Improving Routine Immunization Coverage Through Optimally Designed Predictive Models

作     者：Fareeha Sameen Abdul Momin Kazi Majida Kazmi Munir A Abbasi Saad Ahmed Qazi Lampros K Stergioulas 

作者机构：Faculty of Electrical and Computer EngineeringNED University of Engineering and TechnologyKarachi75270Pakistan Department of Pediatrics and Child HealthAga Khan UniversityKarachi74800Pakistan Surrey Business SchoolUniversity of SurreyGuildfordGU27XHUnited Kingdom Neurocomputation LabNational Centre of Artificial IntelligenceNED University of Engineering&TechnologyKarachi75270Pakistan Faculty of IT and DesignThe Hague University of Applied Sciences2521 ENThe HagueThe Netherlands 

出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)

年 卷 期：2022年第70卷第1期

页      面：375-395页

核心收录：

学科分类：1004[医学-公共卫生与预防医学(可授医学、理学学位)] 100401[医学-流行病与卫生统计学] 10[医学] 

基　　金：This study was funded by GCRF UK and was carried out as part of project CoNTINuE-Capacity building in technology-driven innovation in healthcare 

主　　题：Machine learning predictive models routine immunization vaccine coverage pakistan optimization SMOTE 

摘      要：Routine immunization(RI)of children is the most effective and timely public health intervention for decreasing child mortality rates around the *** being a low-and-middle-income-country(LMIC)has one of the highest child mortality rates in the world occurring mainly due to vaccine-preventable diseases(VPDs).For improving RI coverage,a critical need is to establish potential RI defaulters at an early stage,so that appropriate interventions can be targeted towards such populationwho are identified to be at risk of missing on their scheduled vaccine *** this paper,a machine learning(ML)based predictivemodel has been proposed to predict defaulting and non-defaulting children on upcoming immunization visits and examine the effect of its underlying contributing *** predictivemodel uses data obtained from Paigham-e-Sehat study having immunization records of 3,113 *** design of predictive model is based on obtaining optimal results across accuracy,specificity,and sensitivity,to ensure model outcomes remain practically relevant to the problem *** optimization of predictive model is obtained through selection of significant features and removing data *** machine learning algorithms were applied for prediction of defaulting children for the next immunization *** results showed that the random forest model achieves the optimal accuracy of 81.9%with 83.6%sensitivity and 80.3%*** main determinants of vaccination coverage were found to be vaccine coverage at birth,parental education,and socioeconomic conditions of the defaulting *** information can assist relevant policy makers to take proactive and effective measures for developing evidence based targeted and timely interventions for defaulting children.