Brain Tumor Detection using YOLO

Project Overview

This project implements brain tumor detection using the YOLO (You Only Look Once) object detection model. It identifies different types of brain tumors from MRI scans and provides a bounding box with classification confidence.

Objective

Detect and classify brain tumors from MRI images using YOLO.

Datasets

The project uses a custom dataset with annotated MRI images of brain tumors, including:

• Meningioma
• Glioma
• Pituitary tumors

Techniques

• Data Preprocessing: Cleaning and preparing MRI images for training.
• Exploratory Data Analysis (EDA): Understanding trends, correlations, and key features.
• Model Training: Training YOLO for tumor detection.
• Model Evaluation: Assessing performance using accuracy, precision, recall, and inference speed.

Key Steps

1. Data Preprocessing:
   • Merge annotated MRI images with metadata.
   • Handle missing data and standardize formats.
2. Exploratory Data Analysis (EDA):
   • Analyze trends in tumor occurrence.
   • Visualize MRI images and bounding boxes.
3. Predictive Modeling:
   • Train a YOLO model to detect brain tumors.
   • Fine-tune hyperparameters for better accuracy.
4. Model Evaluation:
   • Measure accuracy and inference speed.
   • Optimize model performance.

Technologies Used

• Python: For data analysis and model training.
• Ultralytics YOLO: For object detection.
• Pandas & NumPy: For data manipulation.
• Scikit-learn: For performance evaluation.
• Matplotlib & OpenCV: For visualization.
• Jupyter Notebook: For interactive experimentation.

Future Improvements

• Train on a larger dataset for higher accuracy.
• Implement real-time detection capabilities.
• Optimize for deployment on edge devices.

How to Use

Clone the Repository

git clone https://github.com/yourusername/brain-tumor-detection-yolo.git

Install Dependencies

pip install -r requirements.txt

Run the Jupyter Notebook to open the project

jupyter notebook

Conclusion

This project presents a deep-learning-based approach to brain tumor detection using YOLO. It enables early diagnosis and enhances medical decision-making through automated MRI analysis.