Multiomics, virtual reality and artificial intelligence in heart failure

Artificial Intelligence and Its Role in Diagnosing Heart Failure: A Narrative Review

Heart failure (HF) is prevalent globally. It is a dynamic disease with varying definitions and classifications due to multiple pathophysiologies and etiologies. The diagnosis, clinical staging, and treatment of HF become complex and subjective, impacting patient prognosis and mortality. Technological advancements, like artificial intelligence (AI), have been significant roleplays in medicine and are increasingly used in cardiovascular medicine to transform drug discovery, clinical care, risk prediction, diagnosis, and treatment. Medical and surgical interventions specific to HF patients rely significantly on early identification of HF. Hospitalization and treatment costs for HF are high, with readmissions increasing the burden. AI can help improve diagnostic accuracy by recognizing patterns and using them in multiple areas of HF management. AI has shown promise in offering early detection and precise diagnoses with the help of ECG analysis, advanced cardiac imaging, leveraging biomarkers, and cardiopulmonary stress testing. However, its challenges include data access, model interpretability, ethical concerns, and generalizability across diverse populations. Despite these ongoing efforts to refine AI models, it suggests a promising future for HF diagnosis. After applying exclusion and inclusion criteria, we searched for data available on PubMed, Google Scholar, and the Cochrane Library and found 150 relevant papers. This review focuses on AI's significant contribution to HF diagnosis in recent years, drastically altering HF treatment and outcomes.

Accelerated DNA methylation age plays a role in the impact of cardiovascular risk factors on the human heart

BACKGROUND

DNA methylation (DNAm) age acceleration (AgeAccel) and cardiac age by 12-lead advanced electrocardiography (A-ECG) are promising biomarkers of biological and cardiac aging, respectively. We aimed to explore the relationships between DNAm age and A-ECG heart age and to understand the extent to which DNAm AgeAccel relates to cardiovascular (CV) risk factors in a British birth cohort from 1946.

RESULTS

We studied four DNAm ages (AgeHannum, AgeHorvath, PhenoAge, and GrimAge) and their corresponding AgeAccel. Outcomes were the results from two publicly available ECG-based cardiac age scores: the Bayesian A-ECG-based heart age score of Lindow et al. 2022 and the deep neural network (DNN) ECG-based heart age score of Ribeiro et al. 2020. DNAm AgeAccel was also studied relative to results from two logistic regression-based A-ECG disease scores, one for left ventricular (LV) systolic dysfunction (LVSD), and one for LV electrical remodeling (LVER). Generalized linear models were used to explore the extent to which any associations between biological cardiometabolic risk factors (body mass index, hypertension, diabetes, high cholesterol, previous cardiovascular disease [CVD], and any CV risk factor) and the ECG-based outcomes are mediated by DNAm AgeAccel. We derived the total effects, average causal mediation effects (ACMEs), average direct effects (ADEs), and the proportion mediated [PM] with their 95% confidence intervals [CIs]. 498 participants (all 60-64 years) were included, with the youngest ECG heart age being 27 and the oldest 90. When exploring the associations between cardiometabolic risk factors and Bayesian A-ECG cardiac age, AgeAccelPheno appears to be a partial mediator, as ACME was 0.23 years [0.01, 0.52] p = 0.028 (i.e., PM≈18%) for diabetes, 0.34 [0.03, 0.74] p = 0.024 (i.e., PM≈15%) for high cholesterol, and 0.34 [0.03, 0.74] p = 0.024 (PM≈15%) for any CV risk factor. Similarly, AgeAccelGrim mediates ≈30% of the relationship between diabetes or high cholesterol and the DNN ECG-based heart age. When exploring the link between cardiometabolic risk factors and the A-ECG-based LVSD and LVER scores, it appears that AgeAccelPheno or AgeAccelGrim mediate 10-40% of these associations.

CONCLUSION

By the age of 60, participants with accelerated DNA methylation appear to have older, weaker, and more electrically impaired hearts. We show that the harmful effects of CV risk factors on cardiac age and health, appear to be partially mediated by DNAm AgeAccelPheno and AgeAccelGrim. This highlights the need to further investigate the potential cardioprotective effects of selective DNA methyltransferases modulators.

