Project: Dynamic Bayesian Networks VS Bayesian Networks: application in the prediction of prostate cancer

Investigators: Ana Sarabando at Universidade do Porto. Created at: 2010-03-24 13:33. Last updated at: 2010-03-24 13:33.

Description:

Databases are expanding with medical knowledge, and there’s the urgent need to explore them. Mortality in cancer is rising all over the world, it’s necessary to take over the medical knowledge in databases to fight against this mortality and improve health care.
Prostate cancer is the second leading cause of mortality in men and the cause(s) of this form of cancer remain to be elucidated.
Prognostic models in medicine are useful to apply in the prediction of cancer survivability in order to collaborate in medical decision throughout the process of care. It also can help finding the major causes of prostate cancer when we explore data of patients with this cancer.
Dynamic Bayesian Networks, in contrast with static Bayesian Networks, are prognostic models that allow for the incorporation of the causal and temporal nature of medical domain knowledge, thereby allowing for detailed prognostic predictions.
In this work we focus our attention on Dynamic Bayesian Networks and in its advantages when used in the prediction of prostate cancer survivability, and in the advantage for the treatment of prostate cancer in using Bayesian Networks to explore the world of databases.

Status: Approved

Operations:

If you are the owner of this project, please Login with your account to access more operations.

Analyses: Refresh