The model architecture used is resnet18 which is trained using PyTorch, and then converted to ONNX format for deployment using Heroku.
Dataset used for training is from Kaggle Chest X-Ray Dataset which which contains over 5800 samples obtained from The Guangzhou Women and Childrenβs Medical Center.
View the notebook here: pneumonia-det.ipynb
The model has been been converted to ONNX format and deployed using Gradio & hosted on Heroku: Pneumonia Detection using chest X-Ray
Predictions on unseen test data: