Estimation Of Expenditure Research Articles

A Dual-Stream Fusion Network for Human Energy Expenditure Estimation with Wearable Sensor

With the increasing awareness of health, using wearable sensors to monitor individual activities and accurately estimate energy expenditure has become a current research focus. However, existing research encounters challenges including low estimation accuracy, a deficiency of frequency domain features, and difficulty in integrating time domain and frequency domain features. To address these issues, we propose an innovative framework called the Dual-Stream Fusion Network (DSFN). This framework combines the Time Domain Encoding (TDE) module, the Frequency Domain Hierarchical-Split Encoding (FDHSE) module, and a Two-Stage Feature Fusion (TSF) module. Specifically, the temporal stream of the framework employs the TDE module to capture deep temporal features that reflect the complex dynamic variations in time-series data. The frequency domain stream introduces the FDHSE module, which extracts frequency domain features using a multi-level, multi-scale approach, ensuring a comprehensive and diverse representation of frequency information. Through this dual-stream architecture, our model effectively learns both time and frequency domain features, addressing the limitations of frequency domain features observed in prior studies. Additionally, we propose the TSF module to fully integrate time and frequency domain features, effectively overcoming the challenge of fusing these two types of features. We conducted experiments on two public datasets, namely the GOTOV dataset (elderly people) and the JSI dataset (young people). Experimental results demonstrate that our method achieves excellent performance across different age groups. Compared to the baseline models, the proposed DSFN significantly improves the accuracy of human energy expenditure estimation.

Lifetime healthcare expenditures across socioeconomic groups

BackgroundA socioeconomic gradient affects healthcare expenditures and longevity in opposite directions as less affluent individuals have higher current healthcare expenditures but simultaneously enjoy shorter lives. Yet, it is unclear whether this cross-sectional healthcare expenditure gradient persists from a lifetime perspective. This paper analyzes lifetime healthcare expenditures across socioeconomic groups using detailed individual-level healthcare expenditure data for the entire Danish population.MethodUsing full population healthcare expenditures from Danish registries, we estimate lifetime healthcare expenditures as age-specific mean healthcare expenditures times the probability of being alive at each age. Our data enables the estimation of lifetime healthcare expenditures by sex, socioeconomic status, and by various types of healthcare expenditure.ResultsOnce we account for mortality differences and all types of healthcare expenditures, all socioeconomic groups spend an almost equal amount on healthcare throughout a lifetime. Lower socioeconomic groups incur the lowest lifetime hospital expenditures, whereas higher socioeconomic groups experience the highest lifetime expenditures on long-term care services. Our findings remain robust across various socioeconomic measures and alternative estimation methodologies.ConclusionImproving the health status of lower socioeconomic groups to align with that of higher socioeconomic groups is costly but may ultimately reduce current healthcare expenditures. Enhanced health outcomes likely increase lifespan, leading to extended periods of healthcare consumption. However, since all socioeconomic groups tend to have similar lifetime healthcare expenditures, this prolonged consumption has limited impact on overall lifetime healthcare costs. Additionally, a significant benefit is the deferment of healthcare expenditures into the future. Overall, our results diminish concerns about socially inequitable utilization of healthcare resources while socioeconomic differences in health and longevity persist, even in a universal healthcare system.

24-h Energy Expenditure Estimation From Accelerometry Assessments: A Cross-validation At Low Levels Of Physical Activity

24-h Energy Expenditure Estimation From Accelerometry Assessments: A Cross-validation At Low Levels Of Physical Activity

National Trends in the Utilization and Expenditure of Chiropractic Care in U.S. Adults: Analysis of the 2007-2016 Medical Expenditure Panel Survey

