Confident interpretation of Bayesian decision tree ensembles for clinical applications
Fieldsend, Jonathan E.
Coats, Timothy J.
Krzanowski, Wojtek J.
Everson, Richard M.
Bailey, Trevor C.
IEEE Transactions on Information Technology in Biomedicine
Institute of Electrical and Electronics Engineers (IEEE)
Bayesian averaging (BA) over ensembles of decision models allows evaluation of the uncertainty of decisions that is of crucial importance for safety-critical applications such as medical diagnostics. The interpretability of the ensemble can also give useful information for experts responsible for making reliable decisions. For this reason, decision trees (DTs) are attractive decision models for experts. However, BA over such models makes an ensemble of DTs uninterpretable. In this paper, we present a new approach to probabilistic interpretation of Bayesian DT ensembles. This approach is based on the quantitative evaluation of uncertainty of the DTs, and allows experts to find a DT that provides a high predictive accuracy and confident outcomes. To make the BA over DTs feasible in our experiments, we use a Markov Chain Monte Carlo technique with a reversible jump extension. The results obtained from clinical data show that in terms of predictive accuracy, the proposed method outperforms the maximum a posteriori (MAP) method that has been suggested for interpretation of DT ensembles.
Copyright © 2007 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
Vol. 11 (3), pp 312-319
Place of publication
Showing items related by title, author, creator and subject.
Rose, DC; Sutherland, WJ; Parker, C; Lobley, M; Winter, M; Morris, C; Twining, S; Ffoulkes, C; Amano, T; Dicks, LV (Elsevier Masson, 2016-09-29)Decision support tools, usually considered to be software-based, may be an important part of the quest for evidence-based decision-making in agriculture to improve productivity and environmental outputs. These tools can ...
British pain clinic practitioners' recognition and use of the bio-psychosocial pain management model for patients when physical interventions are ineffective or inappropriate: results of a qualitative study Harding, G; Campbell, John; Parsons, S; Rahman, A; Underwood, M (BioMed Central, 2010-03-18)BACKGROUND: To explore how chronic musculoskeletal pain is managed in multidisciplinary pain clinics for patients for whom physical interventions are inappropriate or ineffective. METHODS: A qualitative study was undertaken ...
The Development of a Web-based Decision Support System for the Sustainable Management of Contaminated Land Bello-Dambatta, Aisha (University of ExeterCollege of Engineering, Mathematics and Physical Sciences, 2010-12-05)Land is a finite natural resource that is increasingly getting exhausted as a result of land contamination. Land is made up of soil and groundwater, both of which have many functions for which we depend on, including ...