Prediction equation to estimate heart rate at individual ventilatory threshold in female and male obese adults

A Machine Learning Approach to Developing an Accurate Prediction of Maximal Heart Rate During Exercise Testing in Apparently Healthy Adults

PURPOSE

Maximal heart rate (HR max ) continues to be an important measure of adequate effort during an exercise test. The aim of this study was to improve the accuracy of HR max prediction using a machine learning (ML) approach.

METHODS

We used a sample from the Fitness Registry of the Importance of Exercise National Database, which included 17 325 apparently healthy individuals (81% males) who performed a maximal cardiopulmonary exercise test. Two standard formulas for HR max prediction were tested: Formula1 = 220 - age (yr), root-mean-squared error (RMSE) 21.9, relative root-mean-squared error (RRMSE) 1.1; and Formula2 = 209.3 - 0.72 × age (yr), RMSE 22.7 and RRMSE 1.1. For ML model prediction, we used age, weight, height, resting HR, and systolic and diastolic blood pressure. The following ML algorithms to predict HR max were applied: lasso regression (LR), neural networks (NN), support vector machine (SVM) and random forests (RF). An evaluation was performed using cross-validation and by computing the RMSE and RRMSE, Pearson correlation, and Bland-Altman plots. The best predictive model was explained with Shapley Additive Explanations (SHAP).

RESULTS

The HR max for the cohort was 162 ± 20 bpm. All ML models improved HR max prediction and reduced RMSE and RRMSE compared with Formula1 (LR: 20.2%, NN: 20.4%, SVM: 22.2%, and RF: 24.7%). The predictions of all algorithms significantly correlated with HR max ( r = 0.49, 0.51, 0.54, 0.57, respectively; P < .001). Bland-Altman analysis demonstrated lower bias and 95% CI for all ML models in comparison with standard equations. The SHAP explanation showed a high impact of all selected variables.

CONCLUSIONS

Machine learning, particularly the RF model, improved prediction of HR max using readily available measures. This approach should be considered for clinical application to refine HR max prediction.

Prediction of Relevant Training Control Parameters at Individual Anaerobic Threshold without Blood Lactate Measurement

BACKGROUND

Active exercise therapy plays an essential role in tackling the global burden of obesity. Optimizing recommendations in individual training therapy requires that the essential parameters heart rate HR(IAT) and work load (W/kg(IAT) at individual anaerobic threshold (IAT) are known. Performance diagnostics with blood lactate is one of the most established methods for these kinds of diagnostics, yet it is also time consuming and expensive.

METHODS

To establish a regression model which allows HR(IAT) and (W/kg(IAT) to be predicted without measuring blood lactate, a total of 1234 performance protocols with blood lactate in cycle ergometry were analyzed. Multiple linear regression analyses were performed to predict the essential parameters (HR(IAT)) (W/kg(IAT)) by using routine parameters for ergometry without blood lactate.

RESULTS

HR(IAT) can be predicted with an RMSE of 8.77 bpm (p < 0.001), R² = 0.799 (|R| = 0.798) without performing blood lactate diagnostics during cycle ergometry. In addition, it is possible to predict W/kg(IAT) with an RMSE (root mean square error) of 0.241 W/kg (p < 0.001), R² = 0.897 (|R| = 0.897).

CONCLUSIONS

It is possible to predict essential parameters for training management without measuring blood lactate. This model can easily be used in preventive medicine and results in an inexpensive yet better training management of the general population, which is essential for public health.

Automatic Detection of Aerobic Threshold through Recurrence Quantification Analysis of Heart Rate Time Series

During exercise with increasing intensity, the human body transforms energy with mechanisms dependent upon actual requirements. Three phases of the body's energy utilization are recognized, characterized by different metabolic processes, and separated by two threshold points, called aerobic (AerT) and anaerobic threshold (AnT). These thresholds occur at determined values of exercise intensity(workload) and can change among individuals. They are considered indicators of exercise capacities and are useful in the personalization of physical activity plans. They are usually detected by ventilatory or metabolic variables and require expensive equipment and invasive measurements. Recently, particular attention has focused on AerT, which is a parameter especially useful in the overweight and obese population to determine the best amount of exercise intensity for weight loss and increasing physical fitness. The aim of study is to propose a new procedure to automatically identify AerT using the analysis of recurrences (RQA) relying only on Heart rate time series, acquired from a cohort of young athletes during a sub-maximal incremental exercise test (Cardiopulmonary Exercise Test, CPET) on a cycle ergometer. We found that the minima of determinism, an RQA feature calculated from the Recurrence Quantification by Epochs (RQE) approach, identify the time points where generic metabolic transitions occur. Among these transitions, a criterion based on the maximum convexity of the determinism minima allows to detect the first metabolic threshold. The ordinary least products regression analysis shows that values of the oxygen consumption VO₂, heart rate (HR), and Workload correspondent to the AerT estimated by RQA are strongly correlated with the one estimated by CPET (r > 0.64). Mean percentage differences are <2% for both HR and VO₂ and <11% for Workload. The Technical Error for HR at AerT is <8%; intraclass correlation coefficients values are moderate (≥0.66) for all variables at AerT. This system thus represents a useful method to detect AerT relying only on heart rate time series, and once validated for different activities, in future, can be easily implemented in applications acquiring data from portable heart rate monitors.

