Plausible reasoning reflects the "plasticity" element of human reasoning, which, by leveraging the semantics of relevant concepts, allows dealing with incomplete data during decision making. We propose the SEmantics-based Data ANalytics (SeDan) framework that integrates plausible reasoning with expressive, fine-grained biomedical ontologies. Using this framework, an unresolvable query can be rewritten to explore the semantic knowledge graph and infer new knowledge. While the gained insights may be plausible, i.e., not supported by crisp deductive reasoning, they may still aid complex medical decision making by recommending plausible solutions. In this paper, we investigate the efficiency of SeDan in a real-world medical setting to pose intelligent medical queries from BioASQ challenges over the MEDLINE database. Experimental results show that SeDan can expand the query answering coverage by resolving up to 45% of initially unresolvable queries. The correctness of the inferred answers, as well as the underlying plausible reasoning processes, was verified by a domain expert.

Computerizing paper-based CPG and then executing them can provide evidence-informed decision support to physicians at the point of care. Semantic web technologies especially web ontology language (OWL) ontologies have been profusely used to represent computerized CPG. Using semantic web reasoning capabilities to execute OWL-based computerized CPG unties them from a specific custom-built CPG execution engine and increases their shareability as any OWL reasoner and triple store can be utilized for CPG execution. However, existing semantic web reasoning-based CPG execution engines suffer from lack of ability to execute CPG with high levels of expressivity, high cognitive load of computerization of paper-based CPG and updating their computerized versions. In order to address these limitations, we have developed three CPG execution engines based on OWL 1 DL, OWL 2 DL and OWL 2 DL + semantic web rule language (SWRL). OWL 1 DL serves as the base execution engine capable of executing a wide range of CPG constructs, however for executing highly complex CPG the OWL 2 DL and OWL 2 DL + SWRL offer additional executional capabilities. We evaluated the technical performance and medical correctness of our execution engines using a range of CPG. Technical evaluations show the efficiency of our CPG execution engines in terms of CPU time and validity of the generated recommendation in comparison to existing CPG execution engines. Medical evaluations by domain experts show the validity of the CPG-mediated therapy plans in terms of relevance, safety, and ordering for a wide range of patient scenarios.

Non-small cell lung cancer (NSCLC) constitutes the most common type of lung cancer and is frequently diagnosed at advanced stages. Clinical studies have shown that molecular targeted therapies increase survival and improve quality of life in patients. Nevertheless, the realization of personalized therapies for NSCLC faces a number of challenges including the integration of clinical and genetic data and a lack of clinical decision support tools to assist physicians with patient selection. To address this problem, we used frequent pattern mining to establish the relationships of patient characteristics and tumor response in advanced NSCLC. Univariate analysis determined that smoking status, histology, epidermal growth factor receptor (EGFR) mutation, and targeted drug were significantly associated with response to targeted therapy. We applied four classifiers to predict treatment outcome from EGFR tyrosine kinase inhibitors. Overall, the highest classification accuracy was 76.56% and the area under the curve was 0.76. The decision tree used a combination of EGFR mutations, histology, and smoking status to predict tumor response and the output was both easily understandable and in keeping with current knowledge. Our findings suggest that support vector machines and decision trees are a promising approach for clinical decision support in the patient selection for targeted therapy in advanced NSCLC.

Semantic Web technologies have shown to have great potential in many different domains, to facilitate knowledge representation, exchange and reasoning, in a formal and yet both human and machine understandable way. In particular, within the health domain, they enable knowledge integration and understanding by explicitly defining and linking concepts and relationships using ontologies to information within clinical knowledge bases. This additional metadata also allows for automated decision support and semantic based analytics to be implemented, that facilitate improved healthcare at a lower cost. Unfortunately many existing datasets in healthcare environments are still stored in relational databases, as opposed to using semantic technologies. Due to this, the link with explicit metadata is often lacking or non-existent. Furthermore, both the databases and the clinical terminologies can be considerably large, making the mapping and subsequent uses of the information a difficult process. In a full fledged decision support system the level and accuracy of the mapping can greatly influence the effectiveness of any subsequent analysis and decision support tasks. This is especially true in clinical scenarios, where very large and complex sets of terms need to be mapped to relational databases. In this paper we aim to provide a general approach for interlinking relational data with clinical ontology based metadata that allows for a fine grade evaluation, with respect to the mapping's impact on analytics. We evaluate our approach by mapping information from clinical terminologies, such as SNOMED CT, to a large laboratory dataset contained in a relational database, with the goal of creating a full fledged, semantically enabled, analytics and decision support system.

