User requirements were gathered from home care clinicians to understand what parameters are critical in monitoring the health of home care clients. Once solicited, the most appropriate graphical user interface (GUI) features were determined in order to conduct a usability test and qualitative analysis of two GUI prototypes. A few key findings include the need to display data trends, client personal targets and alerts for emergent situations.
Visual data mining combines data visualization and data mining. Therefore, in order to populate the GUI with home care clients' data and support decision making, data mining was also explored. Two data mining techniques, a segmentation algorithm and a feed-forward neural network, were evaluated for their ability to detect trends from simulated data. Results indicate that the segmentation algorithm is more accurate with the given data sets but the network is more robust with varying levels of noise.