ABSTRACT
Infant mortality is an important health measure in a population as a crude indicator of the poverty and socioeconomic level. It also shows the availability and quality of health services and medical technology in a specific region. Although improvements have been observed in the last decades, the implementation of actions to reduce infant mortality is still a concern in many countries. To address such an important problem, this paper proposes a new support decision approach to classify newborns according to their neonatal mortality risk. Using features related to mother, newborn, and socio-demographic, we model the problem using a data-driven classification model able to provide the probability of a newborn dying until 28 t h days of life. More than a theoretical study, decision support tools as the one proposed here is relevant in countries in development as Brazil, because it aims at identifying risky neonates that may die to raise the attention of medical practitioners so that they can work harder to reduce the overall neonatal mortality. Overcoming an AUC of 96%, the proposed method is able to provide not just the probability of death risk but also an explicable interpretation of most important features for model decision, which is paramount in public health applications. Furthermore, we provide an extensive analysis across different rounds of experiments, including an analysis of pre and post partum features influence over data-driven model. Finally, different from previously conducted studies which rely on databases with less than 100,000 samples, our model takes advantage from a new proposed database, constructed using more than 1,400,000 samples comprising births and deaths extracted from public records in São Paulo-Brazil from 2012 to 2018.
ABSTRACT
SPNeodeath dataset includes births and deaths of infants during the neonatal period from São Paulo city between 2012 and 2018, containing more than 1.4 million records. The dataset was created from SINASC and SIM Brazilian information systems for births and deaths respectively. SINASC comprises information about demographic and epidemiological data for the infant, mother, prenatal care and childbirth. SIM collects information about mortality, and it is used as the basis for the calculation of vital statistics, such as neonatal mortality rate. SIM was only used to label records from SINASC, when the death happened until 28 days of life. SPNeodeath has 23 variables with socioeconomic maternal condition features, maternal obstetrics features, newborn related features and previous care related features, besides a label feature describing if the subject survived, or not, after 28 days of life. In order to build the dataset, DBF files were downloaded from DATASUS ftp repository and converted to CSV format, the R programming language, and then the CSV files were processed using Python programming language. Features with incorrect values and unknowing information were removed.
