD), hospital readmission owing to heart failure, and implantable cardioverter defibrillator (ICD) treatment for ventricular arrhythmias.

RESULTS

A total of 109 HCM participants (70 males) were included. During the follow-up (23 ± 7 months), the patients in the high-risk group had higher LGE entropy (p < 0.001) and LGE mass% (p < 0.001) than those in the low-risk group, and patients with endpoint events had higher LGE entropy (p < 0.001) and LGE mass% (p < 0.001) than those without endpoint events. In all participants, there was a link between LGE entropy and LGE mass%, according to the Spearman rank correlation analysis (p < 0.001; r = 0.667). In ROC analysis, the area under the curve (AUC) of LGE entropy was 0.893 (95% CI, 0.794-0.993; P<0.001), AUC of LGE mass% was 0.826 (95% CI, 0.737-0.914; P<0.001), AUC of LVEF was 0.610 (95% CI, 0.473-0.748; P = 0.117) and AUC of 2020 AHA/ACC guideline for risk stratification was 0.716 (95% CI, 0.617-0.815; P = 0.002). According to Kaplan-Meier curves, HCM with a higher LGE entropy (≥cutoff value (<5.873) or ≥ thied tertile (5.540)) were more likely to experience the endpoint events. Following adjustment for the 2020 AHA/ACC guideline for risk categorization, LGE mass%, or decreased LVEF, Cox analysis showed that LGE entropy was independently linked with endpoint events.

CONCLUSIONS

The variability and extent of LGE pictures can be reflected by LGE entropy, which is a reliable, usable, and repeatable metric for risk classification in HCM. It is a prognostic indicator of endpoint events that is independent of other risk indicators.

Strategies for Sudden Cardiac Death Prevention

Sudden cardiac death (SCD) represents a major challenge in modern medicine. The prevention of SCD orbits on two levels, the general population level and individual level. Much research has been done with the aim to improve risk stratification of SCD, although no radical changes in evidence and in therapeutic strategy have been achieved. Artificial intelligence (AI), and in particular machine learning (ML) models, represent novel technologic tools that promise to improve predictive ability of fatal arrhythmic events. In this review, firstly, we analyzed the electrophysiological basis and the major clues of SCD prevention at population and individual level; secondly, we reviewed the main research where ML models were used for risk stratification in other field of cardiology, suggesting its potentiality in the field of SCD prevention.

Quality of science and reporting for radiomics in cardiac magnetic resonance imaging studies: a systematic review

OBJECTIVES

To evaluate the quality of radiomics studies using cardiac magnetic resonance imaging (CMR) according to the radiomics quality score (RQS), Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) guidelines, and the standards defined by the Image Biomarker Standardization Initiative (IBSI) and identify areas needing improvement.

MATERIALS AND METHODS

PubMed and Embase were searched to identify radiomics studies using CMR until March 10, 2021. Of the 259 identified articles, 32 relevant original research articles were included. Studies were scored according to the RQS, TRIPOD guidelines, and IBSI standards by two cardiac radiologists.

RESULTS

The mean RQS was 14.3% of the maximum (5.16 out of 36). RQS were low for the demonstration of validation (-60.6%), calibration statistics (1.6%), potential clinical utility (3.1%), and open science (3.1%) items. No study conducted a phantom study or cost-effectiveness analysis. The adherence to TRIPOD guidelines was 55.9%. Studies were deficient in reporting title (3.1%), stating objective in abstract and introduction (6.3% and 9.4%), missing data (0%), discrimination/calibration (3.1%), and how to use the prediction model (3.1%). According to the IBSI standards, non-uniformity correction, image interpolation, grey-level discretization, and signal intensity normalization were performed in two (6.3%), four (12.5%), six (18.8%), and twelve (37.5%) studies, respectively.

CONCLUSION

The quality of radiomics studies using CMR is suboptimal. Improvements are needed in the areas of validation, calibration, clinical utility, and open science. Complete reporting of study objectives, missing data, discrimination/calibration, how to use the prediction model, and preprocessing steps are necessary.

KEY POINTS

• The quality of science in radiomics studies using CMR is currently inadequate. • RQS were low for validation, calibration, clinical utility, and open science; no study conducted a phantom study or cost-effectiveness analysis. • In stating the study objective, missing data, discrimination/calibration, how to use the prediction model, and preprocessing steps, improvements are needed.

Predictive impacts of chronic kidney disease and cardiac sympathetic nervous activity on lethal arrhythmic events in chronic heart failure

Background

The clinical implications of chronic kidney disease (CKD) and cardiac sympathetic nervous activity (CSNA) regarding lethal arrhythmic events have not yet been fully elucidated in patients with chronic heart failure (CHF). We hypothesized that the combination of CKD and abnormal CSNA, assessed by ¹²³I‐metaiodobenzylguanidine (¹²³I‐MIBG) scintigraphy, may provide useful prognostic information for lethal arrhythmic events.

Methods

We studied 165 consecutive hospitalized CHF patients without dialysis. Cardiac ¹²³I‐MIBG scintigraphy was performed in a clinically stable condition, and abnormal CSNA was defined as a late heart‐to‐mediastinum ratio of <1.6. CKD was defined as an estimated glomerular filtration rate of <60 ml/min/1.73 m². We then investigated the incidence of lethal arrhythmic events (sustained ventricular tachyarrhythmia, appropriate implantable cardioverter‐defibrillator therapy, or sudden cardiac death).

Results

During a median follow‐up of 5.3 years, lethal arrhythmic events were observed in 40 patients (24.2%). The patients were divided into four groups according to the presence of CKD and CSNA abnormality: non‐CKD/normal CSNA (n = 52), CKD/normal CSNA (n = 39), non‐CKD/abnormal CSNA (n = 33), and CKD/abnormal CSNA (n = 41). Kaplan–Meier analysis showed that CKD/abnormal CSNA had the highest event rate (log‐rank p = .004). Additionally, the Cox proportional hazard analysis revealed that CKD/abnormal CSNA was a predictor for lethal arrhythmic events compared with non‐CKD/normal CSNA (hazard ratio, 5.368, p = .001). However, the other two groups did not show significant differences compared with the non‐CKD/normal CSNA group.

Conclusions

The combination of CKD and abnormal CSNA, assessed by ¹²³I‐MIBG scintigraphy, had a high predictive value for lethal arrhythmic events in patients with CHF.

Radiomics side experiments and DAFIT approach in identifying pulmonary hypertension using Cardiac MRI derived radiomics based machine learning models

Side experiments are performed on radiomics models to improve their reproducibility. We measure the impact of myocardial masks, radiomic side experiments and data augmentation for information transfer (DAFIT) approach to differentiate patients with and without pulmonary hypertension (PH) using cardiac MRI (CMRI) derived radiomics. Feature extraction was performed from the left ventricle (LV) and right ventricle (RV) myocardial masks using CMRI in 82 patients (42 PH and 40 controls). Various side study experiments were evaluated: Original data without and with intraclass correlation (ICC) feature-filtering and DAFIT approach (without and with ICC feature-filtering). Multiple machine learning and feature selection strategies were evaluated. Primary analysis included all PH patients with subgroup analysis including PH patients with preserved LVEF (≥ 50%). For both primary and subgroup analysis, DAFIT approach without feature-filtering was the highest performer (AUC 0.957-0.958). ICC approaches showed poor performance compared to DAFIT approach. The performance of combined LV and RV masks was superior to individual masks alone. There was variation in top performing models across all approaches (AUC 0.862-0.958). DAFIT approach with features from combined LV and RV masks provide superior performance with poor performance of feature filtering approaches. Model performance varies based upon the feature selection and model combination.