Expenditure Prediction Research Articles

Prediction of activity-related energy expenditure under free-living conditions using accelerometer-derived physical activity

The purpose of the study was to develop prediction models to estimate physical activity (PA)-related energy expenditure (AEE) based on accelerometry and additional variables in free-living adults. In 50 volunteers (20–69 years) PA was determined over 2 weeks using the hip-worn Actigraph GT3X + as vector magnitude (VM) counts/minute. AEE was calculated based on total daily EE (measured by doubly-labeled water), resting EE (indirect calorimetry), and diet-induced thermogenesis. Anthropometry, body composition, blood pressure, heart rate, fitness, sociodemographic and lifestyle factors, PA habits and food intake were assessed. Prediction models were developed by context-grouping of 75 variables, and within-group stepwise selection (stage I). All significant variables were jointly offered for second stepwise regression (stage II). Explained AEE variance was estimated based on variables remaining significant. Alternative scenarios with different availability of groups from stage I were simulated. When all 11 significant variables (selected in stage I) were jointly offered for stage II stepwise selection, the final model explained 70.7% of AEE variance and included VM-counts (33.8%), fat-free mass (26.7%), time in moderate PA + walking (6.4%) and carbohydrate intake (3.9%). Alternative scenarios explained 53.8–72.4% of AEE. In conclusion, accelerometer counts and fat-free mass explained most of variance in AEE. Prediction was further improved by PA information from questionnaires. These results may be used for AEE prediction in studies using accelerometry

Defining a risk-adjustment formula for the introduction of population-based payments for primary care in France

Novel risk-adjusted payment models for financing primary care are currently being experimented in France. In particular, pilot schemes including shared-savings contracts or prospectively allocated capitation payments are implemented for voluntary primary care structures. Such payment mechanisms require defining a risk-adjustment formula to accurately estimate expected expenditure while maintaining appropriate efficiency incentives. We used nationwide data from the French national health data system (SNDS) to compare the performance of different prospective models for total and outpatient expenditure prediction among more than 8 million individuals aged 65 or more and their application at an aggregate level. We focused on the characterization of morbidity status and on the contextual characteristics to include in the formula. We proposed a set of practical routinely available predictors with fair performance for patient-level expenditure prediction (explaining 32% of variance) that could be used to risk-adjust prospective payments in the French setting. Morbidity information was the strongest predictor but could lead to considerable error in predicted expenditures if introduced as independent binary variables in multiplicative models, underlining the importance of summary morbidity measures and of using the appropriate metric to assess model performance. Distribution of aggregate-level allocations was greatly modified according to the method to account for contextual characteristics. Our work informs the introduction of risk-adjusted models in France and underlines efficiency and fairness issues raised.

Patient expenditure prediction using deep learning framework

Measurement of patient expenditure in healthcare is a critical task that has a variety of applications, including provider profiling, accountable care management, and capitated medical payment adjustment. Currently available methods rely on manually built features and linear regression-based models, both of which need a significant amount of medical domain expertise. and have low prediction accuracy. This study develops a multi-view deep learning system for forecasting future healthcare costs at the individual level based on prior claims data. Our multi-view technique efficiently models heterogeneous data such patient demographics, medical codes, medication usages, and facility usage. To execute spending forecasting tasks, we employed a real-world paediatric dataset with approximately 450,000 patients. According to the empirical data, our proposed technique beats all baselines for predicting medical cost. In the sphere of healthcare, these insights help to improve preventative and responsible care.

Patient Expenditure Prediction in Healthcare Sectors

Patient Expenditure Prediction in Healthcare Sectors

Benchmarking Neuromorphic Hardware and Its Energy Expenditure.

We propose and discuss a platform overarching benchmark suite for neuromorphic hardware. This suite covers benchmarks from low-level characterization to high-level application evaluation using benchmark specific metrics. With this rather broad approach we are able to compare various hardware systems including mixed-signal and fully digital neuromorphic architectures. Selected benchmarks are discussed and results for several target platforms are presented revealing characteristic differences between the various systems. Furthermore, a proposed energy model allows to combine benchmark performance metrics with energy efficiency. This model enables the prediction of the energy expenditure of a network on a target system without actually having access to it. To quantify the efficiency gap between neuromorphics and the biological paragon of the human brain, the energy model is used to estimate the energy required for a full brain simulation. This reveals that current neuromorphic systems are at least four orders of magnitude less efficient. It is argued, that even with a modern fabrication process, two to three orders of magnitude are remaining. Finally, for selected benchmarks the performance and efficiency of the neuromorphic solution is compared to standard approaches.

