Anna Hristoskova

Ontology-driven monitoring of patient's vital signs enabling personalized medical detection and alert.

A major challenge related to caring for patients with chronic conditions is the early detection of exacerbations of the disease. Medical personnel should be contacted immediately in order to intervene in time before an acute state is reached, ensuring patient safety. This paper proposes an approach to an ambient intelligence (AmI) framework supporting real-time remote monitoring of patients diagnosed with congestive heart failure (CHF). Its novelty is the integration of: (i) personalized monitoring of the patients health status and risk stage; (ii) intelligent alerting of the dedicated physician through the construction of medical workflows on-the-fly; and (iii) dynamic adaptation of the vital signs’ monitoring environment on any available device or smart phone located in close proximity to the physician depending on new medical measurements, additional disease specifications or the failure of the infrastructure. The intelligence lies in the adoption of semantics providing for a personalized and automated emergency alerting that smoothly interacts with the physician, regardless of his location, ensuring timely intervention during an emergency. It is evaluated on a medical emergency scenario, where in the case of exceeded patient thresholds, medical personnel are localized and contacted, presenting ad hoc information on the patients condition on the most suited device within the physicians reach.

Dynamic Composition of Semantically Annotated Web Services through QoS-Aware HTN Planning Algorithms

This paper presents a Dynamic Composer for Web services. The services are enriched with semantic descriptions in OWL-S, based on which the Composer automatically creates a combination of services reaching a specified goal. As an example, a trip planning use case is chosen where the goal ranges from booking of a single flight to planning of an entire trip including flight, hotel, transport, etc. The composition is achieved using local and global algorithms satisfying specific quality of service (QoS) constraints and requirements such as the execution time or cost of the invoked Web services. At the same time a more extended HTN planning algorithm is discussed, matching not only service outputs to inputs but also satisfying service preconditions through effects. In addition to the automatic composition, the paper also proposes a recovery mechanism in case of unavailable services. When executing the composition of flight services, unavailable services are dynamically replaced by equivalent services or a new composition achieving the needed result. The presented platform and planning algorithms are put through extensive performance and scalability tests for typical trip booking scenarios, in which basic services are composed to a complex trip planning service.

Ontology-Driven Monitoring of Patient's Vital Signs Enabling Personalized Medical Detection and Alert

A major challenge related to caring for patients with chronic conditions is the early detection of exacerbations of the disease that may be of great significance. The dedicated clinical personnel should be contacted immediately and possibly intervene in time before an acute state is reached, by changing medication, or any other interventions, in order to ensure patient safety. This paper presents an Ambient Intelligence (AmI) framework supporting real-time remote monitoring of patients diagnosed with congestive heart failure. The remote monitoring environment, enhanced with semantic technologies, provides a personalized, accurate and fully automated emergency alerting system that smoothly interacts with the personal physician, regardless his/her physical location in order to ensure in time intervention in case of an emergency. The proposed framework is able to change context at runtime in case new medical services are registered, new rules are defined, or in case of network overload and failure situations.

A formal concept analysis approach to consensus clustering of multi-experiment expression data

BackgroundPresently, with the increasing number and complexity of available gene expression datasets, the combination of data from multiple microarray studies addressing a similar biological question is gaining importance. The analysis and integration of multiple datasets are expected to yield more reliable and robust results since they are based on a larger number of samples and the effects of the individual study-specific biases are diminished. This is supported by recent studies suggesting that important biological signals are often preserved or enhanced by multiple experiments. An approach to combining data from different experiments is the aggregation of their clusterings into a consensus or representative clustering solution which increases the confidence in the common features of all the datasets and reveals the important differences among them.ResultsWe propose a novel generic consensus clustering technique that applies Formal Concept Analysis (FCA) approach for the consolidation and analysis of clustering solutions derived from several microarray datasets. These datasets are initially divided into groups of related experiments with respect to a predefined criterion. Subsequently, a consensus clustering algorithm is applied to each group resulting in a clustering solution per group.These solutions are pooled together and further analysed by employing FCA which allows extracting valuable insights from the data and generating a gene partition over all the experiments. In order to validate the FCA-enhanced approach two consensus clustering algorithms are adapted to incorporate the FCA analysis. Their performance is evaluated on gene expression data from multi-experiment study examining the global cell-cycle control of fission yeast. The FCA results derived from both methods demonstrate that, although both algorithms optimize different clustering characteristics, FCA is able to overcome and diminish these differences and preserve some relevant biological signals.ConclusionsThe proposed FCA-enhanced consensus clustering technique is a general approach to the combination of clustering algorithms with FCA for deriving clustering solutions from multiple gene expression matrices. The experimental results presented herein demonstrate that it is a robust data integration technique able to produce good quality clustering solution that is representative for the whole set of expression matrices.

