Ebrahim Yousefi

Objectives: Reducing infant mortality in the whole world is one of the millennium development goals.The aim of this study was to determine the factors related to infant mortality using data mining algorithms. Methods: This population-based case-control study was conducted in eight provinces of Iran. A sum of 2,386 mothers (1,076 cases and 1,310 controls) enrolled in this study. Data were extracted from health records of mothers and filled with checklists in health centers. We employed several data mining algorithms such as AdaBoost classifier, Support Vector Machine, Artificial Neural Networks, Random Forests, K-nearest neighborhood, and Naïve Bayes in order to recognize the important predictors of infant death; binary logistic regression model was used to clarify the role of each selected predictor. Results: In this study, 58.7% of infant mortalities occurred in rural areas, that 55.6% of them were boys. Moreover, Naïve Bayes and Random Forest were highly capable of predicting related factors among data mining models. Also, the results showed that events during pregnancy such as dental disorders, high blood pressure, loss of parents, factors related to infants such as low birth weight, and factors related to mothers like consanguineous marriage and gap of pregnancy (< 3 years) were all risk factors while the age of pregnancy (18 - 35 year) and a high degree of education were protective factors. Conclusions: Infant mortality is the consequence of a variety of factors, including factors related to infants themselves and their mothers and events during pregnancy. Owing to the high accuracy and ability of modern modeling compared to traditional modeling, it is recommended to use machine learning tools for indicating risk factors of infant mortality.

We developed an unsupervised machine learning algorithm and applied it to big corneal parameters to identify and monitor keratoconus stages. A big dataset of corneal swept source optical coherence tomography (OCT) images of 12,242 eyes acquired from SS-1000 CASIA OCT Imaging Systems in multiple centers across Japan was assembled. A total of 3,156 eyes with valid Ectasia Status Index (ESI) between zero and 100% were selected for the downstream analysis. Four hundred and twenty corneal topography, elevation, and pachymetry parameters (excluding ESI Keratoconus indices) were selected. The algorithm included three major steps. 1) Principal component analysis (PCA) was used to linearly reduce the dimensionality of the input data from 420 to eight significant principal components. 2) Manifold learning was used to further reducing the selected principal components nonlinearly to two eigen-parameters. 3) Finally, a density-based clustering was applied to the eigen-parameters to identify eyes with keratoconus. Visualization of clusters in 2-D space was used to validate the quality of learning subjectively and ESI was used to assess the accuracy of the identified clusters objectively. The proposed method identified four clusters; I: a cluster composed of mostly normal eyes (224 eyes with ESI equal to zero, 23 eyes with ESI between five and 29, and nine eyes with ESI greater than 29), II: a cluster composed of mostly healthy eyes and eyes with forme fruste keratoconus (1772 eyes with ESI equal to zero, 698 eyes with ESI between five and 29, and 117 eyes with ESI greater than 29), III: a cluster composed of mostly eyes with mild keratoconus stage (184 eyes with ESI greater than 29, 74 eyes with ESI between five and 29, and 6 eyes with ESI equal to zero), and IV: a cluster composed of eyes with mostly advanced keratoconus stage (80 eyes had ESI greater than 29 and 1 eye had ESI between five and 29). We found that keratoconus status and severity can be well identified using unsupervised machine learning algorithms along with linear and non-linear corneal data transformation. The proposed method can better identify and visualize the keratoconus stages.