Validation of Resting Energy Expenditure Equations in Older Adults with Obesity

It is unclear which energy expenditure prediction equation should guide weight loss interventions in older adults with obesity. We ascertained the validity of four equations commonly used in practice in a series of weight loss studies of adults aged ≥65 with a body mass index ≥30kg/m2 using indirect calorimetry data. Diagnostic accuracy was defined as <10% discrepancy between predicted and measured resting metabolic rate (RMR). Mean was 73.4 years. RMR using the ReeVue was 1,643 kCal. With 59.0% accuracy, the WHO equation demonstrated the highest accuracy while the Harris-Benedict yielded 53.5% accuracy. The Owens equation demonstrated the least variability (21.5% overprediction, 27.8% underprediction) with 50.7% accuracy. A SECA bioimpedance analyzer noted the second lowest accuracy of 49.6%. Only 43.1% of measurements were within 10% of the gold-standard indirect calorimetry value using the Mifflin equation. All equations demonstrated <60% accuracy suggesting a great need for estimating energy needs.

Oxygen pulse best predicts energy expenditure during stair ascent and descent in individuals with chronic stroke.

To determine whether minute ventilation-to-carbon dioxide production (VE/VCO2), oxygen pulse (VO2/HR), and rate pressure product (RPP: VO2 * HR/1000) can explain energy expenditure during stair ascent/descent and whether energy expenditure during stair ascent/descent can discriminate between walking abilities in individuals with chronic stroke. Regression analysis of cross-sectional data from 50 individuals between 1 and 4 years post-stroke was carried out to investigate the prediction of energy expenditure during stair ascent/descent. In addition, discriminant analysis was carried out to investigate the discrimination between walking abilities for energy expenditure: community (walking speed ≥ 0.8 m/s) and non-community (walking speed < 0.8 m/s) walkers. Oxygen pulse and rate pressure product were retained in the model. Oxygen pulse alone explained 70% of the variance in energy expenditure during stair ascent/descent. By adding rate pressure product, 79% of the variance was explained. Energy expenditure was able to discriminate the community from the non-community walkers, with a cutoff value of 13.8 ml∙kg-1∙min-1 and correctly classified 62% of the non-community and 78% of the community walkers. Oxygen pulse and rate pressure product significantly predicted energy expenditure during stair ascent/descent in individuals with chronic stroke. Energy expenditure during stair ascent/descent discriminated community from non-community walkers.

Energy expenditure prediction equations in paediatric population: a systematic review

Rationale: The determination of energy requirements is of vital importance in the pediatric population, currently various equations have been designed and modified for their prediction. In this study, the equations, their components and metric characteristics that have been designed for the prediction of energy expenditure at baseline and at rest in healthy and sick pediatric patients were identified.

Prediction of outpatient visits and expenditure under the Universal Coverage Scheme in Bangkok using subscriber's attributes: A random forest analysis

ObjectivesThere is limited evidence on methods to allocate budgets to healthcare providers under capitation schemes. The objective of this study was to construct and test models that predict outpatient visits and expenditure for each healthcare facility using subscriber data from the preceding year. Study designWe used the database of the Universal Coverage Scheme in Bangkok, Thailand that stores subscriber information and healthcare service utilization data. One-percent and ten-percent random samples of subscribers were selected as training and testing groups, respectively. MethodsUsing data of the training group, we constructed a model using a random forest algorithm to predict outpatient visits and expenditure in 2017 from the 2016 data. The model was applied to the testing group and facility-level predicted number of visits and expenditure were compared with actual data. ResultsThe identically-structured training and testing groups consisted of 37,259 and 371,650 subscribers, respectively. Approximately 25% of subscribers utilized outpatient services. The R2 for models predicting facility-level utilization rate (visits/subscribers) and expenditure per subscriber in 2017 were 0.85 and 0.75, respectively. ConclusionsThe model to predict outpatient visits and expenditure performed well. Such a prediction model may be useful for allocating budgets to healthcare facilities under capitation systems.

Comparison of the Validity and Generalizability of Machine Learning Algorithms for the Prediction of Energy Expenditure: Validation Study.

BackgroundAccurate solutions for the estimation of physical activity and energy expenditure at scale are needed for a range of medical and health research fields. Machine learning techniques show promise in research-grade accelerometers, and some evidence indicates that these techniques can be applied to more scalable commercial devices.ObjectiveThis study aims to test the validity and out-of-sample generalizability of algorithms for the prediction of energy expenditure in several wearables (ie, Fitbit Charge 2, ActiGraph GT3-x, SenseWear Armband Mini, and Polar H7) using two laboratory data sets comprising different activities.MethodsTwo laboratory studies (study 1: n=59, age 44.4 years, weight 75.7 kg; study 2: n=30, age=31.9 years, weight=70.6 kg), in which adult participants performed a sequential lab-based activity protocol consisting of resting, household, ambulatory, and nonambulatory tasks, were combined in this study. In both studies, accelerometer and physiological data were collected from the wearables alongside energy expenditure using indirect calorimetry. Three regression algorithms were used to predict metabolic equivalents (METs; ie, random forest, gradient boosting, and neural networks), and five classification algorithms (ie, k-nearest neighbor, support vector machine, random forest, gradient boosting, and neural networks) were used for physical activity intensity classification as sedentary, light, or moderate to vigorous. Algorithms were evaluated using leave-one-subject-out cross-validations and out-of-sample validations.ResultsThe root mean square error (RMSE) was lowest for gradient boosting applied to SenseWear and Polar H7 data (0.91 METs), and in the classification task, gradient boost applied to SenseWear and Polar H7 was the most accurate (85.5%). Fitbit models achieved an RMSE of 1.36 METs and 78.2% accuracy for classification. Errors tended to increase in out-of-sample validations with the SenseWear neural network achieving RMSE values of 1.22 METs in the regression tasks and the SenseWear gradient boost and random forest achieving an accuracy of 80% in classification tasks.ConclusionsAlgorithms trained on combined data sets demonstrated high predictive accuracy, with a tendency for superior performance of random forests and gradient boosting for most but not all wearable devices. Predictions were poorer in the between-study validations, which creates uncertainty regarding the generalizability of the tested algorithms.

Energy Expenditure of Level Overground Walking in Young Adults: Comparison With Prediction Equations

The purpose of this study was to investigate the accuracy of the published prediction equations for determining level overground walking energy cost in young adults. In total, 148 healthy young adults volunteered to participate in this study. Resting metabolic rate and energy expenditure variables at speeds of 4, 5, and 6km/h were measured by indirect calorimetry, walking energy expenditure was estimated by 3 published equations. The gross and net metabolic rate per mile of level overground walking increased with increased speed (all P < .01). Females were less economical than males. The present findings revealed that the American College of Sports Medicine and Pandolf etal equations significantly underestimated the energy cost of overground walking at all speeds (all P < .01) in young adults. The percentage mean bias for American College of Sports Medicine, Pandolf etal, and Weyand etal was 12.4%, 16.8%, 1.4% (4km/h); 21.6%, 15.8%, 7.1% (5km/h); and 27.6%, 12%, 6.6% (6km/h). Bland-Altman plots and prediction error analysis showed that the Weyand etal was the most accurate in 3 existing equations. The Weyand etal equation appears to be the most suitable for the prediction of overground walking energy expenditure in young adults.

A supervised clustering MCMC methodology for large categorical feature spaces.

There is a well-established tradition within the statistics literature that explores different techniques for reducing the dimensionality of large feature spaces. The problem is central to machine learning and it has been largely explored under the unsupervised learning paradigm. We introduce a supervised clustering methodology that capitalizes on a Metropolis Hastings algorithm to optimize the partition structure of a large categorical feature space tailored towards minimizing the test error of a learning algorithm. This is a general methodology that can be applied to any supervised learning problem with a large categorical feature space. We show the benefits of the algorithm by applying this methodology to the problem of risk adjustment in competitive health insurance markets. We use a large claims data set that records ICD-10 codes, a large categorical feature space. We aim at improving risk adjustment by clustering diagnostic codes into risk groups suitable for health expenditure prediction. We test the performance of our methodology against common alternatives using panel data from a representative sample of twenty three million citizens in Colombian Healthcare System. Our results outperform common alternatives and suggest that it has potential to improve risk adjustment.