In this paper we pursue the standardization of care, for a specific medical condition, across multiple institutions through the alignment of institutionspecific clinical pathways. Our approach is to represent institution-specific Clinical Pathways (CP) using an ontological model and then align the common care activities between multiple CP, based on their semantic descriptions, to generate a `maximally standardized' care model that merges common care activities whilst allowing to administer unique institution-specific care activities. Our semantic web based approach is used to develop a standardized prostate cancer care model for three Canadian institutions, such that the standardized model can be executed via a web-based system to streamline patient care.

We present a semantic interoperability framework to represent Clinical Pathways (CP) as business process workflows represented using Business Process Modeling Notation (BPMN). We take a knowledge management approach whereby we represent CP as a CP ontology. To represent a CP as a process workflow we have developed a high-level semantic mapping between the CP ontology and the BPMN ontology. The ontology mapping allows the alignment of semantic relations between two ontologies and thus ensures that a clinical process defined in the CP ontology is mapped to a standard BPMN workflow element.

This paper aims to develop an infobutton to automatically retrieve published papers corresponding to a topic-specific online clinical discussion. The knowledge linkages infobutton is designed to supplement online clinical conversations with pertinent medical literature from Pubmed. The project involves three distinct steps: 1) Clinical messages around a specific problem are grouped together into a thread. 2) These threads are processed using Metamap to link the conversations to keywords from the MeSH lexicon. 3) These keywords are used in a novel search strategy to retrieve a set of papers from Pubmed, which are then returned to the user. A pilot study using the messages from 2007 and 2008, was conducted to compare the knowledge linkage search strategy to a vector space model and extended Boolean model. The knowledge linkage model proved to be significantly better in terms of precision ( p = 0.013 and 0.003, respectively) and recall ( p = 0.351 and 0.013). Pertinent papers were returned to over 55% of the threads. This approach has demonstrated how clinicians can supplement their peer communications with evidence based research. Future work should focus on how to improve the threading and keyword-mapping strategies.

Inconsistencies, whether they occur during the process of ontology evolution or appear during the ontology reconciliation process, result in potential harm to an ontology structure, and decrease credibility in representing a consistent and shared vocabulary of an underlying domain. When an inconsistency occurs, there are mainly two ways to deal with it: either resolve it, or reason with the inconsistent ontology. In this paper, we propose our approach for detecting and resolving inconsistencies in ontologies. Our inconsistency detection deals with the identification of contradiction derivations under the integrity constraint rules -- that are given in a logic program. For resolving detected inconsistencies, we generate all possible Minimal Inconsistent Resolve Candidates (MIRCs). Removing an MIRC from the inconsistent ontology will result in a maximal consistent sub-ontology w.r.t. the given logic program. To inform the user about the consequences of removing an MIRC, we also provide a list of all its derived triples. We evaluated our approach on the POKM domain ontology, where we detected all the inconsistencies in this ontology and generated all possible MIRCs for extracting a maximal consistent sub-ontology.

We present an E-Research platform for studying the impact of changes in the ecosystem on oceans and marine life. We have developed the Platform for Ocean Knowledge Management (POKM) that offers a suite of services for researchers to (a) access, share, integrate and operationalize the data, models and knowledge resources available at multiple sites; (b) collaborate in joint scientific research experiments by sharing resources, results, expertise and models; and (c) form a virtual community of researchers. We take a knowledge management approach to establish conceptual, terminological and data level interoperability between ocean and marine life science communities so that they can collaborate to conduct complex experiments. POKM is supported by the CANARIE high bandwidth network enables rapid sharing of high-volume data from distributed repositories to users across the world.

Clinical decision-making involves an active interplay between various medical knowledge modalities. Medical knowledge morphing aims to support clinical decision support by mimicking the interplay between knowledge modalities, as per the problem description, to derive a holistic knowledge-base. We leverage the semantic Web technology suite to pursue knowledge morphing. We present a knowledge morphing framework that involves ontologies to represent the domain and knowledge artifacts, annotation of knowledge artifacts based on the respective ontological model, and ontology mediation activities to knowledge morphing.

