Medical statistics was a key part of the response to the COVID-19 pandemic. Like many teams, we applied our experience with data analysis, real-world evidence and international collaboration to help healthcare systems around the world respond to the health crisis, and learn from it.
\n \n\n \n \nOur lab applies machine learning in medical statistics, allowing us to identify patterns and insights in large amounts of health data. This can lead to improved diagnoses, treatments, and overall healthcare outcomes.
\n \n\n \n \nAs artificial intelligence continues to grow in its power and applications, we are harnessing its power to improve healthcare, making it fair and safe for all.
\n \n\n \n \nSara is an Associate Professor of Health Informatics and Biomedical Data Science within the Centre for Statistics in Medicine. She trained in electrical and biomedical engineering prior to founding the PHI Lab, which studies data science and artificial intelligence for planetary health.
\n \n\n \n \nUsing satellite data, including vegetation levels, nighttime lights, rainfall and temperature, connected to malaria levels in South Asia to improve malaria prediction models.
\n \n\n \n \nUsing de-identified data from fitness trackers connected with localised weather data to study the impact of extreme temperatures on sleep and activity.
\n \n\n \n \nUsing satellite data to track the humanitarian impacts of flooding, particularly on schools, hospitals and roads.
\n \n\n \n \nMeet the faces behind the PHI Team! We'll be adding interviews as we go, so check back to hear more about other team members!
\n \n\n \n \nHear about our research from other outlets, like the Central University's news and our collaborators.
\n \n\n \n \nHear our researchers talk about the research we do and why it matters.
\n \n\n \n \nThis project was co-designed and delivered alongside patient partners, with the aim of ensuring that the research we produced was valuable for patients and the public, with partners helping to guide us at each step of the way.
\n \n\n \n \nAnalysing potential biases in health-specific Large Language Models (LLMs) for global use, trained on real-world patient data.
\n \n\n \n \nAssessing how comorbidities differ between patients with rare cardiometabolic conditions versus common cardiometabolic diseases, looking for sub-groups and how these groups were impacted by the pandemic.
\n \n\n \n \nDetecting clusters of comorbidities in patients diagnosed with osteoarthritis, and how these clusters impact patients' 10-year mortality.
\n \n\n \n \n