Three CV models to predict pneumonia in X-ray images trained on this repo on kaggle
Problem
In the light of current epidemic machine learning may come in handy to help radiologists, hence I decided to train a few models that would be a useful in this setting. I've set a goal of validation accuracy over 80%.
Data
As per of repository descirption: "The dataset is organized into 3 folders (train, test, val) and contains subfolders for each image category (Pneumonia/Normal). There are 5,863 X-Ray images (JPEG) and 2 categories (Pneumonia/Normal).
Chest X-ray images (anterior-posterior) were selected from retrospective cohorts of pediatric patients of one to five years old from Guangzhou Women and Children’s Medical Center, Guangzhou. All chest X-ray imaging was performed as part of patients’ routine clinical care. "
Models tested
CNN
Efficenet
Mobilenet
Model chosen
As per chosen target - only the CNN has met over >80% validation accuracy, thus being the only project being ported to the ruidmentary implementation in "X_ray_implement.ipynb" via Keras' serialization.