Clinical pathways translate evidence-based recommendations into locally practicable, process-specific algorithms that reduce practice variations and optimize quality of care. Our objective was to abstract practice-oriented knowledge from a cohort of real clinical pathways and represent this knowledge as a clinical pathway ontology. We employed a four step methodology: (1) knowledge source identification and classification of clinical pathways according to variations in setting, stage of care, patient type, outcome and specialty; (2) iterative knowledge abstraction using grounded theory; (3) ontology engineering as adapted from the Model-based Incremental Knowledge Engineering approach; and, (4) ontology evaluation through encoding a sample of real clinical pathways. We present our clinical pathway ontology that offers a detailed ontological model describing the structure and function of clinical pathways. Our ontology can potentially integrate with a healthcare semantic web, and ontologies for clinical practice guidelines, patients and institutions to form the foundational knowledge for generating patient-specific CarePlans.

Clinical decision-making involves an active interplay between various medical knowledge modalities - the spectrum of medical knowledge modalities spanning from tacit knowledge to experiential knowledge to explicit knowledge to data-induced knowledge. The ability to simultaneously access and then integrate multiple knowledge modalities pertaining to a common clinical theme is profound for clinical decision making. In this concept paper we introduce knowledge morphing - a knowledge modeling task that allows the integration of heterogeneous medical knowledge modalities, with respect to a clinical case, to yield a comprehensive knowledge resource for decision-support.

The availability of modern imaging techniques such as confocal scanning laser tomography (CSLT) for capturing high-quality optic nerve images offer the potential for developing automatic and objective methods for supporting clinical decision-making in glaucoma. We present a hybrid approach that features the analysis of CSLT images using moment methods to derive abstract image defining features, and the use of these features to train classifiers for automatically distinguishing CSLT images of healthy and diseased optic nerves. As a first step, in this paper, we present investigations in feature subset selection methods for reducing the relatively large input space produced by the moment methods. Our results demonstrate that our methods discriminate between healthy and glaucomatous optic nerves based on shape information automatically derived from CSLT tomography images.

In this paper, we propose a new approach for combining item-based collaborative filtering (CF) with case based reasoning (CBR) to pursue personalized information filtering in a knowledge sharing context. Functionally, our personalized information filtering approach allows the use of recommendations by peers with similar interests and domain experts to guide the selection of information deemed relevant to an active user's profile. We apply item-based similarity computation in a CF framework to retrieve N information objects based on the user's interests and recommended by peer. The N information objects are then subjected to a CBR based compositional adaptation method to further select relevant information objects from the N retrieved past cases in order to generate a more fine-grained recommendation.

Tacit knowledge of health-care experts is an important source of experiential know-how, yet due to various operational and technical reasons, such health-care knowledge is not entirely harnessed and put into professional practice. Emerging knowledge-management (KM) solutions suggest strategies to acquire the seemingly intractable and nonarticulated tacit knowledge of health-care experts. This paper presents a KM methodology, together with its computational implementation, to 1) acquire the tacit knowledge possessed by health-care experts; 2) represent the acquired tacit health-care knowledge in a computational formalism--i.e., clinical scenarios--that allows the reuse of stored knowledge to acquire tacit knowledge; and 3) crystallize the acquired tacit knowledge so that it is validated for health-care decision-support and medical education systems.

We present a personalized health information generation and delivery system that leverages case-based reasoning techniques to dynamically author a personalized health information package based on an individual's current health profile. The work features a compositional adaptation approach, whereby relevant health information elements from the solution component of multiple similar past cases are carefully selected and systematically combined to yield a new personalized health information package. We have implemented a generic Java-based case-based reasoning engine that applies a novel compositional adaptation algorithm to author a HTML-based personalized health information package that can be e-mailed to users.

The application of data mining techniques to medical data is certainly beneficial for researchers interested in discerning the complexity of healthcare processes in real-life operational situations. We present a methodology, together with its computational implementation, for the automated extraction of data-defining CNF symbolic rules from medical data-sets comprising both annotated and un-annotated attributes. We propose a hybrid approach for symbolic rule extraction which features a sequence of methods including data clustering, data discretization and eventually symbolic rule discovery via rough set approximation. We present a generic data mining workbench that can generate cluster/class-defining symbolic rules from medical data, such that the resultant symbolic rules are directly applicable to medical rule-based expert systems.

