Optimal prediction of mortality after abdominal aortic aneurysm repair with statistical models Vassilis Georgiou Hadjianastassiou, MRCS, Leonardo Franco, PhD, Jose M. Jerez, PhD, Iordanis E. Evangelou, DPhil, David R. Goldhill, Paris P. Tekkis, Linda J. Hands Journal of Vascular Surgery Volume 43, Issue 3, Pages 467-473.e3 (March 2006) DOI: 10.1016/j.jvs.2005.11.022 Copyright © 2006 The Society for Vascular Surgery Terms and Conditions
Fig 1 Calibration chart for the predictive models as applied to the validation set of observations. Bars represent the observed and predicted in-hospital percent mortality risk. Journal of Vascular Surgery 2006 43, 467-473.e3DOI: (10.1016/j.jvs.2005.11.022) Copyright © 2006 The Society for Vascular Surgery Terms and Conditions
Fig 2 Subgroup analysis by Chronic Health and operative urgency status for the predictive models (validation set). Error bars represent the 95% confidence interval of the in-hospital percent mortality risk. Journal of Vascular Surgery 2006 43, 467-473.e3DOI: (10.1016/j.jvs.2005.11.022) Copyright © 2006 The Society for Vascular Surgery Terms and Conditions
Appendix I Fig 1 Basic concepts of an Artificial Neural Network Journal of Vascular Surgery 2006 43, 467-473.e3DOI: (10.1016/j.jvs.2005.11.022) Copyright © 2006 The Society for Vascular Surgery Terms and Conditions
Appendix IV Fig 2 Calibration chart for the predictive models as derived from the development set of observations. Bars represent the observed and predicted in-hospital % mortality risk. Journal of Vascular Surgery 2006 43, 467-473.e3DOI: (10.1016/j.jvs.2005.11.022) Copyright © 2006 The Society for Vascular Surgery Terms and Conditions