Overview / Usage

Malaria is a life-threatening disease. It's typically transmitted through the bite of an infected Anopheles mosquito. Infected mosquitoes carry the Plasmodium parasite. When this mosquito bites you, the parasite is released into your bloodstream.

World Health Organization estimates that India has 15 million cases of malaria with 19,500–20,000 deaths annually vs. ∼2 million cases and 1,000 deaths reported (WHO SEARO website). Therefore we are using deep learning technologies and Intel Open Vino toolkit to be able to detect two classes Parasitize and Uninfected.

The Dataset is taken from : [https://ceb.nlm.nih.gov/repositories/malaria-datasets/]

Created a custom CNN model to classify between two classes with Test_Accuracy:- 0.9568058076225046.

Methodology / Approach

1. Data Preparation: We will make data and labels list where data will be image to array implementatation which contains RGB values of each image. and label will be class of cells here I will be taking 0 and 1 for two classes.
2. Saving and Loading the data we prepared so next time we can load it from saved .npy file.
3. Do Train/Test Split of data and labels that prepared in early section. Classes are defined as the unique labels in the data. Here it will be 2 as Parasitized:0 and Uninfected:1, here 0 and 1 are the mapping in labels for these two classes.
4. Compile the model with loss as categorical_crossentropy and using adam optimizer you can test result by trying RMSProp as well as Momentum.
5. Check the accuracy on Test data.
6. Convert the .h5 file to .pb file.
7. Optimize our model to create an *.xml and *.bin file.
8. Now we can use sample images for the inference using Open Vino toolkit.

​

Technologies Used

Hardware:

1. Intel(R)Core i3-4005U CPU @1.70GHz 1.70 GHz
2. RAM: 4GB.
3. Intel NCS.

Technology:

1. Tensorflow 2
2. Python
3. Intel Open Vino toolkit.

​

Repository

https://github.com/pranabsarkar/Malaria-Detection-V1/blob/master/infer.ipynb

Other links

• Intel Software Innovators Medium Publication.