A survey of evolutionary algorithms for decision-tree induction

This paper presents a survey of evolutionary algorithms that are designed for decision-tree induction. In this context, most of the paper focuses on approaches that evolve decision trees as an alternate heuristics to the traditional top-down divide-and-conquer approach. Additionally, we present some alternative methods that make use of evolutionary algorithms to improve particular components of decision-tree classifiers. The paper's original contributions are the following. First, it provides an up-to-date overview that is fully focused on evolutionary algorithms and decision trees and does not concentrate on any specific evolutionary approach. Second, it provides a taxonomy, which addresses works that evolve decision trees and works that design decision-tree components by the use of evolutionary algorithms. Finally, a number of references are provided that describe applications of evolutionary algorithms for decision-tree induction in different domains. At the end of this paper, we address some important issues and open questions that can be the subject of future research.

Birth in Brazil: national survey into labour and birth

Background: Caesarean section rates in Brazil have been steadily increasing. In 2009, for the first time, the number of children born by this type of procedure was greater than the number of vaginal births. Caesarean section is associated with a series of adverse effects on the women and newborn, and recent evidence suggests that the increasing rates of prematurity and low birth weight in Brazil are associated to the increasing rates of Caesarean section and labour induction. Methods: Nationwide hospital-based cohort study of postnatal women and their offspring with follow-up at 45 to 60 days after birth. The sample was stratified by geographic macro-region, type of the municipality and by type of hospital governance. The number of postnatal women sampled was 23,940, distributed in 191 municipalities throughout Brazil. Two electronic questionnaires were applied to the postnatal women, one baseline face-to-face and one follow-up telephone interview. Two other questionnaires were filled with information on patients' medical records and to assess hospital facilities. The primary outcome was the percentage of Caesarean sections (total, elective and according to Robson's groups). Secondary outcomes were: post-partum pain; breastfeeding initiation; severe/near miss maternal morbidity; reasons for maternal mortality; prematurity; low birth weight; use of oxygen use after birth and mechanical ventilation; admission to neonatal ICU; stillbirths; neonatal mortality; readmission in hospital; use of surfactant; asphyxia; severe/near miss neonatal morbidity. The association between variables were investigated using bivariate, stratified and multivariate model analyses. Statistical tests were applied according to data distribution and homogeneity of variances of groups to be compared. All analyses were taken into consideration for the complex sample design. Discussion: This study, for the first time, depicts a national panorama of labour and birth outcomes in Brazil. Regardless of the socioeconomic level, demand for Caesarean section appears to be based on the belief that the quality of obstetric care is closely associated to the technology used in labour and birth. Within this context, it was justified to conduct a nationwide study to understand the reasons that lead pregnant women to submit to Caesarean sections and to verify any association between this type of birth and it's consequences on postnatal health.

Survey of Trachoma within School Students in the State of Roraima, Brazil

Purpose: Trachoma, a blinding conjunctivitis, is the result of repeated infection with Chlamydia trachomatis. There are no recent data for the state of Roraima, Brazil, where it was thought that trachoma no longer existed. These data are derived from school children sampled in this state, with additional data collected from the contacts of children with trachoma. Design: A population-based cross-sectional study with random sampling of students in grades 1 through 4 of all public schools within municipalities where the human development index was less than the national average in 2003. The sample was stratified according to population size. Participants: A sample size of 7200 was determined and a total of 6986 (93%) students were examined, along with an additional 2152 contacts. Methods: All students were examined for trachoma according to World Health Organization criteria. Demographic data and contact information also was collected. The family and school contacts of students with trachoma then were located and examined. Main Outcome Measures: Prevalence and grade of trachoma, age, gender, race, and municipality location. Results: The overall prevalence of trachoma was 4.5% (95% confidence interval [CI], 3.7%–5.3%), but there were municipalities within the state where the prevalence of inflammatory trachoma was more than 10%. The prevalence was greater in rural areas (4.9%; 95% CI, 3.7%–6.0%) compared with urban areas (3.9%; 95% CI, 2.9%–4.9%). Living in indigenous communities was associated with trachoma (odds ratio, 1.6; 95% CI, 0.9 –2.6). An additional 2152 contacts were examined, and the overall trachoma prevalence was 9.3% (95% CI, 8.1–10.5). Conclusions: Trachoma continues to exist in Roraima, Brazil, where there are municipalities with a significant prevalence of disease. The indigenous population is highly mobile, crossing state and international borders, raising the possibility of trachoma in neighboring countries. Trachoma prevalence among the contacts of students with trachoma was higher than the school population, highlighting the importance of contact tracing.