ObjectiveThis study aimed to examine the trends in the utilization and expenditure of chiropractic care in a representative sample of US adults, aged ≥18 years. MethodsSerial cross-sectional data (2007-2016) from the Medical Expenditure Panel Survey (MEPS) were examined. Weighted descriptive statistics were analyzed to obtain national estimates of chiropractic utilization and expenditure, and time-series linear regression was used to assess trends over time. Socio-demographic characteristics and musculoskeletal diagnoses associated with chiropractic use were also reported. ResultsA statistically significantly increasing trend was observed for the number of adults receiving chiropractic care (p < .05), number of visits (p < .05) and utilization rate (P < .05) from 2007 to 2016. A similar trend was not found for chiropractic expenditure during this period (P > .05). The mean number of visits was 8.3 visits per year, with a mean expenditure of $86.94 USD per visit and $721.43 USD per person per year. Mean age of adult chiropractic users each year ranged between 48.6 and 51.2 years old, and users were primarily female (56.3%-60.4%), White persons (90.1%-93.5%), and privately insured (77.3%-82.8%). The most prevalent musculoskeletal diagnoses associated with chiropractic use were low back conditions (45.4%-58.1%), inflammatory/degenerative disc or joint conditions (21.2%-26.8%) and head and neck complaints (9.8%-13.7%). ConclusionThe findings showed statistically significant increasing trends for the number of adult chiropractic users, the number of visits, and percent utilization of chiropractic care from 2007 to 2016. Conversely, we found a statistically null trend for the total annual expenditure on chiropractic care during the same time period. Findings from this analysis imply that, while chiropractic care appears to be growing, it may still be under-utilized, and more efforts should be devoted to ensuring sustained growth and a larger role in the management of musculoskeletal health.

Time-energy budgets outperform dynamic body acceleration in predicting daily energy expenditure in kittiwakes, and estimate a very low cost of gliding flight relative to flapping flight.

Energy is a common currency for any living organism, yet estimating energy expenditure in wild animals is challenging. Accelerometers are commonly used to estimate energy expenditure, via a dynamic body acceleration (DBA) or time-energy budget approach. The dynamic body acceleration approach estimates energy expenditure directly from acceleration but may lead to erroneous estimates during inactivity when acceleration is zero but energy expenditure is not. The time-energy budget approach uses accelerometers and other data streams to assign a behaviour to each time step, and then calculates energy expenditure based on activity-specific metabolic rates assigned to each behaviour. Here, used GPS-accelerometry in breeding black-legged kittiwakes (Rissa tridactyla, n=80) to calculate DBA and time-energy budgets derived from simple biologging metrics (speed, wing beat frequency, GPS position). We then compared these two approaches to estimates of energy expenditure from doubly-labelled water. Energy expenditure estimated from DLW correlated with DBA, but the best model to estimate energy expenditure was based on time-energy budgets. Energy costs of flapping flight were higher than all other kittiwake behaviours (5.54 x basal metabolic rate, BMR). Energetic costs of gliding flight (0.80 x BMR) were the lowest of all behaviours, and equivalent to the cost of resting at the colony. DEE for our birds estimated from our calibration coefficients are similar to DEE for our birds estimated with the model coefficient published using different methods. We conclude that once calibrated with DLW, GPS-accelerometry provides a simple method for measuring energy expenditure in wild kittiwakes based on time-energy budgets.

An Urban Renewal Design Method Based on Carbon Emissions and Carbon Sink Calculations: A Case Study on an Environmental Improvement Project in the Suzhou Industrial Investment Science and Technology Park

