Name
Abstract | ||
|---|---|---|
Data analytics represents a new chance for medical diagnosis and treatment to make it more effective and successful. This expectation is not so easy to achieve as it may look like at a first glance. The medical experts, doctors or general practitioners have their own vocabulary, they use specific terms and type of speaking. On the other side, data analysts have to understand the task and to select the right algorithms. The applicability of the results depends on the effectiveness of the interactions between those two worlds. This paper presents our experiences with various medical data samples in form of SWOT analysis. We identified the most important input attributes for the target diagnosis or extracted decision rules and analysed their interestingness with cooperating doctors, for most promising new cutoff values or an investigation of possible important relations hidden in data sample. In general, this type of knowledge can be used for clinical decision support, but it has to be evaluated on different samples, conditions and ideally in long-term studies. Sometimes, the interaction needed much more time than we expected at the beginning but our experiences are mostly positive. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.3233/978-1-61499-852-5-371 | Studies in Health Technology and Informatics |
Keywords | Field | DocType |
medical data,diagnosis,variables,knowledge | Data science,Data analysis,Knowledge management,Health informatics,Medicine | Conference |
Volume | ISSN | Citations |
247 | 0926-9630 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Frantisek Babic | 1 | 16 | 8.02 |
| Michal Vadovský | 2 | 0 | 0.68 |
| Jan Paralic | 3 | 56 | 13.96 |