Clinical Pathways Scheduling Using Hybrid Genetic Algorithm

Abstract

In order to improve the standard of management in hospitals and effectively control the cost of clinical treatments, this research primarily focuses on optimizing the scheduling of clinical pathways (CPs). A mathematical model for CP scheduling is constructed, and the hybrid genetic algorithm (HGA, combining a genetic algorithm with particle swarm optimization) is proposed for solving this problem so as to distribute medical resources and schedule the treatments of patients reasonably and effectively. The optimal search space can be further enlarged by introducing a new mutation mechanism, which allows a more satisfactory solution to be found. In particular, the relative patient waiting time and relative time efficiency are used as measure indexes, which are more scientific and effective than the usual indexes of absolute time and absolute time efficiency. After comparing absolute waiting time, relative waiting time, utilization of absolute waiting time, and utilization of relative waiting time waiting respectively, the conclusion can confidently be drawn that task scheduling obviously enhances patients' time efficiency, reduces time wastage and therefore promotes patient satisfaction with medical processes. Moreover, the patients can to a certain degree move away from their usual passive role in medical processes by using this scheduling system. In order to further validate the rationality and validity of the proposed method, the heuristic rules for CP scheduling are also tested using the same case. The results demonstrate that the proposed HGA achieves superior performance, in terms of precision, over those heuristic rules for CP scheduling. Therefore, we utilize HGA to optimize CP scheduling, thus providing a decision-making mechanism for medical staff and enhancing the efficiency of medical processes. This research has both theoretical and practical significance for electronic CP management, in particular.

A Survey of Recent Research on Optimization Models and Algorithms for Operations Management from the Process View

“…The majority of the hybrid methods are the combinations of different metaheuristics, such as the combination of Genetic Algorithm and Tabu Search procedures (as discussed by Zhang et al [258], Meeran and Morshed [259], Li and Gao [260], Yu et al [261], and Noori and Ghannadpour (2012)) and the integration of Genetic Algorithm and Simulated Annealing (as discussed by Safari and Sadjadi [262], Rafiei et al [263], and Bettemir and Sonmez [264]). In recent years, the combination of Genetic Algorithm and Particle Swarm Optimization (PSO) had been widely applied in both scheduling and vehicle routing problems (as discussed by Du et al [265], Yu et al [266], Liu et al [267], and Kumar and Vidyarthi [268]). Some other Genetic Algorithm-based hybrid methods were also observed in the literature, such as the hybrid Genetic-Monkey algorithm [269], hybrid Genetic Algorithm combined with the LP-relaxation of the targeted model (as discussed by Mohammad and Ghasem [270]), and the combination of Genetic Algorithm and Local Search procedure with Fuzzy Logic Control, where Fuzzy Logic Control is used to enhance the search ability of the Genetic Algorithm (as discussed by Chamnanlor et al [271]); some Pareto-based hybrid Genetic Algorithms were also developed for dealing with multiobjective problems (as discussed by Zhang et al [272] and Tao et al [273]).…”

Research on Optimization of clinical Pathway Execution based on Multi-agent Simulation

“…Wolf [7] has proposed a mathematical programming model with constraints for clinical pathway and pointed out the difficulty of the clinical path optimization, which involved the trade-off of multiple clinical indicators. Other researchers [8][9][10][11] have introduced hybrid genetic algorithm into clinical pathway scheduling to optimize the execution of the routine diagnostic activities, which has only considered the local human resources constraints and mainly been for single or multiple clinical pathway scheduling in specific departments. Aiming to…”

Online Treatment Compliance Checking for Clinical Pathways

“…1 4 Myocardial ischemia assessment (low-risk patients) c 5 Dual antiplatelet drugs: generally, aspirin and clopidogrel are used simultaneously. c 6 Intravenous injection of GPIIb / IIIa is a considerate choice for patients in medium-risk level or high-risk level who plan to perform PCI surgery · · · · · · Surgery in 0-3 days after admission (if it is necessary to undergo a surgery). …”

“…As results, the compliance rules ζ 3 and ζ 4 are ready to fire. Taking ζ 3 as an example, the compliance checker automatically deduces and fills in the 6 System prototype screenshot on the compliance checking in the pathway workflow execution antecedent axioms of ζ 3 using Pellet, and check if they are satisfied with the partial patient trace σ . If the antecedent axioms of ζ 3 are satisfied, it is fired and the the data property obeyRBTIn3DaysAfterAdmission of σ is set as true.…”

“…Health-care organizations seek for a better control of their medical services not only to satisfy treatment requirements but also to leverage cost-saving opportunities through standardization of treatment behaviors in clinical pathways (CPs) [1][2][3][4][5][6]. With this respect, compliance checking is important as it ensures that treatment behaviors of a healthcare organization act in accordance with the established CP specifications [7].…”