Applications of traditional machine learning and deep learning algorithms in obesity prediction or classification: a systematic review of comparative performance
8 affiliations
- 1Department of Nutritional Sciences, School of Nutritional Sciences and Food Technology, Kermanshah University of Medical Sciences, Kermanshah, Iran
- 2Drug Applied Research Center, Tabriz University of Medical Sciences, Tabriz, Iran
- 3Bio Environmental Health Hazards Research Center, Jiroft University of Medical Sciences, Jiroft, Iran
- 4Clinical Research Development Unit, Imam Khomeini Hospital, Jiroft University of Medical Sciences, Jiroft, Iran
- 5ECA College of Health Sciences, Education Centre of Australia, Parramatta, NSW Australia
- 6School of Human Nutrition, McGill University, Montréal, QC Canada
- 7School of Pharmacy, MCPHS University, Boston, MA USA
- 8Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA USA
Abstract
This systematic review evaluated traditional machine learning (TML) and deep learning (DL) approaches for obesity prediction in longitudinal studies and obesity classification in cross-sectional studies. PubMed, Scopus, and Web of Science were searched from inception to June 15, 2025. Studies applying these algorithms and reporting performance metrics were included. Seventeen studies (three longitudinal and fourteen cross-sectional) met the inclusion criteria. Only one of the 17 included studies performed independent external validation, and calibration reporting was extremely limited, highlighting substantial methodological barriers to clinical implementation. Longitudinal studies showed modest performance, with only one reporting discrimination (AUC = 0.81), limiting comparison between TML and DL for future risk prediction. In contrast, cross-sectional studies reported high apparent performance, with DL models reaching AUCs up to 0.979 and well-tuned TML models also performing strongly; however, many estimates were influenced by methodological limitations, including circular predictor–outcome relationships and incomplete performance reporting. Both approaches performed well for obesity classification, but evidence for clinically meaningful future obesity risk prediction remains limited. Longitudinal designs and independent external validation are needed to improve clinical relevance. Until such evidence is available, AI-based obesity models should be viewed as exploratory decision-support tools rather than routinely implemented instruments for clinical or public health risk assessment.
Related
- A Documentation and Comparative Study of the Traditional "Ba-zerān" Song-Ritual: Form, Performance, and Contextual Evolution
- Comparative evaluation of fractal analysis and canny edge-detection algorithms for the quantification and classification of bone trabecular patterns in cone-beam computed tomography
- “Comparative performance evaluation of ZnO-coconut oil nanofluids stabilized with PVP, SDS and EBT surfactants for sustainable applications”
- A Comparative Evaluation of Missing Data Imputation Techniques and Their Impact on Classification Performance in Healthcare
- Performance of GPT-4o in CT-based crista galli morphological classification: image-only versus morphometric dataset-driven approaches: a controlled comparative study
BibTeX
@article{jamali2026applications,
title = {{Applications of traditional machine learning and deep learning algorithms in obesity prediction or classification: a systematic review of comparative performance}},
author = {Mehrdad Jamali and Meysam Zarezadeh and Mohammad Vesal Bideshki and Mohamed Khalifa and Michelle Cavaleri and Ahmad Saedisomeolia and Rania A. Mekary},
journal = {NPJ Digital Medicine},
year = {2026},
month = jul,
doi = {10.1038/s41746-026-02986-8},
url = {https://doi.org/10.1038/s41746-026-02986-8},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Applications of traditional machine learning and deep learning algorithms in obesity prediction or classification: a systematic review of comparative performance
AU - Mehrdad Jamali
AU - Meysam Zarezadeh
AU - Mohammad Vesal Bideshki
AU - Mohamed Khalifa
AU - Michelle Cavaleri
AU - Ahmad Saedisomeolia
AU - Rania A. Mekary
PY - 2026
DA - 2026/07/11
JO - NPJ Digital Medicine
DO - 10.1038/s41746-026-02986-8
UR - https://doi.org/10.1038/s41746-026-02986-8
AB - This systematic review evaluated traditional machine learning (TML) and deep learning (DL) approaches for obesity prediction in longitudinal studies and obesity classification in cross-sectional studies. PubMed, Scopus, and Web of Science were searched from inception to June 15, 2025. Studies applying these algorithms and reporting performance metrics were included. Seventeen studies (three longitudinal and fourteen cross-sectional) met the inclusion criteria. Only one of the 17 included studies performed independent external validation, and calibration reporting was extremely limited, highlighting substantial methodological barriers to clinical implementation. Longitudinal studies showed modest performance, with only one reporting discrimination (AUC = 0.81), limiting comparison between TML and DL for future risk prediction. In contrast, cross-sectional studies reported high apparent performance, with DL models reaching AUCs up to 0.979 and well-tuned TML models also performing strongly; however, many estimates were influenced by methodological limitations, including circular predictor–outcome relationships and incomplete performance reporting. Both approaches performed well for obesity classification, but evidence for clinically meaningful future obesity risk prediction remains limited. Longitudinal designs and independent external validation are needed to improve clinical relevance. Until such evidence is available, AI-based obesity models should be viewed as exploratory decision-support tools rather than routinely implemented instruments for clinical or public health risk assessment.
SN - 2398-6352
ER -