The Use of Machine Learning for the Care of Hypertension and Heart Failure

The Usefulness of Strain Echocardiography as Diagnostic and Prognostic Index of Cardiac Dysfunction in Septic Patients in Correlation with Cardiac Biomarkers

Background

Compared to standard echocardiography, speckle tracking echocardiography (STE) looks more accurate for the detection of subclinical dysfunction of the myocardium. The aim of our work was to assess the value of STE in the diagnosis of subclinical ventricular dysfunction and as a prognostic index in sepsis patients.

Patients and Methods

An observational prospective study involving critically ill patients aged ≥ 18 years diagnosed with sepsis or septic shock. All patients were subjected to full history-taking, clinical assessment, and scoring system, including Acute Physiology and Chronic Health (APACHE) II score and quick sequential organ failure assessment score. Investigations were done for all patients, including laboratory (complete blood count, C-reactive protein, N-terminal pro-brain natriuretic peptide [NT-proBNP], and troponin-I and serum lactate level), ECG, and echocardiographic examination (conventional and speckle tracking) for measurement of global left ventricular strain.

Results

This study involved 50 patients, nine patients with sepsis and 41 patients with septic shock. Regarding cardiac biomarkers, the mean value of troponin-I was 0.18 ± 0.05 ng/L and for NT-proBNP was 1228.2 ± 832.9 pmol/L. All patients in the study had elevated lactate levels. There was a significant correlation between global longitudinal strain (GLS) and troponin I, NT-proBNP, and lactate levels after 3 days of admission. GLS, lactate, NT-proBNP, troponin levels, and APACHE II Score were significant predictors of mortality with a sensitivity of 76.5%, 88.2%, 88.2%, 76.5%, and 88.2%, respectively.

Conclusion

GLS measured by speckle tracking echocardiography looks to be a sensitive diagnostic tool for early detection of subclinical left ventricular dysfunction in patients with sepsis in addition to be a sensitive predictor of in-hospital mortality.

Precision Hypertension

Hypertension affects >1 billion people worldwide. Complications of hypertension include stroke, renal failure, cardiac hypertrophy, myocardial infarction, and cardiac failure. Despite the development of various antihypertensive drugs, the number of people with uncontrolled hypertension continues to rise. While the lack of compliance associated with frequent side effects to medication is a contributory issue, there has been a failure to consider the diverse nature of hypertensive populations. Instead, we propose that hypertension can only be truly managed by precision. A precision medicine approach would consider each patient's unique factors. In this review, we discuss the progress toward precision medicine for hypertension with more predictiveness and individualization of treatment. We will highlight the advances in data science, omics (genomics, metabolomics, proteomics, etc), artificial intelligence, gene therapy, and gene editing and their application to precision hypertension.

Shapely additive values can effectively visualize pertinent covariates in machine learning when predicting hypertension

Machine learning methods are widely used within the medical field to enhance prediction. However, little is known about the reliability and efficacy of these models to predict long-term medical outcomes such as blood pressure using lifestyle factors, such as diet. The authors assessed whether machine-learning techniques could accurately predict hypertension risk using nutritional information. A cross-sectional study using data from the National Health and Nutrition Examination Survey (NHANES) between January 2017 and March 2020. XGBoost was used as the machine-learning model of choice in this study due to its increased performance relative to other common methods within medical studies. Model prediction metrics (e.g., AUROC, Balanced Accuracy) were used to measure overall model efficacy, covariate Gain statistics (percentage each covariate contributes to the overall prediction) and SHapely Additive exPlanations (SHAP, method to visualize each covariate) were used to provide explanations to machine-learning output and increase the transparency of this otherwise cryptic method. Of a total of 9650 eligible patients, the mean age was 41.02 (SD = 22.16), 4792 (50%) males, 4858 (50%) female, 3407 (35%) White patients, 2567 (27%) Black patients, 2108 (22%) Hispanic patients, and 981 (10%) Asian patients. From evaluation of model gain statistics, age was found to be the single strongest predictor of hypertension, with a gain of 53.1%. Additionally, demographic factors such as poverty and Black race were also strong predictors of hypertension, with gain of 4.33% and 4.18%, respectively. Nutritional Covariates contributed 37% to the overall prediction: Sodium, Caffeine, Potassium, and Alcohol intake being significantly represented within the model. Machine Learning can be used to predict hypertension.

The role of artificial intelligence in hypertensive disorders of pregnancy: towards personalized healthcare

INTRODUCTION

Guidelines advise ongoing follow-up of patients after hypertensive disorders of pregnancy (HDP) to assess cardiovascular risk and manage future patient-specific pregnancy conditions. However, there are limited tools available to monitor patients, with those available tending to be simple risk assessments that lack personalization. A promising approach could be the emerging artificial intelligence (AI)-based techniques, developed from big patient datasets to provide personalized recommendations for preventive advice.

