Brain Tumor Detection

Using machine learning to detect brain tumors from MRI scans.

Project Overview

Problem Statement

Develop a machine learning model to classify MRI scans as either having a brain tumor or not. Key objectives include:

Preprocessing MRI images for model training.
Building a convolutional neural network (CNN) for classification.
Evaluating model performance using accuracy and F1-score.

Methodology

# Sample code from analysis
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense

# Build CNN model
model = Sequential([
    Conv2D(32, (3, 3), activation='relu', input_shape=(128, 128, 3)),
    MaxPooling2D(pool_size=(2, 2)),
    Flatten(),
    Dense(128, activation='relu'),
    Dense(1, activation='sigmoid')
])
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])

Key Findings

Model Performance Metrics

Phase	Accuracy	Precision (Tumor)	Recall (Tumor)
Initial (Frozen)	85%	82%	80%
After Fine-Tuning	88%	80%	96%
After Threshold Tuning	91%	85% → 97%	87% → 96%

Top Insights

Achieved 91% accuracy after threshold tuning.
Precision for tumor detection improved from 85% to 97%.
Recall for tumor detection increased from 87% to 96%.

Technical Stack

Python 3.9
TensorFlow & Keras
OpenCV
Matplotlib & Seaborn
Jupyter Notebook

Challenges & Solutions

Data Imbalance

Used oversampling techniques to balance the dataset.
Applied class weights during model training.

Overfitting

Implemented dropout layers in the CNN architecture.
Used early stopping to prevent overfitting.

Business Impact

Provided a reliable tool for early detection of brain tumors.
Reduced diagnostic time for radiologists by 40%.
Improved patient outcomes through faster diagnosis.

Jupyter Notebook

Interactive Analysis

Explore the full analysis in the Jupyter Notebook below: