Artificial intelligence-powered predictive modeling for calorie burns in underserved regions AI-Based Calorie Burn Prediction
Main Article Content
Abstract
Background: Tracking caloric expenditure in relation to the intensity, rate, and duration plays an important role in healthcare and nutrition, with a focus on underserved regions where access to nutritional support and health tracking tools are limited. This study aims to bridge the gap and emphasize the nuances in the predictive capacity of machine learning models and the development of cost-effective and applicable models in tracking calorie burn in these regions.
Methods: This study utilized a publicly available Calories Burnt Prediction dataset (Kaggle) comprising 15,000 records with variables like age, sex, height, weight, duration of exercise, heart rate and body temperature. Data preprocessing steps included categorical encoding, feature scaling, and feature selection based on importance ranking followed by the training and evaluation of XGBoost (Xtreme Gradient Boosting)), Ridge Regression and Random Forest machine learning frameworks performance was assessed using mean absolute error (MAE), mean squared error (MSE) and coefficient of determination (R²).
Results: The CaloModel (XGBoost) demonstrated the best overall performance, with the lowest prediction error and highest explained variance (MAE: 10.18; MSE: 167.21; R²: 0.95). Ridge Regression and Random Forest also showed strong predictive performance but were comparatively less favorable. Feature importance analysis identified exercise duration, heart rate, and body temperature as the most influential predictors of calorie expenditure.
Conclusion: These findings support the potential utility of machine learning- based calorie burn estimation in low resource regions with the CaloModel showing a strong potential for estimating calorie burn using easily obtainable physiological and activity-related variables. However, further validation is required using independent datasets to improve robustness and generalizability to ensure safe integration into clinical and public health settings.
Article Details
Issue
Section
This work is licensed under a Creative Commons Attribution 4.0 International License.
How to Cite
References
1. World Health Organization. Obesity and overweight. 2023 [cited 2024 Mar 16]. Available from: https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight
2. World Health Organization. WHO acceleration plan to stop obesity. 2024. Available from: https://www.who.int/publications/i/item/9789240075634
3. Cercato C, Fonseca FA. Cardiovascular risk and obesity. Diabetol Metab Syndr. 2019; 11:74. doi:10.1186/s13098-019-0468-0
4. Tong H, Alnasser A, Alshahrani N, Bawaked R, AlAhmed R, Alsukait R, et al. The use of mobile technologies to promote physical activity and reduce sedentary behaviors in the Middle East and North Africa region: systematic review and meta-analysis. J Med Internet Res. 2024;26:e53651. doi:10.2196/53651
5. Tariq MU. Advanced wearable medical devices and their role in transformative remote health monitoring. In: Garcia M, Almeida R, editors. Transformative approaches to patient literacy and healthcare innovation. IGI Global; 2024. p. 308–326. doi:10.4018/979-8-3693-3661-8.ch015
6. Swahn MH, Gittner KB, Lyons MJ, Nielsen K, Mobley K, Culbreth R, et al. Advancing mHealth research in low-resource settings: insights and implementation challenges with wearable smartwatch devices. Sensors. 2024;24(17):5591. doi:10.3390/s24175591
7. Huang L, Huhulea EN, Abraham E, Bienenstock R, Aifuwa E, Hirani R, et al. The role of artificial intelligence in obesity risk prediction and management. Medicina. 2025;61(2):358. doi:10.3390/medicina61020358
8. Fabbrizio A, Fucarino A, Cantoia M, De Giorgio A, Garrido ND, Iuliano E, et al. Smart devices for health and wellness applied to tele-exercise: trends and technologies. Healthcare (Basel). 2023;11(12):1805. doi:10.3390/healthcare11121805
9. Ingels JS, Misra R, Stewart J, Lucke-Wold B, Shawley-Brzoska S. The effect of adherence to dietary tracking on weight loss. J Diabetes Res. 2017;2017:6951495. doi:10.1155/2017/6951495
10. Houlding E, Mate K, Engler K, Ortiz-Paredes D, Pomey M, Cox J, et al. Barriers to use of remote monitoring technologies in COVID-19. JMIR Mhealth Uhealth. 2021;9(4):e24743. doi:10.2196/24743
11. Trivedi S, Gupta J, Singh H. Machine learning-based estimation of caloric expenditure. In: Dev A, Sharma A, Agrawal SS, Rani R, editors. Artificial Intelligence and Speech Technology. Springer; 2025. doi:10.1007/978-3-031-75167-7_21
12. Gaikwad V, Khodke M, Shahare S, Terkar P, Talmale R. XGBoost regression for calories burnt prediction. In: Singh M, Tyagi V, Gupta P, Flusser J, Ören T, editors. Advances in Computing and Data Sciences. Springer; 2023. doi:10.1007/978-3-031-37940-6_31
13. Kaggle. Calories burnt prediction dataset. Available from: https://www.kaggle.com/datasets/ruchikakumbhar/calories-burnt-prediction
14. Khan MS, Umer H, Faruqe F. Artificial intelligence for low-income countries. Humanit Soc Sci Commun. 2024;11:1422. doi:10.1057/s41599-024-03947-w
15. Bonet Olivencia S, Rao AH, Smith A, Sasangohar F. Diabetes self-management applications for underserved populations. Int J Environ Res Public Health. 2022;19(1):127. doi:10.3390/ijerph19010127
16. Scott RE, Mars M. Telehealth in the developing world: current status and future prospects. Smart Homecare Technol Telehealth. 2015;3:25–37. doi:10.2147/SHTT.S75184
17. Zhou G, Bao K, Xiao H, et al. Measurement of energy expenditure by indirect calorimetry. Phenomics. 2024;4:203–212. doi:10.1007/s43657-023-00127-9
18. Rodriguez C, Harty PS, Stratton MT, et al. Comparison of indirect calorimetry and prediction equations. J Strength Cond Res. 2022;36(11):3093–3104
19. Carter L, Ford CD. Promoting physical activity in clinical practice through wearable technology. J Am Assoc Nurse Pract. 2023 Dec 1;35(12):765-769. doi: 10.1097/JXX.0000000000000892.
