Zhang L H, Wang J, Yao B, Chu X Y, Sun Z D, Ma C F
Shaanxi Provincial Blood Center, Xi'an Central Blood Station, Xi'an 710061, China.
Zhonghua Yu Fang Yi Xue Za Zhi. 2024 Aug 6;58(8):1213-1218. doi: 10.3760/cma.j.cn112150-20231226-00496.
To construct a prediction model for the clinical supply of blood components in Xi'an City from 2023 to 2025. Based on the blood supply data of the Blood Management Information System of Shaanxi Provincial Blood Center from January 2013 to December 2022, a gray prediction model and an exponential curve fitting model were used to construct the prediction model, and the optimal prediction model was determined according to the error parameters of the relevant indicators of the model. The supply of blood components in Xi'an from 2023 to 2025 was predicted. The fitting equations of the exponential curve fitting model to predict the supply of suspended red blood cells, platelets and cryoprecipitate in Xi'an were, （+1）=1.16e，（+1）=1.04e and （+1）=1.01e, respectively. The mean absolute errors (mean relative errors) of the exponential curve fitting model in predicting the supply of suspended red blood cells, platelets and cryoprecipitate in Xi'an were 10 488.7 (0.05%), 2 114.9 (0.08%) and 3 089.6 (0.07%), respectively, which were lower than those of the gray prediction model, about 10 488.7 (3.44%), 2 152.78 (8.20%) and 3 441.35 (7.92%), respectively. The exponential curve fitting model predicted that the clinical supply of blood components in Xi'an would increase year by year from 2023 to 2025, and the clinical supply of suspended red blood cells, platelets, and cryoprecipitate in Xi'an would increase to 409 467 U, 69 818 therapeutic volume and 94 724 U, respectively by 2025. The exponential curve fitting model can make a good prediction of the clinical supply of blood components in Xi'an City.
构建西安市2023年至2025年血液成分临床供应预测模型。基于陕西省血液中心血液管理信息系统2013年1月至2022年12月的供血数据，采用灰色预测模型和指数曲线拟合模型构建预测模型，并根据模型相关指标的误差参数确定最优预测模型。对西安市2023年至2025年血液成分供应进行预测。预测西安市悬浮红细胞、血小板和冷沉淀供应的指数曲线拟合模型的拟合方程分别为（+1）=1.16e、（+1）=1.04e和（+1）=1.01e。指数曲线拟合模型预测西安市悬浮红细胞、血小板和冷沉淀供应的平均绝对误差（平均相对误差）分别为10488.7（0.05%）、2114.9（0.08%）和3089.6（0.07%），低于灰色预测模型的相应误差，灰色预测模型的误差分别约为10488.7（3.44%）、2152.78（8.20%）和3441.35（7.92%）。指数曲线拟合模型预测，西安市2023年至2025年血液成分临床供应将逐年增加，到2025年，西安市悬浮红细胞、血小板和冷沉淀的临床供应量将分别增至409467U、69818治疗量和94724U。指数曲线拟合模型能够对西安市血液成分临床供应进行良好预测。
