Atalan Abdulkadir, Şahin Hasan, Atalan Yasemin Ayaz
Faculty of Engineering, Gaziantep Islam Science and Technology University, Gaziantep 27260, Turkey.
Faculty of Engineering, Bursa Technical University, Bursa 16310, Turkey.
Healthcare (Basel). 2022 Sep 30;10(10):1920. doi: 10.3390/healthcare10101920.
A healthcare resource allocation generally plays a vital role in the number of patients treated () and the patient waiting time () in healthcare institutions. This study aimed to estimate and as output variables by considering the number of healthcare resources employed and analyze the cost of health resources to the hospital depending on the cost coefficient () in an emergency department (ED). The integration of the discrete-event simulation (DES) model and machine learning (ML) algorithms, namely random forest (RF), gradient boosting (GB), and AdaBoost (AB), was used to calculate the estimation of the output variables depending on the of resources cost. The AB algorithm performed best in almost all scenarios based on the results of the analysis. According to the AB algorithm based on the , , , and , the accuracy data were calculated as 0.9838, 0.9843, 0.9838, and 0.9846 for ; 0.9514, 0.9517, 0.9514, and 0.9514 for , respectively in the training stage. The GB algorithm had the best performance value, except for the results of the (AB had a better accuracy at 0.8709 based on the value of for ) in the test stage. According to the AB algorithm based on the , , , and , the accuracy data were calculated as 0.7956, 0.9298, 0.8288, and 0.7394 for ; 0.8820, 0.8821, 0.8819, and 0.8818 for in the training phase, respectively. All scenarios created by the coefficient should be preferred for ED since the income provided by the value to the hospital was more than the cost of healthcare resources. On the contrary, the estimation results of ML algorithms based on the coefficient differed. Although values in all ML algorithms with and coefficients reduced the cost of the hospital, values based on and increased the cost of the hospital.
医疗资源分配通常在医疗机构的治疗患者数量（）和患者等待时间（）方面发挥着至关重要的作用。本研究旨在通过考虑所使用的医疗资源数量来估计和作为输出变量，并根据急诊科（ED）中的成本系数（）分析医院的卫生资源成本。离散事件模拟（DES）模型与机器学习（ML）算法（即随机森林（RF）、梯度提升（GB）和AdaBoost（AB））相结合，用于根据资源成本的计算输出变量的估计值。根据分析结果，AB算法在几乎所有场景中表现最佳。根据基于、、和的AB算法，在训练阶段，的准确率数据分别计算为0.9838、0.9843、0.9838和0.9846；的准确率数据分别计算为0.9514、0.9517、0.9514和0.9514。在测试阶段，除了（基于的值，AB的准确率更高，为0.8709）的结果外，GB算法具有最佳性能值。根据基于、、和的AB算法，在训练阶段，的准确率数据分别计算为0.7956、0.9298、0.8288和0.7394；的准确率数据分别计算为0.8820、0.8821、0.8819和0.8818。由于值给医院带来的收入超过了医疗资源成本，因此急诊科应优先选择由系数创建的所有场景。相反，基于系数的ML算法的估计结果不同。尽管所有具有和系数的ML算法中的值都降低了医院成本，但基于和的的值却增加了医院成本。
