Lack Of Semantic Interoperability Research Articles

Multiview Incomplete Knowledge Graph Integration with application to cross-institutional EHR data harmonization

Objective:The growing availability of electronic health records (EHR) data opens opportunities for integrative analysis of multi-institutional EHR to produce generalizable knowledge. A key barrier to such integrative analyses is the lack of semantic interoperability across different institutions due to coding differences. We propose a Multiview Incomplete Knowledge Graph Integration (MIKGI) algorithm to integrate information from multiple sources with partially overlapping EHR concept codes to enable translations between healthcare systems. Methods:The MIKGI algorithm combines knowledge graph information from (i) embeddings trained from the co-occurrence patterns of medical codes within each EHR system and (ii) semantic embeddings of the textual strings of all medical codes obtained from the Self-Aligning Pretrained BERT (SAPBERT) algorithm. Due to the heterogeneity in the coding across healthcare systems, each EHR source provides partial coverage of the available codes. MIKGI synthesizes the incomplete knowledge graphs derived from these multi-source embeddings by minimizing a spherical loss function that combines the pairwise directional similarities of embeddings computed from all available sources. MIKGI outputs harmonized semantic embedding vectors for all EHR codes, which improves the quality of the embeddings and enables direct assessment of both similarity and relatedness between any pair of codes from multiple healthcare systems. Results:With EHR co-occurrence data from Veteran Affairs (VA) healthcare and Mass General Brigham (MGB), MIKGI algorithm produces high quality embeddings for a variety of downstream tasks including detecting known similar or related entity pairs and mapping VA local codes to the relevant EHR codes used at MGB. Based on the cosine similarity of the MIKGI trained embeddings, the AUC was 0.918 for detecting similar entity pairs and 0.809 for detecting related pairs. For cross-institutional medical code mapping, the top 1 and top 5 accuracy were 91.0% and 97.5% when mapping medication codes at VA to RxNorm medication codes at MGB; 59.1% and 75.8% when mapping VA local laboratory codes to LOINC hierarchy. When trained with 500 labels, the lab code mapping attained top 1 and 5 accuracy at 77.7% and 87.9%. MIKGI also attained best performance in selecting VA local lab codes for desired laboratory tests and COVID-19 related features for COVID EHR studies. Compared to existing methods, MIKGI attained the most robust performance with accuracy the highest or near the highest across all tasks. Conclusions:The proposed MIKGI algorithm can effectively integrate incomplete summary data from biomedical text and EHR data to generate harmonized embeddings for EHR codes for knowledge graph modeling and cross-institutional translation of EHR codes.

An Ontology – based Contextual Approach for Cross-domain Applications in Internet of Things

The Internet of Things is an ecosystem which enables objects and devices such as sensors or actuators to communicate and exchange information with each other without human intervention. One of the main challenges in the Internet of Things is the lack of semantic interoperability; devices cannot understand the meaning of raw data, due to the diversity and heterogeneity in data formats from different sources. In order to deal with semantic interoperability, the ontologies are the one way to integrate semantics to raw data; they describe an IoT system and represent the data in a standardized way. The IoT devices provide a great deal of IoT data, mainly used for specific IoT applications such as smart home, smart farming, smart cities or healthcare. Therefore, existing applications became isolated in vertical silos, each one of them use independently their own model (i.e. ontology), which makes this ontologies also limited to a specific domain. Our approach has the goal of breaking down these vertical silos and achieves a semantic interoperability across IoT domains in cross-domain applications. In this paper, we have proposed a development of a single cross-domain ontology named CDOnto, it is considered to be a generic across different IoT domains, which can be extended by domain-specific ontologies. The proposed model follows a contextual approach to organize and distinguish the combined domains (i.e. contexts) representations. In addition, the ontology allows reasoning across overlapping domains and infers a complementary and new knowledge required in cross-domain applications.

Industry Commons: an ecosystem approach to horizontal enablers for sustainable cross-domain industrial innovation (a positioning paper)