In the process of urban renewal, the high carbon emissions caused by pollution from construction waste and the consumption of materials and energy via “demolition and construction” present significant problems. Calculating carbon emissions and sinks is a prerequisite to carving out a low-carbon path via urban renewal. In this paper, we take an environmental improvement project in the Suzhou Industrial Investment Science and Technology Park as an example. Under the framework of the entire life cycle, we define the “renewal phase”, which includes “demolition” and “construction”, as the calculation boundary, and use the emission coefficient method as the primary calculation approach to assess carbon emissions and sinks. We construct a design-oriented carbon revenue and expenditure estimation system based on the BIM model. After its application in empirical cases, the results show the following: ① The building carbon emissions of Scheme A are 342 t higher than those of Scheme B, due to the fact that the “demolition and construction” works in the former increased by 6000 m2 compared to the latter. ② The landscape carbon emissions of Scheme B are 269 t higher than those of Scheme A, due to the addition of a 2500 m2 reinforced concrete overhead walking platform. ③ The annual carbon sink of Scheme A is six times higher than that of Scheme B, due to the fact that the number of trees and shrubs in the former is five-to-six times greater than in the latter. The number of trees planted plays a decisive role in enhancing the carbon sink benefit. ④ In the year during which the renewal project was implemented, the difference in carbon emissions between Scheme A and Scheme B was 72.9 t. By the 16th year, the difference in carbon emissions between the two schemes approached zero, and the carbon reduction advantages of Scheme A became more pronounced over the entire life cycle thereafter. Finally, this article summarizes four low-carbon design strategies for industrial park renewal projects: rational construction and demolition, three-dimensional parking, low-carbon materials, and plant carbon sequestration. In the context of inventory renewal, this article makes a significant contribution in terms of carbon footprint control and the “dual-carbon” goal.

Composite activity type and stride-specific energy expenditure estimation model for thigh-worn accelerometry

BackgroundAccurately measuring energy expenditure during physical activity outside of the laboratory is challenging, especially on a large scale. Thigh-worn accelerometers have gained popularity due to the possibility to accurately detect physical activity types. The use of machine learning techniques for activity classification and energy expenditure prediction may improve accuracy over current methods. Here, we developed a novel composite energy expenditure estimation model by combining an activity classification model with a stride specific energy expenditure model for walking, running, and cycling.MethodsWe first trained a supervised deep learning activity classification model using pooled data from available adult accelerometer datasets. The composite energy expenditure model was then developed and validated using additional data based on a sample of 69 healthy adult participants (49% female; age = 25.2 ± 5.8 years) who completed a standardised activity protocol with indirect calorimetry as the reference measure.ResultsThe activity classification model showed an overall accuracy of 99.7% across all five activity types during validation. The composite model for estimating energy expenditure achieved a mean absolute percentage error of 10.9%. For running, walking, and cycling, the composite model achieved a mean absolute percentage error of 6.6%, 7.9% and 16.1%, respectively.ConclusionsThe integration of thigh-worn accelerometers with machine learning models provides a highly accurate method for classifying physical activity types and estimating energy expenditure. Our novel composite model approach improves the accuracy of energy expenditure measurements and supports better monitoring and assessment methods in non-laboratory settings.

A retrospective observational study of mNGS test utilization to examine the role of diagnostic stewardship at two academic medical centers.

Metagenomic next-generation sequencing (mNGS) stewardship practices remain poorly standardized. This study aims to provide actionable insights for institutions that seek to reduce the unnecessary usage of mNGS. Importantly, we highlight that clinical impact remains challenging to measure without standardized guidelines, and we provide an actual cost estimate of microbiology expenditures on individuals undergoing mNGS.

Estimating and planning hospital costs of public hospitals in Brazil.

While the estimate of hospital costs concerns the past, its planning focuses on the future. However, in many low and middle-income countries, public hospitals do not have robust accounting health systems to evaluate and project their expenses. In Brazil, public hospitals are funded based on government estimates of available hospital infrastructure, historical expenditures and population needs. However, these pieces of information are not always readily available for all hospitals. To solve this challenge, we propose a flexible simulation-based optimisation algorithm that integrates this dual task: estimating and planning hospital costs. The method was applied to a network of 17 public hospitals in Brazil to produce the estimates. Setting the model parameters for population needs and future hospital infrastructure can be used as a cost-projection tool for divestment, maintenance, or investment. Results show that the method can aid health managers in hospitals' global budgeting and policymakers in improving fairness in hospitals' financing.