Heart Rate Modeling and Prediction Using Autoregressive Models and Deep Learning

Physiological time series are affected by many factors, making them highly nonlinear and nonstationary. As a consequence, heart rate time series are often considered difficult to predict and handle. However, heart rate behavior can indicate underlying cardiovascular and respiratory diseases as well as mood disorders. Given the importance of accurate modeling and reliable predictions of heart rate fluctuations for the prevention and control of certain diseases, it is paramount to identify models with the best performance in such tasks. The objectives of this study were to compare the results of three different forecasting models (Autoregressive Model, Long Short-Term Memory Network, and Convolutional Long Short-Term Memory Network) trained and tested on heart rate beats per minute data obtained from twelve heterogeneous participants and to identify the architecture with the best performance in terms of modeling and forecasting heart rate behavior. Heart rate beats per minute data were collected using a wearable device over a period of 10 days from twelve different participants who were heterogeneous in age, sex, medical history, and lifestyle behaviors. The goodness of the results produced by the models was measured using both the mean absolute error and the root mean square error as error metrics. Despite the three models showing similar performance, the Autoregressive Model gave the best results in all settings examined. For example, considering one of the participants, the Autoregressive Model gave a mean absolute error of 2.069 (compared to 2.173 of the Long Short-Term Memory Network and 2.138 of the Convolutional Long Short-Term Memory Network), achieving an improvement of 5.027% and 3.335%, respectively. Similar results can be observed for the other participants. The findings of the study suggest that regardless of an individual's age, sex, and lifestyle behaviors, their heart rate largely depends on the pattern observed in the previous few minutes, suggesting that heart rate can be reasonably regarded as an autoregressive process. The findings also suggest that minute-by-minute heart rate prediction can be accurately performed using a linear model, at least in individuals without pathologies that cause heartbeat irregularities. The findings also suggest many possible applications for the Autoregressive Model, in principle in any context where minute-by-minute heart rate prediction is required (arrhythmia detection and analysis of the response to training, among others).

Exercise as a drug for glucose management and prevention in type 2 diabetes mellitus

Physical inactivity and sedentary behavior are risk factors for type 2 diabetes mellitus (T2DM). Therefore, physical exercise (PE) together with medical treatment might be considered as a key strategy to counteract T2DM. Glycemic control is a central objective in the prevention and management of T2DM, and PE might be able to substantially affect the processes that determine it. Just like a drug, exercise can be dosed based on the characteristics of the individual to increase its benefits and reduce side effects. In this brief review, the mechanisms underlying the effects of PE on glucose metabolism in muscle are illustrated, and the effects of modulation of the parameters characterizing this atypical "drug" on glucose homeostasis are described.

Walking for subjects with type 2 diabetes: a systematic review and joint AMD/SID/SISMES evidence-based practical guideline

Aims

Regular exercise is considered a cornerstone in the management of type 2 diabetes mellitus (T2DM). It improves glucose control and cardiovascular risk factors, contributes to weight loss, and also improves general well-being, likely playing a role in the prevention of chronic complications of diabetes. However, compliance to exercise recommendations is generally inadequate in subjects with T2DM. Walking is the most ancestral form of physical activity in humans, easily applicable in daily life. It may represent, in many patients, a first simple step towards lifestyle changes. Nevertheless, while most diabetic patients do not engage in any weekly walking, exercise guidelines do not generally detail how to improve its use. The aims of this document are to conduct a systematic review of available literature on walking as a therapeutic tool for people with T2DM, and to provide practical, evidence-based clinical recommendations regarding its utilization in these subjects.

Data synthesis

Analysis of available RCTs proved that regular walking training, especially when supervised, improves glucose control in subjects with T2DM, with favorable effects also on cardiorespiratory fitness, body weight, and blood pressure. Moreover, some recent studies have shown that even short bouts of walking, used for breaking prolonged sitting, can ameliorate glucose profiles in diabetic patients with sedentary behavior.

Conclusions

There is sufficient evidence to recognize that walking is a useful therapeutic tool for people with T2DM. This document discusses theoretical and practical issues for improving its use.

Walking for subjects with type 2 diabetes: A systematic review and joint AMD/SID/SISMES evidence-based practical guideline

AIMS

Regular exercise is considered a cornerstone in the management of type 2 diabetes mellitus (T2DM). It improves glucose control and cardiovascular risk factors, contributes to weight loss, and also improves general well-being, likely playing a role in the prevention of chronic complications of diabetes. However, compliance to exercise recommendations is generally inadequate in subjects with T2DM. Walking is the most ancestral form of physical activity in humans, easily applicable in daily life. It may represent, in many patients, a first simple step towards lifestyle changes. Nevertheless, while most diabetic patients do not engage in any weekly walking, exercise guidelines do not generally detail how to improve its use. The aims of this document are to conduct a systematic review of available literature on walking as a therapeutic tool for people with T2DM, and to provide practical, evidence-based clinical recommendations regarding its utilization in these subjects.