This paper introduces the background, concept and definition of the Industry Commons. It initiates a discussion on the positioning of the Industry Commons Ecosystem (ICE) with respect to current research directions in advanced manufacturing and production systems that shape advances in engineering and technology, novel business models and innovation breakthroughs. The potential value of data sharing across industrial domains is estimated at over $100 billion, particularly in view of optimising manufacturing processes. Data sharing across domains however faces a series of well-documented challenges associated with the lack of semantic interoperability and related standards, management of trust and sustainability. Solving bottlenecks in data sharing requires a systemic approach to data management, which can account for all aspects of data use, levels of application, attribution and dynamic exchanges. In this paper we propose a high-level ecosystem approach that integrates societal values with digital affordances of industry’s cognitive-assisted processes, remote interfacing, hybrid applications and large-scale value networks. Early development of an Ontology Commons EcoSystem (OCES) is presented as the key enabling framework for Industry Commons interoperability and a series of enabling frameworks form the basis of future research directions in Trusted Data Sharing and Closed-Loop Lifecycle Management for greater sustainability. Abbreviations: AI – Artificial Intelligence; AIOTI – Alliance of Internet-of-Things Innovation; ALM – Asset Lifecycle Management; ALO – Application-Level Ontology; AP – Application Protocol; API – Application Programming Interface; B2B – Business-to-Business; B2C – Business-to-Customer; CDE – Cross-Domain Ecosystem; CDEI – Cross-Domain Ecosystem Interoperability; CL2M – Closed-Loop Lifecycle Management; CNO – Collaborative Networked Organisations; CPS – Cyber-Physical Systems; CSR – Corporate Social Responsibility; DLO – Domain-Level Ontology; DLT – Distributed Ledger Technology; EM – Enterprise Modelling; FAIR – Findable, Accessible, Interoperable and Reusable; GUI – Graphical User Interface; ICE, Industry Commons Ecosystem; IOF – Industrial Ontology Foundry; IP – Intellectual Property; IPR – Intellectual Property Rights; ISN – Intertwined Supply Network; MIR – Music Information Retrieval; MLO – Middle-Level Ontology; MO – Meta-Ontology; OCES – Ontology Commons EcoSystem; PI – Physical Internet; PLM – Product Lifecycle Management; ROI – Return-on-Investment; SC – Supply Chain; SCM – Supply Chain Management; SOS – System-of-Systems; TDS – Trusted Data Sharing; TLO – Top-Level Ontology; TRO – Top Reference Ontology; TUI – Tangible User Interface.

Metadata Schemas and Ontologies for Building Energy Applications: A Critical Review and Use Case Analysis

Digital and intelligent buildings are critical to realizing efficient building energy operations and a smart grid. With the increasing digitalization of processes throughout the life cycle of buildings, data exchanged between stakeholders and between building systems have grown significantly. However, a lack of semantic interoperability between data in different systems is still prevalent and hinders the development of energy-oriented applications that can be reused across buildings, limiting the scalability of innovative solutions. Addressing this challenge, our review paper systematically reviews metadata schemas and ontologies that are at the foundation of semantic interoperability necessary to move toward improved building energy operations. The review finds 40 schemas that span different phases of the building life cycle, most of which cover commercial building operations and, in particular, control and monitoring systems. The paper’s deeper review and analysis of five popular schemas identify several gaps in their ability to fully facilitate the work of a building modeler attempting to support three use cases: energy audits, automated fault detection and diagnosis, and optimal control. Our findings demonstrate that building modelers focused on energy use cases will find it difficult, labor intensive, and costly to create, sustain, and use semantic models with existing ontologies. This underscores the significant work still to be done to enable interoperable, usable, and maintainable building models. We make three recommendations for future work by the building modeling and energy communities: a centralized repository with a search engine for relevant schemas, the development of more use cases, and better harmonization and standardization of schemas in collaboration with industry to facilitate their adoption by stakeholders addressing varied energy-focused use cases.

Conceptualizing Small and Medium-Sized Smart Cities in the Mediterranean Region

Despite the remarkable interest in smart cities, noticed during the last decade, a consistent comprehension of the concept is not yet fully realized. Various definitions, ranging from exclusively technology-oriented perceptions to broader views, have been introduced, establishing a definitional polyphony and causing lack of semantic interoperability. Empirical evidence witnesses the prevalence of technology-pushed smart city initiatives as well as their failure to meet expectations in several urban domains. When planning “smart,” the relevance of ICT and their applications should be in alignment with spatial and other urban peculiarities and sub-systems' interactions, implying the need for getting deep insight into the city's ontology. The paper focuses on the extension/enrichment of an existing smart city ontology, with concepts and relationships stemming from Mediterranean small and medium-sized cities, in an attempt to outline their main key drivers and their interrelationships and fully grasp the smart city concept in the particular spatial context.

Toward early diagnosis decision support for breast cancer: Ontology-based semantic interoperability.