Semantic reasoning for intelligent emergency response applications

Emergency response applications require the processing of large amounts of data, generated by a diverse set of sensors and devices, in order to provide for an accurate and concise view of the situation at hand. The adoption of semantic technologies allows for the definition of a formal domain model and intelligent data processing and reasoning on this model based on generated device and sensor measurements. This paper presents a novel approach to emergency response applications, such as fire fighting, integrating a formal semantic domain model into an event-based decision support system, which supports reasoning on this model. The developed model consists of several generic ontologies describing concepts and properties which can be applied to diverse context-aware applications. These are extended with emergency response specific ontologies. Additionally, inference on the model performed by a reasoning engine is dynamically synchronized with the rest of the architectural components. This allows to automatically trigger events based on predefined conditions. The proposed ontology and developed reasoning methodology is validated on two scenarios, i.e. (i) the construction of an emergency response incident and corresponding scenario and (ii) monitoring of the state of a fire fighter during an emergency response.

Symbiotic Service Composition in Distributed Sensor Networks

To cope with the evergrowing number of colocated networks and the density they exhibit, we introduce symbiotic networks—networks that intelligently share resources and autonomously adapt to the dynamicity thereof. By allowing the software services provided in such networks to operate in an equally symbiotic manner, new opportunities for the so-called service compositions arise, which take advantage of the multitude of services and combine them to achieve goals set out by the individual networks. To accommodate services in large-scale symbiotic networks, including wireless sensor networks, we propose a software platform which autonomously constructs and orchestrates such compositions. Furthermore, upon changes in the infrastructure, the platform responds by adapting compositions to reflect the changed context. To enable the interaction between services offered by arbitrary partners, the platform deploys ontologies to achieve a common vocabulary and semantic rules to express the policies imposed by the networks involved. By applying the platform to typical scenarios from the field of sensor-augmented cargo transportation and logistics, we illustrate its applicability and, through performance evaluation, show a significant increase in process efficiency. Additionally, by means of a generic problem generator, we quantify the scalability of our platform and show the importance of an appropriate priority function, one of the core constituents of our service composition approach.

Clustering of multiple DNA microarrays through combination of particle swarm intelligence and K-means

In this article we propose a hybrid approach for clustering of gene expression data across multiple experiments, based on Particle Swarm Optimization and k-means clustering. In the proposed algorithm, each experiment identifies a particle initialized with the result of the k-means algorithm applied over the experiment. The final clustering solution is found by updating the particles using the information about the best clustering solution generated by each experiment and the entire set of experiments. The performance of the proposed cluster algorithm is evaluated on time series expression data obtained from a study examining the global cell-cycle control of gene expression in fission yeast Schizosaccharomyces pombe. The obtained experimental results demonstrate that the hybrid algorithm is able to produce good quality clustering solution, which is representative for the whole test compendium and at the same time adequately reflects the specific characteristics of the individual experiments.

An integrative clustering approach combining particle swarm optimization and formal concept analysis

In this article we propose an integrative clustering approach for analysis of gene expression data across multiple experiments, based on Particle Swarm Optimization (PSO) and Formal Concept Analysis (FCA). In the proposed algorithm, the available microarray experiments are initially divided into groups of related datasets with respect to a predefined criterion. Subsequently, a hybrid clustering algorithm, based on PSO and k-means clustering, is applied to each group of experiments separately. This produces a list of different clustering solutions, one per each group. These clustering solutions are pooled together and further analyzed by employing FCA which allows to extract valuable insights from the data and generate a gene partition over the whole set of experiments. The performance of the proposed clustering algorithm is evaluated on time series expression data obtained from a study examining the global cell-cycle control of gene expression in fission yeast Schizosaccharomyces pombe. The obtained experimental results demonstrate that the proposed integrative algorithm allows to generate a unique and robust gene partition over several different microarray datasets.

Ontology-Driven dynamic discovery and distributed coordination of a robot swarm

Swarm robotic systems rely heavily on dynamic interactions to provide interoperability between the different autonomous robots. In current systems, interactions between robots are programmed into the applications controlling them. Incorporating service discovery into these applications allows the robots to dynamically discover other devices. However, since most of these mechanisms use syntax-based matching, the robots cannot reason about the offered functionality. Moreover, as contextual information is often not included in the matching process, it is impossible for robots to select the most suitable device under the current context. This paper aims to tackle these issues by proposing a framework for semantic service discovery in a dynamically changing environment. A semantic layer was added to an existing discovery protocol, offering a semantic interface. Using this framework, services can be searched based on what they offer, with services best suiting the current context yielding the highest matching scores.

Design of a Framework for Automated Service Mashup Creation and Execution Based on Semantic Reasoning

Instead of building self-contained silos, applications are being broken down in independent structures able to offer a scoped service using open communication standards and encoding. Nowadays there is no automatic environment for the construction of new mashups from these reusable services. At the same time the designer of the mashup needs to establish the actual locations for deployment of the different components. This paper introduces the development of a framework focusing on the dynamic creation and execution of service mashups. By enriching the available building blocks with semantic descriptions, new service mashups are automatically composed through the use of planning algorithms. The composed mashups are automatically deployed on the available resources making optimal use of bandwidth, storage and computing power of the network and server elements. The system is extended with dynamic recovery from resource and network failures. This enrichment of business components and services with semantics, reasoning, and distributed deployment is demonstrated by means of an e-shop use case.