| <div style="display: flex; flex-direction: row"> | |
| <div> | |
| <h3>How does it work?</h3> | |
| <p>The model is a Convolutional Neural Network (CNN) trained on the retinal images to predict the disease.</p> | |
| <p>Dataset used for training is MEDMNIST dataset which contains retinal images of 4 different diseases.</p> | |
| <p>It uses PyTorch framework for training and prediction.</p> | |
| </div> | |
| <vr/> | |
| <div> | |
| <h3>How to use?</h3> | |
| <p>Upload an image of the retina and click on 'Submit' to get the prediction.</p> | |
| <p>It will show the predicted disease based on the input image.</p> | |
| <p>It can predict one of the following diseases:</p> | |
| <ul> | |
| <li>Choroidal Neovascularization</li> | |
| <li>Diabetic Macular Edema</li> | |
| <li>Drusen</li> | |
| <li>Normal</li> | |
| </ul> | |
| </div> | |
| </div> | |
| <h3>How accurate is it?</h3> | |
| <p>The model has an accuracy of 75% on the test dataset.</p> |