Skip to main content
Fig. 1 | BMC Medical Informatics and Decision Making

Fig. 1

From: A pattern-discovery-based outcome predictive tool integrated with clinical data repository: design and a case study on contrast related acute kidney injury

Fig. 1

Pattern discovery based visual analytics tool using the in-mode of knowledge incorporation: age in pattern changed by clinician on-the-fly. Note all the prediction metrics are for the training data. The left panel displays the pattern where the modified variable Age [AKI] highlighted in blue illustrates the clinician’s domain knowledge incorporation (in-mode). In the pattern, the prediction target (AKI-Yes) is shown at the top. Pattern variables were connected via arcs indicating statistical significances of Chi-square test of independence [13]. A click on a variable removes an attribute (dimming the blue vertical bar). A click on “Add attribute” shows a pop-up list of variables that could be added by users. The top right panel shows the training predictive metrics of the pattern once “Update pattern” is clicked. The bottom right pattern shows the pattern history summary, where the last pattern is generated in the pre-mode. A click on “Export results” exports the current pattern for post-mode refinement

Back to article page