Challa-Rajesh/Malaria-Disease-Prediction-and-Stage-Classification-using-CNN

This project focuses on building a Convolutional Neural Network (CNN) model to classify microscopic images of red blood cells as either Parasitized (infected with malaria) or Uninfected. Additionally, it extends to predicting the stage of malaria infection if applicable. # **Malaria Cell Image Classification Using CNN and LIME Test** *This animation illustrates the emergence of the parasites from the infected cell. For more information on the growth of human malaria parasites in their dormant form, visit MIT News.* ## Overview **Malaria** is a life-threatening disease caused by parasites that are transmitted to people through the bites of infected female Anopheles mosquitoes. It's preventable and curable, but it can be fatal if not treated promptly. The Plasmodium falciparum parasite is the most dangerous, with the highest rates of complications and mortality. Early diagnosis and treatment are critical in reducing mortality rates and preventing the spread of the disease. According to global health organizations, hundreds of thousands succumb to malaria annually, with a significant impact on child mortality rates. The annual death toll is certainly in the hundreds of thousands, but estimates differ between different global health organizations. The World Health Organization (WHO) estimates that **558,000** people died because of malaria in **2019**; the Institute of Health Metrics and Evaluation (IHME) puts this estimate at **643,000**. Most victims are children. It is one of the leading causes of child mortality. Every twelfth child that died in **2017**, died because of malaria. [1] This project aims to assist in the fight against malaria by automating the detection of infected cells using deep learning. --- [1] Source Link ## Project Objective The goal of this project is to develop a **Convolutional Neural Network (CNN)** that can classify microscopic images of cells as either infected with malaria or uninfected. ## Methodology We've employed a robust workflow to train our model: - **Data Preprocessing**: Standardizing the cell images for model input. - **Data Augmentation**: Enhancing the training data to prevent overfitting and improve model robustness. - **Model Architecture**: Designing a CNN that learns fe …