https://tedebc.ufma.br/jspui/handle/tede/883 2026-08-31T01:07:47Z https://tedebc.ufma.br/jspui/handle/tede/7189 Título: PREDIÇÃO DE NASCIMENTO PRÉ-TERMO EXTREMO EM DUAS CIDADES BRASILEIRAS Autor: BRITO, Emmanuelle Novaes de Vasconcelos Primeiro orientador: THOMAZ, Erika Bárbara Abreu Fonseca Abstract: Introduction: Extreme preterm birth (EPTB) is a major public health problem associated with high neonatal and infant morbidity and mortality. Its occurrence is influenced by multiple clinical, sociodemographic, and behavioral factors, making prevention particularly challenging. Although relatively uncommon, EPTB has substantial consequences for maternal and child health. In this context, machine learning (ML) techniques offer a promising approach for identifying pregnant women at increased risk of EPTB. Objective: To develop and evaluate machine learning-based predictive models to identify predictors of extreme preterm birth among live births from two Brazilian cities. Methods: This analytical cross-sectional study was nested within the 2010 BRISA prenatal and birth cohorts conducted in São Luís, Maranhão, and Ribeirão Preto, São Paulo, Brazil. The inclusion of these cohorts allowed the investigation of maternal and child health in distinct socioeconomic and demographic settings. The study included singleton pregnancies receiving care in public and private healthcare services, comprising 6,414 mother–child pairs from São Luís and 8,273 from Ribeirão Preto, totaling 14,687 observations. A sensitivity analysis was performed using the variable threatened preterm labor during the current pregnancy, which represents ongoing pathophysiological changes potentially related to the onset of the outcome and is considered an important predictor. Six analytical scenarios were developed, corresponding to the combined cohort and the São Luís and Ribeirão Preto cohorts, each analyzed with and without this variable. Based on expert knowledge and previous literature, 119 variables common to both cohorts, 121 variables in São Luís, and 119 variables in Ribeirão Preto were selected as potential predictors, including socioeconomic, demographic, behavioral, reproductive, and maternal healthcare characteristics. The BORUTA feature selection algorithm identified 50 relevant variables, from which the eight most important predictors were retained. Fourteen machine learning algorithms were evaluated in each scenario: penalized logistic regression (PLR), support vector machines (svmLinear and svmRadial), random forest (rf, ranger, and Rborist), boosting methods (glmboost and AdaBoost.M1), nearest neighbor methods (Knn e KKnn), artificial neural networks (nnet), tree-based partitioning models (cforest and ctree2) and generalized additive models (GAM). Model performance was assessed using accuracy, sensitivity, specificity, precision, F1-score, and the area under the receiver operating characteristic curve (AUC-ROC). Results: All analytical scenarios showed class imbalance between EPTB and non-EPTB births. Model performance varied across the six scenarios. When threatened preterm labor was included, kNN showed the best overall performance in the combined cohort (sensitivity = 0.774, AUC-ROC = 0.806, F1-score = 0.049), svmRadial performed best in São Luís (sensitivity = 0.653, AUC-ROC = 0.732, F1-score = 0.063), and ranger achieved the highest performance in Ribeirão Preto (AUC-ROC = 0.860, sensitivity = 0.771, F1-score = 0.053). When this variable was excluded, nnet performed best in the combined cohort (sensitivity = 0.561, AUC-ROC = 0.648, F1-score = 0.019), GAM in São Luís (sensitivity = 0.623, AUCROC = 0.587, F1-score = 0.024), and Rborist in Ribeirão Preto (AUC-ROC = 0.629, sensitivity = 0.573, F1-score = 0.016). Overall, kNN and nnet demonstrated the best predictive performance in scenarios with and without the threatened preterm labor variable, respectively. The eight most important predictors were threatened preterm labor during the current pregnancy, threatened miscarriage during the current pregnancy, maternal occupation, per capita income, alcohol consumption during pregnancy, self-reported maternal race/skin color, planned pregnancy, and hypertension diagnosed before or during pregnancy. Conclusion: Machine learning models can predict extreme preterm birth with good predictive performance using maternal health, fetal, socioeconomic, behavioral, and prenatal care variables. Early identification of pregnancies at increased risk may support timely interventions and improve maternal and neonatal outcomes. These findings provide evidence for the potential application of machine