BIG DATA AND ARTIFICIAL INTELLIGENCE
Patient-level data to understand patients' care
We have a long history of using large national datasets to understand and improve clinical care. In particular, we have experience in longitudinal data analysis along clinical pathways and across multiple treatment pathways. We lead the GlioCova project which gives a national view of brain tumour care, variation, costs and outcomes.
We have developed a range of imaging-based AI tools to help analyse clinical data at scale. Our focus is on useful, robust applications of AI in clinical care. Recent examples include our development of an end-to-end pipeline to determine muscle mass in brain tumor patients, and showing that this predicts patient survival, and scalable approaches to clustering and predicting outcomes in patients.
We are now applying this to using AI to improve clinical care pathways.
To learn more, visit the websites below:
GLIOCOVA: TREATMENT AND HOSPITAL ADMISSIONS FOR PATIENTS WITH GBM IN ENGLAND
Radvile Mauricaite, Kerlann Le Calvez, Matthew Williams
P14.27 EXPLORING END-OF-LIFE CARE IN THE GLIOCOVA NATIONAL BRAIN TUMOUR PATIENT COHORT
R Mauricaite, K Le Calvez, J Droney, M Caldano, M Alam, M Williams
OS14.6.A GLIOCOVA: DEFINING PATIENT SAFETY EVENTS FOR BRAIN TUMOUR PATIENTS UNDERGOING NEUROSURGERY
R Mauricaite, K Le Calvez, A Brodbelt, A Bottle, M Williams
P14.18 PATIENT AND PUBLIC INVOLVEMENT TO DEFINE PATIENT-CENTRED OUTCOMES FROM NATIONAL CANCER DATASETS
L Pakzad-Shahabi, C Cherrington, N Brassil, P Even, D Gardner, W Fulcher, K Le Calvez, R Mauricaite, M Williams