en-es
en
en-zh
en-fr
en-pt
en-de
en-sl
0.25
0.5
0.75
1.25
1.5
1.75
2
Dynamic prediction of survival with clinical and genomic data
Published on 2011-10-244545 Views
An important clinical application of biostatistics is the development of statistical models for the prognosis of a patient at the moment of diagnosis. In cancer the usual way of giving a prognosis is
Related categories
Presentation
Dynamic prediction of survival with clinical and genomic data00:00
Summary01:54
Talk based on ...06:27
Crash-course survival analysis07:02
Prediction model08:08
The data are from ... (1)10:35
The data are from ... (2)13:00
Re-analyzed in ...13:46
Information on survival and censoring14:35
Clinical information15:33
Genomic information16:50
Major problem: How to handle so many predictors?18:42
Penalized Cox regression using genomic data21:42
Results23:35
Relation between Ridge and Lasso24:14
It is hard to see the difference25:40
However26:15
Genomic versus clinical predictor28:13
Dynamic prediction based on Landmarking31:20
Supermodel33:50
5-year prediction34:47
Could we do better?37:12
Degenerates for clinical predictor40:35
Conclusion/discussion40:56
References (1)46:15
References (2)47:19