The COVID-19 pandemic still having devastating effect on the health and well-being of the global population. A critical step in the fight against COVID-19 is effective screening of infected patients, with one of the key screening approaches being radiology examination using chest radiography. Deep-learning artificial intelligent (AI) methods have the potential to help improve diagnostic efficiency and accuracy for reading portable CXRs.
This repository is only POC implementation of the idea discussed above. Needs more solid clinical trials to back this idea and approvals from regulatory bodies to be used in production.
| Feature | Brief Explanation |
|---|---|
| Base Model Architecture | Resnet18 from ResNet family, implemetation from PyTorch |
| Learning Rate Finder | Learning rate finder implemetation from FastAI |
| Learning rate and Momentum scheduler | One cycle policy implemetation from FastAI to achieve superconvergence |
| Explainability | Implemented gradcam |
| Dataset | Dataset COVID-19 Radiography Database from Tawsifur Rahman |
| Model | Metrics(Accuracy) | Epochs |
|---|---|---|
| Resnet18 | 96.14% | 8 |
This project is licensed under the MIT License