Artificial Intelligence in Medicine RSS feed: Current Issue. Artificial Intelligence in Medicine publishes original articles from a wide variety of interdisciplinary perspectives concerning the theory and practice of artificial intelligence (AI) in medicine, human biology, and health care. Particular attention is given to: • AI-based clinical decision making • medical knowledge engineering • knowledge-based and agent-based systems • computational intelligence in bio- and clinical medicine • intelligent medical information systems • AI in medical education • intelligent devices and instruments • automated reasoning and metareasoning in medicine • methodological, philosophical, ethical, and social issues of AI in medicine AIIM features: • original research contributions • methodological reviews • survey papers • special issue articles • position papers • historical perspectives • editorials • guest editorials • letters to the editor • book reviews This journal is included in CITE - The Computational Intelligence Website ! http://www.aiimjournal.com/?rss=yesElsevier Inc.en © 2012 Published by Elsevier Inc. All rights reserved. Artificial Intelligence in Medicine0933-3657551May 2012 © 2012 Published by Elsevier Inc. All rights reserved. healthpermissions@elsevier.comhttp://www.aiimjournal.com/article/PIIS0933365712000371/abstract?rss=yesEditorial Board10.1016/S0933-3657(12)00037-1Artificial Intelligence in Medicine 55, 1 (2012)2012-05-01Artificial Intelligence in Medicine2012-05-01551S0933-3657(12)X0004-6CO2CO2http://www.aiimjournal.com/article/PIIS0933365711001503/abstract?rss=yesAbstract: Objective: Setting up clinical reports within hospital information systems makes it possible to record a variety of clinical presentations. Directed acyclic graphs (Dags) offer a useful way of representing causal relations in clinical problem domains and are at the core of many probabilistic models described in the medical literature, like Bayesian networks. However, medical practitioners are not usually trained to elicit Dag features. Part of the difficulty lies in the application of the concept of direct causality before selecting all the causal variables of interest for a specific patient. We designed an automated interview to tutor medical doctors in the development of Dags to represent their understanding of clinical reports.Methods: Medical notions were analyzed to find patterns in medical reasoning that can be followed by algorithms supporting the elicitation of causal Dags. Clinical relevance was defined to help formulate only relevant questions by driving an expert's attention towards variables causally related to nodes already inserted in the graph. Key procedural features of the proposed interview are described by four algorithms.Results: The automated interview comprises questions on medical notions, phrased in medical terms. The first elicitation session produces questions concerning the patient's chief complaints and the outcomes related to diseases serving as diagnostic hypotheses, their observable manifestations and risk factors. The second session focuses on questions that refine the initial causal paths by considering syndromes, dysfunctions, pathogenic anomalies, biases and effect modifiers. A case study concerning a gastro-enterological problem and one dealing with an infected patient illustrate the output produced by the algorithms, depending on the answers provided by the doctor.Conclusions: The proposed elicitation framework is characterized by strong consistency with medical background and by a progressive introduction of relevant medical topics. Revision and testing of the subjectively elicited Dag is performed by matching the collected answers with the evidence included in accepted sources of biomedical knowledge.Automated interviews on clinical case reports to elicit directed acyclic graphsDavide Luciani, Federico M. Stefanini10.1016/j.artmed.2011.11.007Artificial Intelligence in Medicine 55, 1 (2012)2012-02-10Artificial Intelligence in Medicine2012-02-10551S0933-3657(12)X0004-6Research Articles111http://www.aiimjournal.com/article/PIIS0933365711001400/abstract?rss=yesAbstract: Purpose: Clinicians rely on a large amount of medical knowledge when performing clinical work. In clinical environment, clinical organizations must exploit effective methods of seeking and recommending appropriate medical knowledge in order to help clinicians perform their work.Method: Aiming at supporting medical knowledge search more accurately and realistically, this paper proposes a collaboration-based medical knowledge recommendation approach. In particular, the proposed approach generates clinician trust profile based on the measure of trust factors implicitly from clinicians’ past rating behaviors on knowledge items. And then the generated clinician trust profile is incorporated into collaborative filtering techniques to improve the quality of medical knowledge recommendation, to solve the information-overload problem by suggesting knowledge items of interest to clinicians.Results: Two case studies are conducted at Zhejiang Huzhou Central Hospital of China. One case study is about the drug recommendation hold in the endocrinology department of the hospital. The experimental dataset records 16 clinicians’ drug prescribing tracks in six months. This case study shows a proof-of-concept of the proposed approach. The other case study addresses the problem of radiological computed tomography (CT)-scan report recommendation. In particular, 30 pieces of CT-scan examinational reports about cerebral hemorrhage patients are collected from electronic medical record systems of the hospital, and are evaluated and rated by 19 radiologists of the radiology department and 7 clinicians of the neurology department, respectively. This case study provides some confidence the proposed approach will scale up.Conclusion: The experimental