Preterm or premature birth refers to the delivery of a baby before completing 37 weeks of gestation, poses significant health risks to newborns and remains a global concern. This paper addresses the need for an accurate predictive model for preterm birth using a comprehensive set of parameters from both laboratory and scan reports during pregnancy journey. Our study extends this scope to include a broader range of parameters. Our proposed system involves the integration of various lab and scan parameters, such as RBC count, WBC count, hemoglobin, TSH, abdominal circumference, cervix length, placenta position. Unlike previous studies, which focus on a limited set of parameters, our research aims to build a machine learning model leveraging a diverse range of factors from lab and scan reports. The study includes the application of traditional statistical techniques like (Logistic Regression, Naïve Bayes) and ensembling learners (Random Forest, Gradient Boost, XG Boost). Our end goal is to explore the association between various parameters and preterm labor, utilizing a dataset of approximately 4000 patient records for lab reports and 7000 patient records for scan reports. The predictive model is designed to assess the likelihood of preterm birth occurring at the 35th week of gestation, allowing sufficient time for preventive measures. This research contributes to advancing our understanding of preterm birth predictors and facilitates timely intervention to reduce associated health complications and emotional burdens.