AI in Health

In this project, we analyse publications from the European Heart Journal using generative AI and map cardiovascular research to the International Classification of Diseases (ICD).

Our goal is to trace the prevalence of diseases and other health problems explored in cardiovascular research over the past decade and identify how different health issues have been identified as related topics to cardiovascular health concerns. 

This project examines the various ways AI is used in healthcare and shows how these technologies have developed over time. This exploratory study employs AI to collect information from literature on AI and health – resulting in a structured list of topics encompassing AI technologies in health research and practice.

The project’s outcome is a novel data capability that offers insights into cutting-edge AI health innovations, emerging digital health solutions, and the development of augmented intelligence in health.

This project is designed to support decision-making within complex health systems by delivering key information to health planners. MChart provides crucial insights into mental health service provision – offering valuable information about services, costs and effectiveness.

With its user-friendly interface, MChart will enable health planners to explore various scenarios and visualise the potential impacts of different service interventions. For example, a health planner in Canberra could use MChart to identify areas with high rates of loneliness and assess the availability of related mental health services.

This project leverages AI technology for curating information about health services and explores the applications of generative AI in expediting data collection and processing. 

Research publications aimed at understanding the various aspects of coronaviruses, particularly COVID-19, have significantly shaped our knowledge base. While the urgency to monitor COVID-19 in real-time has decreased, the continual influx of new research of monthly articles underscores the importance of systematic review and analysis to deepen our understanding of the pandemic's broad impact.

To explore research trends and innovations in this space, we developed a pipeline using natural language processing techniques. This pipeline systematically catalogues and synthesises the vast array of research articles – leading to the creation of a dataset with more than 800,000 articles from July 2002 to May 2024. This paper describes the content of this dataset and provides the necessary information to make this dataset accessible and reusable for future research.

Our approach aggregates and organises global research related to coronaviruses into thematic clusters such as vaccine development, public health strategies, infection mechanisms, mental health issues, and economic consequences. Also, we have leveraged the contribution of health experts to review and revise the dataset.