Breast Cancer Classification Using Neural Networks

Overview

This project implements a Deep Learning model for Breast Cancer Classification using TensorFlow and Keras. The model predicts whether a tumor is Benign or Malignant based on features from the Wisconsin Breast Cancer Diagnostic Dataset.

The project demonstrates the complete machine learning workflow, including:

• Data loading and exploration
• Data preprocessing
• Feature standardization
• Train-test splitting
• Neural network development
• Model training and evaluation
• Prediction on new samples

---

Dataset

The dataset used in this project is the Breast Cancer Wisconsin (Diagnostic) Dataset available through Scikit-Learn.

Dataset Information

• Total Samples: 569

• Features: 30 numerical features

• Classes:

  • Malignant (0)
  • Benign (1)

The features are computed from digitized images of breast mass cell nuclei.

---

Technologies Used

• Python
• NumPy
• Pandas
• Matplotlib
• Scikit-Learn
• TensorFlow
• Keras

---

Project Workflow

1. Data Loading

The dataset is loaded directly from Scikit-Learn:

import sklearn.datasets

breast_cancer_dataset = sklearn.datasets.load_breast_cancer()

2. Data Preprocessing

• Converted dataset into a Pandas DataFrame
• Checked for missing values
• Added target labels
• Split features and target variables

3. Data Standardization

from sklearn.preprocessing import StandardScaler

scaler = StandardScaler()
X_train_std = scaler.fit_transform(X_train)
X_test_std = scaler.transform(X_test)

4. Neural Network Architecture

model = keras.Sequential([
    keras.layers.Flatten(input_shape=(30,)),
    keras.layers.Dense(20, activation='relu'),
    keras.layers.Dense(2, activation='sigmoid')
])

5. Model Compilation

model.compile(
    optimizer='adam',
    loss='sparse_categorical_crossentropy',
    metrics=['accuracy']
)

6. Model Training

history = model.fit(
    X_train_std,
    Y_train,
    validation_split=0.1,
    epochs=10
)

---

Results

Model Performance

• Training Accuracy: ~96%
• Validation Accuracy: ~93%
• Test Accuracy: 95.61%

loss, accuracy = model.evaluate(X_test_std, Y_test)
print(accuracy)

Output:

0.9561

---

Visualization

The project includes:

• Training Accuracy vs Validation Accuracy
• Training Loss vs Validation Loss

These visualizations help monitor model performance and detect overfitting.

---

Prediction

The trained model can predict whether a tumor is:

• Benign
• Malignant

for new patient data after applying the same preprocessing steps.

---

Repository Structure

Breast-Cancer-Classification/
│
├── Breast_Cancer_Classification.ipynb
├── README.md
├── LICENSE
└── screenshots/

---

Future Improvements

• Add Confusion Matrix
• Precision, Recall, and F1 Score
• ROC Curve Analysis
• Hyperparameter Tuning
• Model Deployment using Streamlit or Flask
• Save and Load Trained Models

---

Author

krushna mandhare

Machine Learning and Deep Learning Enthusiast

---

License

This project is licensed under the MIT License.