GC-MS analysis of exhaled gas for fine detection of inflammatory diseases

Exhaled gas analysis is a non-invasive test ideal for continuous monitoring of biological metabolic information. We analyzed the exhaled gas of patients with inflammatory diseases for trace gas components that could serve as biomarkers that enable early detection of inflammatory diseases and assessment of treatment efficacy. Furthermore, we examined the clinical potential of this method. We enrolled 34 patients with inflammatory disease and 69 healthy participants. Volatile components from exhaled gas were collected and analyzed by a gas chromatography-mass spectrometry system, and the data were examined for gender, age, inflammatory markers, and changes in markers before and after treatment. The data were tested for statistical significance through discriminant analysis by Volcano plot, Analysis of variance test, principal component analysis, and cluster analysis comparing healthy and patient groups. There were no significant differences in the trace components of exhaled gas by gender or age. However, we found differences in some components of the exhaled gas between healthy and untreated patients. In addition, after treatment, gas patterns including the patient-specific components changed to a state closer to the inflammation-free status. We identified trace components in the exhaled gas of patients with inflammatory diseases and found that some of these regressed after treatment.

Physicians and Machine-Learning Algorithm Performance in Predicting Left-Ventricular Systolic Dysfunction from a Standard 12-Lead-Electrocardiogram

Early detection of left ventricular systolic dysfunction (LVSD) may prompt early care and improve outcomes for asymptomatic patients. Standard 12-lead ECG may be used to predict LVSD. We aimed to compare the performance of Machine Learning Algorithms (MLA) and physicians in predicting LVSD from a standard 12-lead ECG. By utilizing a dataset of 13,820 pairs of ECGs and echocardiography, a deep residual convolutional neural network was trained for predicting LVSD (ejection fraction (EF) < 50%) from ECG. The ECGs of the test set (n = 850) were assessed for LVSD by the MLA and six physicians. The performance was compared using sensitivity, specificity, and C-statistics. The interobserver agreement between the physicians for the prediction of LVSD was moderate (κ = 0.50), with average sensitivity and specificity of 70%. The C-statistic of the MLA was 0.85. Repeating this analysis with LVSD defined as EF < 35% resulted in an improvement in physicians’ average sensitivity to 84% but their specificity decreased to 57%. The MLA C-statistic was 0.88 with this threshold. We conclude that although MLA outperformed physicians in predicting LVSD from standard ECG, prior to robust implementation of MLA in ECG machines, physicians should be encouraged to use this approach as a simple and readily available aid for LVSD screening.

Connections for Matters of the Heart: Network Medicine in Cardiovascular Diseases

Cardiovascular diseases (CVD) are diverse disorders affecting the heart and vasculature in millions of people worldwide. Like other fields, CVD research has benefitted from the deluge of multiomics biomedical data. Current CVD research focuses on disease etiologies and mechanisms, identifying disease biomarkers, developing appropriate therapies and drugs, and stratifying patients into correct disease endotypes. Systems biology offers an alternative to traditional reductionist approaches and provides impetus for a comprehensive outlook toward diseases. As a focus area, network medicine specifically aids the translational aspect of in silico research. This review discusses the approach of network medicine and its application to CVD research.

Application of Metabolomics to Identify Potential Biomarkers for the Early Diagnosis of Coronary Heart Disease

Coronary heart disease (CHD) is one of the leading causes of deaths globally. Identification of serum metabolic biomarkers for its early diagnosis is thus much desirable. Serum samples were collected from healthy controls (n = 86) and patients with CHD (n = 166) and subjected to untargeted and targeted metabolomics analyses. Subsequently, potential biomarkers were detected and screened, and a clinical model was developed for diagnosing CHD. Four dysregulated metabolites, namely PC(17:0/0:0), oxyneurine, acetylcarnitine, and isoundecylic acid, were identified. Isoundecylic acid was not found in Human Metabolome Database, so we could not validate differences in its relative abundance levels. Further, the clinical model combining serum oxyneurine, triglyceride, and weight was found to be more robust than that based on PC(17:0/0:0), oxyneurine, and acetylcarnitine (AUC = 0.731 vs. 0.579, sensitivity = 83.0 vs. 75.5%, and specificity = 64.0 vs. 46.5%). Our findings indicated that serum metabolomics is an effective method to identify differential metabolites and that serum oxyneurine, triglyceride, and weight appear to be promising biomarkers for the early diagnosis of CHD.

