Analysis on the management effect based on PSO-LSTM algorithm for medical consumables
10.3969/j.issn.1672-8270.2025.02.025
- VernacularTitle:基于粒子群优化长短期记忆算法的医用耗材管理效果分析
- Author:
Fangyuan YIN
1
;
Qing HAN
;
Yang ZHOU
;
Dongfeng HAN
Author Information
1. 中国药科大学附属南京天印山医院护理部 南京 211100
- Publication Type:Journal Article
- Keywords:
Particle swarm optimization long Short-term memory(PWO-LSTM)algorithm;
Medical consumables management;
Prediction model;
Effect analysis
- From:
China Medical Equipment
2025;22(2):138-142
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To construct prediction model based on particle swarm optimization long short-term memory(PSO-LSTM)algorithm for medical consumables,so as to predict the consumption of medical consumables and realize the fine management for medical consumables.Methods:The consumption of medical consumables was analyzed through constructed the PSO-LSTM prediction model,and then,the consumption of medical consumables was assessed.A total of 900 medical consumables which were clinically used in the Center of Endoscopy of Nanjing Tianyinshan Hospital Affiliated to China Pharmaceutical University from March to August 2024 were selected,and they were randomly divided into conventional management mode group(n=450 consumables)and PSO-LSTM management mode group(n=450 consumables)by random number table method,so as to conduct management for consumables.The management time,management timeliness,situation of monthly management of medical consumables,and predicted accuracy of consumption amount of medical consumables of two management mode in single diagnosis and treatment were compared.Results:The prepared time,delivered time,response time of temporary application,and the time of returning storehouse of medical consumables in PSO-LSTM model management mode were respectively(3.36±1.66),(9.54±2.69),(3.69±1.65)and(3.47±1.87)min,which were all less than those in conventional management mode,and the differences were statistically significant(t=7.76,13.811,28.147,39.105,P<0.05).The timeliness of supplementing medical consumables,shortage warning and proposing treatment methods of using PSO-LSTM model management mode were respectively(0.87±0.23),(3.26±0.87)and(14.98±2.65)min,which were all less than those of using conventional management mode,and the differences of them were statistically significant(t=4.228,15.482,13.420,P<0.05).In 450 medical consumables of the center of endoscopy,the amounts of the monthly average emergency purchasing and monthly average shortage were respectively 20 and 22 in using PSO-LSTM model management mode,accounting for 4.44%and 4.89%,all of which were lower than those of conventional management mode,and the differences of them were statistically significant(x2=10.364,11.835,P<0.05).The prediction results of using PSO-LSTM model management mode for disposable infusion device and disposable auxiliary kit of gastroscopy were closer to the actual consumption than the conventional management mode,while the differences of them were not significant(P>0.05).Conclusion:The application of the management mode of predictive model of medical consumables based on PSO-LSTM algorithm in the management of medical consumables of the center of endoscopy in hospital can accurately assess the consumption of medical consumables,and provide scientific basis for the procurement decision of medical consumables,and achieve fine management for various kinds of medical consumables,and reduce the cost of medical consumables in the center of endoscopy.
