Malaria detection using PyTorch

Malaria is a major cause of death in tropical and sub-tropical countries, killing each year over 1 million people globally; 90% of fatalities occur in African children. Although effective ways to manage malaria now exist, the number of malaria cases is still increasing, due to several factors.

In this emergency situation, prompt and effective diagnostic methods are essential for the management and control of malaria. A clinical diagnosis of malaria is traditional among medical doctors. This method is least expensive and most widely practiced. A clinical diagnosis of malaria is still challenging because of the non-specific nature of the signs and symptoms, which overlap considerably with other common, as well as potentially life-threatening diseases, e.g. common viral or bacterial infections.

Hence the purpose of the project is to implementation a solution for easy and malaria diagnosis with high accuracy.

Prerequisites

Below are the libraries used in the notebook.

Pytorch, Numpy, Pandas, Matplotlib

Dataset for this project can be found here

How to run this project

1. (Easiest) Fork our Kaggle notebook and try it out.

2. You can run it using Colab.

3. Or clone/download the repository and run the Notebook using the command prompt.

   Run the command: jupyter notebook and then select the file kernel.ipynb from the browser.

Note: You will have to download the dataset for step 2 and 3.

Output

Authors

1. Sachin Dev S
2. P S Vishnu

License

This project is licensed under the MIT License - see the License

Acknowledgments

1. Kaggle
2. Downloading kaggle data to colab