e18072 Background: Breast cancer is ranked as the most common cancer among women in China, cases in China account for about 300,000 of all newly diagnosed breast cancers each year. Developing Clinical Decision Supporting System (CDSS) to assist early diagnosis breast cancer is valuable, however, lack of semantic interoperability in CDSS has been identified as the main obstacle for broad adoption of CDSS. Ontology is considered to be one of the effective approaches to bridge the terminology gap between various clinical systems and data sources. In this study, we describe our efforts in transforming breast cancer clinical knowledge into computable ontology to support breast cancer early diagnosis. Methods: We have built a breast cancer ontology (BCO) to support our CDSS pipeline. A number of NCCN clinical practice guidelines in Oncology for Breast Cancer and the breast cancer diagnosis and treatment guidelines of China are the main knowledge sources of the BCO. BCO is a manually curated resource that contains concepts and relations of breast cancer clinical findings, demographics, laboratory tests, imaging results, treatments, pathologies, cancer stages, body structure and follow-up information. These curated knowledges are annotated through interoperable standard vocabulary SNOMED CT. Two physicians reviewed and evaluated the BCO. Results: Concepts in BCO currently contains 79 clinical findings, 8 demographics, 42 laboratory tests, 105 imaging results, 332 treatments, 141 pathology and cancer stages, 30 body structures and 3 follow-ups. Relations are defined such as finding site, associated with, etc. The concepts hierarchies and relationships were built by using Protégé to support OWL representation. As an initial evaluation results, most of concepts could be mapped to SNOMED CT, but there are some concepts could not be exactly mapped to SNOMED CT. We created local Chinese Vocabularies for these local terms. Conclusions: Based on interoperable BCO, a group of computable rules are developed by adding statements according to early diagnosis criteria of breast cancer. Diagnosis assistance around complex patients is supported though connecting the Electronic Health Record (EHR). This is an ongoing development that the BCO is continuingly enriched with standard concepts, relations and integrate with individual instances to support broader adoptability.

Labeling for Big Data in radiation oncology: The Radiation Oncology Structures ontology.

PurposeLeveraging Electronic Health Records (EHR) and Oncology Information Systems (OIS) has great potential to generate hypotheses for cancer treatment, since they directly provide medical data on a large scale. In order to gather a significant amount of patients with a high level of clinical details, multicenter studies are necessary. A challenge in creating high quality Big Data studies involving several treatment centers is the lack of semantic interoperability between data sources. We present the ontology we developed to address this issue.MethodsRadiation Oncology anatomical and target volumes were categorized in anatomical and treatment planning classes. International delineation guidelines specific to radiation oncology were used for lymph nodes areas and target volumes. Hierarchical classes were created to generate The Radiation Oncology Structures (ROS) Ontology. The ROS was then applied to the data from our institution.ResultsFour hundred and seventeen classes were created with a maximum of 14 children classes (average = 5). The ontology was then converted into a Web Ontology Language (.owl) format and made available online on Bioportal and GitHub under an Apache 2.0 License. We extracted all structures delineated in our department since the opening in 2001. 20,758 structures were exported from our “record-and-verify” system, demonstrating a significant heterogeneity within a single center. All structures were matched to the ROS ontology before integration into our clinical data warehouse (CDW).ConclusionIn this study we describe a new ontology, specific to radiation oncology, that reports all anatomical and treatment planning structures that can be delineated. This ontology will be used to integrate dosimetric data in the Assistance Publique—Hôpitaux de Paris CDW that stores data from 6.5 million patients (as of February 2017).

Success and Failure Factors in the Regional Health Information System Design Process

To identify success and failure factors in the design process of a regional health information system. A constructive evaluation study including interviews, observations, usability study and document analysis. Modelling was found to be a key element for the successful implementation of a health information system. The developed service chain model helped to define use cases and to implement seamless service chains. User participation in the design process was a success factor resulting in good user acceptance and signs of positive impacts on work practices. Evaluation study also helped system developers to guide the system's further development. An important failure factor identified was the lack of semantic interoperability of the system components. The results emphasize the socio-technical nature of health information systems. The starting point for development should be thorough insight into the health care work practices where the information systems are to be used. Successful system design should start from modelling of work processes, data and information flows and definition of concepts and their relations. Health informatics as a scientific discipline provides theories and models for the design and development process.

Domain ontologies for data sharing–an example from environmental monitoring using field GIS

Different geospatial information communities, public authorities as well as private institutions, recognize increasingly the World Wide Web as a medium to distribute their data. With the occurrence of national laws that push authorities to make environmental data accessible, Internet-based services have to be developed to enable the public to obtain information digitally. Dissemination of data is only one side of the coin. The other side is the use of such data. The use requires mechanisms to share data via networks. Lack of semantic interoperability has been identified as the main obstacle for data sharing. Research, however, must develop methods to overcome the problems of sharing data considering their semantics. Ontologies are considered to be one approach to support data sharing. This paper describes the use of ontologies via the Internet based on an example from field GIS supported environmental monitoring. The basic idea is that the members of different information communities get access to the meaning of data if they can approach the ontologies that have been developed by those who collected the data. This might be possible by applying the resource definition framework (RDF) and RDF/Schema. RDF can be used to define and structure terms and vocabulary used in a specific information community. The goal of the paper is to examine the role of ontologies based on the study of a particular application domain, namely stream surveying. The use of RDF/Schema is described related to the example.