A machine learning project that uses a neural network to detect and classify brain tumors into one of four categories: Glioma, Meningioma, No Tumor, or Pituitary.

The dataset consists of 1300 MRI images sourced from Kaggle. It has been preprocessed to improve model performance.

I preprocessed the dataset by applying computer vision and 2D image manipulation concepts, such as convoluting, scaling, rotating.

The model consists of a 3-layer neural network trained for 50 epochs, achieving an accuracy of 96%.

I created a visual display of epoch accuracy over time.

I also used Flask to create a front-end, where users can input their own brain tumor MRI images.

• Python - Core programming language
• Pandas - Data handling and preprocessing
• TensorFlow & Keras - Neural network implementation
• Flask - Used for the front-end to create an interactive UI