In this paper the authors describe the identification of variable interaction networks based on the analysis of medical data. The main goal is to generate mathematical models for medical parameters... Show More

In this paper the authors describe the identification of variable interaction networks based on the analysis of medical data. The main goal is to generate mathematical models for medical parameters using other available parameters in this data set. For each variable the authors identify those features that are most relevant for modeling it; the relevance of a variable can in this context be defined via the frequency of its occurrence in models identified by evolutionary machine learning methods or via the decrease in modeling quality after removing it from the data set. Several data based modeling approaches implemented in HeuristicLab have been applied for identifying estimators for selected continuous as well as discrete medical variables and cancer diagnoses: Genetic programming, linear regression, k-nearest-neighbor regression, support vector machines (optimized using evolutionary algorithms), and random forests. In the empirical section of this paper the authors describe interaction networks identified for a medical data base storing data of more than 600 patients. The authors see that whatever modeling approach is used, it is possible to identify the most important influence factors and display those in interaction networks which can be interpreted without domain knowledge in machine learning or informatics in general.

This paper presents the development of an Ultrasonography Simulator Training Tool. Ultrasonography is a user-dependent technology; operators must possess both physical and cognitive ultrasonography... Show More

This paper presents the development of an Ultrasonography Simulator Training Tool. Ultrasonography is a user-dependent technology; operators must possess both physical and cognitive ultrasonography capability to include mechanical manipulation of the probe, image capture, and interpretation of pathology. There are three major challenges in developing a simulator training tool that provides both the physical (hardware) and cognitive (software-visualization) learning experience: 1) graphical user interface design, 2) hardware utilization and interface, and 3) the integration of software for simulating volume of ultrasound beams and visualization of images (imagery data). This tool is being being developed with educator input from conceptual design to final product validation. The initial output of this effort is a prototype real-time, dynamic ultrasound simulator training tool specific for cardiac interrogation. The tool has been crafted to accommodate a dedicated teaching component that is extendable with additional anatomical imagery and teaching materials to include instructor observation capability.

Health Technology Assessment (HTA) provides information for decision-makers in the health care system. The consortium IFEDH (Innovative Framework for Evidence-based Decision Support in Healthcare)... Show More

Health Technology Assessment (HTA) provides information for decision-makers in the health care system. The consortium IFEDH (Innovative Framework for Evidence-based Decision Support in Healthcare) was formed in order to design a new strategy for the integration of modelling and simulation into the HTA process within limited resources whilst staying open to emerging new technologies. This aim resulted in a framework that consists of a standardized workflow, tasks, methods and good practices. Beginning with formalization of the underlying problem and identification of research questions, the process is followed by data acquisition and modelling and finally ends by generating results for decision support. Visualization and documentation are crucial tasks that support the whole process. Additionally, standards for modelling were defined and new general modelling concepts were developed. All these developments serve as a basis for fast, credible, evidence-based decision support for comparing various technologies in the health system.

Segmentation of morphology in medical image data is a highly context specific and differs from various imaging modalities, necessitating the use of sophisticated mathematical models and algorithms... Show More

Segmentation of morphology in medical image data is a highly context specific and differs from various imaging modalities, necessitating the use of sophisticated mathematical models and algorithms to achieve good results. In this work an algorithm is presented for pre-segmentation of general medical input data, based on a watershed-segmentation strategy utilizing both, original intensities and derived gradient magnitudes for region growing. The number of resulting pre-classified regions is iteratively reduced to a user-defined threshold using merge metrics, accounting for the similarity of intensity profiles of two neighboring regions to merge, as well as the height of the gradient barriers to overcome and geometric aspects like sphericity and size of the border area with respect to the total region size. Based on such a context-independent pre-segmentation, the resulting manageable number of regions can be further merged and classified, utilizing texture features and a priori statistical models. Results are presented from brainweb database.

Transportation has been identified as a major barrier to healthcare access, particularly, within vulnerable population groups. The level of healthcare access that most population segments have in... Show More

Transportation has been identified as a major barrier to healthcare access, particularly, within vulnerable population groups. The level of healthcare access that most population segments have in traditional transit systems may be increased with new initiatives that involve complex and large investments in transit oriented developments (TOD) projects. However, the increasing attractiveness of neighborhoods affected by TOD initiatives may result in the gentrification of vulnerable population segments. These vulnerable segments are likely to be relocated into less attractive neighborhoods characterized by inadequate transit systems. This relocation increases the probabilities of reducing healthcare access for these underserved groups leading to an increase in health disparities. The present discussion calls for research to explore relevant factors that affects these dynamics. A framework that enables the identification of individual factors that affect gentrification processes under TOD initiatives as well as quantifying the effects from these processes is suggested in this paper. A system dynamics framework that allows the understanding of the dynamics associated with this system is suggested in this paper. Critical areas for empirical research are highlighted. These are prerequisites for the effective deployment of initiatives that ensure the mitigation of possible negative impacts on vulnerable populations.