DATA SYNTHESIS

Analysis of available RCTs proved that regular walking training, especially when supervised, improves glucose control in subjects with T2DM, with favorable effects also on cardiorespiratory fitness, body weight and blood pressure. Moreover, some recent studies have shown that even short bouts of walking, used for breaking prolonged sitting, can ameliorate glucose profiles in diabetic patients with sedentary behavior.

CONCLUSIONS

There is sufficient evidence to recognize that walking is a useful therapeutic tool for people with T2DM. This document discusses theoretical and practical issues for improving its use. This article is co-published in the journals Sport Sciences for Health and Nutrition, Metabolism and Cardiovascular Diseases.

Recurrence quantification analysis of heart rate variability during continuous incremental exercise test in obese subjects

The present paper concerns a new description of changing in metabolism during incremental exercises test that permit an individually tailored program of exercises for obese subjects. We analyzed heart rate variability from RR interval time series (tachogram) with an alternative approach, the recurrence quantification analysis, that allows a description of a time series in terms of its dynamic structure and is able to identify the phase transitions. A transition in cardiac signal dynamics was detected and it perfectly reflects the aerobic threshold, as identified by gas exchange during an incremental exercise test, revealing the coupling from the respiratory system toward the heart. Moreover, our analysis shows that, in the recurrence plot of RR interval, it is possible to identify a specific pattern that allows to identify phase transitions between different dynamic regimes. The perfect match of the occurrence of the phase transitions with changes observed in the VO₂ consumption, the gold standard approach to estimate thresholds, strongly supports the possibility of using our analysis of RR interval to detect metabolic threshold. In conclusion, we propose a novel nonlinear data analysis method that allows for an easy and personalized detection of thresholds both from professional and even from low-cost wearable devices, without the need of expensive gas analyzers.

Relationship between individual ventilatory threshold and maximal fat oxidation (MFO) over different obesity classes in women

Objective

The use of the Individual Ventilatory Threshold (IVT), as parameter to prescribe exercise intensity in individuals with obesity, has become more frequent during the last years. This study aimed to evaluate the relationship between IVT and Maximal Fat Oxidation (MFO) in women with obesity.

Methods

Fifty-two obese female adults (age = 43.6±10.9 years; BMI = 38.5±5.2 kg/m²) were included in this study. According to the BMI classification, subjects were divided into three groups: Obese Class I (OBI, n = 16); Obese Class II (OBII, n = 20) and Obese Class III (OBIII, n = 16). All subjects performed an incremental graded exercise test to evaluate peak oxygen uptake (VO_2peak), IVT and MFO. MFO was evaluated using a stoichiometric equation. Fat max zone was determined for each subject within 10% of fat oxidation rates at MFO. For each HR, %HR_max, VO₂ and %VO_2peak variable, Pearson’s correlation test was done between IVT and MFO exercise intensity. When statistical correlation was found we used a comparative statistical analysis to assess differences between IVT and MFO. Statistical significance was set at P ≤ 0.05.

Results

For each HR, %HR_max, VO₂ and %VO_2peak variable there was a positive significant correlation (P<0.01) between IVT and MFO. No significant differences were found for HR, %HR_max, and VO₂ between IVT and MFO. %VO_2peak was significantly higher at IVT than at MFO (P = 0.03).

MFO rates were significantly higher in OBIII women than in women of the other two classes. In all subjects, IVT was within the fat max zone.

Conclusion

The use of HR and VO₂ corresponding to IVT could be a useful parameter not only to improve cardiorespiratory fitness but also to prescribe physical activity that maximize fat oxidation in obese subjects.

Effects of body weight loss program on parameters of muscle performance in female obese adults

BACKGROUND

Body weight loss program may lead to a decrease in lean body mass affecting negatively muscle performance in obese subjects. Thus, the aim of this study was to examine the effect of weight loss on muscle performance in female obese subjects.

METHODS

Eighty obese female adults were enrolled for a 2-month unsupervised aerobic training (UAT) plus nutritional program. In the pre- and postintervention body composition was evaluated by hand-to-foot bioelectrical impedance method, body strength using handgrip test, and lower muscle power was assessed by a 5-repetition chair stand test (CST) and 30-s chair stand test (30sCST) wearing a dynamometer.

RESULTS

Thirty-six subjects completed the protocol, 39% had high compliance (HC), while 61% had low compliance (LC) to exercise prescription. HC group showed a significant decreased body weight, percent of fat mass and lean muscle mass after training. Both groups significantly increased CST performance while only HC significantly increased 30sCST. No differences were found in CST muscle power in both groups between pre- and post-training. However, evaluation of muscle power during 30sCST showed significantly higher value in HC group than LC group after training.

CONCLUSIONS

The results of our study show that although total lean muscle mass decreased after UAT, lower body muscle efficiency increased while muscle power did not change suggesting that in obese patients UAT can help to optimize weight loss and body efficiency. The data might be helpful for exercise professionals to evaluate correctly the muscle performance in obese adults after weight loss programs.