Muscle, endocrine, and immunological markers of frailty in older people.

To analyze muscle, endocrine, and immunological markers that influence frailty in older people assisted in primary care. Cross-sectional, analytical, and probabilistic study were linked to the institutional research "Integrated Health Care for Older People." The study population consisted of males and females aged 60years or more and assisted in primary health care. The research protocol included an interview and physical examination to evaluate the frailty criteria. Analysis of the following were done: serum calcium and creatinine as muscle markers; vitamin D, parathyroid hormone, and insulin-like growth factor-1 as endocrine markers; and interleukin-6, C-reactive protein, leukocytes, and neutrophil-lymphocyte ratio as immunological markers. Statistical analysis included the Mann-Whitney test to compare means, and linear regression to analyze the relationship between dependent and independent variables. There was a relationship between creatinine and prediction of weight loss (p<0.001), leukocytes and prediction of handgrip strength (p=0.022), interleukin-6 and prediction of energy expenditure (p=0.026), and vitamin D and prediction of gait time (p=0.036). Also, sex influenced handgrip strength (p<0.001), and age influenced handgrip strength (p<0.001), gait time (p<0.001) and energy expenditure (p<0.001). The joint use of muscle, endocrine, and immunological markers may be useful to diagnose frailty and to propose resolutive interventions to reduce negative outcomes for older people.

A new doubly labelled water anthropometry-based equation for prediction of total daily energy expenditure in older people from low- and middle-income countries.

For community-living older people, the ability to estimate total daily energy expenditure (TDEE) with validated predictive equations based on anthropometry is limited. To our knowledge no studies exist for non-Caucasian populations OBJECTIVE: To design and validate an anthropometry-based equation to estimate TDEE using doubly-labelled water (DLW) as the criterion measure, and to assess the performance of three other published equations in community-living older people from rural and urban areas of Brazil, Chile, Guatemala, Senegal, Cuba, and Mexico METHODS: This cross-sectional study measured anthropometry and TDEE using DLW in 69 men and 43 women aged 60-89 years. TDEE was also estimated with an anthropometry-based equation derived from the sub-sample of Mexico (n = 38) and with three other published equations. Predictive accuracy of the equations was tested by an external validation procedure RESULTS: TDEE by DLW in the six country sample was 2411 ± 41 kcal/day (mean ± SE) in men and 1939 ± 51 kcal/day in women. The best new Mexican equation was TDEE, kcal/d = [223.4 + (27.9 × weight, kg) + (239.7 × sex)]; where sex: Man = 1 and Woman = 0; having high precision; R2 = 0.89, lowest RMSE = 149.2, and Cp value of 2.0. This new Mexican equation estimated TDEE accurately in the five country sample and at country level after correction for Guatemalan older men, while the published equations performed poorly CONCLUSIONS: The Mexican equation performed better that other published equations and is recommended to accurately estimate energy requirements for community-living older people in five Latin American and one African country.

A nonlinear mixed model approach to predict energy expenditure from heart rate

Objective. Heart rate (HR) monitoring provides a convenient and inexpensive way to predict energy expenditure (EE) during physical activity. However, there is a lot of variation among individuals in the EE-HR relationship, which should be taken into account in predictions. The objective is to develop a model that allows the prediction of EE based on HR as accurately as possible and allows an improvement of the prediction using calibration measurements from the target individual. Approach. We propose a nonlinear (logistic) mixed model for EE and HR measurements and an approach to calibrate the model for a new person who does not belong to the dataset used to estimate the model. The calibration utilizes the estimated model parameters and calibration measurements of HR and EE from the person in question. We compare the results of the logistic mixed model with a simpler linear mixed model for which the calibration is easier to perform. Main results. We show that the calibration is beneficial already with only one pair of measurements on HR and EE. This is an important benefit over an individual-level model fitting, which requires a larger number of measurements. Moreover, we present an algorithm for calculating the confidence and prediction intervals of the calibrated predictions. The analysis was based on up to 11 pairs of EE and HR measurements from each of 54 individuals of a heterogeneous group of people, who performed a maximal treadmill test. Significance. The proposed method allows accurate energy expenditure predictions based on only a few calibration measurements from a new individual without access to the original dataset, thus making the approach viable for example on wearable computers.

