Study Finds Gaps Between AI Research and Infectious Disease Burden
16 Oct 2026
Artificial intelligence is becoming an important tool in infectious disease research, but research attention does not always match the needs of the regions hardest hit. A study by researchers from the Aerospace Information Research Institute of the Chinese Academy of Sciences (AIRCAS), in collaboration with the Chinese Center for Disease Control and Prevention, has found a persistent mismatch between the geographic distribution of the research attention and the burden of infectious diseases. Published in npj Digital Medicine, the study traces how the field has evolved over three decades.
Advances in machine learning, deep learning and large language models are transforming infectious disease research from parameter-driven approaches toward data-driven applications. These technologies are helping researchers forecast outbreaks, support diagnosis and track the spread of disease. However, it has been less clear how these research efforts are distributed across regions and diseases, and whether they reflect global public-health needs.
The researchers analyzed 9,026 publications from 1995 to 2026, tracking publication trends, international collaboration and changing research topics. Using GPT-4o-mini alongside rule-based methods, they identified study locations and disease categories in the literature. They then compared research attention with disability-adjusted life year (DALY) indicators, which measure healthy years of life lost to illness and premature death. Research attention referred to the places studied, rather than simply where the author worked.
The analysis identified three stages in the development of AI-related infectious disease research: early exploration, methodological expansion and rapid growth. China and the United States emerged as the two main hubs of international collaboration, while India, the Republic of Korea and Saudi Arabia have become increasingly active participants. New collaboration clusters have also emerged in Africa.
Research themes shifted from monitoring, classification and predicting disease toward using deep learning to support diagnosis and to understand pandemics. The COVID-19 pandemic accelerated work on natural language processing, and the analysis of social-media posts and public sentiment. These approaches expand AI's role to include tracking public concerns, informing communication during health crises and evaluating policy.
The geographic analysis revealed an uneven distribution of research attention. Research attention focused mainly on countries with strong research systems and computing infrastructure, while heavy infectious disease burdens were largely located in sub-Saharan Africa and parts of South and Southeast Asia.
This mismatch does not mean that research in lower-burden countries has little value elsewhere. Tools for diagnosis, outbreak forecasting and decision-support can benefit high-burden areas when they are shared, adapted to local capacity and put into practice. The key challenge is to build the partnerships and local capacity needed to bring those benefits to communities facing the greatest burdens.
By examining what studies investigate and which places they cover, the study goes beyond counting publications and mapping authors' affiliations. It offers a new approach for evaluating how closely AI research aligns with global public-health priorities.
Global distribution of infectious disease burden and research attention by disease category. (Image by AIRCAS)
Research News
Study Finds Gaps Between AI Research and Infectious Disease Burden
Artificial intelligence is becoming an important tool in infectious disease research, but research attention does not always match the needs of the regions hardest hit. A study by researchers from the Aerospace Information Research Institute of the Chinese Academy of Sciences (AIRCAS), in collaboration with the Chinese Center for Disease Control and Prevention, has found a persistent mismatch between the geographic distribution of the research attention and the burden of infectious diseases. Published in npj Digital Medicine, the study traces how the field has evolved over three decades.
Advances in machine learning, deep learning and large language models are transforming infectious disease research from parameter-driven approaches toward data-driven applications. These technologies are helping researchers forecast outbreaks, support diagnosis and track the spread of disease. However, it has been less clear how these research efforts are distributed across regions and diseases, and whether they reflect global public-health needs.
The researchers analyzed 9,026 publications from 1995 to 2026, tracking publication trends, international collaboration and changing research topics. Using GPT-4o-mini alongside rule-based methods, they identified study locations and disease categories in the literature. They then compared research attention with disability-adjusted life year (DALY) indicators, which measure healthy years of life lost to illness and premature death. Research attention referred to the places studied, rather than simply where the author worked.
The analysis identified three stages in the development of AI-related infectious disease research: early exploration, methodological expansion and rapid growth. China and the United States emerged as the two main hubs of international collaboration, while India, the Republic of Korea and Saudi Arabia have become increasingly active participants. New collaboration clusters have also emerged in Africa.
Research themes shifted from monitoring, classification and predicting disease toward using deep learning to support diagnosis and to understand pandemics. The COVID-19 pandemic accelerated work on natural language processing, and the analysis of social-media posts and public sentiment. These approaches expand AI's role to include tracking public concerns, informing communication during health crises and evaluating policy.
The geographic analysis revealed an uneven distribution of research attention. Research attention focused mainly on countries with strong research systems and computing infrastructure, while heavy infectious disease burdens were largely located in sub-Saharan Africa and parts of South and Southeast Asia.
This mismatch does not mean that research in lower-burden countries has little value elsewhere. Tools for diagnosis, outbreak forecasting and decision-support can benefit high-burden areas when they are shared, adapted to local capacity and put into practice. The key challenge is to build the partnerships and local capacity needed to bring those benefits to communities facing the greatest burdens.
By examining what studies investigate and which places they cover, the study goes beyond counting publications and mapping authors' affiliations. It offers a new approach for evaluating how closely AI research aligns with global public-health priorities.
Global distribution of infectious disease burden and research attention by disease category. (Image by AIRCAS)