Potential sources of inaccuracy in the Apple watch series 4 energy expenditure estimation algorithm during wheelchair propulsion

BackgroundThe Apple Watch (AW) was the first smartwatch to provide wheelchair user (WCU) specific information on energy expenditure (EE), but was found to be inaccurate (i.e., it underestimated) and imprecise (i.e., the underestimation was variable). Insight is therefore needed into where these inaccuracies/imprecisions originate. Accordingly, the aim of this study was to investigate how much of the variation in AW EE is explained by heart rate (HR), in addition to other factors such as body mass and height, sex, age, physical activity level and disability.MethodsForty participants (20 WCU, 20 non-disabled) performed three 4-min treadmill wheelchair propulsion stages at different speed-incline combinations, on three separate days, while wearing an AW series 4 (setting: “outdoor push walking pace”). Linear mixed model analyses investigated how much of the variation in AW EE (kcal·min−1) is explained by the fixed effects AW HR (beats·min−1), body mass and height, sex, age, physical activity level and disability. Participant-ID was included as random-intercept effect. The same mixed model analyses were conducted for criterion EE and HR. Marginal R2 (R2m; fixed effects only) and conditional R2 (R2c; fixed and random effects) values were computed. An R2m close to zero indicates that the fixed effects alone do not explain much variation.ResultsAlthough criterion HR explained a significant amount of variation in criterion EE (R2m: 0.44, R2c: 0.92, p < 0.001), AW HR explained little variation in AW EE (R2m: 0.06, R2c: 0.86, p < 0.001). In contrast, body mass and sex explained a significant amount of variation in AW EE (R2m: 0.74, R2c: 0.79, p < 0.001). No further improvements in fit were achieved by adding body height, age, physical activity level or disability to the AW EE model (R2m: 0.75, R2c: 0.79, p = 0.659).ConclusionOur results remain inconclusive on whether AW heart rate is used as factor to adjust for exercise intensity in the black box AW EE estimation algorithms. In contrast, body mass explained much of the variation in AW EE, indicating that the AW EE estimation algorithm is very reliant on this factor. Future investigations should explore better individualization of EE estimation algorithms.

Preliminary field validity of ActiGraph-based energy expenditure estimation in wheelchair users with spinal cord injury.

Cross-sectional validation study. To develop a raw acceleration signal-based random forest (RF) model for predicting total energy expenditure (TEE) in manual wheelchair users (MWUs) and evaluate the preliminary field validity of this new model, along with four existing models published in prior literature, using the Doubly Labeled Water (DLW) method. General community and research institution in Pittsburgh, USA. A total of 78 participants' data from two previous studies were used to develop the new RF model. A seven-day cross-sectional study was conducted to collect participants' free-living physical activity and TEE data, resting metabolic rate, demographics, and anthropometrics. Ten MWUs with spinal cord injury (SCI) completed the study, with seven participants having valid data for evaluating the preliminary field validity of the five models. The RF model achieved a mean absolute error (MAE) of 0.59 ± 0.60 kcal/min and a mean absolute percentage error (MAPE) of 23.6% ± 24.3% on the validation set. For preliminary field validation, the five assessed models yielded MAE from 136 kcal/day to 1141 kcal/day and MAPE from 6.1% to 50.2%. The model developed by Nightingale et al. in 2015 achieved the best performance (MAE: 136 ± 96 kcal/day, MAPE: 6.1% ± 4.7%), while the RF model achieved comparable performance (MAE: 167 ± 99 kcal/day, MAPE: 7.4% ± 5.1%). Two existing models and our newly developed RF model showed good preliminary field validity for assessing TEE in MWUs with SCI and the potential to detect lifestyle change in this population. Future large-scale field validation studies and model iteration are recommended.