Competitiveness of Timber Floor Elements: An Assessment of Structural Properties, Production, Costs, and Carbon Emissions

Abstract As long-spanning timber floor elements attempt to achieve a meaningful market share, proof of serviceability continues to be a demanding task as international consensus remains unsettled. Initiatives to improve vibration levels are achievable, but a lack of confidence in the market is resulting in increases in margins for both manufacturers and contractors. State-of-the-art concrete alternatives are offered at less than half the price, and even though timber floors offer reduced completion costs and low carbon emissions, the market is continuously reserved. Cost reductions for timber floor elements to competitive levels must be pursued throughout the product details and in the stages of manufacturing. As new wood products are introduced to the market, solution space is increased to levels that demand computerized optimization models, which require accurate expenditure predictions. To meet this challenge, a method called item-driven activity-based consumption (IDABC) has been developed and presented in this study. The method establishes an accurate relationship between product specifications and overall resource consumption linked to finished manufactured products. In addition to production time, method outcomes include cost distributions, including labor costs, and carbon emissions for both accrued materials and production-line activities. A novel approach to resource estimation linked to assembly friendliness is also presented. IDABC has been applied to a timber component and assembly line operated by a major manufacturer in Norway and demonstrates good agreement with empirical data.

Prediksi Belanja Pemerintah Indonesia Menggunakan Long Short-Term Memory (LSTM)

Abstract &#x0D; Estimates of government expenditure for the next period are very important in the government, for instance for the Ministry of Finance of the Republic of Indonesia, because this can be taken into consideration in making policies regarding how much money the government should bear and whether there is sufficient availability of funds to finance it. As is the case in the health, education and social fields, modeling technology in machine learning is expected to be applied in the financial sector in government, namely in making modeling for spending predictions. In this study, it is proposed the application of Long Short-Term Memory (LSTM) Model for expenditure predictions. Experiments show that LSTM model using three hidden layers and the appropriate hyperparameters can produce Mean Square Error (MSE) performance of 0.2325, Root Mean Square Error (RMSE) of 0.4820, Mean Average Error (MAE) of 0.3292 and Mean Everage Presentage Error (MAPE) of 0.4214. This is better than conventional modeling using the Auto Regressive Integrated Moving Average (ARIMA) as a comparison model.

Ecological and economic aspects of the efficiency of Polissia organic plant models

Purpose. The objective of the article is to investigate theoretically the possibility of harmonization of ecological and economic aspects of functioning of organic agricultural products production models under natural and climatic conditions of crop enterprises of Polissia zone in the Region of Chernihiv.&#x0D; Methodology / approach. The study was conducted at the Institute of Agricultural Microbiology and Agro-Industrial Manufacture of the National Academy of Agrarian Sciences of Ukraine during 2011–2020. The analysis of ecological and economic efficiency of organic crop rotations was made based on the results of the proprietary research on the example of generalization of predicted results of the developed standard models of organic crop rotations at crop enterprises of Polissia zone. Main methods: modelling (development, in particular, of standard models of crop rotations, modelling of the size of fertilizer potential over time by years); predicting (prediction of receipt and expenditure of nutrients, organic matter), balance (calculation of balances of humus and key nutrients), calculation and design (calculation of the basic parameters of economic efficiency); abstract and logical (formulation of the provisions of the article based on the analysis of studies of other authors, drawing conclusions based on the results of studies conducted).&#x0D; Results. Developed typical crop rotation models are characterized by positive predictive humus balances and acceptable key nutrient balances (NPK). With the implementation of ‘organic’ prices for agricultural products, most crops achieve an acceptable level of profitability. However, in order to achieve a competitive level of profitability of organic production, it must be provided with an appropriate economic mechanism of its functioning, in particular higher prices compared to traditional agricultural products, etc. Harmonization of economic and ecological aspects in simulated crop rotations is achieved primarily by selecting crops with high yields, which are mainly characterized by insufficient or low level of ecological efficiency, and crops with high or medium ecological efficiency, which having insufficient or low economic efficiency, compensate mostly negative ecological consequences of growing crops with high economic efficiency. This is accompanied by achievement of the ecological balance of agrocenoses and an acceptable level of economic efficiency of organic models. The practical aspect of the activities of specific farms in relation to the choice of the appropriate model of organic management should take into account not only considerations of ecological and economic efficiency, but also the level of resource provision within these limits.&#x0D; Originality / scientific novelty. The possibility of harmonization of economic and ecological aspects and achievement of their acceptable level in crop rotations of organic agriculture for crop enterprises has been theoretically substantiated. For the first time, the methodological approach developed by the authors on the distribution of costs for the production and application of organic fertilizers between crops during crop rotation in proportion to the duration and amount of fertilization was implemented.&#x0D; Practical value / implications. Typical models of crop rotations of organic farming for crop enterprises in Polissia were developed. Methods of choosing a certain model were proposed both according to the criteria of ecological and economic efficiency, and taking into account the organizational and economic capabilities and resource provision of a particular business entity.

