BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Predicting recurrence of prostate cancer: a Bayesian approach - Ro
 ger Sewell
DTSTART:20250410T181500Z
DTEND:20250410T203000Z
UID:TALK216745@talks.cam.ac.uk
CONTACT:Peter Watson
DESCRIPTION:We establish the extent to which predictions of recurrence of 
 prostate cancer (relapse) taken using preoperative biomarkers could be imp
 roved upon using Bayesian methodology. We analyse thedataset of Shariat et
  al to compare the improvement in prediction of relapse times using biomar
 kers with models which omitthem. Using half the dataset for training and t
 he other half for testing\, predictions of relapse time by a Bayesian appr
 oach using a skew-Student mixture model are compared to those using the tr
 aditional Cox model. The predictions from the Bayesian model are found to 
 outperform those of the Cox model but the overall yield of predictive info
 rmation leaves plenty of scope for improvement in the range of biomarkers 
 in use. The Bayesian model presented here is the first such model for pros
 tate cancer to consider the variation of relapse hazard with biomarker con
 centrations to be smooth\, as is intuitively believable. It is also the fi
 rst model to be shown to provide improved quality  of prediction over the 
 Cox model and indeed the first to be shown to provide positive apparent Sh
 annon information relative to an exponential prior.
LOCATION:MRC Cognition and Brain Sciences Unit\, 15 Chaucer Road\, Cambrid
 ge CB2 7EF
END:VEVENT
END:VCALENDAR