Foundations of Biomedical Knowledge Representation

The development and use of computerized decision-support systems in the domain of breast cancer has the potential to facilitate the early detection of disease as well as spare healthy women unnecessary interventions. Despite encouraging trends, there is much room for improvement in the capabilities of such systems to further alleviate the burden of breast cancer. One of the main challenges that current systems face is integrating and translating multi-scale variables like patient risk factors and imaging features into complex management recommendations that would supplement and/or generalize similar activities provided by subspecialty-trained clinicians currently. In this chapter, we discuss the main types of knowledge—objectattribute, spatial, temporal and hierarchical—present in the domain of breast image analysis and their formal representation using two popular techniques from artificial intelligence—Bayesian networks and first-order logic. In particular, we demonstrate (i) the explicit representation of uncertain relationships between low-level image features and high-level image findings (e.g., mass, microcalcifications) by probability distributions in Bayesian networks, and (ii) the expressive power of logic to generally represent the dynamic number of objects in the domain. By concrete examples with patient data we show the practical application of both formalisms and their potential for use in decision-support systems.

[1]  Peter J. F. Lucas,et al.  Critiquing Knowledge Representation in Medical Image Interpretation Using Structure Learning , 2010, KR4HC.

[2]  Christopher E. Hann,et al.  A minimal model of lung mechanics and model-based markers for optimizing ventilator treatment in ARDS patients , 2009, Comput. Methods Programs Biomed..

[3]  Peter J. F. Lucas,et al.  Constructing Bayesian Networks for Linear Dynamic Systems , 2011 .

[4]  J. Krinsley,et al.  Severe hypoglycemia in critically ill patients: Risk factors and outcomes* , 2007, Critical care medicine.

[5]  G. V. Berghe,et al.  Intensive insulin therapy in critically ill patients. , 2001, The New England journal of medicine.

[6]  Stuart J. Russell,et al.  Modelling Glycaemia in ICU Patients - A Dynamic Bayesian Network Approach , 2010, BIOSIGNALS.

[7]  Christopher E. Hann,et al.  Model-based glycaemic control in critical care - A review of the state of the possible , 2006, Biomed. Signal Process. Control..

[8]  Paolo Barbini,et al.  A comparative analysis of predictive models of morbidity in intensive care unit after cardiac surgery – Part II: an illustrative example , 2007, BMC Medical Informatics Decis. Mak..

[9]  S Andreassen,et al.  DIAS--the diabetes advisory system: an outline of the system and the evaluation results obtained so far. , 1997, Computer methods and programs in biomedicine.

[10]  Rinaldo Bellomo,et al.  Variability of Blood Glucose Concentration and Short-term Mortality in Critically Ill Patients , 2006, Anesthesiology.

[11]  Thomas Desaive,et al.  Physiological modeling, tight glycemic control, and the ICU clinician: what are models and how can they affect practice? , 2011, Annals of intensive care.

[12]  L. Lind,et al.  Impaired glucose and lipid metabolism seen in intensive care patients is related to severity of illness and survival. , 1994, Clinical intensive care : international journal of critical & coronary care medicine.

[13]  Christopher E. Hann,et al.  Model-Based Insulin Sensitivity as a Sepsis Diagnostic in Critical Care , 2008, Journal of diabetes science and technology.

[14]  M. J. Rice,et al.  Challenges to Glycemic Measurement in the Perioperative and Critically Ill Patient: A Review , 2009, Journal of diabetes science and technology.

[15]  S. Akhtar,et al.  Creating a Perioperative Glycemic Control Program , 2011, Anesthesiology research and practice.

[16]  Johnny T. Ottesen,et al.  Applied Mathematical Models in Human Physiology , 2004 .

[17]  Stuart J. Russell,et al.  Why are DBNs sparse? , 2010, AISTATS.

[18]  R Bellazzi,et al.  Drug delivery optimization through Bayesian networks. , 1992, Proceedings. Symposium on Computer Applications in Medical Care.

[19]  Keiji Kanazawa,et al.  A model for reasoning about persistence and causation , 1989 .

[20]  Kevin B. Korb,et al.  Incorporating expert knowledge when learning Bayesian network structure: A medical case study , 2011, Artif. Intell. Medicine.

[21]  D. Kansagara,et al.  Intensive Insulin Therapy in Hospitalized Patients: A Systematic Review , 2011, Annals of Internal Medicine.

[22]  Michael G. Madden,et al.  Bayesian networks for mathematical models: Techniques for automatic construction and efficient inference , 2013, Int. J. Approx. Reason..

[23]  Christopher E. Hann,et al.  Tight glycemic control in critical care - The leading role of insulin sensitivity and patient variability: A review and model-based analysis , 2011, Comput. Methods Programs Biomed..

[24]  Brian P. Kavanagh,et al.  Glycemic Control in the ICU , 2010 .

[25]  Stuart J. Russell,et al.  Dynamic bayesian networks: representation, inference and learning , 2002 .

[26]  Marek J. Druzdzel,et al.  Learning Why Things Change: The Difference-Based Causality Learner , 2010, UAI.

[27]  R. Hovorka,et al.  A simulation model of glucose regulation in the critically ill , 2008, Physiological measurement.