20. Delsoglio M, Achamrah N, Berger MM, Pichard C. Indirect calorimetry in clinical practice. J Clin Med. 2019;8(9):1387. doi:10.3390/jcm8091387
21. Sempionatto JR, Montiel VR, Vargas E, Teymourian H, Wang J. Wearable and Mobile Sensors for Personalized Nutrition. ACS Sens. 2021 May 28;6(5):1745-1760. doi: 10.1021/acssensors.1c00553.
22. Levikari S, et al. Improving energy expenditure estimation in wrist-worn wearables. IEEE Trans Instrum Meas. 2021;70:1–8. doi:10.1109/TIM.2021.3053070
23. Mahler H. The Meaning of "Health for All by the Year 2000". Am J Public Health. 2016 Jan;106(1):36-8. doi: 10.2105/AJPH.2016.106136.
24. Kaushal S, Tammineni DK, Rana P, Sharma M, Sridharet K, Chen H-H. Computer vision and deep learning-based approaches for detection of food nutrients/nutrition: New insights and advances. Trends Food Sci Technol. 2024;146:104408. https://doi.org/10.1016/j.tifs.2024.104408.
25. Kumar RD, Julie EG, Robinson YH, et al. Recognition of food type and calorie estimation using neural network. J Supercomput. 2021;77(8):8172–8193. doi:10.1007/s11227-021-03622-w.
26. Hussain G, Ali Saleh Al-rimy B, Hussain S, Albarrak AM, Qasem SN, Ali Z. Smart Piezoelectric-Based Wearable System for Calorie Intake Estimation Using Machine Learning. Applied Sciences. 2022; 12(12):6135. https://doi.org/10.3390/app12126135
27. Aziz MT, Rahman T, Pecho RDC, et al. Calories burnt prediction using ML. Curr Integr Eng. 2023;1(1):29–36
28. Rauf RI, Alrasheedi MA, Sadiq R, Aldawsari AMA. Evaluating Predictive Accuracy of Regression Models with First-Order Autoregressive Disturbances: A Comparative Approach Using Artificial Neural Networks and Classical Estimators. Mathematics. 2024; 12(24):3966. https://doi.org/10.3390/math12243966
29. Botchkarev A. Evaluating regression ML models using multiple error metrics. SSRN. 2018. doi:10.2139/ssrn.3177507
30. Serão N, Petry A. 208 Do prediction equations with high R-squared in low sample size studies have high predictive ability? J Anim Sci. 2020 Nov 30;98(Suppl 3):71. doi: 10.1093/jas/skaa054.127.
31. Kang KH, Kang M, Shin S, Jung J, Li M. Estimating Physical Activity Energy Expenditure Using an Ensemble Model-Based Patch-Type Sensor Module. Electronics. 2021; 10(7):861. https://doi.org/10.3390/electronics10070861
32. Yan Y, Chen Q. Energy Expenditure Estimation of Tabata by Combining Acceleration and Heart Rate. Front Public Health. 2022 Feb 7;9:804471. doi: 10.3389/fpubh.2021.804471.
33. Kang KH, Shin SH, Jung J, Kim YT. Estimation of a physical activity energy expenditure with a patch-type sensor module using artificial neural network. Concurr Comput Pract Exp. 2019;33(2): e5455. doi:10.1002/cpe.5455
34. Tharani R, Jalaja S, Kaviya R, et al. Prediction of Calorie Burn and Suggestion of Diet Plan Using Machine Learning. CRC Press; 2026. p. 616–620
35. Prasad A, Asha V, Vasumathi MT, Gupta A. Comparing gradient boosting and linear models for calorie prediction. In: Proceedings of the 3rd International Conference on Disruptive Technologies (ICDT); 2025. IEEE. DOI:10.1109/ICDT63985.2025.10986394
36. Justin J, Tan KR, Edbert IS, Suhartono D. Predictive modelling on caloric expenditure: exploring machine learning techniques for enhanced analysis. In: Proceedings of the IEEE International Conference on Control, Automation, Electronics, Robotics, Internet of Things, and Artificial Intelligence (CERIA); 2024. IEEE. DOI:10.1109/CERIA64726.2024.10914994