suyash5800/fetalai_using_machine_learning_to_predict_and_monitor_fetal_health

FetalAI is a machine learning-based solution designed to predict and monitor fetal health using Cardiotocogram (CTG) data. This project aims to assist healthcare professionals by automating the classification of fetal health status, helping in the early detection of risks and promoting timely intervention, particularly in low-resource settings FetalAI: Using Machine Learning to Predict and Monitor Fetal Health FetalAI is a machine learning-based solution designed to predict and monitor fetal health using Cardiotocogram (CTG) data. This project aims to assist healthcare professionals by automating the classification of fetal health status, helping in the early detection of risks and promoting timely intervention, particularly in low-resource settings. Table of Contents Introduction Project Overview Dataset Features Project Architecture Installation Usage Model Evaluation Screenshots Advantages & Disadvantages Future Scope License Introduction FetalAI leverages machine learning to analyze CTG data and predict fetal health status in real-time. The solution classifies the health status into three categories: Normal Pathological Suspect This system assists healthcare workers by providing quick, automated assessments, crucial for preventing child and maternal mortality. Project Overview Cardiotocograms (CTGs) provide valuable data about fetal health, including fetal heart rate (FHR), fetal movements, and uterine contractions. Traditional methods of interpreting CTG data can be subjective and prone to error, making automated tools highly beneficial. This project aims to: Improve fetal health monitoring accuracy using machine learning. Provide an automated, cost-effective solution for low-resource settings. Facilitate early diagnosis, reducing maternal and child mortality. Dataset The dataset used in this project consists of CTG data, which contains fetal heart rate, uterine contraction, and other related features. The dataset has been preprocessed to remove missing values, outliers, and to balance the class distribution using SMOTE . Classes: Normal Pathological Suspect Features 1. Handling Missing Values No missing values were found in the dataset. 2. Handling Imbalanced Data We applied SMOTE (Synthetic Minority Over-sampling Technique) to balance the class distribution in the dataset. 3. …