Browse Datasets

Malaria Cell Images

Parasitized vs Uninfected Blood Cell Images (NIH)

image-classification computer-vision healthcare 27,558 images (~339MB) Public Domain (NIH)
Access Dataset

Description

Segmented blood-cell images split evenly between parasitized and uninfected cells, sourced from the US National Institutes of Health for automated malaria screening. A clean, classic benchmark for getting started with medical image classification.

Quick Start

Python 
import kagglehub

path = kagglehub.dataset_download("iarunava/cell-images-for-detecting-malaria")
print(path)  # folders: Parasitized/ and Uninfected/

Project Ideas

Baseline CNN Classifier

Train a ResNet or Inception model to classify parasitized vs uninfected cells.

Beginner

Edge Diagnostic Demo

Quantize the classifier and serve it through a local Streamlit app to simulate a clinic edge tool.

Intermediate

Ethical Considerations

  • This is a sanitized dataset; real blood smears carry artefacts and noise absent here.
  • False negatives in malaria screening carry serious clinical risk.