Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
ChatGPT for obesity management: a review of evidence, potential challenges, and clinical implications
0
Zitationen
3
Autoren
2026
Jahr
Abstract
The global rise in the prevalence of obesity highlights the need for accessible and effective solutions for obesity management. ChatGPT, one of the fastest-growing artificial intelligence (AI) tools, is gaining attention for its potential to offer interactive support in managing complex conditions such as obesity. In this Review, we used a thematic synthesis approach to examine ChatGPT's effect on obesity management. We analysed studies from PubMed, Web of Science, and supplementary sources published between Dec 1, 2022, and Oct 31, 2025, focusing on the applications, effectiveness, and limitations of ChatGPT in obesity management, including its roles in patient support, education, behavioural guidance, and clinical decision making. 37 studies (29 original and eight review-based studies) were included in this Review. Among the original studies, ChatGPT showed high accuracy in nine (75%) of 12 obesity-related lifestyle and nutrition studies and five (50%) of ten bariatric surgery studies compared with expert or guideline standards. Compared with other AI tools (eg, DeepSeek, Copilot, Gemini, Bing, Bard, and DALL·E 3), ChatGPT showed higher overall accuracy; however, evidence from comparisons with specialised applications or algorithms remains scarce. Quality assessment ratings were mostly moderate, and only ten (27%) of the total 37 studies were rated as high confidence. Overall, ChatGPT offers opportunities in eight areas of obesity management: lifestyle support, user engagement, clinical empowerment, medication guidance, virtual assessment, surgical guidance, predictive modelling, and research assistance. However, its implementation is limited by challenges in six key domains: accuracy and reliability, algorithmic bias, cultural sensitivity, transparency and accountability, excessive dependence, and ethical and legal concerns. Addressing these challenges is essential before ChatGPT can be considered for clinical deployment to safeguard patient safety and optimise its effectiveness in obesity management.
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.646 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.554 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 8.071 Zit.
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
2019 · 6.851 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.781 Zit.