Malaria-Detector-Streamlit

A Neural Network to Detect Malaria Parasites in Blood Samples

Background

Plasmodium falciparum is a common type of malaria found across Africa. It is also the most deadly form of malaria. You can detect P. falciparum by taking blood samples and inspecting them under a microscope. Infected red blood cells may have a darker patch which is the parasite. To train this model, I used a dataset made in Uganda by J. Quinn et al.

Steps

1. Download the dataset from the link above.
2. Simply use the tensorflow object detection API to train a model
3. Export the inference graph and call it frozen_inference_graph.pb
4. Move the website.py and correct.py files into the object_detection directory
5. On line 76 of correct.py change the MODEL_NAME to the name of your exported model
6. Move both of the images into the object_detection directory

Demo