Malnutrition is defined as a lack of adequate nutrition, leading to altered body composition and impaired physical and mental function; it includes, in all its forms, undernutrition (e.g. wasting, stunting, underweight), inadequate vitamins or minerals, overweight, obesity. A person with malnutrition experiences both diet-related non-communicable diseases and infectious diseases.
The World Health Organization (WHO) has prioritized healthy ageing from 2016 to 2030, addressing global malnutrition, particularly in older populations. Today, 1 in 10 people are over 65 years of age, and this is expected to increase to 1 in 6 by 2050. Malnutrition affects not only older people but also children, with WHO reporting that 148 million children under 5 suffered from stunting, 45 million from wasting and 37 million from overweight globally in 2023.
Treatment of malnutrition is multidisciplinary and can place a strain on health systems, leading to undiagnosed cases and economic consequences.
Artificial intelligence (AI) is emerging as a powerful tool in the fight against malnutrition, offering new ways to analyse data, create personalized interventions, and optimize health systems. Its potential spans multiple domains, from early identification of at-risk populations to improving food distribution and supporting public health initiatives.
One of the key roles that AI can play is in early diagnosis and risk assessment of malnutrition. By analysing large data sets, including medical, nutritional, socioeconomic, and environmental factors, AI can identify individuals or populations that may be vulnerable to malnutrition. In particular, machine learning models can recognize subtle indicators of malnutrition that traditional screening methods may miss; for example, AI systems can analyse growth patterns, nutrient intake, and biomarkers to predict malnutrition in children or the elderly, allowing for earlier and more targeted interventions.
In addition to early detection, AI can help in developing personalized nutritional interventions, recommending specific dietary changes or nutritional supplements, according to the person’s unique characteristics (e.g. weight, health conditions, and metabolic rate). This personalized approach is particularly valuable in populations where standardized solutions may not be effective. Besides, AI’s ability to integrate genetic and microbiome data further enhances its capacity to tailor nutritional care to the individual, optimizing treatment for malnutrition.
AI can also improve the efficiency of food distribution and resource allocation, addressing the root causes of malnutrition in regions most in need. By analysing factors such as poverty levels, climate, and agricultural trends, AI can predict where malnutrition is most likely to occur and drive governments or organizations in distributing food aid more effectively. AI can also track and optimize food supply chains, reducing waste and ensuring that nutritious food reaches vulnerable communities on time.
Another area where AI shows promise is in enhancing diagnostic tools for malnutrition. AI can automate and refine the analysis of health indicators commonly used in malnutrition assessments. By integrating more complex data, such as imaging or biometric information, AI can provide a more detailed understanding of a patient’s nutritional status. This not only improves diagnostic accuracy but also helps distinguish malnutrition from related conditions such as sarcopenia or cachexia. Additionally, AI and decision support systems (DSS) are emerging as tools to assist healthcare professionals, although clinical nutrition DSS remain underdeveloped. Despite slow progress, AI-enhanced DSS show promise for improving the diagnosis and treatment of malnutrition, with further research needed for practical implementation.
Remote monitoring and telemedicine are other areas where AI can have a significant impact, especially in underserved areas. AI-based systems, often integrated into mobile apps or wearable devices, can track an individual’s nutrient intake, physical activity, and health markers in real-time. This data allows healthcare providers to continuously monitor patients and intervene if necessary, ensuring that those at risk of malnutrition receive timely support.
In addition to individual care, AI can also support public health through predictive modelling. In fact, by analysing demographic, environmental, and health data, AI can anticipate future malnutrition trends and potential outbreaks, allowing policymakers to plan preventive measures, such as nutrition programs or agricultural initiatives, in advance. Furthermore, AI models can assess the impact of environmental changes, such as droughts or floods, on food security, helping to mitigate their effects on malnutrition.
AI also has the potential to accelerate nutrition research and inform public policy. It can rapidly process large amounts of scientific data to identify emerging trends in malnutrition and highlight successful interventions. Governments and non-governmental organizations can use this information to design evidence-based policies to reduce malnutrition. Additionally, AI can simulate the outcomes of various interventions, providing policymakers with valuable information to make informed decisions.
Finally, AI can play a role in increasing public awareness and nutritional literacy on malnutrition prevention through AI-based platforms, such as chatbots, mobile apps, or virtual assistants, that can provide personalized advice on healthy and sustainable eating and food choices.
References:
- Malnutrition – World Health Organization. Available at: https://www.who.int/news-room/fact-sheets/detail/malnutrition
- Janssen, S. M., Bouzembrak, Y., & Tekinerdogan, B. (2024). Artificial Intelligence in Malnutrition: A Systematic Literature Review. Advances in nutrition (Bethesda, Md.), 15(9), 100264. https://doi.org/10.1016/j.advnut.2024.100264
- An, R., & Wang, X. (2023). Artificial Intelligence Applications to Public Health Nutrition. Nutrients, 15(19), 4285. https://doi.org/10.3390/nu15194285
- Sharma, S., Rawal, R., & Shah, D. (2023). Addressing the challenges of AI-based telemedicine: Best practices and lessons learned. Journal of education and health promotion, 12, 338. https://doi.org/10.4103/jehp.jehp_402_23