Effects of canagliflozin on NT-proBNP stratified by left ventricular diastolic function in patients with type 2 diabetes and chronic heart failure: a sub analysis of the CANDLE trial

Background

Identification of the effective subtypes of treatment for heart failure (HF) is an essential topic for optimizing treatment of the disorder. We hypothesized that the beneficial effect of SGLT2 inhibitors (SGLT2i) on the levels of N-terminal pro-brain natriuretic peptide (NT-proBNP) might depend on baseline diastolic function. To elucidate the effects of SGLT2i in type 2 diabetes mellitus (T2DM) and chronic HF we investigated, as a post-hoc sub-study of the CANDLE trial, the effects of canagliflozin on NT-proBNP levels from baseline to 24 weeks, with the data stratified by left ventricular (LV) diastolic function at baseline.

Methods

Patients (n = 233) in the CANDLE trial were assigned randomly to either an add-on canagliflozin (n = 113) or glimepiride treatment groups (n = 120). The primary endpoint was a comparison between the two groups of the changes from baseline to 24 weeks in NT-pro BNP levels, stratified according to baseline ventricular diastolic function.

Results

The change in the geometric mean of NT-proBNP level from baseline to 24 weeks was 0.98 (95% CI 0.89–1.08) in the canagliflozin group and 1.07 (95% CI 0.97–1.18) in the glimepiride group. The ratio of change with canagliflozin/glimepiride was 0.93 (95% CI 0.82–1.05). Responder analyses were used to investigate the response of an improvement in NT-proBNP levels. Although the subgroup analyses for septal annular velocity (SEP-e′) showed no marked heterogeneity in treatment effect, the subgroup with an SEP-e′ < 4.7 cm/s indicated there was an association with lower NT-proBNP levels in the canagliflozin group compared with that in the glimepiride group (ratio of change with canagliflozin/glimepiride (0.83, 95% CI 0.66–1.04).

Conclusions

In the subgroup with a lower LV diastolic function, canagliflozin showed a trend of reduced NT-pro BNP levels compared to that observed with glimepiride. This study suggests that the beneficial effects of canagliflozin treatment may be different in subgroups classified by the severity of LV diastolic dysfunction.

Predicting peri-operative troponin elevation by advanced electrocardiography

BACKGROUND

Peri-operative mortality remains a global problem and an improved pre-operative risk assessment identifying those at highest risk for peri-operative myocardial injury might improve postsurgical outcomes.

AIMS

To determine whether pre-operative measures of advanced electrocardiography (A-ECG) could predict elevated serum troponin T (TnT) in patients undergoing elective, major non-cardiac surgery.

MATERIAL AND METHODS

This observational cohort study included 257 surgical patients who underwent elective major non-cardiac surgery between the years 2012-2013 and 2015-2016 at Karolinska University Hospital. All selected patients were ≥ 18 years of age [median age 70 (63-75) years], had a pre-operative digital 12‑lead ECG < 6 months prior to the procedure and a postoperative high-sensitivity cardiac TnT (hs-cTnT) sample. A-ECG confounders including atrial fibrillation or flutter, abundant premature atrial or ventricular contractions, bundle branch blocks, QRS duration >110 ms, heart rate > 100 beats/min and paced rhythms were excluded. Previously validated A-ECG diagnostic scores that detect cardiovascular pathologies were calculated and compared in patients with and without peri-operative myocardial injury, defined as hs-cTnT >14 ng l^-1.

RESULTS

Pre-operative left ventricular systolic dysfunction by A-ECG was more probable in patients with than without peri-operative myocardial injury (p = 0.03).

CONCLUSIONS

While a pre-operative A-ECG score for LVSD was able to differentiate between patients with versus without elevated peri-operative TnT levels, it did not add any further utility to standard clinical parameters for predicting troponin-related events in the studied population.