Limits...
Personalized Modeling for Prediction with Decision-Path Models.

Visweswaran S, Ferreira A, Ribeiro GA, Oliveira AC, Cooper GF - PLoS ONE (2015)

Bottom Line: Deriving predictive models in medicine typically relies on a population approach where a single model is developed from a dataset of individuals.We introduce three personalized methods that derive personalized decision-path models.Two of the three personalized methods performed statistically significantly better on area under the ROC curve (AUC) and Brier skill score compared to CART.

View Article: PubMed Central - PubMed

Affiliation: Department of Biomedical Informatics, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of America; The Intelligent Systems Program, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of America.

ABSTRACT
Deriving predictive models in medicine typically relies on a population approach where a single model is developed from a dataset of individuals. In this paper we describe and evaluate a personalized approach in which we construct a new type of decision tree model called decision-path model that takes advantage of the particular features of a given person of interest. We introduce three personalized methods that derive personalized decision-path models. We compared the performance of these methods to that of Classification And Regression Tree (CART) that is a population decision tree to predict seven different outcomes in five medical datasets. Two of the three personalized methods performed statistically significantly better on area under the ROC curve (AUC) and Brier skill score compared to CART. The personalized approach of learning decision path models is a new approach for predictive modeling that can perform better than a population approach.

No MeSH data available.


Related in: MedlinePlus

An example population decision tree and a personalized decision path.Panel (a) gives the names of the 21 variables and panel (b) gives their values for a test (current) patient whose outcome we want to predict. Panel (c) shows a population decision tree (derived by CART) and the path used for performing inference, and panel (d) shows a personalized decision path (derived by the DP-BAY method that is described later) for the patient in (b).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4476684&req=5

pone.0131022.g001: An example population decision tree and a personalized decision path.Panel (a) gives the names of the 21 variables and panel (b) gives their values for a test (current) patient whose outcome we want to predict. Panel (c) shows a population decision tree (derived by CART) and the path used for performing inference, and panel (d) shows a personalized decision path (derived by the DP-BAY method that is described later) for the patient in (b).

Mentions: In some circumstances, it is possible to derive the same type of predictive model using either a population or a personalized approach. For example, a probabilistic decision tree is a population model that is popular in biomedicine [5, 6]. A tree consists of interior nodes that represent variables and leaf nodes that represent predictions for the target variable (such as a clinical outcome). A path from the root node of the tree to a leaf node represents conjunctions of features (variable values), and the leaf node at the end of the path represents a probability distribution over the target variable (see Fig 1). Decision trees can also be derived using a personalized approach; such a tree is a single path consisting of a conjunction of features that are present in the current person and a leaf node at the end of the path that represents a probability distribution over the target variable. We call such a model a decision-path model, which is derived specifically for the current person.


Personalized Modeling for Prediction with Decision-Path Models.

Visweswaran S, Ferreira A, Ribeiro GA, Oliveira AC, Cooper GF - PLoS ONE (2015)

An example population decision tree and a personalized decision path.Panel (a) gives the names of the 21 variables and panel (b) gives their values for a test (current) patient whose outcome we want to predict. Panel (c) shows a population decision tree (derived by CART) and the path used for performing inference, and panel (d) shows a personalized decision path (derived by the DP-BAY method that is described later) for the patient in (b).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4476684&req=5

pone.0131022.g001: An example population decision tree and a personalized decision path.Panel (a) gives the names of the 21 variables and panel (b) gives their values for a test (current) patient whose outcome we want to predict. Panel (c) shows a population decision tree (derived by CART) and the path used for performing inference, and panel (d) shows a personalized decision path (derived by the DP-BAY method that is described later) for the patient in (b).
Mentions: In some circumstances, it is possible to derive the same type of predictive model using either a population or a personalized approach. For example, a probabilistic decision tree is a population model that is popular in biomedicine [5, 6]. A tree consists of interior nodes that represent variables and leaf nodes that represent predictions for the target variable (such as a clinical outcome). A path from the root node of the tree to a leaf node represents conjunctions of features (variable values), and the leaf node at the end of the path represents a probability distribution over the target variable (see Fig 1). Decision trees can also be derived using a personalized approach; such a tree is a single path consisting of a conjunction of features that are present in the current person and a leaf node at the end of the path that represents a probability distribution over the target variable. We call such a model a decision-path model, which is derived specifically for the current person.

Bottom Line: Deriving predictive models in medicine typically relies on a population approach where a single model is developed from a dataset of individuals.We introduce three personalized methods that derive personalized decision-path models.Two of the three personalized methods performed statistically significantly better on area under the ROC curve (AUC) and Brier skill score compared to CART.

View Article: PubMed Central - PubMed

Affiliation: Department of Biomedical Informatics, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of America; The Intelligent Systems Program, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of America.

ABSTRACT
Deriving predictive models in medicine typically relies on a population approach where a single model is developed from a dataset of individuals. In this paper we describe and evaluate a personalized approach in which we construct a new type of decision tree model called decision-path model that takes advantage of the particular features of a given person of interest. We introduce three personalized methods that derive personalized decision-path models. We compared the performance of these methods to that of Classification And Regression Tree (CART) that is a population decision tree to predict seven different outcomes in five medical datasets. Two of the three personalized methods performed statistically significantly better on area under the ROC curve (AUC) and Brier skill score compared to CART. The personalized approach of learning decision path models is a new approach for predictive modeling that can perform better than a population approach.

No MeSH data available.


Related in: MedlinePlus