New paper: Identifying COPD sub-types using data-driven approaches in primary care population

A group of researchers led by Maria Pikoula from the Institute of Health Informatics, UCL and Jennifer Kathleen Quint from Imperial College London, have recently had their research on the topic of Chronic Obstructive Pulmonary Disease (COPD) subtypes published in BMC Medical Informatics and Decision Making.

The group sought to discover, describe and validate COPD subtypes using cluster analysis on data derived from electronic health records. They applied two unsupervised learning algorithms in 30,961 current and former smokers diagnosed with COPD. This enabled them to identify and characterize five COPD patient clusters with distinct patient characteristics. The subgroups have differing risk factors, comorbidities and prognosis.

These findings are in line with previous related studies and draw attention to anxiety and depression as important drivers of COPD in young, female patients.

Pikoula M, Quint J K, Nissen F, Hemingway H, Smeeth L, Denaxas S (2019). Identifying clinically important COPD sub-types using data-driven approaches in primary care population based electronic health records. BMC Medical Informatics and Decision Making 19:86. doi: 10.1186/s12911-019-0805-0

For more information on each of the researcher’s current work, please visit their university research portals.

Maria Pikoula: https://iris.ucl.ac.uk/iris/browse/profile?upi=MPIKO26

Jennifer Quint: https://www.imperial.ac.uk/people/j.quint

Francis Nissen: http://ehr.lshtm.ac.uk/team/francis-nissen/

Harry Hemingway: https://iris.ucl.ac.uk/iris/browse/profile?upi=HHEMI65

Liam Smeeth: https://www.lshtm.ac.uk/aboutus/people/smeeth.liam

Spiros Denaxas: https://iris.ucl.ac.uk/iris/browse/profile?upi=SDENA57