Dataset: https://www.kaggle.com/datasets/paultimothymooney/chest-xray-pneumonia
What it does ?
Our project is a web application that uses machine learning to detect pneumonia and cataracts in medical images. Users can upload an X-ray scan or an eye image, and our model analyzes the image to determine the presence of pneumonia or cataracts.
How we built it ?
We built the web application using a Flask framework for the backend and HTML, CSS, and JavaScript for the front end. The machine learning models were developed using TensorFlow and Keras. We trained the models on large datasets of medical images to ensure accurate detection.
Challenges we ran into:
One of the main challenges we faced was acquiring a diverse and labeled dataset for training our models.Our models required many rounds of training and were constantly overfitting. We spent a lot of time tweaking and fine-tuning the parameters. Additionally, optimizing the models for real-time predictions and integrating them into the web application posed technical challenges that required careful implementation.
Accomplishments that we're proud of:
We are proud of developing a functional web application that accurately detects pneumonia and cataracts in medical images. Our models achieved high accuracy rates, and we successfully integrated them into the user-friendly interface. We also implemented features like image preview and drag-and-drop functionality to enhance the user experience.
What we learned?
Throughout this project, we gained valuable experience in web development, machine learning, and data preprocessing. We learned how to train deep learning models using large datasets and how to deploy them in a web application. Additionally, we improved our collaboration and problem-solving skills as a team.
Built With
Flask
HTML
CSS
JavaScript
TensorFlow
Keras
Python
Note please run the Train python notebook to obtain the model weights