learning in prenatal risk stratification and encourage further validation in different populations and clinical settings. Instituição: Universidade Federal do Maranhão Tipo do documento: Dissertação 2026-05-26T00:00:00Z https://tedebc.ufma.br/jspui/handle/tede/7081 Título: Predição de risco de atraso no desenvolvimento neuropsicomotor nos primeiros mil dias de vida utilizando aprendizado de máquina: coorte brisa Autor: SILVA, Katia Susana Azevedo Primeiro orientador: SOUZA, Bruno Feres de Abstract: Introduction: The first thousand days of life represent a critical window of brain plasticity, in which neuropsychomotor development (NPMD) is sensitive to biological and psychosocial influences. Early prediction of risks during this period is fundamental for timely interventions, although it constitutes a complex challenge in community-based populations. Objective: To develop and evaluate Machine Learning (ML) models to predict the risk of NPMD delay in children during the first thousand days of life. Methods: Prospective study with data from 972 mother-child pairs from the BRISA cohort (São Luís – MA). The outcome was the risk of delay assessed in the cognitive, communication and motor domains evaluated by the Bayley III scale. The algorithms Penalized Logistic Regression (PLR), Random Forest (RF), Support Vector Machines (SVM) with radial kernel and Artificial Neural Networks (ANN), and the null classifier were tested. Preprocessing included imputation of missing data, encoding of categorical variables by one-shot encoding and standardization of numerical variables. For the evaluation of the generated ML models, K-fold cross-validation (k = 5) with 10 repetitions was used. As performances metrics, the area under the AUC-ROC curve (AUC), sensitivity, specificity, accuracy, precision and F1-score were used. The interpretability of the models was analyzed using the SHapley Additive Explanations (SHAP) technique. Results: The highest prevalence of risk of delay occurred in expressive communication (44.1%). The models showed moderate performance with AUC ranging between 0.533 and 0.603, with PLR being the most stable algorithm, reaching an AUC of 0.603 in the fine motor domain. SHAP analysis identified as main predictors: family history of epilepsy, child's sex, maternal depressive symptoms, psychological violence and social support, maternal education and family income. Conclusion: ML models demonstrated moderate discrimination ability in heterogeneous population samples, showing that increased algorithmic complexity did not surpass logistic regression. The findings reinforce the multifactorial nature of development, highlighting the impact of psychosocial and biological determinants, contributing to the development of screening tools applicable to Primary Health Care. Instituição: Universidade Federal do Maranhão Tipo do documento: Dissertação 2026-02-27T00:00:00Z https://tedebc.ufma.br/jspui/handle/tede/7050 Título: PERCEPÇÕES DE PROFISSIONAIS DA ATENÇÃO PRIMÁRIA SOBRE A SAÚDE DE PESSOAS LGBTQIA+ Autor: ABREU, Carla Michelle Rodrigues Primeiro orientador: LIMA, Sara Fiterman Abstract: Introduction: Despite advances in the creation of the National Policy for Comprehensive Health of Lesbians, Gays, Bisexuals, Transvestites, and Transsexuals, the health needs of the LGBTQIA+ population are still unknown to a large part of health professionals. Without adequate care, lesbians, gays, bisexuals, and transgender people tend to resist seeking qualified support. In this context, it becomes fundamental that professionals develop skills and competencies to understand and value the real needs of this population. Objective: To analyze the perceptions and attitudes of Primary Care professionals regarding the health of the LGBTQIA+ community. Method: This is a qualitative, descriptive, and analytical study, carried out between October 2025 and January 2026 at the Bezerra de Menezes Primary Health Care Unit, São Francisco Primary Health Care Unit, Centro Primary Health Care Unit, and Liberdade Primary Health Care Unit, located in São Luís – MA, with professionals from the Family Health Strategy. Data collection was carried out through a focus group, using a structured questionnaire for sociodemographic information and professional trajectory, and a semi-structured interview guide. The interviews were transcribed and analyzed according to Thematic Analysis, following the perspective of Braun and Clarke. Results: From the analysis of the statements, three main categories emerged: the first, Conceptual understanding of gender identity and sexual orientation, is composed of