Machine learning algorithms for energy efficiency: Mitigating carbon dioxide emissions and optimizing costs in a hospital infrastructure

Hospital buildings incur significant energy consumption due to their expensive operations, advanced medical equipment, strict sanitation protocols, and compliance with environmental standards in various weather conditions. For this study, focusing on two distinct climate regions in Turkey—Aksaray and Bursa provinces—a total of 1,440 unique scenarios were generated. These scenarios utilized materials from the Revit Building Information Modeling (BIM) software library, incorporating variations in thermal transmittance coefficients, solar heat gain coefficients, and building orientation angles. The research employed machine learning algorithms to predict energy consumption, carbon dioxide (CO2) emissions, total expenditures, and life cycle costs. While the Revit BIM software library provides numerous combinations of building materials, it’s important to note that only a limited subset undergoes practical testing during construction. The main objective is to streamline the estimation of energy consumption, CO2 emissions, overall costs, and life cycle expenditures in similar architectural settings and under comparable climate conditions, eliminating the need for drafting and energy calculation software. Across all machine learning algorithms, predictions generated by artificial neural networks closely aligned with actual values. Notably, the R-squared (R2) values, a critical evaluation metric, produced the following outcomes: both R2 values for energy were 0.95; total costs achieved 0.93 and 0.97; CO2 emissions attained 0.94 and 0.97; and life cycle costs garnered 0.95 and 0.94 for validation and test datasets, respectively. These results suggest that the successes observed in the two provinces can be extrapolated to data from all regions in Turkey. Moreover, our model is capable of predicting energy consumption, costs, and CO2 emissions based on thermal transmittance values complying with the Turkish standard TS 825 and building orientation angles.

Energy expenditure estimation during activities of daily living in middle-aged and older adults using an accelerometer integrated into a hearing aid.

Accelerometers were traditionally worn on the hip to estimate energy expenditure (EE) during physical activity but are increasingly replaced by products worn on the wrist to enhance wear compliance, despite potential compromises in EE estimation accuracy. In the older population, where the prevalence of hearing loss is higher, a new, integrated option may arise. Thus, this study aimed to investigate the accuracy and precision of EE estimates using an accelerometer integrated into a hearing aid and compare its performance with sensors simultaneously worn on the wrist and hip. Sixty middle-aged to older adults (average age 64.0 ± 8.0 years, 48% female) participated. They performed a 20-min resting energy expenditure measurement (after overnight fast) followed by a standardized breakfast and 13 different activities of daily living, 12 of them were individually selected from a set of 35 activities, ranging from sedentary and low intensity to more dynamic and physically demanding activities. Using indirect calorimetry as a reference for the metabolic equivalent of task (MET), we compared the EE estimations made using a hearing aid integrated device (Audéo) against those of a research device worn on the hip (ZurichMove) and consumer devices positioned on the wrist (Garmin and Fitbit). Class-estimated and class-known models were used to evaluate the accuracy and precision of EE estimates via Bland-Altman analyses. The findings reveal a mean bias and 95% limit of agreement for Audéo (class-estimated model) of -0.23 ± 3.33 METs, indicating a slight advantage over wrist-worn consumer devices (Garmin: -0.64 ± 3.53 METs and Fitbit: -0.67 ± 3.40 METs). Class-know models reveal a comparable performance between Audéo (-0.21 ± 2.51 METs) and ZurichMove (-0.13 ± 2.49 METs). Sub-analyses show substantial variability in accuracy for different activities and good accuracy when activities are averaged over a typical day's usage of 10 h (+61 ± 302 kcal). This study shows the potential of hearing aid-integrated accelerometers in accurately estimating EE across a wide range of activities in the target demographic, while also highlighting the necessity for ongoing optimization efforts considering precision limitations observed across both consumer and research devices.

Allocation of the household food budget among shopping basket items: How is it influenced by promotions?

