Department of Digital Systems, University of Piraeus, 80, Karaoli and Dimitriou Str, Piraeus, 18534, Piraeus, Greece.
BioAssist, 1 Mpakou str, 11524, Athens, Greece.
J Med Syst. 2019 Feb 5;43(3):62. doi: 10.1007/s10916-019-1183-y.
Current healthcare services promise improved life-quality and care. Nevertheless, most of these entities operate independently due to the ingested data' diversity, volume, and distribution, maximizing the challenge of data processing and exchange. Multi-site clinical healthcare organizations today, request for healthcare data to be transformed into a common format and through standardized terminologies to enable data exchange. Consequently, interoperability constraints highlight the need of a holistic solution, as current techniques are tailored to specific scenarios, without meeting the corresponding standards' requirements. This manuscript focuses on a data transformation mechanism that can take full advantage of a data intensive environment without losing the realistic complexity of health, confronting the challenges of heterogeneous data. The developed mechanism involves running ontology alignment and transformation operations in healthcare datasets, stored into a triple-based data store, and restructuring it according to specified criteria, discovering the correspondence and possible transformations between the ingested data and specific Health Level 7 (HL7) Fast Healthcare Interoperability Resources (FHIR) through semantic and ontology alignment techniques. The evaluation of this mechanism results into the fact that it should be used in scenarios where real-time healthcare data streams emerge, and thus their exploitation is critical in real-time, since it performs better and more efficient in comparison with a different data transformation mechanism.
目前的医疗保健服务承诺提高生活质量和护理水平。然而，由于摄入数据的多样性、数量和分布，大多数这些实体都是独立运作的，这使得数据处理和交换的挑战最大化。今天的多站点临床医疗保健组织要求将医疗保健数据转换为通用格式，并通过标准化术语实现数据交换。因此，互操作性限制突出了需要整体解决方案的必要性，因为当前的技术针对特定场景进行了定制，而不符合相应标准的要求。本文档专注于一种数据转换机制，该机制可以充分利用数据密集型环境，而不会丢失健康的现实复杂性，同时应对异构数据的挑战。所开发的机制涉及在医疗保健数据集中运行本体对齐和转换操作，这些数据集存储在基于三元组的数据存储中，并根据指定的标准对其进行重构，通过语义和本体对齐技术发现摄入数据与特定健康水平 7 (HL7) 快速医疗互操作性资源 (FHIR) 之间的对应关系和可能的转换。对该机制的评估结果表明，它应该在实时医疗保健数据流出现的场景中使用，因此在实时环境中对其进行利用至关重要，因为与不同的数据转换机制相比，它的性能更好，效率更高。
