Workshop #1 : Knowledge Representation for Health Care and Process-oriented Information Systems in Healthcare (KR4HC’13 / ProHealth’13)
Workshop #2: Agents Applied in Health Care
Tutorial #1 : An Introduction to Agent-Based Modeling
Tutorial #2 : Bayesian Networks in Computational Neuroscience
Tutorial #3 : Evaluating Prognostic models in Medicine
Healthcare organizations are facing the challenge of delivering high quality services at affordable costs. These challenges become more prominent with the growth in the aging population with chronic diseases and the rise of healthcare costs. High degree of specialization of medical disciplines, huge amounts of medical knowledge and patient data, and the need for personalized healthcare are prevalent trends in this information-intensive domain. The emerging situation necessitates computer-based support of healthcare process & knowledge management as well as clinical decision-making.
To address these challenges, this workshop brings together researchers from two communities. The knowledge-representation for healthcare community has been focusing on knowledge representation and reasoning to support knowledge management and clinical decision-making. This community has been developing efficient representations, technologies, and tools for integrating all the important elements that health care providers work with: Electronic Medical Records (EMRs) and healthcare information systems, clinical practice guidelines, and standardized medical vocabularies. The process-oriented information systems in healthcare community has been studying ways to adopt business process technology in order to provide effective solutions for healthcare process management. Their challenges include supporting high degree of flexibility, integration with EMRs, and the need for tight cooperation and communication among medical care teams.
This joint workshop shall elaborate both the potential and the limitations of the two approaches for supporting healthcare process & healthcare knowledge management as well as clinical decision-making. It shall further provide a forum wherein challenges, paradigms, and tools for optimized knowledge-based clinical process support can be debated.
Call for papers: PDF
Intelligent agent-based systems constitute one of the most exciting research areas in Artificial Intelligence. Due to the growing interest in the application of agent-based systems, a number of applications addressing clinical problems are already based in agent technology. The 8th workshop on Agents Applied in Health Care provides a forum to the specialists in the field to meet and report on the results achieved in this area, to discuss the benefits (and drawbacks) that agent-based systems may bring to the medical domain, and also to provide a list of the research topics that should be tackled in the near future. The organising committee will pay special attention to papers describing applications which are not just academic, but that are already deployed and running in a real medical environment.
Current topics of research in this area include, among others, agent-based medical decision support systems, personalized health systems for remote and autonomous tele-assistance, communication and co-operation between distributed intelligent agents to manage patient care, information agents that retrieve medical information from distributed repositories, intelligent and distributed health care data mining, agent-based medical tutoring systems and multi-agent systems that assist medical professionals in the tasks of monitoring and diagnosis.
- John H. Holmes (firstname.lastname@example.org)
- Center for Clinical Epidemiology and Biostatistics
- Perelman School of Medicine University of Pennsylvania
- Philadelphia, PA, USA
Agent-based models offer an approach to simulation that is at once mathematical and social, and as such they offer a unique way to
view and model the behavior of intelligent agents over a range of investigator-created environmental and agent conditions. This tutorial will introduce
attendees to the concepts of agent-based modeling, including identifying suitable problems for modeling with agents, creating simulated environments,
imbuing agents with intelligence. parameterization, running temporal models, evaluating agent behavior, and applying agent-based models to a variety of
biomedical domains, such as infectious disease, violence, immunology, and social networks. Throughout the tutorial, attendees will gain practical experience in
learning the concepts through developing and implementing an agent-based model using NetLogo, a well-known and intuitive modeling package that is freely
available online http://ccl.northwestern.edu/netlogo/. Level: 50% basic, 50% intermediate. Attendees are not expected to have any prior experience in NetLogo or in
- Pedro Larranaga (email@example.com) and Concha Bielza
- Computational Intelligence Group
- Department of Artificial Intelligence
- School of Computer Science
- Universidad Politécnica de Madrid
- Madrid, Spain
Neuroscience faces challenging problems that require new machine learning methods. In this tutorial we will present some real-world examples
where Bayesian network models fit after some adaptation. These problems include: (a) neuroanatomy issues, like modeling and simulation of dendritic trees,
inferring electrophysiological behavior from morphological neuron characteristics, and classifying neuron and spine types based on morphological features;
(b) neurodegenerative diseases, like predicting health-related quality of life in Parkinson's disease, classification of dementia stages in Parkinsonís disease and
searching for genetic biomarkers in Alzheimer's disease.
- Ameen Abu-Hanna (firstname.lastname@example.org) and Niels Peek
- Department of Medical Informatics
- Academic Medical Center
- University of Amsterdam
- Amsterdam, The Netherlands
The reliable prediction of outcomes from disease and disease treatment is becoming increasingly
important in the delivery and organization of health care, and has a wide range of applications in clinical care, biomedical research, and
healthcare management. The standard methodology for obtaining objective outcome predictions is to build a prognostic model from a set of
observed patient data and clinical outcomes, and to apply that model to data from new patients. This tutorial focuses on methodologies for
quantitative assessment of the performance of prognostic models. The key to quantitative evaluation is the use of reliable methods for obtaining
valid performance measures with well-defined characteristics. The tutorial will clarify the relevant methods and the relationship between them using
conceptual and mathematical frameworks. All methods are illustrated with real-world examples from the domains of cardiac surgery and intensive care medicine.
We will also specifically address a framework for how to develop, evaluate,and report on prognostic modelling. This framework will be illustrated by various published