Explore the vast range of titles available.

This study evaluates and compares the performance of different machine learning techniques on predicting the individuals at risk of developing hypertension, and who are likely to benefit most from interventions, using the cardiorespiratory fitness data. The dataset of this study contains information of 23,095 patients who underwent clinician- referred exercise treadmill stress testing at Henry Ford Health Systems between 1991 and 2009 and had a complete 10-year follow-up. The variables of the dataset include information on vital signs, diagnosis and clinical laboratory measurements. Six machine learning techniques were investigated: LogitBoost (LB), Bayesian Network classifier (BN), Locally Weighted Naive Bayes (LWB), Artificial Neural Network (ANN), Support Vector Machine (SVM) and Random Tree Forest (RTF). Using different validation methods, the RTF model has shown the best performance (AUC = 0.93) and outperformed all other machine learning techniques examined in this study. The results have also shown that it is critical to carefully explore and evaluate the performance of the machine learning models using various model evaluation methods as the prediction accuracy can significantly differ.

Background Although complex machine learning models are commonly outperforming the traditional simple interpretable models, clinicians find it hard to understand and trust these complex models due to the lack of intuition and explanation of their predictions. The aim of this study to demonstrate the utility of various model-agnostic explanation techniques of machine learning models with a case study for analyzing the outcomes of the machine learning random forest model for predicting the individuals at risk of developing hypertension based on cardiorespiratory fitness data. Methods The dataset used in this study contains information of 23,095 patients who underwent clinician-referred exercise treadmill stress testing at Henry Ford Health Systems between 1991 and 2009 and had a complete 10-year follow-up. Five global interpretability techniques (Feature Importance, Partial Dependence Plot, Individual Conditional Expectation, Feature Interaction, Global Surrogate Models) and two local interpretability techniques (Local Surrogate Models, Shapley Value) have been applied to present the role of the interpretability techniques on assisting the clinical staff to get better understanding and more trust of the outcomes of the machine learning-based predictions. Results Several experiments have been conducted and reported. The results show that different interpretability techniques can shed light on different insights on the model behavior where global interpretations can enable clinicians to understand the entire conditional distribution modeled by the trained response function. In contrast, local interpretations promote the understanding of small parts of the conditional distribution for specific instances. Conclusions Various interpretability techniques can vary in their explanations for the behavior of the machine learning model. The global interpretability techniques have the advantage that it can generalize over the entire population while local interpretability techniques focus on giving explanations at the level of instances. Both methods can be equally valid depending on the application need. Both methods are effective methods for assisting clinicians on the medical decision process, however, the clinicians will always remain to hold the final say on accepting or rejecting the outcome of the machine learning models and their explanations based on their domain expertise.

There is increasing interest to make primary data from published research publicly available. We aimed to assess the current status of making research data available in highly-cited journals across the scientific literature. We reviewed the first 10 original research papers of 2009 published in the 50 original research journals with the highest impact factor. For each journal we documented the policies related to public availability and sharing of data. Of the 50 journals, 44 (88%) had a statement in their instructions to authors related to public availability and sharing of data. However, there was wide variation in journal requirements, ranging from requiring the sharing of all primary data related to the research to just including a statement in the published manuscript that data can be available on request. Of the 500 assessed papers, 149 (30%) were not subject to any data availability policy. Of the remaining 351 papers that were covered by some data availability policy, 208 papers (59%) did not fully adhere to the data availability instructions of the journals they were published in, most commonly (73%) by not publicly depositing microarray data. The other 143 papers that adhered to the data availability instructions did so by publicly depositing only the specific data type as required, making a statement of willingness to share, or actually sharing all the primary data. Overall, only 47 papers (9%) deposited full primary raw data online. None of the 149 papers not subject to data availability policies made their full primary data publicly available. A substantial proportion of original research papers published in high-impact journals are either not subject to any data availability policies, or do not adhere to the data availability instructions in their respective journals. This empiric evaluation highlights opportunities for improvement.