three core meanings: Lack of knowledge of the acronym LGBTQIA+ and its meanings, generating misunderstanding about sexual and gender diversity; The first part, "Attitudes and Positions of Professionals Towards the LGBTQIA+ Population," is composed of three core meanings: Perception and Recognition of Prejudice in the Care of the LGBTQIA+ Population; Influence of Personal, Religious, Moral, or Cultural Beliefs; Recognition of Ethical-Professional Duty in Care. The second part, "Experiences in the Care of the LGBTQIA+ Population in Primary Health Care," is composed of four core meanings: Programmed Invisibility: Accounts of Little or No Experience; Clinical Reductionism: Care Centered on Complaint and Pathologization; Conducts Marked by Insecurity, Hesitation, and Exclusionary Attitudes; Primary Health Care as an Entry Point: Potentialities and Limitations in Reception. Final Considerations: The participants' statements reveal conceptual confusion, embarrassment when addressing the topic, and resistance in the correct use of pronouns and social names, revealing weaknesses both in the technical field and in the fulfillment of human rights. legal. The influence of personal moral and religious values proves to be a limiting element for professional practice guided by ethical principles and the universality of the SUS (Brazilian Unified Health System). Instituição: Universidade Federal do Maranhão Tipo do documento: Dissertação 2026-02-27T00:00:00Z https://tedebc.ufma.br/jspui/handle/tede/7002 Título: Associação entre consumo de alimentos segundo o grau de processamento e dificuldades comportamentais em adolescentes de 10 a 13 anos da coorte brisa Autor: OLIVEIRA, Lucia Regina Moreira de Primeiro orientador: VIOLA, Poliana Cristina de Almeida Fonseca Abstract: Introduction: Adolescence is a critical period for the consolidation of dietary habits and mental health development. Growing evidence suggests that diet quality, particularly the high consumption of Ultra-Processed Foods (UPF), can negatively influence behavior and mental health, given their low nutrient density and pro-inflammatory potential. Studying this relationship, while controlling for socioeconomic and behavioral factors, is essential for formulating specific interventions. Objective: To investigate the association between the degree of food processing and behavioral difficulties in adolescents (10 to 13 years old) participating in the BRISA cohort. Methods: This was a cross-sectional cohort study conducted with adolescents aged 10-12 years from the BRISA Cohort, São Luís, Maranhão (N=2293). The exposure variable was food consumption, categorized according to the NOVA classification and measured in energy percentage (%Kcal) using a 24-Hour Dietary Recall (24hDR). The primary outcome was the Total Difficulties Score, assessed by the Strengths and Difficulties Questionnaire (SDQ). Sociodemographic, socioeconomic, and behavioral variables (Sex, Age, Brazilian Economic Class, Physical Activity, Screen Time, Sleep Duration) were used as confounding factors, as directed by a Directed Acyclic Graph (DAG). The association was tested through bivariate analyses and adjusted Multiple Linear Regression. A significance level of 5% was adopted. Results: The average consumption of UPF corresponded to 30.3% of the Total Energy Value of the diet. The SDQ analysis revealed that 26% of adolescents were classified as at risk (borderline + clinical) in the total score. Bivariate analysis indicated a significant association between UPF consumption and sex, with female adolescents being more prevalent in the higher consumption tertiles (p=0.002). Multiple Linear Regression demonstrated that, even after adjustment for all confounders, a higher consumption of UPF was independently associated with higher behavioral difficulties scores (β=0.300; 95% CI: 0.05; 0.55). Complementarily, an increase in the consumption of Unprocessed or Minimally Processed Foods (UMPF) was associated with a reduction in the Total Difficulties Score (β = - 0.410; 95% CI: -0.70; -0.12). Other risk factors included sedentary behavior/inactivity and high screen time. Conclusion: The consumption of ultra-processed foods acts as a fundamental determinant of increased behavioral difficulties in adolescence. The findings reinforce the urgent need for comprehensive public policies that promote the substitution of ultra-processed foods with diets based on unprocessed and minimally processed foods, aiming to protect the mental health of Brazilian adolescents. Instituição: Universidade Federal do Maranhão Tipo do documento: Dissertação 2026-02-25T00:00:00Z