Cardiovascular diseases are considered to be the main cause of death in developed countries. Due to limitations resulting from in-vivo measurements of velocity, the analysis and evaluation of... Show More

Cardiovascular diseases are considered to be the main cause of death in developed countries. Due to limitations resulting from in-vivo measurements of velocity, the analysis and evaluation of hemodynamic parameters by means of computational simulations have become the only efficient solution, especially in cases of pathological ventricle or of assisted conditions. A Left Ventricular Assist Device (LVAD) has been used to provide hemodynamic support to patients with critical cardiac failure. The approach used in this work is to develop a Computational Fluid Dynamics (CFD) model of the aorta, including a mechanical support similar to the LVAD (Jarvik 2000 Heart Inc, New York, NY) in order to assess its effects on local haemodynamics. Its purpose is to analyse the hemodynamic effects of a continuous flow LVAD, evaluated in three different working conditions, taking into account the outflow-graft anastomosis in two different locations: thoracic and ascending aorta.

This research work is aimed at proposing a simulation model based on Intelligent Agents devoted to reproduce human behavior influence over the evolution and impact of obesity epidemics. Indeed... Show More

This research work is aimed at proposing a simulation model based on Intelligent Agents devoted to reproduce human behavior influence over the evolution and impact of obesity epidemics. Indeed, obesity is a real big problem for both USA and European countries, so it is necessary to take under control this phenomenon and, above all, to support Agencies and Nations with simulation models in order to promote specific actions, to guarantee population healthy and to reduce the related social costs. To this end, taking advantage of previous experiences on Human Behavior Models, a Library including Intelligent Agents for Computer Generated Forces (IA-CGF Libraries) has been developed. This library is conceived to reproduce complex scenarios with particular attention to non-conventional frameworks on the progression of obesity epidemics in the world where human behaviors play a crucial role. As for the simulation models test, calibration and validation, two scenarios with different underlying social and cultural conditions have been considered and compared, namely: Italy (obesity prevalence ~10%) and U.S.A. (obesity prevalence ~35%). This way, it has been possible to gain fruitful insights about how simulation models evolve over different social and cultural conditions in different countries.

Markov cohort models are one of the standard methodologies for conducting cost-effectiveness analyses of health care interventions of chronic diseases. The system dynamics (SD) approach, where a... Show More

Markov cohort models are one of the standard methodologies for conducting cost-effectiveness analyses of health care interventions of chronic diseases. The system dynamics (SD) approach, where a model consists of stocks and their interconnecting flows, is rarely used, but in principle it can be used to conduct the same type of analysis. The authors show that a simple transformation from the transition probabilities of a Markov model to relative rates always leads to an equivalent system dynamics model. Here the stocks match the Markov states, because they store the same number of patients as in the corresponding state. The authors demonstrate an approach on an exemplary cost-effectiveness analysis for a smoking cessation programme for chronic obstructive pulmonary disease (COPD) with a simplified Markov model based on Menn (2009). Both the Markov model and the system dynamics model lead to nearly identical results. However, the latter offers more flexibility as it can incorporate interactions between different patient groups.

The proposed research work introduces an innovative framework based on Serious Games and agent-driven models for supporting and renewing educational processes in Health Care. To this end the... Show More

The proposed research work introduces an innovative framework based on Serious Games and agent-driven models for supporting and renewing educational processes in Health Care. To this end the proposed solution, called MARIA (Model for Advanced and Realistic patient simulation driven by Intelligent Agents) includes new models of virtual humans driven by intelligent agents to be used as patients in a distributed simulation game together with students. MARIA’s main goal is to teach concepts related to the continuous care of patients in order to create a strong relationship among medical doctor, students and their set of personal virtual patients. Therefore it is able to reproduce the patient life cycles and teach subjects such as pathology evolution, preventive actions, symptoms identification and human relationships to the student. These features make the proposed solution an innovation in the educational programs for health students enabled by new technologies.