Atrial fibrillation (AF) is a significant health care problem for patients with obstructive sleep apnea (OSA). Continuous positive airway pressure (CPAP) as a therapy for OSA is underused, and it is unknown if CPAP might reduce rates of AF. We systematically reviewed the published reports on CPAP use and risk of AF. MEDLINE, EMBASE, CINAHL, Web of Science, meeting abstracts, and Cochrane databases were searched from inception to June 2015. Studies needed to report the rates of AF in participants who were and were not on CPAP. Data were extracted by 2 authors. A total of 8 studies on OSA were identified (1 randomized controlled trial) with 698 CPAP users and 549 non-CPAP users. In a random effects model, patients treated with CPAP had a 42% decreased risk of AF (pooled risk ratio, 0.58; 95% confidence interval, 0.47 to 0.70; p <0.001). There was low heterogeneity in the results (I2 = 30%). In metaregression analysis, benefits of CPAP were stronger for younger, obese, and male patients (p <0.05). An inverse relationship between CPAP therapy and AF recurrence was observed. Results suggest that more patients with AF also should be tested for OSA.

Exercise capacity (EC) is inversely related to the risk of cardiovascular disease and incident heart failure (HF) in healthy subjects. However, there are no present studies that exclusively evaluate EC and the risk of incident HF in patients with known coronary heart disease (CHD). We aimed to determine the relation between EC and incident HF in patients with an established clinical diagnosis of CHD. We retrospectively identified 8,387 patients (age 61 ± 12 years; 30% women; 33% non-White) with a history of myocardial infarction (MI) or coronary revascularization procedure and no history of HF at the time of a clinically indicated exercise stress test completed between 1991 and 2009. EC was quantified in metabolic equivalents of task (METs) estimated from treadmill testing. Incident HF was identified through June 2010 from administrative databases based on ≥3 encounters with International Classification of Diseases, Ninth Revision 428.x. Cox regression analysis was used to evaluate the risk of incident HF associated with METs. Covariates included age; gender; race; hypertension, diabetes, hyperlipidemia, smoking, and MI; medications for CHD and lung diseases; and clinical indication for treadmill testing. During a median follow-up of 8.2 years (interquartile range 4.7 to 12.4 years) after the exercise test, 23% of the cohort experienced a new HF diagnosis. Lower EC categories were associated with higher HF incidence compared with METs ≥12, with nearly fourfold greater adjusted risk among patients with METs <6. Per unit increase in METs of EC was associated with a 12% lower adjusted risk for HF. There was no significant interaction based on race (p = 0.06), gender (p = 0.88), age ≤61 years (p = 0.60), history of MI (p = 0.31), or diabetes (p = 0.38). This study reveals that among men and women with CHD and no history of HF, EC is independently and inversely related to the risk of future HF.

Heart rate (HR) at rest is associated with adverse cardiovascular events; however, the biologic mechanism for the relation is unclear. We hypothesized a strong association between HR at rest and subclinical inflammation, given their common interrelation with the autonomic nervous system. HR at rest was recorded at baseline in the Multi-Ethnic Study of Atherosclerosis, a cohort of 4 racial or ethnic groups without cardiovascular disease at baseline and then divided into quintiles. Subclinical inflammation was measured using high-sensitivity C-reactive protein, interleukin-6, and fibrinogen. We used progressively adjusted regression models with terms for physical activity and atrioventricular nodal blocking agents in the fully adjusted models. We examined inflammatory markers as both continuous and categorical variables using the clinical cut point of ≥3 mg/L for high-sensitivity C-reactive protein and the upper quartiles of fibrinogen (≥389 mg/dl) and interleukin-6 (≥1.89 pg/ml). Participants had a mean age of 62 years (SD 9.7), mean resting heart rate of 63 beats/min (SD 9.6) and were 47% men. Increased HR at rest was significantly associated with higher levels of all 3 inflammatory markers in both continuous (p for trend <0.001) and categorical (p for trend <0.001) models. Results were similar among all 3 inflammatory markers, and there was no significant difference in the association among the 4 racial or ethnic groups. In conclusion, an increased HR at rest was associated with a higher level of inflammation among an ethnically diverse group of subjects without known cardiovascular disease.

Purpose of Review In this review, we explore the development of digital PET scanners and describe the mechanism by which they work. We dive into some technical details on what differentiates a digital PET from a conventional PET scanner and how such differences lead to better imaging characteristics. Additionally, we summarize the available evidence on the improvements in the images acquired by digital PET as well as the remaining pitfalls. Finally, we report the comparative studies available on how digital PET compares to conventional PET, particularly in the quantification of coronary blood flow. Recent Findings The advent of digital PET offers high sensitivity and time-of-flight (TOF), which allow lower activity and scan times, with much less risk of detector saturation. This allows faster patient throughput, scanning more patients per generator, and acquiring more consistent image quality across patients. The higher sensitivity captures more of the potential artifacts, particularly motion-related ones, which presents a current challenge that still needs to be tackled. Summary The digital silicon photomultiplier (SiPM) positron emission tomography (PET) machine has been an important development in the technological advancements of non-invasive nuclear cardiovascular imaging. It has enhanced the utility for PET myocardial perfusion imaging (MPI) and myocardial blood flow (MBF) quantification.

