Managing different time granularities of clinical information by an interval-based temporal data model

In the field of databases, time management at different levels of granularity has been an issue for several years, for instance when dealing with clinical information from different databases using different time units, dealing with natural language expressions, or when dealing with temporal uncertainty. A temporal data model is proposed to manage the temporal aspect of data, presented at various and mixed levels of granularity. The concept of temporal assertions shapes the entire temporal information. The model provides a temporal dimension to the data by using intervals that can be specified at different granularities. The model supports a three-valued logic, where True, False and Undefined are the truth values. The temporal data model allows to manage some degrees of uncertainty when establishing temporal relationships between intervals or between temporal assertions, expressed at different granularities. The logical connectives and quantifiers can manage each of the three truth-values. We applied the temporal data model by implementing an object-oriented database system for managing follow-up clinical data from patients who underwent percutaneous transluminal coronary angioplasty