[28]  Bart De Moor,et al.  A minimal model for glycemia control in critically ill patients , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[29]  Gil Alterovitz,et al.  An artificial intelligence tool to predict fluid requirement in the intensive care unit: a proof-of-concept study , 2008, Critical care.

[30]  Brenda G Fahy,et al.  Glucose control in the intensive care unit , 2009, Critical care medicine.

[31]  Michael G. Madden,et al.  Clinical Time Series Data Analysis Using Mathematical Models and DBNs , 2011, AIME.

[32]  Oliver Lundqvist,et al.  Numerical Methods for Ordinary Differential Equations , 2013, An Introduction to Numerical Methods and Analysis 3e.

[33]  A Hart,et al.  Evaluating black-boxes as medical decision aids: issues arising from a study of neural networks. , 1990, Medical informatics = Medecine et informatique.

[34]  Malene Højbjerre,et al.  A Bayesian Approach to Bergman's Minimal Model , 2003, AISTATS.

[35]  N. Britton.  Essential Mathematical Biology , 2004 .

[36]  Qiao Li,et al.  Artificial arterial blood pressure artifact models and an evaluation of a robust blood pressure and heart rate estimator , 2009, Biomedical engineering online.

[37]  Jürgen Hesser,et al.  Virtual Intensive Care Unit (ICU): Real-Time Simulation Environment Applying Hybrid Approach Using Dynamic Bayesian Networks and ODEs , 2009, MMVR.

[38]  Peter Norvig,et al.  Artificial Intelligence: A Modern Approach , 1995 .

[39]  Peter E. Kloeden.  Stochastic Differential Equations , 2011, International Encyclopedia of Statistical Science.

[40]  Johan Decruyenaere,et al.  A novel approach for prediction of tacrolimus blood concentration in liver transplantation patients in the intensive care unit through support vector regression , 2007, Critical care.

[41]  Peter J. F. Lucas,et al.  Bayesian networks in biomedicine and health-care , 2004, Artif. Intell. Medicine.

[42]  Alexander J. Hartemink,et al.  Non-stationary dynamic Bayesian networks , 2008, NIPS.

[43]  R W Morris.  Mathematical models of biological systems. , 1989, Reproductive toxicology.

[44]  A. Iserles.  A First Course in the Numerical Analysis of Differential Equations: Stiff equations , 2008 .

[45]  Brian Hutton,et al.  Reliability of point-of-care testing for glucose measurement in critically ill adults* , 2005, Critical care medicine.

[46]  Woochang Lee,et al.  Evaluation of the Blood Gas Analyzer GEM Premier 4000 , 2009 .

[47]  Jeffrey C. Lagarias,et al.  Convergence Properties of the Nelder-Mead Simplex Method in Low Dimensions , 1998, SIAM J. Optim..

[48]  R. Bergman,et al.  Physiologic evaluation of factors controlling glucose tolerance in man: measurement of insulin sensitivity and beta-cell glucose sensitivity from the response to intravenous glucose. , 1981, The Journal of clinical investigation.

[49]  Pedro M. Domingos.  Toward knowledge-rich data mining , 2007, Data Mining and Knowledge Discovery.

[50]  Kevin P. Murphy,et al.  Learning the Structure of Dynamic Probabilistic Networks , 1998, UAI.

[51]  H. Y. Chen,et al.  Prediction of tacrolimus blood levels by using the neural network with genetic algorithm in liver transplantation patients. , 1999, Therapeutic drug monitoring.

[52]  Moritz Diehl,et al.  Nonlinear model predictive control with moving horizon state and disturbance estimation - Application to the normalization of blood glucose in the critically ill , 2008 .

[53]  Giuseppe De Nicolao,et al.  Dynamic Probabilistic Networks for Modelling and Identifying Dynamic Systems: a MCMC Approach , 1997, Intell. Data Anal..

[54]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[55]  N. Gordon,et al.  Novel approach to nonlinear/non-Gaussian Bayesian state estimation , 1993 .

[56]  C. Hanson,et al.  Artificial intelligence applications in the intensive care unit , 2001, Critical care medicine.

[57]  Bart De Moor,et al.  Prediction Performance Comparison between three Intensive Care Unit Glucose Models , 2009 .

[58]  Peter J. F. Lucas,et al.  A dynamic Bayesian network for diagnosing ventilator-associated pneumonia in ICU patients , 2009, Expert Syst. Appl..

[59]  George C. Verghese,et al.  Bayesian Networks for Cardiovascular Monitoring , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[60]  Catherine G. Enright,et al.  A Probabilistic Framework Based on Mathematical Models with Application to Medical Data Streams , 2012 .

[61]  Stuart J. Russell,et al.  Probabilistic detection of short events, with application to critical care monitoring , 2008, NIPS.

[62]  Edward H. Shortliffe,et al.  Rule Based Expert Systems: The Mycin Experiments of the Stanford Heuristic Programming Project (The Addison-Wesley series in artificial intelligence) , 1984 .

[63]  Nir Friedman,et al.  Probabilistic Graphical Models - Principles and Techniques , 2009 .

[64]  Paolo Barbini,et al.  A comparative analysis of predictive models of morbidity in intensive care unit after cardiac surgery – Part I: model planning , 2007, BMC Medical Informatics Decis. Mak..