Mobile phone-based pattern recognition and data analysis for patients with type 1 diabetes

Abstract

Background: Persons with type 1 diabetes who use electronic self-help tools, most commonly blood glucose meters, record a large amount of data about their personal condition. Mobile phones are powerful and ubiquitous computers that have a potential for data analysis, and the purpose of this study is to explore how self-gathered data can help users improve their blood glucose management.

Subjects and methods: Thirty patients with insulin-regulated type 1 diabetes were equipped with a mobile phone application for 3-6 months, recording blood glucose, insulin, dietary information, physical activity, and disease symptoms. The data were analyzed in terms of usage of the different modules and which data processing and visualization tools could be constructed to support the use of these data.

Results: Eighteen patients (denoted "adopters") recorded complete data for over 80 consecutive days, up to 247 days. Among those who withdrew or did not use the application extensively, the most common reasons given were outdated or difficult-to-use phone. Data analysis using period finding and scale-space trends was found to yield significant patterns for most adopters. Pattern recognition methods to predict low or high blood glucose were found to be performing poorly.

Conclusions: Minimally intrusive mobile applications enable users with type 1 diabetes to record data that can provide data-driven feedback to the user, potentially providing relevant insight into their disease.

Figures

FIG. 1.

Screenshots from the Few Touch Application. From left to right, (A) the main screen with all basic options, (B) recording of insulin dose with editable time (default current time), (C) Registration of food and drink into six categories, (D) overview list of last activities, with normal blood glucose in green, high glucose in yellow, low glucose in red, food or drink in blue, and insulin in white, and (E) graph with the last glucose measurements.

FIG. 2.

Usage statistics for all recruited patients who contributed any data. Each column is, from the bottom to the top, the number of recordings in the logs for blood glucose (n.BG), insulin, activity, and food, respectively. The patients classified as adopters are denoted with a dot at the low end of the column. The patients are grouped into three groups according to overall period of usage time T as measured from first to last entry in the Few Touch Application. Five users with no data are not shown. Color graphics available online at www.liebertonline.com/dia

FIG. 3.

Receiver operating characteristics plot for classification of low or high blood glucose using support vector machines with a radial kernel function. Gray dots indicate high glucose (80% quantile), black dots low glucose (20% quantile), and each dot corresponds to one patient. The diagonal line shows the random classifier, points closer to the upper left corner than the diagonal line mean improved classification over random guessing. Color graphics available online at www.liebertonline.com/dia

FIG. 4.

Significant scale-space trends for three selected patients. Gray dots are blood glucose measurements, the y-axis is cropped for visibility, and black vertical lines indicate measurements above the selected range. Black smooth lines are kernel regression functions using linear least squares regression with a quartic kernel function and bandwidths in hours as indicated in the individual panels. Red segments show a significant decreasing trend at the given scale, and blue segments indicate a significant increasing trend. The scales shown are chosen based on the full c-SiZer plots such that most significant trends are shown, although some details on the very smallest scales are omitted for clarity. Bandwidth is here defined as half of the support of the kernel.