2016/08/30 by Santosh Tirunagari, Simon Bull, Tirunagari, Santosh +7
Computer Science · #Data Stream Mining Techniques #Time Series Analysis and Forecasting
paper · pdf · doi:10.48550/arxiv.1609.05716
Due to the chronic nature of diabetes, patient self-care factors play an\nimportant role in any treatment plan. In order to understand the behaviour of\npatients in response to medical advice on self-care, clinicians often conduct\ncross-sectional surveys. When analysing the survey data, statistical machine\nlearning methods can potentially provide additional insight into the data\neither through deeper understanding of the patterns present or making\ninformation available to clinicians in an intuitive manner. In this study, we\nuse self-organising maps (SOMs) to visualise the responses of patients who\nshare similar responses to survey questions, with the goal of helping\nclinicians understand how patients are managing their treatment and where\naction should be taken. The principle behavioural patterns revealed through\nthis are that: patients who take the correct dose of insulin also tend to take\ntheir injections at the correct time, patients who eat on time also tend to\ncorrectly manage their food portions and patients who check their blood glucose\nwith a monitor also tend to adjust their insulin dosage and carry snacks to\ncounter low blood glucose. The identification of these positive behavioural\npatterns can also help to inform treatment by exploiting their negative\ncorollaries.\n