Pre-Term-Birth-Prediction

Now a days in urban areas as well as rural areas, most of the pregnant women's don aware of preterm birth. Pre term birth means baby born before 37 weeks of pregnancy due to alcohol drinking, no caring and unavoidable circumstances.

Preterm Birth Prediction using Machine Learning

Overview:

Nowadays, in both urban and rural areas, many pregnant women are unaware of the risks associated with preterm birth. Preterm birth refers to the delivery of a baby before 37 weeks of pregnancy due to factors such as alcohol consumption, lack of prenatal care, and unavoidable circumstances.

Dataset Description:

The dataset used in this project is extracted from Chinese laboratories by analyzing electrohistogram signals (EHS) from 60 pregnant women. Through thorough examination and analysis of these signals, a dataset was constructed with the following features:

Count Contraction: No. of times the contraction has occurred. Length of Contraction: Length of contraction in terms of womb. STD: Standard Deviation of EH signal from its normal distribution. Entropy: The randomness of molecules present in contraction or in the womb. Contraction times: The number of intervals during pregnancy.

Exploratory Data Analysis:

Various exploratory data analysis techniques have been employed, including Histogram, Boxplot, Bar Graph, Violin Plot, Pie chart, and Correlation coefficients, which can be found in the repository notebook.

Algorithms Used:

• Decision Tree Classifier
• Random Forest Classifier
• Support Vector Classifier
• Stacked Classifier

Applications/Future Scope:

The results of our study suggest that the prediction model based on Random Forest and Support Vector Classifier shows potential for predicting preterm birth in the early stages of pregnancy. There is a need for further development of methods and enhancement of existing ones to address the challenges faced by healthcare providers during pregnancy.

The ultimate goal of this expert system development is to predict preterm labor risk for medium and rural families. The purpose of this study is to assess the feasibility of using machine learning to generate expert system knowledge-based rules for predicting preterm delivery, particularly in healthcare settings.