The internet is gaining popularity as a means of delivering employee-based cardiovascular (CV) wellness interventions though little is known about the cardiovascular health outcomes of these programs. In this review, we examined the effectiveness of internet-based employee cardiovascular wellness and prevention programs. We conducted a systematic review by searching PubMed, Web of Science and Cochrane library for all published studies on internet-based programs aimed at improving CV health among employees up to November 2012. We grouped the outcomes according to the American Heart Association (AHA) indicators of cardiovascular wellbeing--weight, BP, lipids, smoking, physical activity, diet, and blood glucose. A total of 18 randomized trials and 11 follow-up studies met our inclusion/exclusion criteria. Follow-up duration ranged from 6-24 months. There were significant differences in intervention types and number of components in each intervention. Modest improvements were observed in more than half of the studies with weight related outcomes while no improvement was seen in virtually all the studies with physical activity outcome. In general, internet-based programs were more successful if the interventions also included some physical contact and environmental modification, and if they were targeted at specific disease entities such as hypertension. Only a few of the studies were conducted in persons at-risk for CVD, none in blue-collar workers or low-income earners. Internet based programs hold promise for improving the cardiovascular wellness among employees however much work is required to fully understand its utility and long term impact especially in special/at-risk populations.

Background Prior studies have demonstrated that cardiorespiratory fitness (CRF) is a strong marker of cardiovascular health. Machine learning (ML) can enhance the prediction of outcomes through classification techniques that classify the data into predetermined categories. The aim of this study is to present an evaluation and comparison of how machine learning techniques can be applied on medical records of cardiorespiratory fitness and how the various techniques differ in terms of capabilities of predicting medical outcomes (e.g. mortality). Methods We use data of 34,212 patients free of known coronary artery disease or heart failure who underwent clinician-referred exercise treadmill stress testing at Henry Ford Health Systems Between 1991 and 2009 and had a complete 10-year follow-up. Seven machine learning classification techniques were evaluated: Decision Tree (DT), Support Vector Machine (SVM), Artificial Neural Networks (ANN), Naïve Bayesian Classifier (BC), Bayesian Network (BN), K-Nearest Neighbor (KNN) and Random Forest (RF). In order to handle the imbalanced dataset used, the Synthetic Minority Over-Sampling Technique (SMOTE) is used. Results Two set of experiments have been conducted with and without the SMOTE sampling technique. On average over different evaluation metrics, SVM Classifier has shown the lowest performance while other models like BN, BC and DT performed better. The RF classifier has shown the best performance (AUC = 0.97) among all models trained using the SMOTE sampling. Conclusions The results show that various ML techniques can significantly vary in terms of its performance for the different evaluation metrics. It is also not necessarily that the more complex the ML model, the more prediction accuracy can be achieved. The prediction performance of all models trained with SMOTE is much better than the performance of models trained without SMOTE. The study shows the potential of machine learning methods for predicting all-cause mortality using cardiorespiratory fitness data.

To determine whether sex modifies the relationship between fitness and mortality. We included 57,284 patients without coronary artery disease or heart failure who completed a routine treadmill exercise test between 1991 and 2009. We determined metabolic equivalent tasks (METs) and linked patient records with mortality data via the Social Security Death Index. Multivariable Cox regression was used to determine the association between sex, fitness, and all-cause mortality. There were 29,470 men (51.4%) and 27,814 women (48.6%) with mean ages of 53 and 54 years, respectively. Overall, men achieved 1.7 METs higher than women (P<.001). During median follow-up of 10 years, there were 6402 deaths. The mortality rate for men in each MET group was similar to that for women, who achieved an average of 2.6 METs lower (P=.004). Fitness was inversely associated with mortality in both men (hazard ratio [HR], 0.84 per 1 MET; 95% CI, 0.83-0.85) and women (HR, 0.83 per 1 MET; 95% CI, 0.81-0.84). This relationship did not plateau at high or low MET values. Although men demonstrated 1.7 METs higher than women, their survival was equivalent to that of women demonstrating 2.6 METs lower. Furthermore, higher MET values were associated with lower mortality for both men and women across the range of MET values. These findings are useful for tailoring prognostic information and lifestyle guidance to men and women undergoing stress testing.