Xudong Lu1, Zhengxing Huang, Huilong Duan. 1. College of Biomedical Engineering and Instrument Science of Zhejiang University, The Key Laboratory of Biomedical Engineering, Ministry of Education, China
Abstract
OBJECTIVES: Efficient clinical treatment processes is considered a key factor of medical quality control. Current IT solutions are far away from this perspective since they typically have difficulty supporting the variances occurring in clinical practices, and providing adequate flexible support of clinical processes. METHODS: This paper proposes a hybrid approach based on rough set theory and case-based reasoning to allow physicians to rapidly adjust patients' treatment processes to changes of patients' clinical states. In detail, the proposed approach recommends appropriate treatment plans in clinical process execution by adopting a similarity measure to select appropriate clinical treatment plans executed on patients who presented similar features to the current one. Such clinical treatment plans are then applied to suggest which actions to perform next in clinical treatment process execution. RESULTS: As a motivating scenario, this study performs the experiments of type 2 diabetes patient's treatment process. The results show that the proposed approach is feasible to recommend suitable clinical treatment plans in clinical process execution, which makes adaptive clinical treatment processes possible.
OBJECTIVES: Efficient clinical treatment processes is considered a key factor of medical quality control. Current IT solutions are far away from this perspective since they typically have difficulty supporting the variances occurring in clinical practices, and providing adequate flexible support of clinical processes. METHODS: This paper proposes a hybrid approach based on rough set theory and case-based reasoning to allow physicians to rapidly adjust patients' treatment processes to changes of patients' clinical states. In detail, the proposed approach recommends appropriate treatment plans in clinical process execution by adopting a similarity measure to select appropriate clinical treatment plans executed on patients who presented similar features to the current one. Such clinical treatment plans are then applied to suggest which actions to perform next in clinical treatment process execution. RESULTS: As a motivating scenario, this study performs the experiments of type 2 diabetespatient's treatment process. The results show that the proposed approach is feasible to recommend suitable clinical treatment plans in clinical process execution, which makes adaptive clinical treatment processes possible.