Computational Ontologies in Orthopaedic Surgery

Background Information Technology (IT) plays an important role in storing and collating the vast amounts of healthcare data. However, analyzing and integrating this data to extract useful information is difficult due to the heterogeneous, siloed, disparate, and unstructured nature of the data. Where are we now? Attempts to standardize data reporting by establishing reporting standards, checklists and guidelines have not been optimal [ 3 , 11 , 19 ]. Moreover, efforts to integrate data through the use of registries, data sharing networks, vocabularies and data standards have also yielded limited results. These efforts, when applied to orthopaedics, where theoretical knowledge is scattered over subspecialties, make it a cognitively challenging and tedious process. Where do we need to go? Implementing data standardization is an important step towards homogenizing the data so that it can be integrated. Once integrated, the next step would be data analysis for information extraction. This information would be useful in answering important questions, especially in orthopaedic clinical practice and research, and could even help optimize methodologies in the education field. How do we get there? With the ability to describe concepts in a standardized manner and define existing interrelationships, ontologies are a potential solution. They assist in standardizing and integrating data and also impart strong inferential capabilities at a granular level. When applied to orthopaedics, they can standardize data collection, link data sources, generate knowledge based on the assumptions present in the interlinked data, thus answering important questions regarding orthopaedic clinical practice, research and education [ 22 , 28 , 30 ].