Semantic Web Health Care and Life Sciences Interest Group Charter

Motivations

Biological Science

An understanding of the fundamental underlying principles of biology forms the basis on which all biomedical research relies. Biological research investigates phenomena at a range of scales: molecules, cells, cell populations, tissues, organs, organisms, populations, and ecosystems and employs an astounding variety of experimental techniques, instruments and reagents. This research (e.g. gene expression, phenotypes, chemical screening, ...) generates data and conclusions from which new hypotheses are drawn which subsequently propel new studies at an ever increasing rate. The resulting proliferation of isolated databases hampers efforts to combine results.

The HCLSIG focus area will aid this enterprise and provide ways for researchers in the applied fields to make the best use of the richness of this rapidly accumulating knowledge. HCLSIG activities in the this area will include working with key data repositories towards their semantic integration by advocating for and assisting labs, database creators and publishers who would make information accessible using Semantic Web technologies. The group will apply ontologies to the integration of heterogeneous data, show how common analysis tools can use data from and publish to the Semantic Web, and explore other ways that Semantic Web technologies might further facilitate this scientific work.

Translational Medicine

Recent advances in biological understanding are allowing pharmaceutical companies to begin to develop tailored therapeutics, thereby allowing patients to receive the right drug, at the right dose, and at the right time. However, in order for such treatments to be developed, companies need to be able to better link data from the laboratory to the clinic (bench to bedside). This concept is frequently referred to as translational medicine.

This HCLSIG activity will provide resources that demonstrate the value of Semantic Web technologies to translational medicine. Activities in this area will focus on connecting pre-clinical and clinical trial data with clinical decision support knowledge in order to assess drug efficacy and safety. Examples of potential activities include the integration of health outcome data for identification of safety signals, aggregation of clinical trials data for identification of studies of interest, demonstration of the minimal costs required for the integration of unforeseen data sets into an existing data model to enable answering unanticipated scientific questions, and the creation of dashboards that show how heterogeneous and disparate data can be integrated to aid decision making.

Health Care

Within the larger domain of health care there has been a dramatic increase in the demand for information systems that capture expressive clinical data, host rich clinical knowledge, and are able to deliver robust decision support on behalf of healthcare quality improvement and clinical research. This HCLSIG activity will focus on applying the strengths of Semantic Web technologies directly relevant to meeting this demand. A primary goal will be to aid in efforts to unify the collection of data for the purpose of both primary care (electronic medical records) and clinical research (patient recruitment, study management, outcomes-based longitudinal analysis, etc.).

This interest group will work towards promoting this goal in several ways. First, it will attempt to work with ongoing efforts to standardize and harmonize the acquisition and exchange of medical data led by standard bodies such as the CDISC, Health Level Seven (HL7) and BRIDG, enabling the use of these standards with Semantic Web technologies. Second, it will collaborate with efforts focused on building formal ontologies for clinical medicine and investigations expressed in Semantic Web languages such as OWL and RDFS. Finally, it will explore enabling interoperability through the documentation of mappings between terminologies.

Another topic of interest is building innovative clinical decision support capabilities into patient record systems. Our task in this area is to identify best practices for clinical guideline representation in such a way that standards-based reasoning systems and knowledge sources can be leveraged in these patient record systems.