DIAGNOSIS AND SEGMENTATION OF BRAIN TUMORS

💡 Inspiration

As we know how expensive procedure to detect the brain tumor can be and how time-consuming it can be. So we build this project so that the doctor can easily detect the brain tumor and the patient don't have to pay much expensive fee.

❓ What it does

Machine Learning (ML) and Artificial Intelligence (AI)-based automated classification and segmentation approaches have consistently outperformed manual classification. As a result, presenting a system that uses Deep Learning Algorithms such as Convolution Neural Network (CNN), Federated Deep Network (FDN), and Transfer Learning (TL) to conduct detection and classification will be beneficial to doctors all over the world. Brain tumors are difficult to understand. There are numerous variations in the sizes and locations of brain tumors (s). This makes fully comprehending the tumor's nature extremely challenging. For MRI analysis, a competent Neurosurgeon is also necessary. The dearth of skilled doctors and understanding about malignancies makes generating results from MRI's extremely difficult and time-consuming in poor nations. As a result, this problem may be solved by a cloud-based automated solution.

🔧 How we built it 🔨

The mobile application is built using flutter. And the web application is build using Streamlit.

🏃‍♂️ Challenges we ran into

Throughout the process, we experienced several obstacles:

Understanding the platform and getting the initial flow to function took some effort, as did getting the dashboard UI to sync. Multiple DevOps tools are integrated across the entire process. When it comes to training the model, finding the right hyperparameters is crucial. Making the website responsive, as well as displaying past projected images for the current forecast. Integrating the Frontend with the Machine Learning Model is the actual problem. We overcame all the obstacles by working on the problem regularly and as a team.

🏆 Accomplishments that we're proud of

we are proud on ourselves that we have completed this project within the time limit

🧠 What we learned

We learned how to make our own models to detect brain tumor. We learned how to build an application using flutter, which we used the first time. Furthermore, we also learned how to use stream lit to built web applications