SOFTWARE & SYSTEMS
PNEUMONIA X-RAY CLASSIFIER
A Keras model (VGG16 with two added dense layers) that distinguishes a healthy chest X-ray from one with pneumonia. Reached 92% accuracy (author-reported; the test split is not documented in the repo). A learning project.
completed
- Framework
- Keras
Chest-image classification study · Concept illustration
KERAS VGG16 TRANSFER LEARNING ON CHEST X-RAYS
01 / The build
Behind the project.
A chest X-ray classifier built in Keras: a pretrained VGG16 with two added dense layers, trained to distinguish healthy torso X-rays from pneumonia. Reached 92% accuracy (author-reported; the dataset split is not documented in the repository). A learning project.
Technical details
Under the hood.
- Framework
- Keras
- Architecture
- VGG16 + Two Dense Layers
- Task
- Healthy vs Pneumonia Chest X-Ray
- Accuracy
- 92% (author-reported)
Notes & outcomes
What came out of it.
- 92% accuracy (author-reported).
- VGG16 transfer learning on chest X-rays.
Keep exploring