I am Chief Applied Scientist in Health AI Innovation at Oracle Health.
I am an academic data scientist and health services researcher by background and have dedicated my career to design, deploy, and evaluate computational techniques to make better decisions about health and healthcare to improve outcomes. To progress towards this goal I focus on two complementary areas of investigation : patient-centered data and predictive modeling.
My labs have shown that patient-reported outcomes measures (PROMs) can be used as a clinical intervention to improve quality of life, identification of need, and patient survival. We have consistently shown that it is possible to dramatically reduce the burden of PROMs without sacrificing assessment accuracy by leveraging computerised adaptive testing (e.g., Harrison and Gibbons, 2022). Work is underway to develop and validate agent-derived assessments (ADAs) from various machine learning models to supplement data collected using PROs.
Our award-winning collaborative research on the development of AI-enhanced screening techniques to identify complete pathological response (pCR) to neoadjuvant chemotherapy for localised breast cancer suggests that it may be possible to safely de-escalate confirmatory surgery in a large proportion of patients. (e.g., J Clin Oncol, Eur J Can).
We strive to develop effective, safe, and unbiased AI tools to inform clinical decision making to improve outcomes, reduce costs, and ensure that appropriate care is provided at the end-of-life. I collaborate with Project Ronin to demonstrate successful implementation of these tools in real world settings.
My current research interests are focused on the development of safe and effective tools built around large language models (LLMs) and AI agents for a wide range of use cases across healthcare. I am currently leading a new study to assess the use of LLMs to facilitate to assessment and reporting of cancer-related burden to create actionable insights for policymakers around the world as part of a Lancet Commission on Cancer and Health Systems.
I maintain links with the University of Cambridge as Director of Health Assessment and Innovation at the Psychometrics Centre and Health Director of the Concerto Platform. I was previously National Institute for Health Research (NIHR) Fellow based at The Healthcare Improvement Studies (THIS) Institute, and College Research Associate at Corpus Christi College, Cambridge. I then moved to Brigham and Women's Hospital and Harvard Medical School as a founding Co-Director of the Patient-Reported Outcomes, Value & Experience (PROVE) Center.
I have been invited to lecture on topics relating to data science and reported outcomes at institutions including NASA, Karolinska Institutet, and Harvard Business School. From 2014 until 2022 I delivered in-person training courses on modern psychometrics, big data, and machine learning at the University of Cambridge.
At MD Anderson, I was the founding Director of the Institute for Data Science in Oncology Fellowship program which seeks to endow post-doctoral and junior medical staff with the skills and critical insight necessary to become leaders in oncologic data science.