Within the confines of a healthcare enterprise memory (HEM), most traditional medical systems do not sufficiently provide the necessary assistance to healthcare practitioners in the handling of critical situations. Furthermore, localized knowledge repositories often lack the required knowledge for problem solving. Therefore, in this paper, we present an agent-based knowledge broker called the Intelligent Healthcare Knowledge Assistant (IHKA) for dynamic knowledge gathering, filtering, adaptation and acquisition from a HEM comprising an amalgamation of (i) databases storing empirical knowledge, (ii) case bases storing experiential knowledge, (iii) scenario bases storing tacit knowledge and (iv) document bases storing explicit knowledge. The featured work leverages intelligent agent techniques for autonomous HEM-wide navigation, approximate content matching, inter- and intra-content correlation, and knowledge adaptation and procurement to meet the user's healthcare knowledge needs.

This paper presents a case for an intelligent agent-based framework for knowledge discovery in a distributed healthcare environment comprising multiple heterogeneous healthcare data repositories. Data-mediated knowledge discovery, especially from multiple heterogeneous data resources, is a tedious process and imposes significant operational constraints on end-users. We demonstrate that autonomous, reactive and proactive intelligent agents provide an opportunity to generate end-user-oriented, packaged, value-added decision-support/strategic planning services for healthcare professionals and managers. We propose the use of intelligent agents to implement a distributed agent-based data mining information structure that provides a suite of healthcare-oriented decision-support/strategic planning services.

Clinical evidence exists in modalities other than published clinical literature, such as: clinical data ranging from patient clinical profiles to clinical trials; clinical experiences of eminent medical practitioners; and medical knowledge bases encapsulating knowledge about patient care, healthcare guidelines and protocols, clinical workflow, and so on. We propose a technology-enriched strategy to exploit advanced computer technologies-knowledge management, data mining, case-based reasoning strategies and Internet technology-within traditional evidence-based medicine systems to derive all-encompassing clinical evidence derived from heterogeneous clinical evidence modalities. The paper features a conceptual overview of an integrated clinical evidence system designed to augment the typical literature-based clinical evidence with additional technology-mediated clinical evidence.

Provides a technology-enriched solution to Web-mediated patient empowerment initiatives via the implementation of an intelligent information structure that features (1) the dynamic composition of personalized healthcare information (PHI) conforming to an individual's EMR (electronic medical record) based health profile; and (2) proactive Internet-based delivery of PHI. Healthcare information personalization is achieved via profile-specific selection and template-specific aggregation of multiple, fine-grained, pre-authored, generic, topic-specific healthcare information content. The use of intelligent constraint satisfaction techniques ensure the medical correctness of the dynamically composed PHI document. The application of Internet-based push technology allows the PHI document to be pushed to the individual's e-mail account, thus guaranteeing the timely availability of high-quality health maintenance information.

The development of medical case-based reasoning (CBR) systems necessitates the active involvement of medical experts. The work featured in this paper aims to minimize the involvement of medical experts in enhancing the knowledge content of medical CBR systems by using causal information contained in a generic electronic patient record (EPR) as an alternate source of CBR-compliant cases. We present an automated case acquisition and transcription information structure that features: (a) an agent to proactively procure XML-based EPRs from Internet-accessible EPR repositories; and (b) a case generation methodology to automatically transform generic EPRs into specialized CBR-compliant clinical cases (CCs). EPR-CC transformation is achieved by establishing a multi-level equivalence between the EPR and CC constructs .e. structural equivalence via meta-data constructs, terminological equivalence via a meta-thesaurus and conceptual equivalence via domain-specific ontologies. The transformed CCs are intended to be seamlessly incorporated within CBR-based medical diagnostic systems.

The knowledge-rich nature of the healthcare domain has made it an ideal environment for the application of knowledge-based techniques. However, the abstract nature of tacit healthcare knowledge has resulted in the under-utilization of such a vital component of the overall healthcare delivery system. Therefore, we present a six-step approach to procure tacit healthcare knowledge using healthcare scenarios. This process is supported by a healthcare scenario composer, which is a key component of a wider framework called the Tacit Knowledge Acquisition Info-Structure (TKAI). This information structure aims to combine the effectiveness of healthcare scenarios, ontologies and artificial intelligence techniques to facilitate the efficient acquisition, representation, refinement and dissemination of expert-quality tacit healthcare knowledge from/to healthcare experts.

We present a technology-enriched, Web-enabled, value-added distance exam preparation and evaluation service that provides: (a) offline execution of fully featured preparatory exercises and evaluation tests in a real-life simulated examination environment; (b) content personalization to address scholastic weakness; and (c) the use of data mining techniques to ensure content effectiveness and the pro-active identification of the academic needs of various student segments. The solution is designed as a client-server architecture featuring Java technology and XML-mediated information exchange over the Internet.