Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Comparative Study
. 2013 Nov 16:2013:413-21.
eCollection 2013.

Decision path models for patient-specific modeling of patient outcomes

Antonio Ferreira  1 , Gregory F Cooper  1 , Shyam Visweswaran  1
Affiliations
Comparative Study

Decision path models for patient-specific modeling of patient outcomes

Antonio Ferreira et al. AMIA Annu Symp Proc. .

Abstract

Patient-specific models are constructed to take advantage of the particular features of the patient case of interest compared to commonly used population-wide models that are constructed to perform well on average on all cases. We introduce two patient-specific algorithms that are based on the decision tree paradigm. These algorithms construct a decision path specific for each patient of interest compared to a single population-wide decision tree with many paths that is applicable to all patients of interest that are constructed by standard algorithms. We applied the patient-specific algorithms to predict five different outcomes in clinical datasets. Compared to the population-wide CART decision tree the patient-specific decision path models had superior performance on area under the ROC curve (AUC) and had comparable performance on balanced accuracy. Our results provide support for patient-specific algorithms being a promising approach for predicting clinical outcomes.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Example of a population-wide decision tree and a patient-specific decision path constructed from a breast cancer dataset. Panel (a) represents a patient whose outcome we desire to predict and for whom the values of the nine predictors A1 to A9 are given by the vector as shown. Panel (b) shows the decision tree constructed by the CART algorithm and panel (c) shows the decision path constructed by the PSDP-BA algorithm. For the patient in (a) the decision tree predicts target value 0 via the path given by the solid arrows and the decision path also predicts 0. Note that the decision path is shorter and contains only two predictors while the path in the tree contains six predictors. Also, compared to the path in the decision tree, the patient-specific decision path predicts outcome 0 with higher confidence due to the larger sample size used in estimating the probability parameters.
Figure 2.
Figure 2.
Psuedocode for the patient-specific decision path using balanced accuracy (PSDP-BA) algorithm.
See this image and copyright information in PMC

References

    1. Abu-Hanna A, Lucas PJ. Prognostic models in medicine. AI and statistical approaches. Methods Inf Med. 2001;40(1):1–5. - PubMed
    1. Cooper GF, et al. Predicting dire outcomes of patients with community acquired pneumonia. J Biomed Inform. 2005;38(5):347–366. - PubMed
    1. Visweswaran S, et al. Learning patient-specific predictive models from clinical data. J Biomed Inform. 2010;43(5):669–685. - PMC - PubMed
    1. Visweswaran S, Cooper GF. Learning instance-specific predictive models. Journal of Machine Learning Research. 2010 Dec;11:3333–3369. - PMC - PubMed
    1. Visweswaran S. Learning patient-specific models from clinical data. Department of Biomedical Informatics [PhD dissertation] 2007. Available from: http://etd.library.pitt.edu/ETD/available/etd-11292007-232406/unrestrict.... - PMC - PubMed

Publication types

LinkOut - more resources