Corpus ID: 3113036

Mining Medical Data for Predictive and Sequential patterns : PKDD 2001

@inproceedings{JensenMiningMD,
  title={Mining Medical Data for Predictive and Sequential patterns : PKDD 2001},
  author={S. Jensen}
}
Data relating to patient information and medical exams connected with thrombosis attacks were analysed using SPSS Clementine data mining workbench. The abilit y to predict the onset and successful diagnosis of thrombosis is key to the intervention of the disease, and sequential patterns of symptoms and lab exams may indicate a trending from a pre-thrombosis to active thrombosis condition. In this report, predictive modelli ng, association rules and sequence detection were used to investigate… CONTINUE READING
18 Citations

Figures and Tables from this paper

Association Rule Mining and Medical Application: A Detailed Survey
  • 10
  • PDF
Discovering sequential patterns in a UK general practice database
  • 9
  • PDF
Inferring characteristic phenotypes via class association rule mining in the bone dysplasia domain
  • 12
  • PDF
The PKDD Discovery Challenges on Thrombosis Data
  • 2
  • PDF

References

7%, 0.619) ThrombosisFlag_Present < = diagnosis_SLE & diagnosis_APS (19: 4.4%, 0.526) ThrombosisFlag_Present < = diagnosis_SJS & diagnosis_APS (7: 1.6%, 0.571)
  • 7%, 0.619) ThrombosisFlag_Present < = diagnosis_SLE & diagnosis_APS (19: 4.4%, 0.526) ThrombosisFlag_Present < = diagnosis_SJS & diagnosis_APS (7: 1.6%, 0.571)