Forecasting future dental health expenditures: Development of a framework using data from 32 OECD countries.

ObjectiveTo (1) develop a framework for forecasting future dental expenditures, using currently available information, and (2) identify relevant research and data gaps such that dental expenditure predictions can continuously be improved in the future.MethodsOur analyses focused on 32 OECD countries. Dependent on the number of predictors, we employed dynamic univariate and multivariate modelling approaches with various model specifications. For univariate modelling, an auto‐regressive (AR) dynamic model was employed to incorporate historical trends in dental expenditures. Multivariate modelling took account of historical trends, as well as of relationships between dental expenditures, dental morbidity, economic growth in terms of gross domestic product and demographic changes.ResultsEstimates of dental expenditures varied substantially across different model specifications. Models relying on dental morbidity as one of the predictors performed worst regardless of their specification. Using the best‐fitted model specification, that is the univariate second‐order autoregression [AR(2)], the forecasted dental expenditures across 32 OECD countries amounted to US$316bn (95% forecasted interval, FI: 258‐387) in 2020, US$434bn (95%FI: 354‐532) in 2030 and US$594bn (95%FI: 485‐728) in 2040. Per capita spending in 2040 was forecasted to be highest in Germany (US$889, 95%FI: 726‐1090) and lowest in Mexico (US$52, 95%FI: 42‐64).ConclusionsThe present study demonstrates the feasibility and challenges in predicting dental expenditures and can serve as a basis for improvement towards more sustainable and resilient health policy and resource planning. Within the limitations of available data sources, our findings suggest that dental expenditures in OECD countries could increase substantially over the next two decades and vary considerably across countries. For more accurate estimation and a better understanding of determinants of dental expenditures, more comprehensive data on dental spending and dental morbidity are urgently needed.

Energy expenditure in a Syme’s amputee triathlete

Energy expenditure prediction equations developed for able-bodied populations may be inaccurate for para-athletes, a population characterized by different types of impairments. Sufficient data do not exist currently to generate or validate energy expenditure predictive equations for para-athletes to establish dietary recommendations to cover the demands of sport. The purpose of this observational study was to assess energy expenditure and substrate usage from a trained para-athlete at rest and while performing the athlete’s specific mode of exercise. One para-athlete recruited from the Challenged Athletes Foundation (female; age 32) participated in the study. The participant was a left leg Syme’s amputee triathlete. Resting metabolic rate was measured for the triathlete. Energy expenditure was also measured at moderate (64-76% of age-predicted HRmax) and vigorous intensity exercise (77-95% of age-predicted HRmax) on a stationary cycle ergometer on a separate day. The triathlete’s energy expenditure during exercise was higher than that predicted by the metabolic equivalent of task (MET) matched by exercise type at a moderate (measured: 7.0 kcal/min, predicted: 5.0 kcal/min) and vigorous intensity (measured: 9.3 kcal/min, predicted: 7.0 kcal/min). The Revised Harris-Benedict and the Mifflin-St Jeor equations underestimated resting metabolic rate in the triathlete (measured: 2,194 kcal/day, predicted: 1,348 ± 39 kcal/day). This study indicates that equations developed for able-bodied populations underestimate energy expenditure in a Syme’s amputee when compared to the measured energy expenditure values assessed by indirect calorimetry at rest and during a workout on both a stationary cycle ergometer and while running on the track.