Machine Learning-based Classifiers for the Prediction of Low Birth Weight.

OBJECTIVES: Low birth weight (LBW) is a global concern associated with fetal and neonatal mortality as well as adverse consequences such as intellectual disability, impaired cognitive development, and chronic diseases in adulthood. Numerous factors contribute to LBW and vary based on the region. The main objectives of this study were to compare four machine learning classifiers in the prediction of LBW and to determine the most important factors related to this phenomenon in Hamadan, Iran. METHODS: We carried out a retrospective cross-sectional study on a dataset collected from Fatemieh Hospital in 2017 that included 741 mother-newborn pairs and 13 potential factors. Decision tree, random forest, artificial neural network, support vector machine, and logistic regression (LR) methods were used to predict LBW, with five evaluation criteria utilized to compare performance. RESULTS: Our findings revealed a 7% prevalence of LBW. The average accuracy of all models was 87% or higher. The LR method provided a sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and accuracy of 74%, 89%, 7.04%, 29%, and 88%, respectively. Using LR, gestational age, number of abortions, gravida, consanguinity, maternal age at delivery, and neonatal sex were determined to be the six most important variables associated with LBW. CONCLUSIONS: Our findings underscore the importance of facilitating timely diagnosis of causes of abortion, providing genetic counseling to consanguineous couples, and strengthening care before and during pregnancy (particularly for young mothers) to reduce LBW.

Identifying factors associated with the hospital readmission rate among patients with major depressive disorder.

BACKGROUND: Major depressive disorder (MDD) is a common recurrent mental disorder and one of the leading causes of disability in the world. The recurrence of MDD is associated with increased psychological and social burden, limitations for the patient, family, and society; therefore, action to reduce and prevent the recurrence of this disorder or hospital readmissions for depression among the patients is essential. METHODS: The data of this retrospective cohort study were extracted from records of 1005 patients with MDD hospitalized in Farshchian hospital in Hamadan city, Iran (2011-2018). The hospital readmissions rate due to depression episodes was modeled using generalized Poisson regression (GPR). Demographic and clinical characteristics of the patients were considered as explanatory variables. SAS v9.4 was used (P < 0.05). RESULTS: A majority of the patients were male (66.37%). The mean (standard deviation) of age at onset of MDD and the average number of hospital readmissions were 32.39 (13.03) years and 0.53 (1.84), respectively (most patients (74.3%) did not experience hospital readmissions). According to the results of the GPR, the lower age at the onset of the disease (IRR = 1.02;P = 0.008), illiteracy (IRR = 2.06;P = 0.003), living in urban areas (IRR = 1.56;P = 0.015), history of psychiatric illnesses in the family (IRR = 1.75;P = 0.004), history of emotional problems (IRR = 1.42;P = 0.028) and having medical disorders (IRR = 1.44;P = 0.035) were positively associated with the number of hospitalizations. CONCLUSION: According to our findings, urbanization, early onset of the disease, illiteracy, family history of mental illness, emotional problems, and medical disorders are among major risk factors associated with an increased number of hospital readmissions of MDD.