AREAS COVERED

In this narrative review, we discuss the impact of integrating AI and big data analysis for personalized cardiovascular care, focusing on the management of HDP.

EXPERT OPINION

The pathophysiological response of women to pregnancy varies, and deeper insight into each response can be gained through a deeper analysis of the medical history of pregnant women based on clinical records and imaging data. Further research is required to be able to implement AI for clinical cases using multi-modality and multi-organ assessment, and this could expand both knowledge on pregnancy-related disorders and personalized treatment planning.

Recent technologies in cardiac imaging

Cardiac imaging allows physicians to view the structure and function of the heart to detect various heart abnormalities, ranging from inefficiencies in contraction, regulation of volumetric input and output of blood, deficits in valve function and structure, accumulation of plaque in arteries, and more. Commonly used cardiovascular imaging techniques include x-ray, computed tomography (CT), magnetic resonance imaging (MRI), echocardiogram, and positron emission tomography (PET)/single-photon emission computed tomography (SPECT). More recently, even more tools are at our disposal for investigating the heart's physiology, performance, structure, and function due to technological advancements. This review study summarizes cardiac imaging techniques with a particular interest in MRI and CT, noting each tool's origin, benefits, downfalls, clinical application, and advancement of cardiac imaging in the near future.

Machine Learning Risk Prediction for Incident Heart Failure in Patients With Atrial Fibrillation

BACKGROUND

Atrial fibrillation (AF) increases the risk of heart failure (HF); however, little focus is placed on the risk stratification for, and prevention of, incident HF in patients with AF.

OBJECTIVES

This study aimed to construct and validate a machine learning (ML) prediction model for HF hospitalization in patients with AF.

METHODS

The Fushimi AF Registry is a community-based prospective survey of patients with AF in Fushimi-ku, Kyoto, Japan. We divided the data set of the registry into derivation (n = 2,383) and validation (n = 2,011) cohorts. An ML model was built to predict the incidence of HF hospitalization using the derivation cohort, and predictive ability was examined using the validation cohort.

RESULTS

HF hospitalization occurred in 606 patients (14%) during a median follow-up period of 4.4 years in the entire registry. Data of transthoracic echocardiography and biomarkers were frequently nominated as important predictive variables across all 6 ML models. The ML model based on a random forest algorithm using 7 variables (age, history of HF, creatinine clearance, cardiothoracic ratio on x-ray, left ventricular [LV] ejection fraction, LV end-systolic diameter, and LV asynergy) had high prediction performance (area under the receiver operating characteristics curve [AUC]: 0.75) and was significantly superior to the Framingham HF risk model (AUC: 0.67; P < 0.001). Based on Kaplan-Meier curves, the ML model could stratify the risk of HF hospitalization during the follow-up period (log-rank; P < 0.001).

CONCLUSIONS

The ML model revealed important predictors and helped us to stratify the risk of HF, providing opportunities for the prevention of HF in patients with AF.

Speckle-tracking echocardiography for detecting myocardial dysfunction in sepsis and septic shock patients: A single emergency department study

BACKGROUND

Septic cardiomyopathy (SCM) occurs in the early stage of sepsis and septic shock, which has implications for treatment strategies and prognosis. Additionally, myocardial involvement in the early stages of sepsis is difficult to identify. Here, we assess subclinical myocardial function using laboratory tests and speckle-tracking echocardiography (STE).

METHODS

Emergency department patients diagnosed with sepsis or septic shock were included for analysis. Those with other causes of acute or pre-existing cardiac dysfunction were excluded. Transthoracic echocardiography (TTE), including conventional echocardiography and STE, were performed for all patients three hours after initial resuscitation. Samples for laboratory tests were taken around the time of TTE.

RESULTS

Left ventricular functions of 60 patients were analyzed, including 21 septic shock patients and 39 sepsis patients. There was no significant difference in global longitudinal strain (GLS), global circumferential strain (GCS), or global radical strain (GRS) between patients with sepsis and septic shock (all with P>0.05). However, GLS and GCS were significantly less negative in patients with abnormal troponin levels or in patients with abnormal left ventricular ejection fraction (LVEF) values (all with P<0.05). There were also moderate correlations between GLS and levels of cTnI (r=0.40, P=0.002) or N-terminal pro-B-type natriuretic peptide (NT-proBNP) (r=0.44, P=0.001) in sepsis and septic shock patients.

CONCLUSION

Myocardial dysfunction, e.g., lower LVEF or less negative GLS in patients with sepsis or septic shock, is more affected by myocardial injury. GLS could be incorporated into mainstream clinical practice as a supplementary LVEF parameter, especially for those with elevated troponin levels.