results show that the proposed approach performs well in recommending medical knowledge items of interest to clinicians, which indicates that the proposed approach is feasible in clinical practice.Collaboration-based medical knowledge recommendationZhengxing Huang, Xudong Lu, Huilong Duan, Chenhui Zhao10.1016/j.artmed.2011.10.002Artificial Intelligence in Medicine 55, 1 (2012)2012-02-10Artificial Intelligence in Medicine2012-02-10551S0933-3657(12)X0004-6Research Articles1324http://www.aiimjournal.com/article/PIIS0933365711001461/abstract?rss=yesAbstract: Objectives: Prediction of prostate cancer pathological stage is an essential step in a patient's pathway. It determines the treatment that will be applied further. In current practice, urologists use the pathological stage predictions provided in Partin tables to support their decisions. However, Partin tables are based on logistic regression (LR) and built from US data. Our objective is to investigate a range of both predictive methods and of predictive variables for pathological stage prediction and assess them with respect to their predictive quality based on UK data.Methods and material: The latest version of Partin tables was applied to a large scale British dataset in order to measure their performances by mean of concordance index (c-index). The data was collected by the British Association of Urological Surgeons (BAUS) and gathered records from over 1700 patients treated with prostatectomy in 57 centers across UK. The original methodology was replicated using the BAUS dataset and evaluated using concordance index. In addition, a selection of classifiers, including, among others, LR, artificial neural networks and Bayesian networks (BNs) was applied to the same data and compared with each other using the area under the ROC curve (AUC). Subsets of the data were created in order to observe how classifiers perform with the inclusion of extra variables. Finally a local dataset prepared by the Aberdeen Royal Infirmary was used to study the effect on predictive performance of using different variables.Results: Partin tables have low predictive quality (c-index=0.602) when applied on UK data for comparison on patients with organ confined and extra prostatic extension conditions, patients at the two most frequently observed pathological stages. The use of replicate lookup tables built from British data shows an improvement in the classification, but the overall predictive quality remains low (c-index=0.610).Comparing a range of classifiers shows that BNs generally outperform other methods. Using the four variables from Partin tables, naive Bayes is the best classifier for the prediction of each class label (AUC=0.662 for OC). When two additional variables are added, the results of LR (0.675), artificial neural networks (0.656) and BN methods (0.679) are overall improved. BNs show higher AUCs than the other methods when the number of variables raisesConclusion: The predictive quality of Partin tables can be described as low to moderate on UK data. This means that following the predictions generated by Partin tables, many patients would received an inappropriate treatment, generally associated with a deterioration of their quality of life. In addition to demographic differences between UK and the original US population, the methodology and in particular LR present limitations. BN represents a promising alternative to LR from which prostate cancer staging can benefit. Heuristic search for structure learning and the inclusion of more variables are elements that further improve BN models quality.Machine learning for improved pathological staging of prostate cancer: A performance comparison on a range of classifiersOlivier Regnier-Coudert, John McCall, Robert Lothian, Thomas Lam, Sam McClinton, James N’Dow10.1016/j.artmed.2011.11.003Artificial Intelligence in Medicine 55, 1 (2012)2012-02-10Artificial Intelligence in Medicine2012-02-10551S0933-3657(12)X0004-6Research Articles2535http://www.aiimjournal.com/article/PIIS0933365711001497/abstract?rss=yesAbstract: Objectives: Many classification problems must deal with data that contains missing values. In such cases data imputation is critical. This paper evaluates the performance of several statistical and machine learning imputation methods, including our novel multiple imputation ensemble approach, using different datasets.Materials and methods: Several state-of-the-art approaches are compared using different datasets. Some state-of-the-art classifiers (including support vector machines and input decimated ensembles) are tested with several imputation methods. The novel approach proposed in this work is a multiple imputation method based on random subspace, where each missing value is calculated considering a different cluster of the data. We have used a fuzzy clustering approach for the clustering algorithm.Results: Our experiments have shown that the proposed multiple imputation approach based on clustering and a random subspace classifier outperforms several other state-of-the-art approaches. Using the Wilcoxon signed-rank test (reject the null hypothesis, level of significance 0.05) we have shown that the proposed best approach is outperformed by the classifier trained using the original data (i.e., without missing values) only when >20% of the data are missed. Moreover, we have shown that coupling an imputation method with our cluster based imputation we outperform the base method (level of significance ∼0.05).Conclusion: Starting from the assumptions that the feature set must be partially redundant and that the redundancy is distributed randomly over the feature set, we have proposed a method that works quite well even when a large percentage of the features is missing (≥30%). Our best approach is available (MATLAB code) at bias.csr.unibo.it/nanni/MI.rar.A classifier ensemble approach for the missing