0.25
0.5
0.75
1.25
1.5
1.75
2
Dynamic prediction of survival with clinical and genomic data
Published on Oct 24, 20114534 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
Chapter list
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