Abstract: This talk will introduce a pipeline that exploits recent developments in topological data analysis to identify homogeneous clusters in high-dimensional data. Our approach is based on Mapper, an algorithm that reduces a point cloud into a one-dimensional graph. Written in Python and freely available online, the pipeline offers several advantages over existing clustering techniques. These include the ability to integrate prior knowledge into the clustering process and selection of optimal clusters; the use of the bootstrap to restrict the search to robust topological features; the use of machine learning to inspect clusters; and the ability to incorporate mixed data types.

Minibio: Dr Ewan Carr is a BRC Research Fellow in the Department of Biostatistics and Health Informatics, King's College London. He completed his PhD in Social Statistics at the Cathie Marsh Centre for Census and Survey Research, University of Manchester, and was previously based in Epidemiology and Public Health, UCL. His research interests are varied, but centre around understanding the linkages between mental and physical health. Recent projects have included assessment of measurement overlap in a latent variable model; prediction models for severe COVID outcome at KCH; links between mental health and patient-physician discordance in psoriasis; feasibility trials of digital interventions for mental health; and biomedical applications of topological data analysis.

Publication