Value of information analysis for interventional and counterfactual Bayesian networks in forensic medical sciences

• Published in: Artificial intelligence in medicine Vol. 66; pp. 41 - 52
• Main Authors: Constantinou, Anthony Costa, Yet, Barbaros, Fenton, Norman, Neil, Martin, Marsh, William
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.01.2016
• Subjects: Bayes Theorem; Bayesian analysis; Bayesian networks; Causal inference; Causality; Choice Behavior; Computer Simulation; Counterfactual analysis; Decision making; Decision Support Techniques; Expert systems; Forensic computing; Forensic engineering; Forensic medicine; Forensic Medicine - methods; Gain; Humans; Information Storage and Retrieval - methods; Internal Medicine; Interventional analysis; Medical; Models, Statistical; Neural Networks (Computer); Other; Risk analysis; Risk Assessment; Risk Factors; Uncertainty; Value of Information; Violence - ethnology; Violence - prevention & control; Violence - psychology; Causal inference; Value of Information; Bayesian networks; Forensic medicine; Counterfactual analysis; Interventional analysis
• DOI: 10.1016/j.artmed.2015.09.002
• ISSN: 0933-3657; 1873-2860

•A method based on the underlying principle of Value of Information analysis.•For interventional and counterfactual Bayesian networks in forensic medical sciences.•Examines whether interventional decision making is subject to amendments.•Assesses what further information is worthwhile seeking prior to decision making. Inspired by real-world examples from the forensic medical sciences domain, we seek to determine whether a decision about an interventional action could be subject to amendments on the basis of some incomplete information within the model, and whether it would be worthwhile for the decision maker to seek further information prior to suggesting a decision. The method is based on the underlying principle of Value of Information to enhance decision analysis in interventional and counterfactual Bayesian networks. The method is applied to two real-world Bayesian network models (previously developed for decision support in forensic medical sciences) to examine the average gain in terms of both Value of Information (average relative gain ranging from 11.45% and 59.91%) and decision making (potential amendments in decision making ranging from 0% to 86.8%). We have shown how the method becomes useful for decision makers, not only when decision making is subject to amendments on the basis of some unknown risk factors, but also when it is not. Knowing that a decision outcome is independent of one or more unknown risk factors saves us from the trouble of seeking information about the particular set of risk factors. Further, we have also extended the assessment of this implication to the counterfactual case and demonstrated how answers about interventional actions are expected to change when some unknown factors become known, and how useful this becomes in forensic medical science.