Retailers have been using promotion as a differentiation strategy that influences consumers' expenditures and their shopping basket budgetary allocation. This study assessed the effect of retail promotions on total shopping basket expenditure and determined whether promotions provoke a reallocation of the shopping budget. The analysis was performed on a chain of supermarkets in Catalonia, Spain using a consumer scanner data set from Kantar Worldpanel for 2017. The methodological approach had two steps: prediction of the effect of promotion on household expenditures using an expenditure regression model and estimation of the promotion own- and cross-effect using the censored Exact Affine Stone Index. Promotion had a positive own-effect and mostly a negative asymmetric cross-effect, implying a small but significant budget reallocation.

Resting Energy Expenditure, comparison of predictive formulas with bioimpedance: Peruvian population

ntroduction: In areas with limited access to healthcare systems, Resting Energy Expenditure (REE) estimation is performed using predictive equations to calculate an individual’s caloric requirement. One problem is that these equations were validated in populations with different characteristics from those in Latin America, such as race, height, or body mass, leading to potential errors in the prediction of this parameter. Objective: To determine the REE using predictive formulas compared with bioimpedance in Peruvians. Materials and methods: A comparative analytical cross-sectional study with secondary database analysis of the CRONICAS cohort. Results: we worked with a total of 666 subjects. The Mjeor equation was the one with the highest rating of 0.95, a lower mean absolute percentage error (MAPE) of 4.69%, and equivalence was found with the REE values. In the multiple regression, it was observed that the Mjeor equation was the one that least overestimated the REE, increasing 0.77 Kcal/day (95% CI: 0.769-0.814; p&lt;0.001) for each point that increased the REE determined by bioimpedance. The strength of association between Mjeor and bioimpedance was 0.9037. Furthermore, in the regression of the data (weight, height, age) in the Mjeor equation it was observed that the coefficients obtained were the same as those used in the original equation. Conclusions: The Mjeor equation seems to be the most adequate to estimate the REE in the Peruvian population. Future prospective studies should confirm the usefulness of this formula with potential utility in primary health care.

24-h energy expenditure in people with type 1 diabetes: impact on equations for clinical estimation of energy expenditure.

Type 1 diabetes (T1D) is associated with an increase in resting metabolic rate (RMR), but the impact of T1D on other components of 24-h energy expenditure (24-h EE) is not known. Also, there is a lack of equations to estimate 24-h EE in patients with T1D. The aims of this analysis were to compare 24-h EE and its components in young adults with T1D and healthy controls across the spectrum of body mass index (BMI) and derive T1D-specific equations from clinical variables. Thirty-three young adults with T1D diagnosed ≥1 year prior and 33 healthy controls matched for sex, age and BMI were included in this analysis. We measured 24-h EE inside a whole room indirect calorimeter (WRIC) and body composition with dual x-ray absorptiometry. Participants with T1D had significantly higher 24-h EE than healthy controls (T1D = 2047 ± 23 kcal/day vs control= 1908 ± 23 kcal/day; P < 0.01). We derived equations to estimate 24-h EE with both body composition (fat free mass + fat mass) and anthropometric (weight + height) models, which provided high coefficients of determination (R2 = 0.912 for both). A clinical model that did not incorporate spontaneous physical activity yielded high coefficients of determination as well (R2 = 0.897 and R2 = 0.880 for body composition and anthropometric models, respectively). These results confirm that young adults with established T1D have increased 24-h EE relative to controls without T1D. The derived equations from clinically available variables can assist clinicians with energy prescriptions for weight management in patients with T1D.

A CNN Model for Physical Activity Recognition and Energy Expenditure Estimation from an Eyeglass-Mounted Wearable Sensor.

