Mining temporal classes from time series data
Proceedings of the eleventh international conference on Information and knowledge management
Mining approximate top-k subspace anomalies in multi-dimensional time-series data
VLDB '07 Proceedings of the 33rd international conference on Very large data bases
Clustering Categorical Data Using Silhouette Coefficient as a Relocating Measure
ICCIMA '07 Proceedings of the International Conference on Computational Intelligence and Multimedia Applications (ICCIMA 2007) - Volume 02
Analysis and Visualization of Time Series Data from Consumer-Generated Media and News Archives
WI-IATW '07 Proceedings of the 2007 IEEE/WIC/ACM International Conferences on Web Intelligence and Intelligent Agent Technology - Workshops
Time series analysis of open-source software projects
Proceedings of the 47th Annual Southeast Regional Conference
Discrete wavelet transform-based time series analysis and mining
ACM Computing Surveys (CSUR)
On discovering moving clusters in spatio-temporal data
SSTD'05 Proceedings of the 9th international conference on Advances in Spatial and Temporal Databases
Hi-index | 0.00 |
This paper focuses on assessing the behavior of a patient over time periods for managing type 2 Diabetes. In some cases, patients with type 2 diabetes not only behave differently from other patients, but the severity of a given health problem varies even for an individual patient. We focus on understanding how and when patients differ from other patients. In addition, we also look at the diversity that exists within an individual patient especially over time-periods throughout the day. Our aim is to identify such time intervals when a patient may need more targeted care. Thus, for type 2 Diabetes we identify which time-periods exhibit a mismatch in terms of the blood glucose readings and the insulin doses. For instance, if the blood glucose readings fluctuate and the insulin doses are fixed it may indicate a poor management of the insulin doses and therefore a poor management of Diabetes. Based on such findings a number of factors can be taken into consideration when drawing out a care plan for example diet, lifestyle, and type of treatment, among others. Our study uses a data mining approach, particularly clustering to study the measurements in blood glucose and doses of regular insulin for a selected number of patients. We look at their behavior on an overall days' basis, which we refer to as large-scale binning. Additionally, we study their behavior at specified time intervals throughout the day, which we refer to as small-scale binning. Our findings indicate that we are clearly able to see the trends in blood glucose readings as compared to the insulin doses for different patients indicating a well managed or a poorly managed plan.