A bi-level interactive decision support framework to identify data mining-oriented electronic health record architectures

• We develop ISAW model from a crisp individual single-level environment to a fuzzy group bi-level environment.
• We utilize this bi-level ISAW model to reach medical decision makers to a consensus on a data mining-oriented EHR architecture.
• We consider fuzzy logic and fuzzy sets to represent ambiguous, uncertain or imprecise information.

Nowadays, many healthcares are generating and collecting a huge amount of medical data. Due to the difficulty of analyzing this massive volume of data using traditional methods, medical data mining on Electronic Health Record (EHR) has been a major concern in medical research. Therefore, it is necessary to assess EHR architectures based on the capabilities of extracting useful medical knowledge from a huge amount of EHR databases. In this paper, we develop a bi-level interactive decision support framework to identify data mining-oriented EHR architectures. The contribution of this bi-level framework is fourfold: (1) it develops Interactive Simple Additive Weighting (ISAW) model from an individual single-level environment to a group bi-level environment; (2) it utilizes decision makers’ preferences gradually in the course of interactions to reach to a consensus on an data mining-oriented EHR architecture; (3) it considers fuzzy logic and fuzzy sets to represent ambiguous, uncertain or imprecise information; and (4) it synthesizes a representative outcome based on qualitative and quantitative indicators in the EHR assessment process. A case study demonstrates the applicability of the proposed bi-level interactive framework for benchmarking a national data mining-oriented EHR.

Figure optionsDownload as PowerPoint slide