Multimodal and Multidimensional Artificial Intelligence Technology in Obesity

Although the prevalence of obesity is increasing worldwide, related treatment remains a complex challenge that requires multidimensional approaches. Recent advancements in artificial intelligence (AI) have led to the development of multimodal methods capable of integrating diverse types of data. These AI approaches utilize both multimodal data integration and multidimensional feature representations, enabling personalized, data-driven strategies for obesity management. AI can support obesity management through applications such as risk prediction, clinical decision support systems, large language models, and digital therapeutics. Several studies have shown that these AI-based weight loss programs can achieve significant weight reduction and behavioral changes. These AI systems can induce behavioral modifications through continuous personalized feedback and improve accessibility for people in underserved areas. However, these AI technologies must address issues such as data privacy and security, transparency and accountability, and consider the potential widening health disparities between individuals who have access to AI technology and those who do not, as well as strategies for sustained user engagement. Conducting long-term clinical trials and evaluations of cost-effectiveness across diverse, large-scale populations would facilitate the effective application of AI in obesity management, ultimately contributing to improvements in public health.