The increasing quantity of data in biomedical informatics is leading towards better patient profiling and personalized medicine. Lab tests, medical images, and clinical data represent extraordinary sources for patient characterization. While retrospective studies focus on finding correlations in this sheer volume of data, potential new biomarkers are difficult to identify. A common approach is to observe patient mortality with respect to different clinical variables in what is called survival analysis. Kaplan-Meier plots, also known as survival curves, are generally used to examine patient survival in retrospective and prognostic studies. The plot is very intuitive and hence very popular in the medical domain to disclose evidence of poor or good prognosis. However, the Kaplan-Meier plots are mostly static and the data exploration of the plotted cohorts can be performed only with additional analysis. There is a need to make survival plots interactive and to integrate potential prognostic data that may reveal correlations with disease progression. We introduce SurviVIS, a visual analytics approach for interactive survival analysis and data integration on Kaplan-Meier plots. We demonstrate our work on a melanoma dataset and in the perspective of a potential use case in precision imaging.
|Number of pages||5|
|Publication status||Published - 2019|
|Event||10th International EuroVis Workshop on Visual Analytics - Porto, Portugal|
Duration: 3 Jun 2019 → 3 Jun 2019
|Conference||10th International EuroVis Workshop on Visual Analytics|
|Abbreviated title||EuroVA 2019|
|Period||3/06/19 → 3/06/19|