feature problemLoris Nanni, Alessandra Lumini, Sheryl Brahnam10.1016/j.artmed.2011.11.006Artificial Intelligence in Medicine 55, 1 (2012)2012-02-10Artificial Intelligence in Medicine2012-02-10551S0933-3657(12)X0004-6Research Articles3750http://www.aiimjournal.com/article/PIIS0933365712000176/abstract?rss=yesAbstract: Objectives: This paper presents a model independent feature selection approach to identify a small subset of marker genes.Methods and material: An evaluation measure, minimum expected cost of misclassification (MEMC), is used to estimate the discriminative power of a feature subset without building a model. The MECM measure is combined with sequential forward search for feature selection. This approach was applied to a breast cancer profiling problem, with the goal of identifying a small number of marker genes whose expression can be used to predict cancer molecular subtype (p53 gene status). Furthermore, the method was also applied to find a small set of single-nucleotide polymorphisms (SNPs) that can be used to predict molecular phenotype of a different type, namely alleles (genetic variants) of human leukocyte antigen genes that play an important roles in autoimmunity.Results: Two marker genes were identified based on p53 status, which achieved a p-value of 7.53×10−5 (vs. 6×10−4 with 32 genes identified by previous research) in survival analysis. Six SNP loci were identified that achieved a leave-one-out cross-validation accuracy of 92.8% (vs. 90.6% and 89.5% with 18 SNPs selected using χ2 statistics and information gain, respectively).Conclusion: The MECM-based feature selection approach is capable of identifying a smaller subset of market genes with comparable or even better performance than that obtained using conventional filter methods.Identifying a small set of marker genes using minimum expected cost of misclassificationSamuel H. Huang, Dengyao Mo, Jarek Meller, Michael Wagner10.1016/j.artmed.2012.01.004Artificial Intelligence in Medicine 55, 1 (2012)2012-03-05Artificial Intelligence in Medicine2012-03-05551S0933-3657(12)X0004-6Research Articles5159http://www.aiimjournal.com/article/PIIS0933365712000152/abstract?rss=yesAbstract: Objective: The objective of this study was to determine whether metabolic parameters derived from ex vivo analysis of tissue samples are predictive of biologic characteristics of recurrent low grade gliomas (LGGs). This was achieved by exploring the use of multivariate pattern recognition methods to generate statistical models of the metabolic characteristics of recurrent LGGs that correlate with aggressive biology and poor clinical outcome.Methods: Statistical models were constructed to distinguish between patients with recurrent gliomas that had undergone malignant transformation to a higher grade and those that remained grade 2. The pattern recognition methods explored in this paper include three filter-based feature selection methods (chi-square, gain ratio, and two-way conditional probability), a genetic search wrapper-based feature subset selection algorithm, and five classification algorithms (linear discriminant analysis, logistic regression, functional trees, support vector machines, and decision stump logit boost). The accuracy of each pattern recognition framework was evaluated using leave-one-out cross-validation and bootstrapping.Materials: The population studied included fifty-three patients with recurrent grade 2 gliomas. Among these patients, seven had tumors that transformed to grade 4, twenty-four had tumors that transformed to grade 3, and twenty-two had tumors that remained grade 2. Image-guided tissue samples were obtained from these patients using surgical navigation software. Part of each tissue sample was examined by a pathologist for histological features and for consistency with the tumor grade diagnosis. The other part of the tissue sample was analyzed with ex vivo nuclear magnetic resonance (NMR) spectroscopy.Results: Distinguishing between recurrent low grade gliomas that transformed to a higher grade and those that remained grade 2 was achieved with 96% accuracy, using areas of the ex vivo NMR spectrum corresponding to myoinositol, 2-hydroxyglutarate, hypo-taurine, choline, glycerophosphocholine, phosphocholine, glutathione, and lipid. Logistic regression and decision stump boosting models were able to distinguish between recurrent gliomas that transformed to a higher grade and those that did not with 100% training accuracy (95% confidence interval [93–100%]), 96% leave-one-out cross-validation accuracy (95% confidence interval [87–100%]), and 96% bootstrapping accuracy (95% confidence interval [95–97%]). Linear discriminant analysis, functional trees, and support vector machines were able to achieve leave-one-out cross-validation accuracy above 90% and bootstrapping accuracy above 85%. The three feature ranking methods were comparable in performance.Conclusions: This study demonstrates the feasibility of using quantitative pattern recognition methods for the analysis of metabolic data from brain tissue obtained during the surgical resection of gliomas. All pattern recognition techniques provided good diagnostic accuracies, though logistic regression and decision stump boosting slightly outperform the other classifiers. These methods identified biomarkers that can be used to detect malignant transformations in individual low grade gliomas, and can lead to a timely change in treatment for each patient.Identifying malignant transformations in recurrent low grade gliomas using high resolution magic angle spinning spectroscopyAlexandra Constantin, Adam Elkhaled, Llewellyn Jalbert, Radhika Srinivasan, Soonmee Cha, Susan M. Chang, Ruzena Bajcsy, Sarah J. Nelson10.1016/j.artmed.2012.01.002Artificial Intelligence in Medicine 55, 1 (2012)2012-03-05Artificial Intelligence in Medicine2012-03-05551S0933-3657(12)X0004-6Research Articles6170