Metabolic syndrome poses a significant health challenge worldwide, prompting the need for comprehensive strategies integrating physical activity monitoring and energy expenditure. Wearable sensor devices have been used both for energy intake and energy expenditure (EE) estimation. Traditionally, sensors are attached to the hip or wrist. The primary aim of this research is to investigate the use of an eyeglass-mounted wearable energy intake sensor (Automatic Ingestion Monitor v2, AIM-2) for simultaneous recognition of physical activity (PAR) and estimation of steady-state EE as compared to a traditional hip-worn device. Study data were collected from six participants performing six structured activities, with the reference EE measured using indirect calorimetry (COSMED K5) and reported as metabolic equivalents of tasks (METs). Next, a novel deep convolutional neural network-based multitasking model (Multitasking-CNN) was developed for PAR and EE estimation. The Multitasking-CNN was trained with a two-step progressive training approach for higher accuracy, where in the first step the model for PAR was trained, and in the second step the model was fine-tuned for EE estimation. Finally, the performance of Multitasking-CNN on AIM-2 attached to eyeglasses was compared to the ActiGraph GT9X (AG) attached to the right hip. On the AIM-2 data, Multitasking-CNN achieved a maximum of 95% testing accuracy of PAR, a minimum of 0.59 METs mean square error (MSE), and 11% mean absolute percentage error (MAPE) in EE estimation. Conversely, on AG data, the Multitasking-CNN model achieved a maximum of 82% testing accuracy in PAR, a minimum of 0.73 METs MSE, and 13% MAPE in EE estimation. These results suggest the feasibility of using an eyeglass-mounted sensor for both PAR and EE estimation.

Energy Expenditure Prediction from Accelerometry Data Using Long Short-Term Memory Recurrent Neural Networks.

The accurate estimation of energy expenditure from simple objective accelerometry measurements provides a valuable method for investigating the effect of physical activity (PA) interventions or population surveillance. Methods have been evaluated previously, but none utilize the temporal aspects of the accelerometry data. In this study, we investigated the energy expenditure prediction from acceleration measured at the subjects' hip, wrist, thigh, and back using recurrent neural networks utilizing temporal elements of the data. The acceleration was measured in children (N = 33) performing a standardized activity protocol in their natural environment. The energy expenditure was modelled using Multiple Linear Regression (MLR), stacked long short-term memory (LSTM) networks, and combined convolutional neural networks (CNN) and LSTM. The correlation and mean absolute percentage error (MAPE) were 0.76 and 19.9% for the MLR, 0.882 and 0.879 and 14.22% for the LSTM, and, with the combined LSTM-CNN, the best performance of 0.883 and 13.9% was achieved. The prediction error for vigorous intensities was significantly different (p < 0.01) from those of the other intensity domains: sedentary, light, and moderate. Utilizing the temporal elements of movement significantly improves energy expenditure prediction accuracy compared to other conventional approaches, but the prediction error for vigorous intensities requires further investigation.

Application of Ensemble Learning and Feature Selection Models in Energy Expenditure Estimation of Fitness Tracking Devices

The use of fitness tracking devices to monitor energy expenditure (EE) is becoming increasingly popular among consumers. However, the accuracy of EE estimation still needs to be improved due to the lack of accurate data filtering and more efficient data analysis models. To address this issue, we designed an energy expenditure estimation (EEE) model based on smart tracking devices. This paper proposes an algorithm for data reconstruction called PC-DF-RFECV: firstly, Pearson correlation analysis (PC) is used for initial data screening, then Feature Derivation (DF) is applied to reconstruct new features, and finally, the Recursive Feature Elimination algorithm based on cross-validation (RFECV) is employed to select the new features. Furthermore, the EEE model adopts a weighted average (Wei-Voting) strategy to enhance the robustness of the model by integrating the predictions of multiple base learners (XGBoost, LightGBM, Random Forest). During algorithm optimization, a Bayesian hyperparameter optimization technique is employed to fine-tune the model's hyperparameters. Empirical findings indicate that the EEE model, when applied to the dataset collected from motion tracking devices, attains an score of 0.9985, of 1.033, and of 0.865. These results demonstrate superior performance compared to current state-of-the-art research and conventional algorithms including random forests, gradient boosting, and bagging.

Potential smoke-free dividend across local areas in England: a cross-sectional analysis

BackgroundThe value that might be added to local economies each year through the money that people who smoke tobacco would save if everyone quit smoking is called the ‘smoke-free dividend’....